diff options
Diffstat (limited to 'drivers/net')
83 files changed, 7048 insertions, 467 deletions
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c index 110d6c403bdd..4ffd303c64ea 100644 --- a/drivers/net/dsa/ocelot/felix_vsc9959.c +++ b/drivers/net/dsa/ocelot/felix_vsc9959.c @@ -1745,7 +1745,7 @@ static void vsc9959_psfp_sfi_table_del(struct ocelot *ocelot, u32 index) static void vsc9959_psfp_parse_gate(const struct flow_action_entry *entry, struct felix_stream_gate *sgi) { - sgi->index = entry->gate.index; + sgi->index = entry->hw_index; sgi->ipv_valid = (entry->gate.prio < 0) ? 0 : 1; sgi->init_ipv = (sgi->ipv_valid) ? entry->gate.prio : 0; sgi->basetime = entry->gate.basetime; @@ -1947,7 +1947,7 @@ static int vsc9959_psfp_filter_add(struct ocelot *ocelot, int port, kfree(sgi); break; case FLOW_ACTION_POLICE: - index = a->police.index + VSC9959_PSFP_POLICER_BASE; + index = a->hw_index + VSC9959_PSFP_POLICER_BASE; if (index > VSC9959_PSFP_POLICER_MAX) { ret = -EINVAL; goto err; diff --git a/drivers/net/dsa/sja1105/sja1105_flower.c b/drivers/net/dsa/sja1105/sja1105_flower.c index 72b9b39b0989..7dcdd784aea4 100644 --- a/drivers/net/dsa/sja1105/sja1105_flower.c +++ b/drivers/net/dsa/sja1105/sja1105_flower.c @@ -379,7 +379,7 @@ int sja1105_cls_flower_add(struct dsa_switch *ds, int port, vl_rule = true; rc = sja1105_vl_gate(priv, port, extack, cookie, - &key, act->gate.index, + &key, act->hw_index, act->gate.prio, act->gate.basetime, act->gate.cycletime, diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h index 533b8519ec35..466273b22f0a 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h @@ -898,6 +898,8 @@ #define PCS_V2_WINDOW_SELECT 0x9064 #define PCS_V2_RV_WINDOW_DEF 0x1060 #define PCS_V2_RV_WINDOW_SELECT 0x1064 +#define PCS_V2_YC_WINDOW_DEF 0x18060 +#define PCS_V2_YC_WINDOW_SELECT 0x18064 /* PCS register entry bit positions and sizes */ #define PCS_V2_WINDOW_DEF_OFFSET_INDEX 6 @@ -1030,8 +1032,8 @@ #define XP_PROP_0_PORT_ID_WIDTH 8 #define XP_PROP_0_PORT_MODE_INDEX 8 #define XP_PROP_0_PORT_MODE_WIDTH 4 -#define XP_PROP_0_PORT_SPEEDS_INDEX 23 -#define XP_PROP_0_PORT_SPEEDS_WIDTH 4 +#define XP_PROP_0_PORT_SPEEDS_INDEX 22 +#define XP_PROP_0_PORT_SPEEDS_WIDTH 5 #define XP_PROP_1_MAX_RX_DMA_INDEX 24 #define XP_PROP_1_MAX_RX_DMA_WIDTH 5 #define XP_PROP_1_MAX_RX_QUEUES_INDEX 8 diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-pci.c b/drivers/net/ethernet/amd/xgbe/xgbe-pci.c index 90cb55eb5466..efdcf484a510 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-pci.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-pci.c @@ -278,6 +278,13 @@ static int xgbe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) (rdev->vendor == PCI_VENDOR_ID_AMD) && (rdev->device == 0x15d0)) { pdata->xpcs_window_def_reg = PCS_V2_RV_WINDOW_DEF; pdata->xpcs_window_sel_reg = PCS_V2_RV_WINDOW_SELECT; + } else if (rdev && (rdev->vendor == PCI_VENDOR_ID_AMD) && + (rdev->device == 0x14b5)) { + pdata->xpcs_window_def_reg = PCS_V2_YC_WINDOW_DEF; + pdata->xpcs_window_sel_reg = PCS_V2_YC_WINDOW_SELECT; + + /* Yellow Carp devices do not need cdr workaround */ + pdata->vdata->an_cdr_workaround = 0; } else { pdata->xpcs_window_def_reg = PCS_V2_WINDOW_DEF; pdata->xpcs_window_sel_reg = PCS_V2_WINDOW_SELECT; @@ -460,7 +467,7 @@ static int __maybe_unused xgbe_pci_resume(struct device *dev) return ret; } -static const struct xgbe_version_data xgbe_v2a = { +static struct xgbe_version_data xgbe_v2a = { .init_function_ptrs_phy_impl = xgbe_init_function_ptrs_phy_v2, .xpcs_access = XGBE_XPCS_ACCESS_V2, .mmc_64bit = 1, @@ -475,7 +482,7 @@ static const struct xgbe_version_data xgbe_v2a = { .an_cdr_workaround = 1, }; -static const struct xgbe_version_data xgbe_v2b = { +static struct xgbe_version_data xgbe_v2b = { .init_function_ptrs_phy_impl = xgbe_init_function_ptrs_phy_v2, .xpcs_access = XGBE_XPCS_ACCESS_V2, .mmc_64bit = 1, diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c index 213769054391..2156600641b6 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c @@ -124,10 +124,10 @@ #include "xgbe.h" #include "xgbe-common.h" -#define XGBE_PHY_PORT_SPEED_100 BIT(0) -#define XGBE_PHY_PORT_SPEED_1000 BIT(1) -#define XGBE_PHY_PORT_SPEED_2500 BIT(2) -#define XGBE_PHY_PORT_SPEED_10000 BIT(3) +#define XGBE_PHY_PORT_SPEED_100 BIT(1) +#define XGBE_PHY_PORT_SPEED_1000 BIT(2) +#define XGBE_PHY_PORT_SPEED_2500 BIT(3) +#define XGBE_PHY_PORT_SPEED_10000 BIT(4) #define XGBE_MUTEX_RELEASE 0x80000000 diff --git a/drivers/net/ethernet/broadcom/bcm4908_enet.c b/drivers/net/ethernet/broadcom/bcm4908_enet.c index b07cb9bc5f2d..4a2622b05ee1 100644 --- a/drivers/net/ethernet/broadcom/bcm4908_enet.c +++ b/drivers/net/ethernet/broadcom/bcm4908_enet.c @@ -635,7 +635,6 @@ static int bcm4908_enet_poll_tx(struct napi_struct *napi, int weight) struct bcm4908_enet_dma_ring_bd *buf_desc; struct bcm4908_enet_dma_ring_slot *slot; struct device *dev = enet->dev; - unsigned int bytes = 0; int handled = 0; while (handled < weight && tx_ring->read_idx != tx_ring->write_idx) { @@ -646,7 +645,6 @@ static int bcm4908_enet_poll_tx(struct napi_struct *napi, int weight) dma_unmap_single(dev, slot->dma_addr, slot->len, DMA_TO_DEVICE); dev_kfree_skb(slot->skb); - bytes += slot->len; if (++tx_ring->read_idx == tx_ring->length) tx_ring->read_idx = 0; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 2b06d78baa08..a19dd6797070 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -1850,6 +1850,14 @@ struct bnx2x { /* Vxlan/Geneve related information */ u16 udp_tunnel_ports[BNX2X_UDP_PORT_MAX]; + +#define FW_CAP_INVALIDATE_VF_FP_HSI BIT(0) + u32 fw_cap; + + u32 fw_major; + u32 fw_minor; + u32 fw_rev; + u32 fw_eng; }; /* Tx queues may be less or equal to Rx queues */ @@ -2525,5 +2533,6 @@ void bnx2x_register_phc(struct bnx2x *bp); * Meant for implicit re-load flows. */ int bnx2x_vlan_reconfigure_vid(struct bnx2x *bp); - +int bnx2x_init_firmware(struct bnx2x *bp); +void bnx2x_release_firmware(struct bnx2x *bp); #endif /* bnx2x.h */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 54a2334dee56..8d36ebbf08e1 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -2365,10 +2365,8 @@ int bnx2x_compare_fw_ver(struct bnx2x *bp, u32 load_code, bool print_err) if (load_code != FW_MSG_CODE_DRV_LOAD_COMMON_CHIP && load_code != FW_MSG_CODE_DRV_LOAD_COMMON) { /* build my FW version dword */ - u32 my_fw = (BCM_5710_FW_MAJOR_VERSION) + - (BCM_5710_FW_MINOR_VERSION << 8) + - (BCM_5710_FW_REVISION_VERSION << 16) + - (BCM_5710_FW_ENGINEERING_VERSION << 24); + u32 my_fw = (bp->fw_major) + (bp->fw_minor << 8) + + (bp->fw_rev << 16) + (bp->fw_eng << 24); /* read loaded FW from chip */ u32 loaded_fw = REG_RD(bp, XSEM_REG_PRAM); diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h index 3f8435208bf4..a84d015da5df 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_fw_defs.h @@ -241,6 +241,8 @@ IRO[221].m2)) #define XSTORM_VF_TO_PF_OFFSET(funcId) \ (IRO[48].base + ((funcId) * IRO[48].m1)) +#define XSTORM_ETH_FUNCTION_INFO_FP_HSI_VALID_E2_OFFSET(fid) \ + (IRO[386].base + ((fid) * IRO[386].m1)) #define COMMON_ASM_INVALID_ASSERT_OPCODE 0x0 /* eth hsi version */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h index 622fadc50316..611efee75834 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h @@ -3024,7 +3024,8 @@ struct afex_stats { #define BCM_5710_FW_MAJOR_VERSION 7 #define BCM_5710_FW_MINOR_VERSION 13 -#define BCM_5710_FW_REVISION_VERSION 15 +#define BCM_5710_FW_REVISION_VERSION 21 +#define BCM_5710_FW_REVISION_VERSION_V15 15 #define BCM_5710_FW_ENGINEERING_VERSION 0 #define BCM_5710_FW_COMPILE_FLAGS 1 diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 651bc1d7a57a..4953f5e1e390 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -74,9 +74,19 @@ __stringify(BCM_5710_FW_MINOR_VERSION) "." \ __stringify(BCM_5710_FW_REVISION_VERSION) "." \ __stringify(BCM_5710_FW_ENGINEERING_VERSION) + +#define FW_FILE_VERSION_V15 \ + __stringify(BCM_5710_FW_MAJOR_VERSION) "." \ + __stringify(BCM_5710_FW_MINOR_VERSION) "." \ + __stringify(BCM_5710_FW_REVISION_VERSION_V15) "." \ + __stringify(BCM_5710_FW_ENGINEERING_VERSION) + #define FW_FILE_NAME_E1 "bnx2x/bnx2x-e1-" FW_FILE_VERSION ".fw" #define FW_FILE_NAME_E1H "bnx2x/bnx2x-e1h-" FW_FILE_VERSION ".fw" #define FW_FILE_NAME_E2 "bnx2x/bnx2x-e2-" FW_FILE_VERSION ".fw" +#define FW_FILE_NAME_E1_V15 "bnx2x/bnx2x-e1-" FW_FILE_VERSION_V15 ".fw" +#define FW_FILE_NAME_E1H_V15 "bnx2x/bnx2x-e1h-" FW_FILE_VERSION_V15 ".fw" +#define FW_FILE_NAME_E2_V15 "bnx2x/bnx2x-e2-" FW_FILE_VERSION_V15 ".fw" /* Time in jiffies before concluding the transmitter is hung */ #define TX_TIMEOUT (5*HZ) @@ -747,9 +757,7 @@ static int bnx2x_mc_assert(struct bnx2x *bp) CHIP_IS_E1(bp) ? "everest1" : CHIP_IS_E1H(bp) ? "everest1h" : CHIP_IS_E2(bp) ? "everest2" : "everest3", - BCM_5710_FW_MAJOR_VERSION, - BCM_5710_FW_MINOR_VERSION, - BCM_5710_FW_REVISION_VERSION); + bp->fw_major, bp->fw_minor, bp->fw_rev); return rc; } @@ -12308,6 +12316,15 @@ static int bnx2x_init_bp(struct bnx2x *bp) bnx2x_read_fwinfo(bp); + if (IS_PF(bp)) { + rc = bnx2x_init_firmware(bp); + + if (rc) { + bnx2x_free_mem_bp(bp); + return rc; + } + } + func = BP_FUNC(bp); /* need to reset chip if undi was active */ @@ -12320,6 +12337,7 @@ static int bnx2x_init_bp(struct bnx2x *bp) rc = bnx2x_prev_unload(bp); if (rc) { + bnx2x_release_firmware(bp); bnx2x_free_mem_bp(bp); return rc; } @@ -13317,16 +13335,11 @@ static int bnx2x_check_firmware(struct bnx2x *bp) /* Check FW version */ offset = be32_to_cpu(fw_hdr->fw_version.offset); fw_ver = firmware->data + offset; - if ((fw_ver[0] != BCM_5710_FW_MAJOR_VERSION) || - (fw_ver[1] != BCM_5710_FW_MINOR_VERSION) || - (fw_ver[2] != BCM_5710_FW_REVISION_VERSION) || - (fw_ver[3] != BCM_5710_FW_ENGINEERING_VERSION)) { + if (fw_ver[0] != bp->fw_major || fw_ver[1] != bp->fw_minor || + fw_ver[2] != bp->fw_rev || fw_ver[3] != bp->fw_eng) { BNX2X_ERR("Bad FW version:%d.%d.%d.%d. Should be %d.%d.%d.%d\n", - fw_ver[0], fw_ver[1], fw_ver[2], fw_ver[3], - BCM_5710_FW_MAJOR_VERSION, - BCM_5710_FW_MINOR_VERSION, - BCM_5710_FW_REVISION_VERSION, - BCM_5710_FW_ENGINEERING_VERSION); + fw_ver[0], fw_ver[1], fw_ver[2], fw_ver[3], + bp->fw_major, bp->fw_minor, bp->fw_rev, bp->fw_eng); return -EINVAL; } @@ -13404,34 +13417,51 @@ do { \ (u8 *)bp->arr, len); \ } while (0) -static int bnx2x_init_firmware(struct bnx2x *bp) +int bnx2x_init_firmware(struct bnx2x *bp) { - const char *fw_file_name; + const char *fw_file_name, *fw_file_name_v15; struct bnx2x_fw_file_hdr *fw_hdr; int rc; if (bp->firmware) return 0; - if (CHIP_IS_E1(bp)) + if (CHIP_IS_E1(bp)) { fw_file_name = FW_FILE_NAME_E1; - else if (CHIP_IS_E1H(bp)) + fw_file_name_v15 = FW_FILE_NAME_E1_V15; + } else if (CHIP_IS_E1H(bp)) { fw_file_name = FW_FILE_NAME_E1H; - else if (!CHIP_IS_E1x(bp)) + fw_file_name_v15 = FW_FILE_NAME_E1H_V15; + } else if (!CHIP_IS_E1x(bp)) { fw_file_name = FW_FILE_NAME_E2; - else { + fw_file_name_v15 = FW_FILE_NAME_E2_V15; + } else { BNX2X_ERR("Unsupported chip revision\n"); return -EINVAL; } + BNX2X_DEV_INFO("Loading %s\n", fw_file_name); rc = request_firmware(&bp->firmware, fw_file_name, &bp->pdev->dev); if (rc) { - BNX2X_ERR("Can't load firmware file %s\n", - fw_file_name); - goto request_firmware_exit; + BNX2X_DEV_INFO("Trying to load older fw %s\n", fw_file_name_v15); + + /* try to load prev version */ + rc = request_firmware(&bp->firmware, fw_file_name_v15, &bp->pdev->dev); + + if (rc) + goto request_firmware_exit; + + bp->fw_rev = BCM_5710_FW_REVISION_VERSION_V15; + } else { + bp->fw_cap |= FW_CAP_INVALIDATE_VF_FP_HSI; + bp->fw_rev = BCM_5710_FW_REVISION_VERSION; } + bp->fw_major = BCM_5710_FW_MAJOR_VERSION; + bp->fw_minor = BCM_5710_FW_MINOR_VERSION; + bp->fw_eng = BCM_5710_FW_ENGINEERING_VERSION; + rc = bnx2x_check_firmware(bp); if (rc) { BNX2X_ERR("Corrupt firmware file %s\n", fw_file_name); @@ -13487,7 +13517,7 @@ request_firmware_exit: return rc; } -static void bnx2x_release_firmware(struct bnx2x *bp) +void bnx2x_release_firmware(struct bnx2x *bp) { kfree(bp->init_ops_offsets); kfree(bp->init_ops); @@ -14004,6 +14034,7 @@ static int bnx2x_init_one(struct pci_dev *pdev, return 0; init_one_freemem: + bnx2x_release_firmware(bp); bnx2x_free_mem_bp(bp); init_one_exit: diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c index 74a8931ce1d1..11d15cd03600 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c @@ -758,9 +758,18 @@ static void bnx2x_vf_igu_reset(struct bnx2x *bp, struct bnx2x_virtf *vf) void bnx2x_vf_enable_access(struct bnx2x *bp, u8 abs_vfid) { + u16 abs_fid; + + abs_fid = FW_VF_HANDLE(abs_vfid); + /* set the VF-PF association in the FW */ - storm_memset_vf_to_pf(bp, FW_VF_HANDLE(abs_vfid), BP_FUNC(bp)); - storm_memset_func_en(bp, FW_VF_HANDLE(abs_vfid), 1); + storm_memset_vf_to_pf(bp, abs_fid, BP_FUNC(bp)); + storm_memset_func_en(bp, abs_fid, 1); + + /* Invalidate fp_hsi version for vfs */ + if (bp->fw_cap & FW_CAP_INVALIDATE_VF_FP_HSI) + REG_WR8(bp, BAR_XSTRORM_INTMEM + + XSTORM_ETH_FUNCTION_INFO_FP_HSI_VALID_E2_OFFSET(abs_fid), 0); /* clear vf errors*/ bnx2x_vf_semi_clear_err(bp, abs_vfid); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c index 1471b6130a2b..d8afcf8d6b30 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c @@ -1962,7 +1962,7 @@ static int bnxt_tc_setup_indr_cb(struct net_device *netdev, struct Qdisc *sch, v void *data, void (*cleanup)(struct flow_block_cb *block_cb)) { - if (!bnxt_is_netdev_indr_offload(netdev)) + if (!netdev || !bnxt_is_netdev_indr_offload(netdev)) return -EOPNOTSUPP; switch (type) { diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c index 0536d2c76fbc..3555c12edb45 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c @@ -1182,7 +1182,7 @@ static int enetc_psfp_parse_clsflower(struct enetc_ndev_priv *priv, } /* parsing gate action */ - if (entryg->gate.index >= priv->psfp_cap.max_psfp_gate) { + if (entryg->hw_index >= priv->psfp_cap.max_psfp_gate) { NL_SET_ERR_MSG_MOD(extack, "No Stream Gate resource!"); err = -ENOSPC; goto free_filter; @@ -1202,7 +1202,7 @@ static int enetc_psfp_parse_clsflower(struct enetc_ndev_priv *priv, } refcount_set(&sgi->refcount, 1); - sgi->index = entryg->gate.index; + sgi->index = entryg->hw_index; sgi->init_ipv = entryg->gate.prio; sgi->basetime = entryg->gate.basetime; sgi->cycletime = entryg->gate.cycletime; @@ -1244,7 +1244,7 @@ static int enetc_psfp_parse_clsflower(struct enetc_ndev_priv *priv, refcount_set(&fmi->refcount, 1); fmi->cir = entryp->police.rate_bytes_ps; fmi->cbs = entryp->police.burst; - fmi->index = entryp->police.index; + fmi->index = entryp->hw_index; filter->flags |= ENETC_PSFP_FLAGS_FMI; filter->fmi_index = fmi->index; sfi->meter_id = fmi->index; diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 0b274d8fa45b..3facb55b7161 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -327,6 +327,16 @@ config ICE_SWITCHDEV If unsure, say N. +config ICE_HWTS + bool "Support HW cross-timestamp on platforms with PTM support" + default y + depends on ICE && X86 + help + Say Y to enable hardware supported cross-timestamping on platforms + with PCIe PTM support. The cross-timestamp is available through + the PTP clock driver precise cross-timestamp ioctl + (PTP_SYS_OFFSET_PRECISE). + config FM10K tristate "Intel(R) FM10000 Ethernet Switch Host Interface Support" default n diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c index 21eff0895a7a..f6d56867f857 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c @@ -143,7 +143,7 @@ s32 fm10k_tlv_attr_put_mac_vlan(u32 *msg, u16 attr_id, * @vlan: location of buffer to store VLAN * * This function pulls the MAC address back out of the attribute and will - * place it in the array pointed by by mac_addr. It will return success + * place it in the array pointed by mac_addr. It will return success * if provided with a valid pointers. **/ s32 fm10k_tlv_attr_get_mac_vlan(u32 *attr, u8 *mac_addr, u16 *vlan) diff --git a/drivers/net/ethernet/intel/ice/ice_cgu_regs.h b/drivers/net/ethernet/intel/ice/ice_cgu_regs.h new file mode 100644 index 000000000000..57abd52386d0 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_cgu_regs.h @@ -0,0 +1,116 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2018-2021, Intel Corporation. */ + +#ifndef _ICE_CGU_REGS_H_ +#define _ICE_CGU_REGS_H_ + +#define NAC_CGU_DWORD9 0x24 +union nac_cgu_dword9 { + struct { + u32 time_ref_freq_sel : 3; + u32 clk_eref1_en : 1; + u32 clk_eref0_en : 1; + u32 time_ref_en : 1; + u32 time_sync_en : 1; + u32 one_pps_out_en : 1; + u32 clk_ref_synce_en : 1; + u32 clk_synce1_en : 1; + u32 clk_synce0_en : 1; + u32 net_clk_ref1_en : 1; + u32 net_clk_ref0_en : 1; + u32 clk_synce1_amp : 2; + u32 misc6 : 1; + u32 clk_synce0_amp : 2; + u32 one_pps_out_amp : 2; + u32 misc24 : 12; + } field; + u32 val; +}; + +#define NAC_CGU_DWORD19 0x4c +union nac_cgu_dword19 { + struct { + u32 tspll_fbdiv_intgr : 8; + u32 fdpll_ulck_thr : 5; + u32 misc15 : 3; + u32 tspll_ndivratio : 4; + u32 tspll_iref_ndivratio : 3; + u32 misc19 : 1; + u32 japll_ndivratio : 4; + u32 japll_iref_ndivratio : 3; + u32 misc27 : 1; + } field; + u32 val; +}; + +#define NAC_CGU_DWORD22 0x58 +union nac_cgu_dword22 { + struct { + u32 fdpll_frac_div_out_nc : 2; + u32 fdpll_lock_int_for : 1; + u32 synce_hdov_int_for : 1; + u32 synce_lock_int_for : 1; + u32 fdpll_phlead_slip_nc : 1; + u32 fdpll_acc1_ovfl_nc : 1; + u32 fdpll_acc2_ovfl_nc : 1; + u32 synce_status_nc : 6; + u32 fdpll_acc1f_ovfl : 1; + u32 misc18 : 1; + u32 fdpllclk_div : 4; + u32 time1588clk_div : 4; + u32 synceclk_div : 4; + u32 synceclk_sel_div2 : 1; + u32 fdpllclk_sel_div2 : 1; + u32 time1588clk_sel_div2 : 1; + u32 misc3 : 1; + } field; + u32 val; +}; + +#define NAC_CGU_DWORD24 0x60 +union nac_cgu_dword24 { + struct { + u32 tspll_fbdiv_frac : 22; + u32 misc20 : 2; + u32 ts_pll_enable : 1; + u32 time_sync_tspll_align_sel : 1; + u32 ext_synce_sel : 1; + u32 ref1588_ck_div : 4; + u32 time_ref_sel : 1; + } field; + u32 val; +}; + +#define TSPLL_CNTR_BIST_SETTINGS 0x344 +union tspll_cntr_bist_settings { + struct { + u32 i_irefgen_settling_time_cntr_7_0 : 8; + u32 i_irefgen_settling_time_ro_standby_1_0 : 2; + u32 reserved195 : 5; + u32 i_plllock_sel_0 : 1; + u32 i_plllock_sel_1 : 1; + u32 i_plllock_cnt_6_0 : 7; + u32 i_plllock_cnt_10_7 : 4; + u32 reserved200 : 4; + } field; + u32 val; +}; + +#define TSPLL_RO_BWM_LF 0x370 +union tspll_ro_bwm_lf { + struct { + u32 bw_freqov_high_cri_7_0 : 8; + u32 bw_freqov_high_cri_9_8 : 2; + u32 biascaldone_cri : 1; + u32 plllock_gain_tran_cri : 1; + u32 plllock_true_lock_cri : 1; + u32 pllunlock_flag_cri : 1; + u32 afcerr_cri : 1; + u32 afcdone_cri : 1; + u32 feedfwrdgain_cal_cri_7_0 : 8; + u32 m2fbdivmod_cri_7_0 : 8; + } field; + u32 val; +}; + +#endif /* _ICE_CGU_REGS_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 157add1268d9..2a1ee60e85f4 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -2189,6 +2189,18 @@ ice_parse_1588_func_caps(struct ice_hw *hw, struct ice_hw_func_caps *func_p, info->clk_freq = (number & ICE_TS_CLK_FREQ_M) >> ICE_TS_CLK_FREQ_S; info->clk_src = ((number & ICE_TS_CLK_SRC_M) != 0); + if (info->clk_freq < NUM_ICE_TIME_REF_FREQ) { + info->time_ref = (enum ice_time_ref_freq)info->clk_freq; + } else { + /* Unknown clock frequency, so assume a (probably incorrect) + * default to avoid out-of-bounds look ups of frequency + * related information. + */ + ice_debug(hw, ICE_DBG_INIT, "1588 func caps: unknown clock frequency %u\n", + info->clk_freq); + info->time_ref = ICE_TIME_REF_FREQ_25_000; + } + ice_debug(hw, ICE_DBG_INIT, "func caps: ieee_1588 = %u\n", func_p->common_cap.ieee_1588); ice_debug(hw, ICE_DBG_INIT, "func caps: src_tmr_owned = %u\n", diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h index a49082485642..d16738a3d3a7 100644 --- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h +++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h @@ -100,6 +100,7 @@ #define PF_SB_ATQT 0x0022FE00 #define PF_SB_ATQT_ATQT_S 0 #define PF_SB_ATQT_ATQT_M ICE_M(0x3FF, 0) +#define PF_SB_REM_DEV_CTL 0x002300F0 #define PRTDCB_GENC 0x00083000 #define PRTDCB_GENC_PFCLDA_S 16 #define PRTDCB_GENC_PFCLDA_M ICE_M(0xFFFF, 16) @@ -440,6 +441,10 @@ #define GLV_UPRCL(_i) (0x003B2000 + ((_i) * 8)) #define GLV_UPTCL(_i) (0x0030A000 + ((_i) * 8)) #define PRTRPB_RDPC 0x000AC260 +#define GLHH_ART_CTL 0x000A41D4 +#define GLHH_ART_CTL_ACTIVE_M BIT(0) +#define GLHH_ART_TIME_H 0x000A41D8 +#define GLHH_ART_TIME_L 0x000A41DC #define GLTSYN_AUX_IN_0(_i) (0x000889D8 + ((_i) * 4)) #define GLTSYN_AUX_IN_0_INT_ENA_M BIT(4) #define GLTSYN_AUX_OUT_0(_i) (0x00088998 + ((_i) * 4)) @@ -452,6 +457,8 @@ #define GLTSYN_ENA_TSYN_ENA_M BIT(0) #define GLTSYN_EVNT_H_0(_i) (0x00088970 + ((_i) * 4)) #define GLTSYN_EVNT_L_0(_i) (0x00088968 + ((_i) * 4)) +#define GLTSYN_HHTIME_H(_i) (0x00088900 + ((_i) * 4)) +#define GLTSYN_HHTIME_L(_i) (0x000888F8 + ((_i) * 4)) #define GLTSYN_INCVAL_H(_i) (0x00088920 + ((_i) * 4)) #define GLTSYN_INCVAL_L(_i) (0x00088918 + ((_i) * 4)) #define GLTSYN_SHADJ_H(_i) (0x00088910 + ((_i) * 4)) @@ -468,6 +475,8 @@ #define GLTSYN_TGT_L_0(_i) (0x00088928 + ((_i) * 4)) #define GLTSYN_TIME_H(_i) (0x000888D8 + ((_i) * 4)) #define GLTSYN_TIME_L(_i) (0x000888D0 + ((_i) * 4)) +#define PFHH_SEM 0x000A4200 /* Reset Source: PFR */ +#define PFHH_SEM_BUSY_M BIT(0) #define PFTSYN_SEM 0x00088880 #define PFTSYN_SEM_BUSY_M BIT(0) #define VSIQF_FD_CNT(_VSI) (0x00464000 + ((_VSI) * 4)) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 865f2231bb24..476533f63e09 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -539,7 +539,7 @@ skip: ice_pf_dis_all_vsi(pf, false); if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) - ice_ptp_release(pf); + ice_ptp_prepare_for_reset(pf); if (hw->port_info) ice_sched_clear_port(hw->port_info); @@ -1063,6 +1063,9 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, if (link_up == old_link && link_speed == old_link_speed) return 0; + if (!ice_is_e810(&pf->hw)) + ice_ptp_link_change(pf, pf->hw.pf_id, link_up); + if (ice_is_dcb_active(pf)) { if (test_bit(ICE_FLAG_DCB_ENA, pf->flags)) ice_dcb_rebuild(pf); @@ -4123,13 +4126,14 @@ static void ice_log_pkg_init(struct ice_hw *hw, enum ice_ddp_state state) break; case ICE_DDP_PKG_LOAD_ERROR: dev_err(dev, "An error occurred on the device while loading the DDP package. The device will be reset.\n"); - /* poll for reset to complete */ - if (ice_check_reset(hw)) - dev_err(dev, "Error resetting device. Please reload the driver\n"); + /* poll for reset to complete */ + if (ice_check_reset(hw)) + dev_err(dev, "Error resetting device. Please reload the driver\n"); break; case ICE_DDP_PKG_ERR: default: dev_err(dev, "An unknown error occurred when loading the DDP package. Entering Safe Mode.\n"); + break; } } @@ -5839,6 +5843,8 @@ static int ice_up_complete(struct ice_vsi *vsi) ice_print_link_msg(vsi, true); netif_tx_start_all_queues(vsi->netdev); netif_carrier_on(vsi->netdev); + if (!ice_is_e810(&pf->hw)) + ice_ptp_link_change(pf, pf->hw.pf_id, true); } /* clear this now, and the first stats read will be used as baseline */ @@ -6239,6 +6245,8 @@ int ice_down(struct ice_vsi *vsi) WARN_ON(!test_bit(ICE_VSI_DOWN, vsi->state)); if (vsi->netdev && vsi->type == ICE_VSI_PF) { + if (!ice_is_e810(&vsi->back->hw)) + ice_ptp_link_change(vsi->back, vsi->back->hw.pf_id, false); netif_carrier_off(vsi->netdev); netif_tx_disable(vsi->netdev); } else if (vsi->type == ICE_VSI_SWITCHDEV_CTRL) { @@ -6685,7 +6693,7 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) * fail. */ if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) - ice_ptp_init(pf); + ice_ptp_reset(pf); /* rebuild PF VSI */ err = ice_vsi_rebuild_by_type(pf, ICE_VSI_PF); @@ -6694,6 +6702,10 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) goto err_vsi_rebuild; } + /* configure PTP timestamping after VSI rebuild */ + if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) + ice_ptp_cfg_timestamp(pf, false); + err = ice_vsi_rebuild_by_type(pf, ICE_VSI_SWITCHDEV_CTRL); if (err) { dev_err(dev, "Switchdev CTRL VSI rebuild failed: %d\n", err); diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c index 0014a1002ed3..d3f65e061a62 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp.c @@ -6,6 +6,8 @@ #define E810_OUT_PROP_DELAY_NS 1 +#define UNKNOWN_INCVAL_E822 0x100000000ULL + static const struct ptp_pin_desc ice_pin_desc_e810t[] = { /* name idx func chan */ { "GNSS", GNSS, PTP_PF_EXTTS, 0, { 0, } }, @@ -281,6 +283,8 @@ static void ice_set_tx_tstamp(struct ice_pf *pf, bool on) else val &= ~PFINT_OICR_TSYN_TX_M; wr32(&pf->hw, PFINT_OICR_ENA, val); + + pf->ptp.tstamp_config.tx_type = on ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; } /** @@ -303,6 +307,9 @@ static void ice_set_rx_tstamp(struct ice_pf *pf, bool on) continue; vsi->rx_rings[i]->ptp_rx = on; } + + pf->ptp.tstamp_config.rx_filter = on ? HWTSTAMP_FILTER_ALL : + HWTSTAMP_FILTER_NONE; } /** @@ -313,18 +320,10 @@ static void ice_set_rx_tstamp(struct ice_pf *pf, bool on) * This function will configure timestamping during PTP initialization * and deinitialization */ -static void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena) +void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena) { ice_set_tx_tstamp(pf, ena); ice_set_rx_tstamp(pf, ena); - - if (ena) { - pf->ptp.tstamp_config.rx_filter = HWTSTAMP_FILTER_ALL; - pf->ptp.tstamp_config.tx_type = HWTSTAMP_TX_ON; - } else { - pf->ptp.tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE; - pf->ptp.tstamp_config.tx_type = HWTSTAMP_TX_OFF; - } } /** @@ -682,6 +681,406 @@ static int ice_ptp_write_adj(struct ice_pf *pf, s32 adj) } /** + * ice_base_incval - Get base timer increment value + * @pf: Board private structure + * + * Look up the base timer increment value for this device. The base increment + * value is used to define the nominal clock tick rate. This increment value + * is programmed during device initialization. It is also used as the basis + * for calculating adjustments using scaled_ppm. + */ +static u64 ice_base_incval(struct ice_pf *pf) +{ + struct ice_hw *hw = &pf->hw; + u64 incval; + + if (ice_is_e810(hw)) + incval = ICE_PTP_NOMINAL_INCVAL_E810; + else if (ice_e822_time_ref(hw) < NUM_ICE_TIME_REF_FREQ) + incval = ice_e822_nominal_incval(ice_e822_time_ref(hw)); + else + incval = UNKNOWN_INCVAL_E822; + + dev_dbg(ice_pf_to_dev(pf), "PTP: using base increment value of 0x%016llx\n", + incval); + + return incval; +} + +/** + * ice_ptp_reset_ts_memory_quad - Reset timestamp memory for one quad + * @pf: The PF private data structure + * @quad: The quad (0-4) + */ +static void ice_ptp_reset_ts_memory_quad(struct ice_pf *pf, int quad) +{ + struct ice_hw *hw = &pf->hw; + + ice_write_quad_reg_e822(hw, quad, Q_REG_TS_CTRL, Q_REG_TS_CTRL_M); + ice_write_quad_reg_e822(hw, quad, Q_REG_TS_CTRL, ~(u32)Q_REG_TS_CTRL_M); +} + +/** + * ice_ptp_check_tx_fifo - Check whether Tx FIFO is in an OK state + * @port: PTP port for which Tx FIFO is checked + */ +static int ice_ptp_check_tx_fifo(struct ice_ptp_port *port) +{ + int quad = port->port_num / ICE_PORTS_PER_QUAD; + int offs = port->port_num % ICE_PORTS_PER_QUAD; + struct ice_pf *pf; + struct ice_hw *hw; + u32 val, phy_sts; + int err; + + pf = ptp_port_to_pf(port); + hw = &pf->hw; + + if (port->tx_fifo_busy_cnt == FIFO_OK) + return 0; + + /* need to read FIFO state */ + if (offs == 0 || offs == 1) + err = ice_read_quad_reg_e822(hw, quad, Q_REG_FIFO01_STATUS, + &val); + else + err = ice_read_quad_reg_e822(hw, quad, Q_REG_FIFO23_STATUS, + &val); + + if (err) { + dev_err(ice_pf_to_dev(pf), "PTP failed to check port %d Tx FIFO, err %d\n", + port->port_num, err); + return err; + } + + if (offs & 0x1) + phy_sts = (val & Q_REG_FIFO13_M) >> Q_REG_FIFO13_S; + else + phy_sts = (val & Q_REG_FIFO02_M) >> Q_REG_FIFO02_S; + + if (phy_sts & FIFO_EMPTY) { + port->tx_fifo_busy_cnt = FIFO_OK; + return 0; + } + + port->tx_fifo_busy_cnt++; + + dev_dbg(ice_pf_to_dev(pf), "Try %d, port %d FIFO not empty\n", + port->tx_fifo_busy_cnt, port->port_num); + + if (port->tx_fifo_busy_cnt == ICE_PTP_FIFO_NUM_CHECKS) { + dev_dbg(ice_pf_to_dev(pf), + "Port %d Tx FIFO still not empty; resetting quad %d\n", + port->port_num, quad); + ice_ptp_reset_ts_memory_quad(pf, quad); + port->tx_fifo_busy_cnt = FIFO_OK; + return 0; + } + + return -EAGAIN; +} + +/** + * ice_ptp_check_tx_offset_valid - Check if the Tx PHY offset is valid + * @port: the PTP port to check + * + * Checks whether the Tx offset for the PHY associated with this port is + * valid. Returns 0 if the offset is valid, and a non-zero error code if it is + * not. + */ +static int ice_ptp_check_tx_offset_valid(struct ice_ptp_port *port) +{ + struct ice_pf *pf = ptp_port_to_pf(port); + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + u32 val; + int err; + + err = ice_ptp_check_tx_fifo(port); + if (err) + return err; + + err = ice_read_phy_reg_e822(hw, port->port_num, P_REG_TX_OV_STATUS, + &val); + if (err) { + dev_err(dev, "Failed to read TX_OV_STATUS for port %d, err %d\n", + port->port_num, err); + return -EAGAIN; + } + + if (!(val & P_REG_TX_OV_STATUS_OV_M)) + return -EAGAIN; + + return 0; +} + +/** + * ice_ptp_check_rx_offset_valid - Check if the Rx PHY offset is valid + * @port: the PTP port to check + * + * Checks whether the Rx offset for the PHY associated with this port is + * valid. Returns 0 if the offset is valid, and a non-zero error code if it is + * not. + */ +static int ice_ptp_check_rx_offset_valid(struct ice_ptp_port *port) +{ + struct ice_pf *pf = ptp_port_to_pf(port); + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + int err; + u32 val; + + err = ice_read_phy_reg_e822(hw, port->port_num, P_REG_RX_OV_STATUS, + &val); + if (err) { + dev_err(dev, "Failed to read RX_OV_STATUS for port %d, err %d\n", + port->port_num, err); + return err; + } + + if (!(val & P_REG_RX_OV_STATUS_OV_M)) + return -EAGAIN; + + return 0; +} + +/** + * ice_ptp_check_offset_valid - Check port offset valid bit + * @port: Port for which offset valid bit is checked + * + * Returns 0 if both Tx and Rx offset are valid, and -EAGAIN if one of the + * offset is not ready. + */ +static int ice_ptp_check_offset_valid(struct ice_ptp_port *port) +{ + int tx_err, rx_err; + + /* always check both Tx and Rx offset validity */ + tx_err = ice_ptp_check_tx_offset_valid(port); + rx_err = ice_ptp_check_rx_offset_valid(port); + + if (tx_err || rx_err) + return -EAGAIN; + + return 0; +} + +/** + * ice_ptp_wait_for_offset_valid - Check for valid Tx and Rx offsets + * @work: Pointer to the kthread_work structure for this task + * + * Check whether both the Tx and Rx offsets are valid for enabling the vernier + * calibration. + * + * Once we have valid offsets from hardware, update the total Tx and Rx + * offsets, and exit bypass mode. This enables more precise timestamps using + * the extra data measured during the vernier calibration process. + */ +static void ice_ptp_wait_for_offset_valid(struct kthread_work *work) +{ + struct ice_ptp_port *port; + int err; + struct device *dev; + struct ice_pf *pf; + struct ice_hw *hw; + + port = container_of(work, struct ice_ptp_port, ov_work.work); + pf = ptp_port_to_pf(port); + hw = &pf->hw; + dev = ice_pf_to_dev(pf); + + if (ice_ptp_check_offset_valid(port)) { + /* Offsets not ready yet, try again later */ + kthread_queue_delayed_work(pf->ptp.kworker, + &port->ov_work, + msecs_to_jiffies(100)); + return; + } + + /* Offsets are valid, so it is safe to exit bypass mode */ + err = ice_phy_exit_bypass_e822(hw, port->port_num); + if (err) { + dev_warn(dev, "Failed to exit bypass mode for PHY port %u, err %d\n", + port->port_num, err); + return; + } +} + +/** + * ice_ptp_port_phy_stop - Stop timestamping for a PHY port + * @ptp_port: PTP port to stop + */ +static int +ice_ptp_port_phy_stop(struct ice_ptp_port *ptp_port) +{ + struct ice_pf *pf = ptp_port_to_pf(ptp_port); + u8 port = ptp_port->port_num; + struct ice_hw *hw = &pf->hw; + int err; + + if (ice_is_e810(hw)) + return 0; + + mutex_lock(&ptp_port->ps_lock); + + kthread_cancel_delayed_work_sync(&ptp_port->ov_work); + + err = ice_stop_phy_timer_e822(hw, port, true); + if (err) + dev_err(ice_pf_to_dev(pf), "PTP failed to set PHY port %d down, err %d\n", + port, err); + + mutex_unlock(&ptp_port->ps_lock); + + return err; +} + +/** + * ice_ptp_port_phy_restart - (Re)start and calibrate PHY timestamping + * @ptp_port: PTP port for which the PHY start is set + * + * Start the PHY timestamping block, and initiate Vernier timestamping + * calibration. If timestamping cannot be calibrated (such as if link is down) + * then disable the timestamping block instead. + */ +static int +ice_ptp_port_phy_restart(struct ice_ptp_port *ptp_port) +{ + struct ice_pf *pf = ptp_port_to_pf(ptp_port); + u8 port = ptp_port->port_num; + struct ice_hw *hw = &pf->hw; + int err; + + if (ice_is_e810(hw)) + return 0; + + if (!ptp_port->link_up) + return ice_ptp_port_phy_stop(ptp_port); + + mutex_lock(&ptp_port->ps_lock); + + kthread_cancel_delayed_work_sync(&ptp_port->ov_work); + + /* temporarily disable Tx timestamps while calibrating PHY offset */ + ptp_port->tx.calibrating = true; + ptp_port->tx_fifo_busy_cnt = 0; + + /* Start the PHY timer in bypass mode */ + err = ice_start_phy_timer_e822(hw, port, true); + if (err) + goto out_unlock; + + /* Enable Tx timestamps right away */ + ptp_port->tx.calibrating = false; + + kthread_queue_delayed_work(pf->ptp.kworker, &ptp_port->ov_work, 0); + +out_unlock: + if (err) + dev_err(ice_pf_to_dev(pf), "PTP failed to set PHY port %d up, err %d\n", + port, err); + + mutex_unlock(&ptp_port->ps_lock); + + return err; +} + +/** + * ice_ptp_link_change - Set or clear port registers for timestamping + * @pf: Board private structure + * @port: Port for which the PHY start is set + * @linkup: Link is up or down + */ +int ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup) +{ + struct ice_ptp_port *ptp_port; + + if (!test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) + return 0; + + if (port >= ICE_NUM_EXTERNAL_PORTS) + return -EINVAL; + + ptp_port = &pf->ptp.port; + if (ptp_port->port_num != port) + return -EINVAL; + + /* Update cached link err for this port immediately */ + ptp_port->link_up = linkup; + + if (!test_bit(ICE_FLAG_PTP, pf->flags)) + /* PTP is not setup */ + return -EAGAIN; + + return ice_ptp_port_phy_restart(ptp_port); +} + +/** + * ice_ptp_reset_ts_memory - Reset timestamp memory for all quads + * @pf: The PF private data structure + */ +static void ice_ptp_reset_ts_memory(struct ice_pf *pf) +{ + int quad; + + quad = pf->hw.port_info->lport / ICE_PORTS_PER_QUAD; + ice_ptp_reset_ts_memory_quad(pf, quad); +} + +/** + * ice_ptp_tx_ena_intr - Enable or disable the Tx timestamp interrupt + * @pf: PF private structure + * @ena: bool value to enable or disable interrupt + * @threshold: Minimum number of packets at which intr is triggered + * + * Utility function to enable or disable Tx timestamp interrupt and threshold + */ +static int ice_ptp_tx_ena_intr(struct ice_pf *pf, bool ena, u32 threshold) +{ + struct ice_hw *hw = &pf->hw; + int err = 0; + int quad; + u32 val; + + ice_ptp_reset_ts_memory(pf); + + for (quad = 0; quad < ICE_MAX_QUAD; quad++) { + err = ice_read_quad_reg_e822(hw, quad, Q_REG_TX_MEM_GBL_CFG, + &val); + if (err) + break; + + if (ena) { + val |= Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M; + val &= ~Q_REG_TX_MEM_GBL_CFG_INTR_THR_M; + val |= ((threshold << Q_REG_TX_MEM_GBL_CFG_INTR_THR_S) & + Q_REG_TX_MEM_GBL_CFG_INTR_THR_M); + } else { + val &= ~Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M; + } + + err = ice_write_quad_reg_e822(hw, quad, Q_REG_TX_MEM_GBL_CFG, + val); + if (err) + break; + } + + if (err) + dev_err(ice_pf_to_dev(pf), "PTP failed in intr ena, err %d\n", + err); + return err; +} + +/** + * ice_ptp_reset_phy_timestamping - Reset PHY timestamping block + * @pf: Board private structure + */ +static void ice_ptp_reset_phy_timestamping(struct ice_pf *pf) +{ + ice_ptp_port_phy_restart(&pf->ptp.port); +} + +/** * ice_ptp_adjfine - Adjust clock increment rate * @info: the driver's PTP info structure * @scaled_ppm: Parts per million with 16-bit fractional field @@ -698,7 +1097,7 @@ static int ice_ptp_adjfine(struct ptp_clock_info *info, long scaled_ppm) int neg_adj = 0; int err; - incval = ICE_PTP_NOMINAL_INCVAL_E810; + incval = ice_base_incval(pf); if (scaled_ppm < 0) { neg_adj = 1; @@ -905,7 +1304,10 @@ static int ice_ptp_cfg_clkout(struct ice_pf *pf, unsigned int chan, start_time = div64_u64(current_time + NSEC_PER_SEC - 1, NSEC_PER_SEC) * NSEC_PER_SEC + phase; - start_time -= E810_OUT_PROP_DELAY_NS; + if (ice_is_e810(hw)) + start_time -= E810_OUT_PROP_DELAY_NS; + else + start_time -= ice_e822_pps_delay(ice_e822_time_ref(hw)); /* 2. Write TARGET time */ wr32(hw, GLTSYN_TGT_L(chan, tmr_idx), lower_32_bits(start_time)); @@ -1088,6 +1490,12 @@ ice_ptp_settime64(struct ptp_clock_info *info, const struct timespec64 *ts) struct ice_hw *hw = &pf->hw; int err; + /* For Vernier mode, we need to recalibrate after new settime + * Start with disabling timestamp block + */ + if (pf->ptp.port.link_up) + ice_ptp_port_phy_stop(&pf->ptp.port); + if (!ice_ptp_lock(hw)) { err = -EBUSY; goto exit; @@ -1104,6 +1512,10 @@ ice_ptp_settime64(struct ptp_clock_info *info, const struct timespec64 *ts) /* Reenable periodic outputs */ ice_ptp_enable_all_clkout(pf); + + /* Recalibrate and re-enable timestamp block */ + if (pf->ptp.port.link_up) + ice_ptp_port_phy_restart(&pf->ptp.port); exit: if (err) { dev_err(ice_pf_to_dev(pf), "PTP failed to set time %d\n", err); @@ -1177,6 +1589,101 @@ static int ice_ptp_adjtime(struct ptp_clock_info *info, s64 delta) return 0; } +#ifdef CONFIG_ICE_HWTS +/** + * ice_ptp_get_syncdevicetime - Get the cross time stamp info + * @device: Current device time + * @system: System counter value read synchronously with device time + * @ctx: Context provided by timekeeping code + * + * Read device and system (ART) clock simultaneously and return the corrected + * clock values in ns. + */ +static int +ice_ptp_get_syncdevicetime(ktime_t *device, + struct system_counterval_t *system, + void *ctx) +{ + struct ice_pf *pf = (struct ice_pf *)ctx; + struct ice_hw *hw = &pf->hw; + u32 hh_lock, hh_art_ctl; + int i; + + /* Get the HW lock */ + hh_lock = rd32(hw, PFHH_SEM + (PFTSYN_SEM_BYTES * hw->pf_id)); + if (hh_lock & PFHH_SEM_BUSY_M) { + dev_err(ice_pf_to_dev(pf), "PTP failed to get hh lock\n"); + return -EFAULT; + } + + /* Start the ART and device clock sync sequence */ + hh_art_ctl = rd32(hw, GLHH_ART_CTL); + hh_art_ctl = hh_art_ctl | GLHH_ART_CTL_ACTIVE_M; + wr32(hw, GLHH_ART_CTL, hh_art_ctl); + +#define MAX_HH_LOCK_TRIES 100 + + for (i = 0; i < MAX_HH_LOCK_TRIES; i++) { + /* Wait for sync to complete */ + hh_art_ctl = rd32(hw, GLHH_ART_CTL); + if (hh_art_ctl & GLHH_ART_CTL_ACTIVE_M) { + udelay(1); + continue; + } else { + u32 hh_ts_lo, hh_ts_hi, tmr_idx; + u64 hh_ts; + + tmr_idx = hw->func_caps.ts_func_info.tmr_index_assoc; + /* Read ART time */ + hh_ts_lo = rd32(hw, GLHH_ART_TIME_L); + hh_ts_hi = rd32(hw, GLHH_ART_TIME_H); + hh_ts = ((u64)hh_ts_hi << 32) | hh_ts_lo; + *system = convert_art_ns_to_tsc(hh_ts); + /* Read Device source clock time */ + hh_ts_lo = rd32(hw, GLTSYN_HHTIME_L(tmr_idx)); + hh_ts_hi = rd32(hw, GLTSYN_HHTIME_H(tmr_idx)); + hh_ts = ((u64)hh_ts_hi << 32) | hh_ts_lo; + *device = ns_to_ktime(hh_ts); + break; + } + } + /* Release HW lock */ + hh_lock = rd32(hw, PFHH_SEM + (PFTSYN_SEM_BYTES * hw->pf_id)); + hh_lock = hh_lock & ~PFHH_SEM_BUSY_M; + wr32(hw, PFHH_SEM + (PFTSYN_SEM_BYTES * hw->pf_id), hh_lock); + + if (i == MAX_HH_LOCK_TRIES) + return -ETIMEDOUT; + + return 0; +} + +/** + * ice_ptp_getcrosststamp_e822 - Capture a device cross timestamp + * @info: the driver's PTP info structure + * @cts: The memory to fill the cross timestamp info + * + * Capture a cross timestamp between the ART and the device PTP hardware + * clock. Fill the cross timestamp information and report it back to the + * caller. + * + * This is only valid for E822 devices which have support for generating the + * cross timestamp via PCIe PTM. + * + * In order to correctly correlate the ART timestamp back to the TSC time, the + * CPU must have X86_FEATURE_TSC_KNOWN_FREQ. + */ +static int +ice_ptp_getcrosststamp_e822(struct ptp_clock_info *info, + struct system_device_crosststamp *cts) +{ + struct ice_pf *pf = ptp_info_to_pf(info); + + return get_device_system_crosststamp(ice_ptp_get_syncdevicetime, + pf, NULL, cts); +} +#endif /* CONFIG_ICE_HWTS */ + /** * ice_ptp_get_ts_config - ioctl interface to read the timestamping config * @pf: Board private structure @@ -1234,7 +1741,6 @@ ice_ptp_set_timestamp_mode(struct ice_pf *pf, struct hwtstamp_config *config) case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: case HWTSTAMP_FILTER_ALL: - config->rx_filter = HWTSTAMP_FILTER_ALL; ice_set_rx_tstamp(pf, true); break; default: @@ -1266,8 +1772,8 @@ int ice_ptp_set_ts_config(struct ice_pf *pf, struct ifreq *ifr) if (err) return err; - /* Save these settings for future reference */ - pf->ptp.tstamp_config = config; + /* Return the actual configuration set */ + config = pf->ptp.tstamp_config; return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? -EFAULT : 0; @@ -1399,6 +1905,26 @@ static void ice_ptp_setup_pins_e810(struct ptp_clock_info *info) } /** + * ice_ptp_set_funcs_e822 - Set specialized functions for E822 support + * @pf: Board private structure + * @info: PTP info to fill + * + * Assign functions to the PTP capabiltiies structure for E822 devices. + * Functions which operate across all device families should be set directly + * in ice_ptp_set_caps. Only add functions here which are distinct for E822 + * devices. + */ +static void +ice_ptp_set_funcs_e822(struct ice_pf *pf, struct ptp_clock_info *info) +{ +#ifdef CONFIG_ICE_HWTS + if (boot_cpu_has(X86_FEATURE_ART) && + boot_cpu_has(X86_FEATURE_TSC_KNOWN_FREQ)) + info->getcrosststamp = ice_ptp_getcrosststamp_e822; +#endif /* CONFIG_ICE_HWTS */ +} + +/** * ice_ptp_set_funcs_e810 - Set specialized functions for E810 support * @pf: Board private structure * @info: PTP info to fill @@ -1437,7 +1963,10 @@ static void ice_ptp_set_caps(struct ice_pf *pf) info->gettimex64 = ice_ptp_gettimex64; info->settime64 = ice_ptp_settime64; - ice_ptp_set_funcs_e810(pf, info); + if (ice_is_e810(&pf->hw)) + ice_ptp_set_funcs_e810(pf, info); + else + ice_ptp_set_funcs_e822(pf, info); } /** @@ -1584,7 +2113,7 @@ s8 ice_ptp_request_ts(struct ice_ptp_tx *tx, struct sk_buff *skb) u8 idx; /* Check if this tracker is initialized */ - if (!tx->init) + if (!tx->init || tx->calibrating) return -1; spin_lock(&tx->lock); @@ -1707,6 +2236,27 @@ ice_ptp_release_tx_tracker(struct ice_pf *pf, struct ice_ptp_tx *tx) } /** + * ice_ptp_init_tx_e822 - Initialize tracking for Tx timestamps + * @pf: Board private structure + * @tx: the Tx tracking structure to initialize + * @port: the port this structure tracks + * + * Initialize the Tx timestamp tracker for this port. For generic MAC devices, + * the timestamp block is shared for all ports in the same quad. To avoid + * ports using the same timestamp index, logically break the block of + * registers into chunks based on the port number. + */ +static int +ice_ptp_init_tx_e822(struct ice_pf *pf, struct ice_ptp_tx *tx, u8 port) +{ + tx->quad = port / ICE_PORTS_PER_QUAD; + tx->quad_offset = tx->quad * INDEX_PER_PORT; + tx->len = INDEX_PER_PORT; + + return ice_ptp_alloc_tx_tracker(tx); +} + +/** * ice_ptp_init_tx_e810 - Initialize tracking for Tx timestamps * @pf: Board private structure * @tx: the Tx tracking structure to initialize @@ -1777,6 +2327,130 @@ static void ice_ptp_periodic_work(struct kthread_work *work) } /** + * ice_ptp_reset - Initialize PTP hardware clock support after reset + * @pf: Board private structure + */ +void ice_ptp_reset(struct ice_pf *pf) +{ + struct ice_ptp *ptp = &pf->ptp; + struct ice_hw *hw = &pf->hw; + struct timespec64 ts; + int err, itr = 1; + u64 time_diff; + + if (test_bit(ICE_PFR_REQ, pf->state)) + goto pfr; + + if (!hw->func_caps.ts_func_info.src_tmr_owned) + goto reset_ts; + + err = ice_ptp_init_phc(hw); + if (err) + goto err; + + /* Acquire the global hardware lock */ + if (!ice_ptp_lock(hw)) { + err = -EBUSY; + goto err; + } + + /* Write the increment time value to PHY and LAN */ + err = ice_ptp_write_incval(hw, ice_base_incval(pf)); + if (err) { + ice_ptp_unlock(hw); + goto err; + } + + /* Write the initial Time value to PHY and LAN using the cached PHC + * time before the reset and time difference between stopping and + * starting the clock. + */ + if (ptp->cached_phc_time) { + time_diff = ktime_get_real_ns() - ptp->reset_time; + ts = ns_to_timespec64(ptp->cached_phc_time + time_diff); + } else { + ts = ktime_to_timespec64(ktime_get_real()); + } + err = ice_ptp_write_init(pf, &ts); + if (err) { + ice_ptp_unlock(hw); + goto err; + } + + /* Release the global hardware lock */ + ice_ptp_unlock(hw); + + if (!ice_is_e810(hw)) { + /* Enable quad interrupts */ + err = ice_ptp_tx_ena_intr(pf, true, itr); + if (err) + goto err; + } + +reset_ts: + /* Restart the PHY timestamping block */ + ice_ptp_reset_phy_timestamping(pf); + +pfr: + /* Init Tx structures */ + if (ice_is_e810(&pf->hw)) { + err = ice_ptp_init_tx_e810(pf, &ptp->port.tx); + } else { + kthread_init_delayed_work(&ptp->port.ov_work, + ice_ptp_wait_for_offset_valid); + err = ice_ptp_init_tx_e822(pf, &ptp->port.tx, + ptp->port.port_num); + } + if (err) + goto err; + + set_bit(ICE_FLAG_PTP, pf->flags); + + /* Start periodic work going */ + kthread_queue_delayed_work(ptp->kworker, &ptp->work, 0); + + dev_info(ice_pf_to_dev(pf), "PTP reset successful\n"); + return; + +err: + dev_err(ice_pf_to_dev(pf), "PTP reset failed %d\n", err); +} + +/** + * ice_ptp_prepare_for_reset - Prepare PTP for reset + * @pf: Board private structure + */ +void ice_ptp_prepare_for_reset(struct ice_pf *pf) +{ + struct ice_ptp *ptp = &pf->ptp; + u8 src_tmr; + + clear_bit(ICE_FLAG_PTP, pf->flags); + + /* Disable timestamping for both Tx and Rx */ + ice_ptp_cfg_timestamp(pf, false); + + kthread_cancel_delayed_work_sync(&ptp->work); + kthread_cancel_work_sync(&ptp->extts_work); + + if (test_bit(ICE_PFR_REQ, pf->state)) + return; + + ice_ptp_release_tx_tracker(pf, &pf->ptp.port.tx); + + /* Disable periodic outputs */ + ice_ptp_disable_all_clkout(pf); + + src_tmr = ice_get_ptp_src_clock_index(&pf->hw); + + /* Disable source clock */ + wr32(&pf->hw, GLTSYN_ENA(src_tmr), (u32)~GLTSYN_ENA_TSYN_ENA_M); + + /* Acquire PHC and system timer to restore after reset */ + ptp->reset_time = ktime_get_real_ns(); +} + +/** * ice_ptp_init_owner - Initialize PTP_1588_CLOCK device * @pf: Board private structure * @@ -1786,27 +2460,16 @@ static void ice_ptp_periodic_work(struct kthread_work *work) */ static int ice_ptp_init_owner(struct ice_pf *pf) { - struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; struct timespec64 ts; - u8 src_idx; - int err; - - wr32(hw, GLTSYN_SYNC_DLAY, 0); - - /* Clear some HW residue and enable source clock */ - src_idx = hw->func_caps.ts_func_info.tmr_index_owned; + int err, itr = 1; - /* Enable source clocks */ - wr32(hw, GLTSYN_ENA(src_idx), GLTSYN_ENA_TSYN_ENA_M); - - /* Enable PHY time sync */ - err = ice_ptp_init_phy_e810(hw); - if (err) - goto err_exit; - - /* Clear event status indications for auxiliary pins */ - (void)rd32(hw, GLTSYN_STAT(src_idx)); + err = ice_ptp_init_phc(hw); + if (err) { + dev_err(ice_pf_to_dev(pf), "Failed to initialize PHC, err %d\n", + err); + return err; + } /* Acquire the global hardware lock */ if (!ice_ptp_lock(hw)) { @@ -1815,7 +2478,7 @@ static int ice_ptp_init_owner(struct ice_pf *pf) } /* Write the increment time value to PHY and LAN */ - err = ice_ptp_write_incval(hw, ICE_PTP_NOMINAL_INCVAL_E810); + err = ice_ptp_write_incval(hw, ice_base_incval(pf)); if (err) { ice_ptp_unlock(hw); goto err_exit; @@ -1832,6 +2495,13 @@ static int ice_ptp_init_owner(struct ice_pf *pf) /* Release the global hardware lock */ ice_ptp_unlock(hw); + if (!ice_is_e810(hw)) { + /* Enable quad interrupts */ + err = ice_ptp_tx_ena_intr(pf, true, itr); + if (err) + goto err_exit; + } + /* Ensure we have a clock device */ err = ice_ptp_create_clock(pf); if (err) @@ -1845,72 +2515,106 @@ static int ice_ptp_init_owner(struct ice_pf *pf) err_clk: pf->ptp.clock = NULL; err_exit: - dev_err(dev, "PTP failed to register clock, err %d\n", err); - return err; } /** - * ice_ptp_init - Initialize the PTP support after device probe or reset + * ice_ptp_init_work - Initialize PTP work threads + * @pf: Board private structure + * @ptp: PF PTP structure + */ +static int ice_ptp_init_work(struct ice_pf *pf, struct ice_ptp *ptp) +{ + struct kthread_worker *kworker; + + /* Initialize work functions */ + kthread_init_delayed_work(&ptp->work, ice_ptp_periodic_work); + kthread_init_work(&ptp->extts_work, ice_ptp_extts_work); + + /* Allocate a kworker for handling work required for the ports + * connected to the PTP hardware clock. + */ + kworker = kthread_create_worker(0, "ice-ptp-%s", + dev_name(ice_pf_to_dev(pf))); + if (IS_ERR(kworker)) + return PTR_ERR(kworker); + + ptp->kworker = kworker; + + /* Start periodic work going */ + kthread_queue_delayed_work(ptp->kworker, &ptp->work, 0); + + return 0; +} + +/** + * ice_ptp_init_port - Initialize PTP port structure + * @pf: Board private structure + * @ptp_port: PTP port structure + */ +static int ice_ptp_init_port(struct ice_pf *pf, struct ice_ptp_port *ptp_port) +{ + mutex_init(&ptp_port->ps_lock); + + if (ice_is_e810(&pf->hw)) + return ice_ptp_init_tx_e810(pf, &ptp_port->tx); + + kthread_init_delayed_work(&ptp_port->ov_work, + ice_ptp_wait_for_offset_valid); + return ice_ptp_init_tx_e822(pf, &ptp_port->tx, ptp_port->port_num); +} + +/** + * ice_ptp_init - Initialize PTP hardware clock support * @pf: Board private structure * - * This function sets device up for PTP support. The first time it is run, it - * will create a clock device. It does not create a clock device if one - * already exists. It also reconfigures the device after a reset. + * Set up the device for interacting with the PTP hardware clock for all + * functions, both the function that owns the clock hardware, and the + * functions connected to the clock hardware. + * + * The clock owner will allocate and register a ptp_clock with the + * PTP_1588_CLOCK infrastructure. All functions allocate a kthread and work + * items used for asynchronous work such as Tx timestamps and periodic work. */ void ice_ptp_init(struct ice_pf *pf) { - struct device *dev = ice_pf_to_dev(pf); - struct kthread_worker *kworker; + struct ice_ptp *ptp = &pf->ptp; struct ice_hw *hw = &pf->hw; int err; - /* PTP is currently only supported on E810 devices */ - if (!ice_is_e810(hw)) - return; - - /* Check if this PF owns the source timer */ + /* If this function owns the clock hardware, it must allocate and + * configure the PTP clock device to represent it. + */ if (hw->func_caps.ts_func_info.src_tmr_owned) { err = ice_ptp_init_owner(pf); if (err) - return; + goto err; } - /* Disable timestamping for both Tx and Rx */ - ice_ptp_cfg_timestamp(pf, false); - - /* Initialize the PTP port Tx timestamp tracker */ - ice_ptp_init_tx_e810(pf, &pf->ptp.port.tx); - - /* Initialize work functions */ - kthread_init_delayed_work(&pf->ptp.work, ice_ptp_periodic_work); - kthread_init_work(&pf->ptp.extts_work, ice_ptp_extts_work); + ptp->port.port_num = hw->pf_id; + err = ice_ptp_init_port(pf, &ptp->port); + if (err) + goto err; - /* Allocate a kworker for handling work required for the ports - * connected to the PTP hardware clock. - */ - kworker = kthread_create_worker(0, "ice-ptp-%s", dev_name(dev)); - if (IS_ERR(kworker)) { - err = PTR_ERR(kworker); - goto err_kworker; - } - pf->ptp.kworker = kworker; + /* Start the PHY timestamping block */ + ice_ptp_reset_phy_timestamping(pf); set_bit(ICE_FLAG_PTP, pf->flags); + err = ice_ptp_init_work(pf, ptp); + if (err) + goto err; - /* Start periodic work going */ - kthread_queue_delayed_work(pf->ptp.kworker, &pf->ptp.work, 0); - - dev_info(dev, "PTP init successful\n"); + dev_info(ice_pf_to_dev(pf), "PTP init successful\n"); return; -err_kworker: +err: /* If we registered a PTP clock, release it */ if (pf->ptp.clock) { - ptp_clock_unregister(pf->ptp.clock); + ptp_clock_unregister(ptp->clock); pf->ptp.clock = NULL; } - dev_err(dev, "PTP failed %d\n", err); + clear_bit(ICE_FLAG_PTP, pf->flags); + dev_err(ice_pf_to_dev(pf), "PTP failed %d\n", err); } /** @@ -1934,6 +2638,8 @@ void ice_ptp_release(struct ice_pf *pf) kthread_cancel_delayed_work_sync(&pf->ptp.work); + ice_ptp_port_phy_stop(&pf->ptp.port); + mutex_destroy(&pf->ptp.port.ps_lock); if (pf->ptp.kworker) { kthread_destroy_worker(pf->ptp.kworker); pf->ptp.kworker = NULL; diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.h b/drivers/net/ethernet/intel/ice/ice_ptp.h index 53c15fc9d996..afd048d69959 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.h +++ b/drivers/net/ethernet/intel/ice/ice_ptp.h @@ -82,6 +82,8 @@ struct ice_tx_tstamp { * @quad_offset: offset into timestamp block of the quad to get the real index * @len: length of the tstamps and in_use fields. * @init: if true, the tracker is initialized; + * @calibrating: if true, the PHY is calibrating the Tx offset. During this + * window, timestamps are temporarily disabled. */ struct ice_ptp_tx { struct kthread_work work; @@ -92,6 +94,7 @@ struct ice_ptp_tx { u8 quad_offset; u8 len; u8 init; + u8 calibrating; }; /* Quad and port information for initializing timestamp blocks */ @@ -101,15 +104,24 @@ struct ice_ptp_tx { /** * struct ice_ptp_port - data used to initialize an external port for PTP * - * This structure contains PTP data related to the external ports. Currently - * it is used for tracking the Tx timestamps of a port. In the future this - * structure will also hold information for the E822 port initialization - * logic. + * This structure contains data indicating whether a single external port is + * ready for PTP functionality. It is used to track the port initialization + * and determine when the port's PHY offset is valid. * * @tx: Tx timestamp tracking for this port + * @ov_work: delayed work task for tracking when PHY offset is valid + * @ps_lock: mutex used to protect the overall PTP PHY start procedure + * @link_up: indicates whether the link is up + * @tx_fifo_busy_cnt: number of times the Tx FIFO was busy + * @port_num: the port number this structure represents */ struct ice_ptp_port { struct ice_ptp_tx tx; + struct kthread_delayed_work ov_work; + struct mutex ps_lock; /* protects overall PTP PHY start procedure */ + bool link_up; + u8 tx_fifo_busy_cnt; + u8 port_num; }; #define GLTSYN_TGT_H_IDX_MAX 4 @@ -127,6 +139,7 @@ struct ice_ptp_port { * @info: structure defining PTP hardware capabilities * @clock: pointer to registered PTP clock device * @tstamp_config: hardware timestamping configuration + * @reset_time: kernel time after clock stop on reset */ struct ice_ptp { struct ice_ptp_port port; @@ -140,6 +153,7 @@ struct ice_ptp { struct ptp_clock_info info; struct ptp_clock *clock; struct hwtstamp_config tstamp_config; + u64 reset_time; }; #define __ptp_port_to_ptp(p) \ @@ -152,9 +166,15 @@ struct ice_ptp { #define ptp_info_to_pf(i) \ container_of(__ptp_info_to_ptp((i)), struct ice_pf, ptp) +#define PFTSYN_SEM_BYTES 4 #define PTP_SHARED_CLK_IDX_VALID BIT(31) +#define TS_CMD_MASK 0xF +#define SYNC_EXEC_CMD 0x3 #define ICE_PTP_TS_VALID BIT(0) +#define FIFO_EMPTY BIT(2) +#define FIFO_OK 0xFF +#define ICE_PTP_FIFO_NUM_CHECKS 5 /* Per-channel register definitions */ #define GLTSYN_AUX_OUT(_chan, _idx) (GLTSYN_AUX_OUT_0(_idx) + ((_chan) * 8)) #define GLTSYN_AUX_IN(_chan, _idx) (GLTSYN_AUX_IN_0(_idx) + ((_chan) * 8)) @@ -175,11 +195,13 @@ struct ice_ptp { #define N_PER_OUT_E810T 3 #define N_PER_OUT_E810T_NO_SMA 2 #define N_EXT_TS_E810_NO_SMA 2 +#define ETH_GLTSYN_ENA(_i) (0x03000348 + ((_i) * 4)) #if IS_ENABLED(CONFIG_PTP_1588_CLOCK) struct ice_pf; int ice_ptp_set_ts_config(struct ice_pf *pf, struct ifreq *ifr); int ice_ptp_get_ts_config(struct ice_pf *pf, struct ifreq *ifr); +void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena); int ice_get_ptp_clock_index(struct ice_pf *pf); s8 ice_ptp_request_ts(struct ice_ptp_tx *tx, struct sk_buff *skb); @@ -188,8 +210,11 @@ void ice_ptp_process_ts(struct ice_pf *pf); void ice_ptp_rx_hwtstamp(struct ice_rx_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, struct sk_buff *skb); +void ice_ptp_reset(struct ice_pf *pf); +void ice_ptp_prepare_for_reset(struct ice_pf *pf); void ice_ptp_init(struct ice_pf *pf); void ice_ptp_release(struct ice_pf *pf); +int ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup); #else /* IS_ENABLED(CONFIG_PTP_1588_CLOCK) */ static inline int ice_ptp_set_ts_config(struct ice_pf *pf, struct ifreq *ifr) { @@ -201,6 +226,7 @@ static inline int ice_ptp_get_ts_config(struct ice_pf *pf, struct ifreq *ifr) return -EOPNOTSUPP; } +static inline void ice_ptp_cfg_timestamp(struct ice_pf *pf, bool ena) { } static inline int ice_get_ptp_clock_index(struct ice_pf *pf) { return -1; @@ -216,7 +242,11 @@ static inline void ice_ptp_process_ts(struct ice_pf *pf) { } static inline void ice_ptp_rx_hwtstamp(struct ice_rx_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, struct sk_buff *skb) { } +static inline void ice_ptp_reset(struct ice_pf *pf) { } +static inline void ice_ptp_prepare_for_reset(struct ice_pf *pf) { } static inline void ice_ptp_init(struct ice_pf *pf) { } static inline void ice_ptp_release(struct ice_pf *pf) { } +static inline int ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup) +{ return 0; } #endif /* IS_ENABLED(CONFIG_PTP_1588_CLOCK) */ #endif /* _ICE_PTP_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h new file mode 100644 index 000000000000..4109aa3b2fcd --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h @@ -0,0 +1,374 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2018-2021, Intel Corporation. */ + +#ifndef _ICE_PTP_CONSTS_H_ +#define _ICE_PTP_CONSTS_H_ + +/* Constant definitions related to the hardware clock used for PTP 1588 + * features and functionality. + */ +/* Constants defined for the PTP 1588 clock hardware. */ + +/* struct ice_time_ref_info_e822 + * + * E822 hardware can use different sources as the reference for the PTP + * hardware clock. Each clock has different characteristics such as a slightly + * different frequency, etc. + * + * This lookup table defines several constants that depend on the current time + * reference. See the struct ice_time_ref_info_e822 for information about the + * meaning of each constant. + */ +const struct ice_time_ref_info_e822 e822_time_ref[NUM_ICE_TIME_REF_FREQ] = { + /* ICE_TIME_REF_FREQ_25_000 -> 25 MHz */ + { + /* pll_freq */ + 823437500, /* 823.4375 MHz PLL */ + /* nominal_incval */ + 0x136e44fabULL, + /* pps_delay */ + 11, + }, + + /* ICE_TIME_REF_FREQ_122_880 -> 122.88 MHz */ + { + /* pll_freq */ + 783360000, /* 783.36 MHz */ + /* nominal_incval */ + 0x146cc2177ULL, + /* pps_delay */ + 12, + }, + + /* ICE_TIME_REF_FREQ_125_000 -> 125 MHz */ + { + /* pll_freq */ + 796875000, /* 796.875 MHz */ + /* nominal_incval */ + 0x141414141ULL, + /* pps_delay */ + 12, + }, + + /* ICE_TIME_REF_FREQ_153_600 -> 153.6 MHz */ + { + /* pll_freq */ + 816000000, /* 816 MHz */ + /* nominal_incval */ + 0x139b9b9baULL, + /* pps_delay */ + 12, + }, + + /* ICE_TIME_REF_FREQ_156_250 -> 156.25 MHz */ + { + /* pll_freq */ + 830078125, /* 830.78125 MHz */ + /* nominal_incval */ + 0x134679aceULL, + /* pps_delay */ + 11, + }, + + /* ICE_TIME_REF_FREQ_245_760 -> 245.76 MHz */ + { + /* pll_freq */ + 783360000, /* 783.36 MHz */ + /* nominal_incval */ + 0x146cc2177ULL, + /* pps_delay */ + 12, + }, +}; + +const struct ice_cgu_pll_params_e822 e822_cgu_params[NUM_ICE_TIME_REF_FREQ] = { + /* ICE_TIME_REF_FREQ_25_000 -> 25 MHz */ + { + /* refclk_pre_div */ + 1, + /* feedback_div */ + 197, + /* frac_n_div */ + 2621440, + /* post_pll_div */ + 6, + }, + + /* ICE_TIME_REF_FREQ_122_880 -> 122.88 MHz */ + { + /* refclk_pre_div */ + 5, + /* feedback_div */ + 223, + /* frac_n_div */ + 524288, + /* post_pll_div */ + 7, + }, + + /* ICE_TIME_REF_FREQ_125_000 -> 125 MHz */ + { + /* refclk_pre_div */ + 5, + /* feedback_div */ + 223, + /* frac_n_div */ + 524288, + /* post_pll_div */ + 7, + }, + + /* ICE_TIME_REF_FREQ_153_600 -> 153.6 MHz */ + { + /* refclk_pre_div */ + 5, + /* feedback_div */ + 159, + /* frac_n_div */ + 1572864, + /* post_pll_div */ + 6, + }, + + /* ICE_TIME_REF_FREQ_156_250 -> 156.25 MHz */ + { + /* refclk_pre_div */ + 5, + /* feedback_div */ + 159, + /* frac_n_div */ + 1572864, + /* post_pll_div */ + 6, + }, + + /* ICE_TIME_REF_FREQ_245_760 -> 245.76 MHz */ + { + /* refclk_pre_div */ + 10, + /* feedback_div */ + 223, + /* frac_n_div */ + 524288, + /* post_pll_div */ + 7, + }, +}; + +/* struct ice_vernier_info_e822 + * + * E822 hardware calibrates the delay of the timestamp indication from the + * actual packet transmission or reception during the initialization of the + * PHY. To do this, the hardware mechanism uses some conversions between the + * various clocks within the PHY block. This table defines constants used to + * calculate the correct conversion ratios in the PHY registers. + * + * Many of the values relate to the PAR/PCS clock conversion registers. For + * these registers, a value of 0 means that the associated register is not + * used by this link speed, and that the register should be cleared by writing + * 0. Other values specify the clock frequency in Hz. + */ +const struct ice_vernier_info_e822 e822_vernier[NUM_ICE_PTP_LNK_SPD] = { + /* ICE_PTP_LNK_SPD_1G */ + { + /* tx_par_clk */ + 31250000, /* 31.25 MHz */ + /* rx_par_clk */ + 31250000, /* 31.25 MHz */ + /* tx_pcs_clk */ + 125000000, /* 125 MHz */ + /* rx_pcs_clk */ + 125000000, /* 125 MHz */ + /* tx_desk_rsgb_par */ + 0, /* unused */ + /* rx_desk_rsgb_par */ + 0, /* unused */ + /* tx_desk_rsgb_pcs */ + 0, /* unused */ + /* rx_desk_rsgb_pcs */ + 0, /* unused */ + /* tx_fixed_delay */ + 25140, + /* pmd_adj_divisor */ + 10000000, + /* rx_fixed_delay */ + 17372, + }, + /* ICE_PTP_LNK_SPD_10G */ + { + /* tx_par_clk */ + 257812500, /* 257.8125 MHz */ + /* rx_par_clk */ + 257812500, /* 257.8125 MHz */ + /* tx_pcs_clk */ + 156250000, /* 156.25 MHz */ + /* rx_pcs_clk */ + 156250000, /* 156.25 MHz */ + /* tx_desk_rsgb_par */ + 0, /* unused */ + /* rx_desk_rsgb_par */ + 0, /* unused */ + /* tx_desk_rsgb_pcs */ + 0, /* unused */ + /* rx_desk_rsgb_pcs */ + 0, /* unused */ + /* tx_fixed_delay */ + 6938, + /* pmd_adj_divisor */ + 82500000, + /* rx_fixed_delay */ + 6212, + }, + /* ICE_PTP_LNK_SPD_25G */ + { + /* tx_par_clk */ + 644531250, /* 644.53125 MHZ */ + /* rx_par_clk */ + 644531250, /* 644.53125 MHz */ + /* tx_pcs_clk */ + 390625000, /* 390.625 MHz */ + /* rx_pcs_clk */ + 390625000, /* 390.625 MHz */ + /* tx_desk_rsgb_par */ + 0, /* unused */ + /* rx_desk_rsgb_par */ + 0, /* unused */ + /* tx_desk_rsgb_pcs */ + 0, /* unused */ + /* rx_desk_rsgb_pcs */ + 0, /* unused */ + /* tx_fixed_delay */ + 2778, + /* pmd_adj_divisor */ + 206250000, + /* rx_fixed_delay */ + 2491, + }, + /* ICE_PTP_LNK_SPD_25G_RS */ + { + /* tx_par_clk */ + 0, /* unused */ + /* rx_par_clk */ + 0, /* unused */ + /* tx_pcs_clk */ + 0, /* unused */ + /* rx_pcs_clk */ + 0, /* unused */ + /* tx_desk_rsgb_par */ + 161132812, /* 162.1328125 MHz Reed Solomon gearbox */ + /* rx_desk_rsgb_par */ + 161132812, /* 162.1328125 MHz Reed Solomon gearbox */ + /* tx_desk_rsgb_pcs */ + 97656250, /* 97.62625 MHz Reed Solomon gearbox */ + /* rx_desk_rsgb_pcs */ + 97656250, /* 97.62625 MHz Reed Solomon gearbox */ + /* tx_fixed_delay */ + 3928, + /* pmd_adj_divisor */ + 206250000, + /* rx_fixed_delay */ + 29535, + }, + /* ICE_PTP_LNK_SPD_40G */ + { + /* tx_par_clk */ + 257812500, + /* rx_par_clk */ + 257812500, + /* tx_pcs_clk */ + 156250000, /* 156.25 MHz */ + /* rx_pcs_clk */ + 156250000, /* 156.25 MHz */ + /* tx_desk_rsgb_par */ + 0, /* unused */ + /* rx_desk_rsgb_par */ + 156250000, /* 156.25 MHz deskew clock */ + /* tx_desk_rsgb_pcs */ + 0, /* unused */ + /* rx_desk_rsgb_pcs */ + 156250000, /* 156.25 MHz deskew clock */ + /* tx_fixed_delay */ + 5666, + /* pmd_adj_divisor */ + 82500000, + /* rx_fixed_delay */ + 4244, + }, + /* ICE_PTP_LNK_SPD_50G */ + { + /* tx_par_clk */ + 644531250, /* 644.53125 MHZ */ + /* rx_par_clk */ + 644531250, /* 644.53125 MHZ */ + /* tx_pcs_clk */ + 390625000, /* 390.625 MHz */ + /* rx_pcs_clk */ + 390625000, /* 390.625 MHz */ + /* tx_desk_rsgb_par */ + 0, /* unused */ + /* rx_desk_rsgb_par */ + 195312500, /* 193.3125 MHz deskew clock */ + /* tx_desk_rsgb_pcs */ + 0, /* unused */ + /* rx_desk_rsgb_pcs */ + 195312500, /* 193.3125 MHz deskew clock */ + /* tx_fixed_delay */ + 2778, + /* pmd_adj_divisor */ + 206250000, + /* rx_fixed_delay */ + 2868, + }, + /* ICE_PTP_LNK_SPD_50G_RS */ + { + /* tx_par_clk */ + 0, /* unused */ + /* rx_par_clk */ + 644531250, /* 644.53125 MHz */ + /* tx_pcs_clk */ + 0, /* unused */ + /* rx_pcs_clk */ + 644531250, /* 644.53125 MHz */ + /* tx_desk_rsgb_par */ + 322265625, /* 322.265625 MHz Reed Solomon gearbox */ + /* rx_desk_rsgb_par */ + 322265625, /* 322.265625 MHz Reed Solomon gearbox */ + /* tx_desk_rsgb_pcs */ + 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + /* rx_desk_rsgb_pcs */ + 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + /* tx_fixed_delay */ + 2095, + /* pmd_adj_divisor */ + 206250000, + /* rx_fixed_delay */ + 14524, + }, + /* ICE_PTP_LNK_SPD_100G_RS */ + { + /* tx_par_clk */ + 0, /* unused */ + /* rx_par_clk */ + 644531250, /* 644.53125 MHz */ + /* tx_pcs_clk */ + 0, /* unused */ + /* rx_pcs_clk */ + 644531250, /* 644.53125 MHz */ + /* tx_desk_rsgb_par */ + 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + /* rx_desk_rsgb_par */ + 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + /* tx_desk_rsgb_pcs */ + 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + /* rx_desk_rsgb_pcs */ + 644531250, /* 644.53125 MHz Reed Solomon gearbox */ + /* tx_fixed_delay */ + 1620, + /* pmd_adj_divisor */ + 206250000, + /* rx_fixed_delay */ + 7775, + }, +}; + +#endif /* _ICE_PTP_CONSTS_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c index 29f947c0cd2e..ec8450f034e6 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c @@ -3,6 +3,8 @@ #include "ice_common.h" #include "ice_ptp_hw.h" +#include "ice_ptp_consts.h" +#include "ice_cgu_regs.h" /* Low level functions for interacting with and managing the device clock used * for the Precision Time Protocol. @@ -29,6 +31,15 @@ * * For E810 devices, the increment frequency is 812.5 MHz * + * For E822 devices the clock can be derived from different sources, and the + * increment has an effective frequency of one of the following: + * - 823.4375 MHz + * - 783.36 MHz + * - 796.875 MHz + * - 816 MHz + * - 830.078125 MHz + * - 783.36 MHz + * * The hardware captures timestamps in the PHY for incoming packets, and for * outgoing packets on request. To support this, the PHY maintains a timer * that matches the lower 64 bits of the global source timer. @@ -37,6 +48,24 @@ * shadow registers are used to prepare the desired initial values. A special * sync command is issued to trigger copying from the shadow registers into * the appropriate source and PHY registers simultaneously. + * + * The driver supports devices which have different PHYs with subtly different + * mechanisms to program and control the timers. We divide the devices into + * families named after the first major device, E810 and similar devices, and + * E822 and similar devices. + * + * - E822 based devices have additional support for fine grained Vernier + * calibration which requires significant setup + * - The layout of timestamp data in the PHY register blocks is different + * - The way timer synchronization commands are issued is different. + * + * To support this, very low level functions have an e810 or e822 suffix + * indicating what type of device they work on. Higher level abstractions for + * tasks that can be done on both devices do not have the suffix and will + * correctly look up the appropriate low level function when running. + * + * Functions which only make sense on a single device family may not have + * a suitable generic implementation */ /** @@ -51,6 +80,2447 @@ u8 ice_get_ptp_src_clock_index(struct ice_hw *hw) return hw->func_caps.ts_func_info.tmr_index_assoc; } +/** + * ice_ptp_read_src_incval - Read source timer increment value + * @hw: pointer to HW struct + * + * Read the increment value of the source timer and return it. + */ +static u64 ice_ptp_read_src_incval(struct ice_hw *hw) +{ + u32 lo, hi; + u8 tmr_idx; + + tmr_idx = ice_get_ptp_src_clock_index(hw); + + lo = rd32(hw, GLTSYN_INCVAL_L(tmr_idx)); + hi = rd32(hw, GLTSYN_INCVAL_H(tmr_idx)); + + return ((u64)(hi & INCVAL_HIGH_M) << 32) | lo; +} + +/** + * ice_ptp_src_cmd - Prepare source timer for a timer command + * @hw: pointer to HW structure + * @cmd: Timer command + * + * Prepare the source timer for an upcoming timer sync command. + */ +static void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) +{ + u32 cmd_val; + u8 tmr_idx; + + tmr_idx = ice_get_ptp_src_clock_index(hw); + cmd_val = tmr_idx << SEL_CPK_SRC; + + switch (cmd) { + case INIT_TIME: + cmd_val |= GLTSYN_CMD_INIT_TIME; + break; + case INIT_INCVAL: + cmd_val |= GLTSYN_CMD_INIT_INCVAL; + break; + case ADJ_TIME: + cmd_val |= GLTSYN_CMD_ADJ_TIME; + break; + case ADJ_TIME_AT_TIME: + cmd_val |= GLTSYN_CMD_ADJ_INIT_TIME; + break; + case READ_TIME: + cmd_val |= GLTSYN_CMD_READ_TIME; + break; + } + + wr32(hw, GLTSYN_CMD, cmd_val); +} + +/** + * ice_ptp_exec_tmr_cmd - Execute all prepared timer commands + * @hw: pointer to HW struct + * + * Write the SYNC_EXEC_CMD bit to the GLTSYN_CMD_SYNC register, and flush the + * write immediately. This triggers the hardware to begin executing all of the + * source and PHY timer commands synchronously. + */ +static void ice_ptp_exec_tmr_cmd(struct ice_hw *hw) +{ + wr32(hw, GLTSYN_CMD_SYNC, SYNC_EXEC_CMD); + ice_flush(hw); +} + +/* E822 family functions + * + * The following functions operate on the E822 family of devices. + */ + +/** + * ice_fill_phy_msg_e822 - Fill message data for a PHY register access + * @msg: the PHY message buffer to fill in + * @port: the port to access + * @offset: the register offset + */ +static void +ice_fill_phy_msg_e822(struct ice_sbq_msg_input *msg, u8 port, u16 offset) +{ + int phy_port, phy, quadtype; + + phy_port = port % ICE_PORTS_PER_PHY; + phy = port / ICE_PORTS_PER_PHY; + quadtype = (port / ICE_PORTS_PER_QUAD) % ICE_NUM_QUAD_TYPE; + + if (quadtype == 0) { + msg->msg_addr_low = P_Q0_L(P_0_BASE + offset, phy_port); + msg->msg_addr_high = P_Q0_H(P_0_BASE + offset, phy_port); + } else { + msg->msg_addr_low = P_Q1_L(P_4_BASE + offset, phy_port); + msg->msg_addr_high = P_Q1_H(P_4_BASE + offset, phy_port); + } + + if (phy == 0) + msg->dest_dev = rmn_0; + else if (phy == 1) + msg->dest_dev = rmn_1; + else + msg->dest_dev = rmn_2; +} + +/** + * ice_is_64b_phy_reg_e822 - Check if this is a 64bit PHY register + * @low_addr: the low address to check + * @high_addr: on return, contains the high address of the 64bit register + * + * Checks if the provided low address is one of the known 64bit PHY values + * represented as two 32bit registers. If it is, return the appropriate high + * register offset to use. + */ +static bool ice_is_64b_phy_reg_e822(u16 low_addr, u16 *high_addr) +{ + switch (low_addr) { + case P_REG_PAR_PCS_TX_OFFSET_L: + *high_addr = P_REG_PAR_PCS_TX_OFFSET_U; + return true; + case P_REG_PAR_PCS_RX_OFFSET_L: + *high_addr = P_REG_PAR_PCS_RX_OFFSET_U; + return true; + case P_REG_PAR_TX_TIME_L: + *high_addr = P_REG_PAR_TX_TIME_U; + return true; + case P_REG_PAR_RX_TIME_L: + *high_addr = P_REG_PAR_RX_TIME_U; + return true; + case P_REG_TOTAL_TX_OFFSET_L: + *high_addr = P_REG_TOTAL_TX_OFFSET_U; + return true; + case P_REG_TOTAL_RX_OFFSET_L: + *high_addr = P_REG_TOTAL_RX_OFFSET_U; + return true; + case P_REG_UIX66_10G_40G_L: + *high_addr = P_REG_UIX66_10G_40G_U; + return true; + case P_REG_UIX66_25G_100G_L: + *high_addr = P_REG_UIX66_25G_100G_U; + return true; + case P_REG_TX_CAPTURE_L: + *high_addr = P_REG_TX_CAPTURE_U; + return true; + case P_REG_RX_CAPTURE_L: + *high_addr = P_REG_RX_CAPTURE_U; + return true; + case P_REG_TX_TIMER_INC_PRE_L: + *high_addr = P_REG_TX_TIMER_INC_PRE_U; + return true; + case P_REG_RX_TIMER_INC_PRE_L: + *high_addr = P_REG_RX_TIMER_INC_PRE_U; + return true; + default: + return false; + } +} + +/** + * ice_is_40b_phy_reg_e822 - Check if this is a 40bit PHY register + * @low_addr: the low address to check + * @high_addr: on return, contains the high address of the 40bit value + * + * Checks if the provided low address is one of the known 40bit PHY values + * split into two registers with the lower 8 bits in the low register and the + * upper 32 bits in the high register. If it is, return the appropriate high + * register offset to use. + */ +static bool ice_is_40b_phy_reg_e822(u16 low_addr, u16 *high_addr) +{ + switch (low_addr) { + case P_REG_TIMETUS_L: + *high_addr = P_REG_TIMETUS_U; + return true; + case P_REG_PAR_RX_TUS_L: + *high_addr = P_REG_PAR_RX_TUS_U; + return true; + case P_REG_PAR_TX_TUS_L: + *high_addr = P_REG_PAR_TX_TUS_U; + return true; + case P_REG_PCS_RX_TUS_L: + *high_addr = P_REG_PCS_RX_TUS_U; + return true; + case P_REG_PCS_TX_TUS_L: + *high_addr = P_REG_PCS_TX_TUS_U; + return true; + case P_REG_DESK_PAR_RX_TUS_L: + *high_addr = P_REG_DESK_PAR_RX_TUS_U; + return true; + case P_REG_DESK_PAR_TX_TUS_L: + *high_addr = P_REG_DESK_PAR_TX_TUS_U; + return true; + case P_REG_DESK_PCS_RX_TUS_L: + *high_addr = P_REG_DESK_PCS_RX_TUS_U; + return true; + case P_REG_DESK_PCS_TX_TUS_L: + *high_addr = P_REG_DESK_PCS_TX_TUS_U; + return true; + default: + return false; + } +} + +/** + * ice_read_phy_reg_e822 - Read a PHY register + * @hw: pointer to the HW struct + * @port: PHY port to read from + * @offset: PHY register offset to read + * @val: on return, the contents read from the PHY + * + * Read a PHY register for the given port over the device sideband queue. + */ +int +ice_read_phy_reg_e822(struct ice_hw *hw, u8 port, u16 offset, u32 *val) +{ + struct ice_sbq_msg_input msg = {0}; + int err; + + ice_fill_phy_msg_e822(&msg, port, offset); + msg.opcode = ice_sbq_msg_rd; + + err = ice_sbq_rw_reg(hw, &msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n", + err); + return err; + } + + *val = msg.data; + + return 0; +} + +/** + * ice_read_64b_phy_reg_e822 - Read a 64bit value from PHY registers + * @hw: pointer to the HW struct + * @port: PHY port to read from + * @low_addr: offset of the lower register to read from + * @val: on return, the contents of the 64bit value from the PHY registers + * + * Reads the two registers associated with a 64bit value and returns it in the + * val pointer. The offset always specifies the lower register offset to use. + * The high offset is looked up. This function only operates on registers + * known to be two parts of a 64bit value. + */ +static int +ice_read_64b_phy_reg_e822(struct ice_hw *hw, u8 port, u16 low_addr, u64 *val) +{ + u32 low, high; + u16 high_addr; + int err; + + /* Only operate on registers known to be split into two 32bit + * registers. + */ + if (!ice_is_64b_phy_reg_e822(low_addr, &high_addr)) { + ice_debug(hw, ICE_DBG_PTP, "Invalid 64b register addr 0x%08x\n", + low_addr); + return -EINVAL; + } + + err = ice_read_phy_reg_e822(hw, port, low_addr, &low); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read from low register 0x%08x\n, err %d", + low_addr, err); + return err; + } + + err = ice_read_phy_reg_e822(hw, port, high_addr, &high); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read from high register 0x%08x\n, err %d", + high_addr, err); + return err; + } + + *val = (u64)high << 32 | low; + + return 0; +} + +/** + * ice_write_phy_reg_e822 - Write a PHY register + * @hw: pointer to the HW struct + * @port: PHY port to write to + * @offset: PHY register offset to write + * @val: The value to write to the register + * + * Write a PHY register for the given port over the device sideband queue. + */ +int +ice_write_phy_reg_e822(struct ice_hw *hw, u8 port, u16 offset, u32 val) +{ + struct ice_sbq_msg_input msg = {0}; + int err; + + ice_fill_phy_msg_e822(&msg, port, offset); + msg.opcode = ice_sbq_msg_wr; + msg.data = val; + + err = ice_sbq_rw_reg(hw, &msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n", + err); + return err; + } + + return 0; +} + +/** + * ice_write_40b_phy_reg_e822 - Write a 40b value to the PHY + * @hw: pointer to the HW struct + * @port: port to write to + * @low_addr: offset of the low register + * @val: 40b value to write + * + * Write the provided 40b value to the two associated registers by splitting + * it up into two chunks, the lower 8 bits and the upper 32 bits. + */ +static int +ice_write_40b_phy_reg_e822(struct ice_hw *hw, u8 port, u16 low_addr, u64 val) +{ + u32 low, high; + u16 high_addr; + int err; + + /* Only operate on registers known to be split into a lower 8 bit + * register and an upper 32 bit register. + */ + if (!ice_is_40b_phy_reg_e822(low_addr, &high_addr)) { + ice_debug(hw, ICE_DBG_PTP, "Invalid 40b register addr 0x%08x\n", + low_addr); + return -EINVAL; + } + + low = (u32)(val & P_REG_40B_LOW_M); + high = (u32)(val >> P_REG_40B_HIGH_S); + + err = ice_write_phy_reg_e822(hw, port, low_addr, low); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d", + low_addr, err); + return err; + } + + err = ice_write_phy_reg_e822(hw, port, high_addr, high); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d", + high_addr, err); + return err; + } + + return 0; +} + +/** + * ice_write_64b_phy_reg_e822 - Write a 64bit value to PHY registers + * @hw: pointer to the HW struct + * @port: PHY port to read from + * @low_addr: offset of the lower register to read from + * @val: the contents of the 64bit value to write to PHY + * + * Write the 64bit value to the two associated 32bit PHY registers. The offset + * is always specified as the lower register, and the high address is looked + * up. This function only operates on registers known to be two parts of + * a 64bit value. + */ +static int +ice_write_64b_phy_reg_e822(struct ice_hw *hw, u8 port, u16 low_addr, u64 val) +{ + u32 low, high; + u16 high_addr; + int err; + + /* Only operate on registers known to be split into two 32bit + * registers. + */ + if (!ice_is_64b_phy_reg_e822(low_addr, &high_addr)) { + ice_debug(hw, ICE_DBG_PTP, "Invalid 64b register addr 0x%08x\n", + low_addr); + return -EINVAL; + } + + low = lower_32_bits(val); + high = upper_32_bits(val); + + err = ice_write_phy_reg_e822(hw, port, low_addr, low); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d", + low_addr, err); + return err; + } + + err = ice_write_phy_reg_e822(hw, port, high_addr, high); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d", + high_addr, err); + return err; + } + + return 0; +} + +/** + * ice_fill_quad_msg_e822 - Fill message data for quad register access + * @msg: the PHY message buffer to fill in + * @quad: the quad to access + * @offset: the register offset + * + * Fill a message buffer for accessing a register in a quad shared between + * multiple PHYs. + */ +static void +ice_fill_quad_msg_e822(struct ice_sbq_msg_input *msg, u8 quad, u16 offset) +{ + u32 addr; + + msg->dest_dev = rmn_0; + + if ((quad % ICE_NUM_QUAD_TYPE) == 0) + addr = Q_0_BASE + offset; + else + addr = Q_1_BASE + offset; + + msg->msg_addr_low = lower_16_bits(addr); + msg->msg_addr_high = upper_16_bits(addr); +} + +/** + * ice_read_quad_reg_e822 - Read a PHY quad register + * @hw: pointer to the HW struct + * @quad: quad to read from + * @offset: quad register offset to read + * @val: on return, the contents read from the quad + * + * Read a quad register over the device sideband queue. Quad registers are + * shared between multiple PHYs. + */ +int +ice_read_quad_reg_e822(struct ice_hw *hw, u8 quad, u16 offset, u32 *val) +{ + struct ice_sbq_msg_input msg = {0}; + int err; + + if (quad >= ICE_MAX_QUAD) + return -EINVAL; + + ice_fill_quad_msg_e822(&msg, quad, offset); + msg.opcode = ice_sbq_msg_rd; + + err = ice_sbq_rw_reg(hw, &msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n", + err); + return err; + } + + *val = msg.data; + + return 0; +} + +/** + * ice_write_quad_reg_e822 - Write a PHY quad register + * @hw: pointer to the HW struct + * @quad: quad to write to + * @offset: quad register offset to write + * @val: The value to write to the register + * + * Write a quad register over the device sideband queue. Quad registers are + * shared between multiple PHYs. + */ +int +ice_write_quad_reg_e822(struct ice_hw *hw, u8 quad, u16 offset, u32 val) +{ + struct ice_sbq_msg_input msg = {0}; + int err; + + if (quad >= ICE_MAX_QUAD) + return -EINVAL; + + ice_fill_quad_msg_e822(&msg, quad, offset); + msg.opcode = ice_sbq_msg_wr; + msg.data = val; + + err = ice_sbq_rw_reg(hw, &msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n", + err); + return err; + } + + return 0; +} + +/** + * ice_read_phy_tstamp_e822 - Read a PHY timestamp out of the quad block + * @hw: pointer to the HW struct + * @quad: the quad to read from + * @idx: the timestamp index to read + * @tstamp: on return, the 40bit timestamp value + * + * Read a 40bit timestamp value out of the two associated registers in the + * quad memory block that is shared between the internal PHYs of the E822 + * family of devices. + */ +static int +ice_read_phy_tstamp_e822(struct ice_hw *hw, u8 quad, u8 idx, u64 *tstamp) +{ + u16 lo_addr, hi_addr; + u32 lo, hi; + int err; + + lo_addr = (u16)TS_L(Q_REG_TX_MEMORY_BANK_START, idx); + hi_addr = (u16)TS_H(Q_REG_TX_MEMORY_BANK_START, idx); + + err = ice_read_quad_reg_e822(hw, quad, lo_addr, &lo); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read low PTP timestamp register, err %d\n", + err); + return err; + } + + err = ice_read_quad_reg_e822(hw, quad, hi_addr, &hi); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read high PTP timestamp register, err %d\n", + err); + return err; + } + + /* For E822 based internal PHYs, the timestamp is reported with the + * lower 8 bits in the low register, and the upper 32 bits in the high + * register. + */ + *tstamp = ((u64)hi) << TS_PHY_HIGH_S | ((u64)lo & TS_PHY_LOW_M); + + return 0; +} + +/** + * ice_clear_phy_tstamp_e822 - Clear a timestamp from the quad block + * @hw: pointer to the HW struct + * @quad: the quad to read from + * @idx: the timestamp index to reset + * + * Clear a timestamp, resetting its valid bit, from the PHY quad block that is + * shared between the internal PHYs on the E822 devices. + */ +static int +ice_clear_phy_tstamp_e822(struct ice_hw *hw, u8 quad, u8 idx) +{ + u16 lo_addr, hi_addr; + int err; + + lo_addr = (u16)TS_L(Q_REG_TX_MEMORY_BANK_START, idx); + hi_addr = (u16)TS_H(Q_REG_TX_MEMORY_BANK_START, idx); + + err = ice_write_quad_reg_e822(hw, quad, lo_addr, 0); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to clear low PTP timestamp register, err %d\n", + err); + return err; + } + + err = ice_write_quad_reg_e822(hw, quad, hi_addr, 0); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to clear high PTP timestamp register, err %d\n", + err); + return err; + } + + return 0; +} + +/** + * ice_read_cgu_reg_e822 - Read a CGU register + * @hw: pointer to the HW struct + * @addr: Register address to read + * @val: storage for register value read + * + * Read the contents of a register of the Clock Generation Unit. Only + * applicable to E822 devices. + */ +static int +ice_read_cgu_reg_e822(struct ice_hw *hw, u32 addr, u32 *val) +{ + struct ice_sbq_msg_input cgu_msg; + int err; + + cgu_msg.opcode = ice_sbq_msg_rd; + cgu_msg.dest_dev = cgu; + cgu_msg.msg_addr_low = addr; + cgu_msg.msg_addr_high = 0x0; + + err = ice_sbq_rw_reg(hw, &cgu_msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read CGU register 0x%04x, err %d\n", + addr, err); + return err; + } + + *val = cgu_msg.data; + + return err; +} + +/** + * ice_write_cgu_reg_e822 - Write a CGU register + * @hw: pointer to the HW struct + * @addr: Register address to write + * @val: value to write into the register + * + * Write the specified value to a register of the Clock Generation Unit. Only + * applicable to E822 devices. + */ +static int +ice_write_cgu_reg_e822(struct ice_hw *hw, u32 addr, u32 val) +{ + struct ice_sbq_msg_input cgu_msg; + int err; + + cgu_msg.opcode = ice_sbq_msg_wr; + cgu_msg.dest_dev = cgu; + cgu_msg.msg_addr_low = addr; + cgu_msg.msg_addr_high = 0x0; + cgu_msg.data = val; + + err = ice_sbq_rw_reg(hw, &cgu_msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write CGU register 0x%04x, err %d\n", + addr, err); + return err; + } + + return err; +} + +/** + * ice_clk_freq_str - Convert time_ref_freq to string + * @clk_freq: Clock frequency + * + * Convert the specified TIME_REF clock frequency to a string. + */ +static const char *ice_clk_freq_str(u8 clk_freq) +{ + switch ((enum ice_time_ref_freq)clk_freq) { + case ICE_TIME_REF_FREQ_25_000: + return "25 MHz"; + case ICE_TIME_REF_FREQ_122_880: + return "122.88 MHz"; + case ICE_TIME_REF_FREQ_125_000: + return "125 MHz"; + case ICE_TIME_REF_FREQ_153_600: + return "153.6 MHz"; + case ICE_TIME_REF_FREQ_156_250: + return "156.25 MHz"; + case ICE_TIME_REF_FREQ_245_760: + return "245.76 MHz"; + default: + return "Unknown"; + } +} + +/** + * ice_clk_src_str - Convert time_ref_src to string + * @clk_src: Clock source + * + * Convert the specified clock source to its string name. + */ +static const char *ice_clk_src_str(u8 clk_src) +{ + switch ((enum ice_clk_src)clk_src) { + case ICE_CLK_SRC_TCX0: + return "TCX0"; + case ICE_CLK_SRC_TIME_REF: + return "TIME_REF"; + default: + return "Unknown"; + } +} + +/** + * ice_cfg_cgu_pll_e822 - Configure the Clock Generation Unit + * @hw: pointer to the HW struct + * @clk_freq: Clock frequency to program + * @clk_src: Clock source to select (TIME_REF, or TCX0) + * + * Configure the Clock Generation Unit with the desired clock frequency and + * time reference, enabling the PLL which drives the PTP hardware clock. + */ +static int +ice_cfg_cgu_pll_e822(struct ice_hw *hw, enum ice_time_ref_freq clk_freq, + enum ice_clk_src clk_src) +{ + union tspll_ro_bwm_lf bwm_lf; + union nac_cgu_dword19 dw19; + union nac_cgu_dword22 dw22; + union nac_cgu_dword24 dw24; + union nac_cgu_dword9 dw9; + int err; + + if (clk_freq >= NUM_ICE_TIME_REF_FREQ) { + dev_warn(ice_hw_to_dev(hw), "Invalid TIME_REF frequency %u\n", + clk_freq); + return -EINVAL; + } + + if (clk_src >= NUM_ICE_CLK_SRC) { + dev_warn(ice_hw_to_dev(hw), "Invalid clock source %u\n", + clk_src); + return -EINVAL; + } + + if (clk_src == ICE_CLK_SRC_TCX0 && + clk_freq != ICE_TIME_REF_FREQ_25_000) { + dev_warn(ice_hw_to_dev(hw), + "TCX0 only supports 25 MHz frequency\n"); + return -EINVAL; + } + + err = ice_read_cgu_reg_e822(hw, NAC_CGU_DWORD9, &dw9.val); + if (err) + return err; + + err = ice_read_cgu_reg_e822(hw, NAC_CGU_DWORD24, &dw24.val); + if (err) + return err; + + err = ice_read_cgu_reg_e822(hw, TSPLL_RO_BWM_LF, &bwm_lf.val); + if (err) + return err; + + /* Log the current clock configuration */ + ice_debug(hw, ICE_DBG_PTP, "Current CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n", + dw24.field.ts_pll_enable ? "enabled" : "disabled", + ice_clk_src_str(dw24.field.time_ref_sel), + ice_clk_freq_str(dw9.field.time_ref_freq_sel), + bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked"); + + /* Disable the PLL before changing the clock source or frequency */ + if (dw24.field.ts_pll_enable) { + dw24.field.ts_pll_enable = 0; + + err = ice_write_cgu_reg_e822(hw, NAC_CGU_DWORD24, dw24.val); + if (err) + return err; + } + + /* Set the frequency */ + dw9.field.time_ref_freq_sel = clk_freq; + err = ice_write_cgu_reg_e822(hw, NAC_CGU_DWORD9, dw9.val); + if (err) + return err; + + /* Configure the TS PLL feedback divisor */ + err = ice_read_cgu_reg_e822(hw, NAC_CGU_DWORD19, &dw19.val); + if (err) + return err; + + dw19.field.tspll_fbdiv_intgr = e822_cgu_params[clk_freq].feedback_div; + dw19.field.tspll_ndivratio = 1; + + err = ice_write_cgu_reg_e822(hw, NAC_CGU_DWORD19, dw19.val); + if (err) + return err; + + /* Configure the TS PLL post divisor */ + err = ice_read_cgu_reg_e822(hw, NAC_CGU_DWORD22, &dw22.val); + if (err) + return err; + + dw22.field.time1588clk_div = e822_cgu_params[clk_freq].post_pll_div; + dw22.field.time1588clk_sel_div2 = 0; + + err = ice_write_cgu_reg_e822(hw, NAC_CGU_DWORD22, dw22.val); + if (err) + return err; + + /* Configure the TS PLL pre divisor and clock source */ + err = ice_read_cgu_reg_e822(hw, NAC_CGU_DWORD24, &dw24.val); + if (err) + return err; + + dw24.field.ref1588_ck_div = e822_cgu_params[clk_freq].refclk_pre_div; + dw24.field.tspll_fbdiv_frac = e822_cgu_params[clk_freq].frac_n_div; + dw24.field.time_ref_sel = clk_src; + + err = ice_write_cgu_reg_e822(hw, NAC_CGU_DWORD24, dw24.val); + if (err) + return err; + + /* Finally, enable the PLL */ + dw24.field.ts_pll_enable = 1; + + err = ice_write_cgu_reg_e822(hw, NAC_CGU_DWORD24, dw24.val); + if (err) + return err; + + /* Wait to verify if the PLL locks */ + usleep_range(1000, 5000); + + err = ice_read_cgu_reg_e822(hw, TSPLL_RO_BWM_LF, &bwm_lf.val); + if (err) + return err; + + if (!bwm_lf.field.plllock_true_lock_cri) { + dev_warn(ice_hw_to_dev(hw), "CGU PLL failed to lock\n"); + return -EBUSY; + } + + /* Log the current clock configuration */ + ice_debug(hw, ICE_DBG_PTP, "New CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n", + dw24.field.ts_pll_enable ? "enabled" : "disabled", + ice_clk_src_str(dw24.field.time_ref_sel), + ice_clk_freq_str(dw9.field.time_ref_freq_sel), + bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked"); + + return 0; +} + +/** + * ice_init_cgu_e822 - Initialize CGU with settings from firmware + * @hw: pointer to the HW structure + * + * Initialize the Clock Generation Unit of the E822 device. + */ +static int ice_init_cgu_e822(struct ice_hw *hw) +{ + struct ice_ts_func_info *ts_info = &hw->func_caps.ts_func_info; + union tspll_cntr_bist_settings cntr_bist; + int err; + + err = ice_read_cgu_reg_e822(hw, TSPLL_CNTR_BIST_SETTINGS, + &cntr_bist.val); + if (err) + return err; + + /* Disable sticky lock detection so lock err reported is accurate */ + cntr_bist.field.i_plllock_sel_0 = 0; + cntr_bist.field.i_plllock_sel_1 = 0; + + err = ice_write_cgu_reg_e822(hw, TSPLL_CNTR_BIST_SETTINGS, + cntr_bist.val); + if (err) + return err; + + /* Configure the CGU PLL using the parameters from the function + * capabilities. + */ + err = ice_cfg_cgu_pll_e822(hw, ts_info->time_ref, + (enum ice_clk_src)ts_info->clk_src); + if (err) + return err; + + return 0; +} + +/** + * ice_ptp_set_vernier_wl - Set the window length for vernier calibration + * @hw: pointer to the HW struct + * + * Set the window length used for the vernier port calibration process. + */ +static int ice_ptp_set_vernier_wl(struct ice_hw *hw) +{ + u8 port; + + for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) { + int err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_WL, + PTP_VERNIER_WL); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to set vernier window length for port %u, err %d\n", + port, err); + return err; + } + } + + return 0; +} + +/** + * ice_ptp_init_phc_e822 - Perform E822 specific PHC initialization + * @hw: pointer to HW struct + * + * Perform PHC initialization steps specific to E822 devices. + */ +static int ice_ptp_init_phc_e822(struct ice_hw *hw) +{ + int err; + u32 regval; + + /* Enable reading switch and PHY registers over the sideband queue */ +#define PF_SB_REM_DEV_CTL_SWITCH_READ BIT(1) +#define PF_SB_REM_DEV_CTL_PHY0 BIT(2) + regval = rd32(hw, PF_SB_REM_DEV_CTL); + regval |= (PF_SB_REM_DEV_CTL_SWITCH_READ | + PF_SB_REM_DEV_CTL_PHY0); + wr32(hw, PF_SB_REM_DEV_CTL, regval); + + /* Initialize the Clock Generation Unit */ + err = ice_init_cgu_e822(hw); + if (err) + return err; + + /* Set window length for all the ports */ + return ice_ptp_set_vernier_wl(hw); +} + +/** + * ice_ptp_prep_phy_time_e822 - Prepare PHY port with initial time + * @hw: pointer to the HW struct + * @time: Time to initialize the PHY port clocks to + * + * Program the PHY port registers with a new initial time value. The port + * clock will be initialized once the driver issues an INIT_TIME sync + * command. The time value is the upper 32 bits of the PHY timer, usually in + * units of nominal nanoseconds. + */ +static int +ice_ptp_prep_phy_time_e822(struct ice_hw *hw, u32 time) +{ + u64 phy_time; + u8 port; + int err; + + /* The time represents the upper 32 bits of the PHY timer, so we need + * to shift to account for this when programming. + */ + phy_time = (u64)time << 32; + + for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) { + /* Tx case */ + err = ice_write_64b_phy_reg_e822(hw, port, + P_REG_TX_TIMER_INC_PRE_L, + phy_time); + if (err) + goto exit_err; + + /* Rx case */ + err = ice_write_64b_phy_reg_e822(hw, port, + P_REG_RX_TIMER_INC_PRE_L, + phy_time); + if (err) + goto exit_err; + } + + return 0; + +exit_err: + ice_debug(hw, ICE_DBG_PTP, "Failed to write init time for port %u, err %d\n", + port, err); + + return err; +} + +/** + * ice_ptp_prep_port_adj_e822 - Prepare a single port for time adjust + * @hw: pointer to HW struct + * @port: Port number to be programmed + * @time: time in cycles to adjust the port Tx and Rx clocks + * + * Program the port for an atomic adjustment by writing the Tx and Rx timer + * registers. The atomic adjustment won't be completed until the driver issues + * an ADJ_TIME command. + * + * Note that time is not in units of nanoseconds. It is in clock time + * including the lower sub-nanosecond portion of the port timer. + * + * Negative adjustments are supported using 2s complement arithmetic. + */ +int +ice_ptp_prep_port_adj_e822(struct ice_hw *hw, u8 port, s64 time) +{ + u32 l_time, u_time; + int err; + + l_time = lower_32_bits(time); + u_time = upper_32_bits(time); + + /* Tx case */ + err = ice_write_phy_reg_e822(hw, port, P_REG_TX_TIMER_INC_PRE_L, + l_time); + if (err) + goto exit_err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_TX_TIMER_INC_PRE_U, + u_time); + if (err) + goto exit_err; + + /* Rx case */ + err = ice_write_phy_reg_e822(hw, port, P_REG_RX_TIMER_INC_PRE_L, + l_time); + if (err) + goto exit_err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_RX_TIMER_INC_PRE_U, + u_time); + if (err) + goto exit_err; + + return 0; + +exit_err: + ice_debug(hw, ICE_DBG_PTP, "Failed to write time adjust for port %u, err %d\n", + port, err); + return err; +} + +/** + * ice_ptp_prep_phy_adj_e822 - Prep PHY ports for a time adjustment + * @hw: pointer to HW struct + * @adj: adjustment in nanoseconds + * + * Prepare the PHY ports for an atomic time adjustment by programming the PHY + * Tx and Rx port registers. The actual adjustment is completed by issuing an + * ADJ_TIME or ADJ_TIME_AT_TIME sync command. + */ +static int +ice_ptp_prep_phy_adj_e822(struct ice_hw *hw, s32 adj) +{ + s64 cycles; + u8 port; + + /* The port clock supports adjustment of the sub-nanosecond portion of + * the clock. We shift the provided adjustment in nanoseconds to + * calculate the appropriate adjustment to program into the PHY ports. + */ + if (adj > 0) + cycles = (s64)adj << 32; + else + cycles = -(((s64)-adj) << 32); + + for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) { + int err; + + err = ice_ptp_prep_port_adj_e822(hw, port, cycles); + if (err) + return err; + } + + return 0; +} + +/** + * ice_ptp_prep_phy_incval_e822 - Prepare PHY ports for time adjustment + * @hw: pointer to HW struct + * @incval: new increment value to prepare + * + * Prepare each of the PHY ports for a new increment value by programming the + * port's TIMETUS registers. The new increment value will be updated after + * issuing an INIT_INCVAL command. + */ +static int +ice_ptp_prep_phy_incval_e822(struct ice_hw *hw, u64 incval) +{ + int err; + u8 port; + + for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) { + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_TIMETUS_L, + incval); + if (err) + goto exit_err; + } + + return 0; + +exit_err: + ice_debug(hw, ICE_DBG_PTP, "Failed to write incval for port %u, err %d\n", + port, err); + + return err; +} + +/** + * ice_ptp_read_port_capture - Read a port's local time capture + * @hw: pointer to HW struct + * @port: Port number to read + * @tx_ts: on return, the Tx port time capture + * @rx_ts: on return, the Rx port time capture + * + * Read the port's Tx and Rx local time capture values. + * + * Note this has no equivalent for the E810 devices. + */ +static int +ice_ptp_read_port_capture(struct ice_hw *hw, u8 port, u64 *tx_ts, u64 *rx_ts) +{ + int err; + + /* Tx case */ + err = ice_read_64b_phy_reg_e822(hw, port, P_REG_TX_CAPTURE_L, tx_ts); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read REG_TX_CAPTURE, err %d\n", + err); + return err; + } + + ice_debug(hw, ICE_DBG_PTP, "tx_init = 0x%016llx\n", + (unsigned long long)*tx_ts); + + /* Rx case */ + err = ice_read_64b_phy_reg_e822(hw, port, P_REG_RX_CAPTURE_L, rx_ts); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read RX_CAPTURE, err %d\n", + err); + return err; + } + + ice_debug(hw, ICE_DBG_PTP, "rx_init = 0x%016llx\n", + (unsigned long long)*rx_ts); + + return 0; +} + +/** + * ice_ptp_one_port_cmd - Prepare a single PHY port for a timer command + * @hw: pointer to HW struct + * @port: Port to which cmd has to be sent + * @cmd: Command to be sent to the port + * + * Prepare the requested port for an upcoming timer sync command. + * + * Note there is no equivalent of this operation on E810, as that device + * always handles all external PHYs internally. + */ +static int +ice_ptp_one_port_cmd(struct ice_hw *hw, u8 port, enum ice_ptp_tmr_cmd cmd) +{ + u32 cmd_val, val; + u8 tmr_idx; + int err; + + tmr_idx = ice_get_ptp_src_clock_index(hw); + cmd_val = tmr_idx << SEL_PHY_SRC; + switch (cmd) { + case INIT_TIME: + cmd_val |= PHY_CMD_INIT_TIME; + break; + case INIT_INCVAL: + cmd_val |= PHY_CMD_INIT_INCVAL; + break; + case ADJ_TIME: + cmd_val |= PHY_CMD_ADJ_TIME; + break; + case READ_TIME: + cmd_val |= PHY_CMD_READ_TIME; + break; + case ADJ_TIME_AT_TIME: + cmd_val |= PHY_CMD_ADJ_TIME_AT_TIME; + break; + } + + /* Tx case */ + /* Read, modify, write */ + err = ice_read_phy_reg_e822(hw, port, P_REG_TX_TMR_CMD, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read TX_TMR_CMD, err %d\n", + err); + return err; + } + + /* Modify necessary bits only and perform write */ + val &= ~TS_CMD_MASK; + val |= cmd_val; + + err = ice_write_phy_reg_e822(hw, port, P_REG_TX_TMR_CMD, val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write back TX_TMR_CMD, err %d\n", + err); + return err; + } + + /* Rx case */ + /* Read, modify, write */ + err = ice_read_phy_reg_e822(hw, port, P_REG_RX_TMR_CMD, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read RX_TMR_CMD, err %d\n", + err); + return err; + } + + /* Modify necessary bits only and perform write */ + val &= ~TS_CMD_MASK; + val |= cmd_val; + + err = ice_write_phy_reg_e822(hw, port, P_REG_RX_TMR_CMD, val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write back RX_TMR_CMD, err %d\n", + err); + return err; + } + + return 0; +} + +/** + * ice_ptp_port_cmd_e822 - Prepare all ports for a timer command + * @hw: pointer to the HW struct + * @cmd: timer command to prepare + * + * Prepare all ports connected to this device for an upcoming timer sync + * command. + */ +static int +ice_ptp_port_cmd_e822(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) +{ + u8 port; + + for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) { + int err; + + err = ice_ptp_one_port_cmd(hw, port, cmd); + if (err) + return err; + } + + return 0; +} + +/* E822 Vernier calibration functions + * + * The following functions are used as part of the vernier calibration of + * a port. This calibration increases the precision of the timestamps on the + * port. + */ + +/** + * ice_phy_get_speed_and_fec_e822 - Get link speed and FEC based on serdes mode + * @hw: pointer to HW struct + * @port: the port to read from + * @link_out: if non-NULL, holds link speed on success + * @fec_out: if non-NULL, holds FEC algorithm on success + * + * Read the serdes data for the PHY port and extract the link speed and FEC + * algorithm. + */ +static int +ice_phy_get_speed_and_fec_e822(struct ice_hw *hw, u8 port, + enum ice_ptp_link_spd *link_out, + enum ice_ptp_fec_mode *fec_out) +{ + enum ice_ptp_link_spd link; + enum ice_ptp_fec_mode fec; + u32 serdes; + int err; + + err = ice_read_phy_reg_e822(hw, port, P_REG_LINK_SPEED, &serdes); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read serdes info\n"); + return err; + } + + /* Determine the FEC algorithm */ + fec = (enum ice_ptp_fec_mode)P_REG_LINK_SPEED_FEC_MODE(serdes); + + serdes &= P_REG_LINK_SPEED_SERDES_M; + + /* Determine the link speed */ + if (fec == ICE_PTP_FEC_MODE_RS_FEC) { + switch (serdes) { + case ICE_PTP_SERDES_25G: + link = ICE_PTP_LNK_SPD_25G_RS; + break; + case ICE_PTP_SERDES_50G: + link = ICE_PTP_LNK_SPD_50G_RS; + break; + case ICE_PTP_SERDES_100G: + link = ICE_PTP_LNK_SPD_100G_RS; + break; + default: + return -EIO; + } + } else { + switch (serdes) { + case ICE_PTP_SERDES_1G: + link = ICE_PTP_LNK_SPD_1G; + break; + case ICE_PTP_SERDES_10G: + link = ICE_PTP_LNK_SPD_10G; + break; + case ICE_PTP_SERDES_25G: + link = ICE_PTP_LNK_SPD_25G; + break; + case ICE_PTP_SERDES_40G: + link = ICE_PTP_LNK_SPD_40G; + break; + case ICE_PTP_SERDES_50G: + link = ICE_PTP_LNK_SPD_50G; + break; + default: + return -EIO; + } + } + + if (link_out) + *link_out = link; + if (fec_out) + *fec_out = fec; + + return 0; +} + +/** + * ice_phy_cfg_lane_e822 - Configure PHY quad for single/multi-lane timestamp + * @hw: pointer to HW struct + * @port: to configure the quad for + */ +static void ice_phy_cfg_lane_e822(struct ice_hw *hw, u8 port) +{ + enum ice_ptp_link_spd link_spd; + int err; + u32 val; + u8 quad; + + err = ice_phy_get_speed_and_fec_e822(hw, port, &link_spd, NULL); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to get PHY link speed, err %d\n", + err); + return; + } + + quad = port / ICE_PORTS_PER_QUAD; + + err = ice_read_quad_reg_e822(hw, quad, Q_REG_TX_MEM_GBL_CFG, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read TX_MEM_GLB_CFG, err %d\n", + err); + return; + } + + if (link_spd >= ICE_PTP_LNK_SPD_40G) + val &= ~Q_REG_TX_MEM_GBL_CFG_LANE_TYPE_M; + else + val |= Q_REG_TX_MEM_GBL_CFG_LANE_TYPE_M; + + err = ice_write_quad_reg_e822(hw, quad, Q_REG_TX_MEM_GBL_CFG, val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write back TX_MEM_GBL_CFG, err %d\n", + err); + return; + } +} + +/** + * ice_phy_cfg_uix_e822 - Configure Serdes UI to TU conversion for E822 + * @hw: pointer to the HW structure + * @port: the port to configure + * + * Program the conversion ration of Serdes clock "unit intervals" (UIs) to PHC + * hardware clock time units (TUs). That is, determine the number of TUs per + * serdes unit interval, and program the UIX registers with this conversion. + * + * This conversion is used as part of the calibration process when determining + * the additional error of a timestamp vs the real time of transmission or + * receipt of the packet. + * + * Hardware uses the number of TUs per 66 UIs, written to the UIX registers + * for the two main serdes clock rates, 10G/40G and 25G/100G serdes clocks. + * + * To calculate the conversion ratio, we use the following facts: + * + * a) the clock frequency in Hz (cycles per second) + * b) the number of TUs per cycle (the increment value of the clock) + * c) 1 second per 1 billion nanoseconds + * d) the duration of 66 UIs in nanoseconds + * + * Given these facts, we can use the following table to work out what ratios + * to multiply in order to get the number of TUs per 66 UIs: + * + * cycles | 1 second | incval (TUs) | nanoseconds + * -------+--------------+--------------+------------- + * second | 1 billion ns | cycle | 66 UIs + * + * To perform the multiplication using integers without too much loss of + * precision, we can take use the following equation: + * + * (freq * incval * 6600 LINE_UI ) / ( 100 * 1 billion) + * + * We scale up to using 6600 UI instead of 66 in order to avoid fractional + * nanosecond UIs (66 UI at 10G/40G is 6.4 ns) + * + * The increment value has a maximum expected range of about 34 bits, while + * the frequency value is about 29 bits. Multiplying these values shouldn't + * overflow the 64 bits. However, we must then further multiply them again by + * the Serdes unit interval duration. To avoid overflow here, we split the + * overall divide by 1e11 into a divide by 256 (shift down by 8 bits) and + * a divide by 390,625,000. This does lose some precision, but avoids + * miscalculation due to arithmetic overflow. + */ +static int ice_phy_cfg_uix_e822(struct ice_hw *hw, u8 port) +{ + u64 cur_freq, clk_incval, tu_per_sec, uix; + int err; + + cur_freq = ice_e822_pll_freq(ice_e822_time_ref(hw)); + clk_incval = ice_ptp_read_src_incval(hw); + + /* Calculate TUs per second divided by 256 */ + tu_per_sec = (cur_freq * clk_incval) >> 8; + +#define LINE_UI_10G_40G 640 /* 6600 UIs is 640 nanoseconds at 10Gb/40Gb */ +#define LINE_UI_25G_100G 256 /* 6600 UIs is 256 nanoseconds at 25Gb/100Gb */ + + /* Program the 10Gb/40Gb conversion ratio */ + uix = div_u64(tu_per_sec * LINE_UI_10G_40G, 390625000); + + err = ice_write_64b_phy_reg_e822(hw, port, P_REG_UIX66_10G_40G_L, + uix); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write UIX66_10G_40G, err %d\n", + err); + return err; + } + + /* Program the 25Gb/100Gb conversion ratio */ + uix = div_u64(tu_per_sec * LINE_UI_25G_100G, 390625000); + + err = ice_write_64b_phy_reg_e822(hw, port, P_REG_UIX66_25G_100G_L, + uix); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write UIX66_25G_100G, err %d\n", + err); + return err; + } + + return 0; +} + +/** + * ice_phy_cfg_parpcs_e822 - Configure TUs per PAR/PCS clock cycle + * @hw: pointer to the HW struct + * @port: port to configure + * + * Configure the number of TUs for the PAR and PCS clocks used as part of the + * timestamp calibration process. This depends on the link speed, as the PHY + * uses different markers depending on the speed. + * + * 1Gb/10Gb/25Gb: + * - Tx/Rx PAR/PCS markers + * + * 25Gb RS: + * - Tx/Rx Reed Solomon gearbox PAR/PCS markers + * + * 40Gb/50Gb: + * - Tx/Rx PAR/PCS markers + * - Rx Deskew PAR/PCS markers + * + * 50G RS and 100GB RS: + * - Tx/Rx Reed Solomon gearbox PAR/PCS markers + * - Rx Deskew PAR/PCS markers + * - Tx PAR/PCS markers + * + * To calculate the conversion, we use the PHC clock frequency (cycles per + * second), the increment value (TUs per cycle), and the related PHY clock + * frequency to calculate the TUs per unit of the PHY link clock. The + * following table shows how the units convert: + * + * cycles | TUs | second + * -------+-------+-------- + * second | cycle | cycles + * + * For each conversion register, look up the appropriate frequency from the + * e822 PAR/PCS table and calculate the TUs per unit of that clock. Program + * this to the appropriate register, preparing hardware to perform timestamp + * calibration to calculate the total Tx or Rx offset to adjust the timestamp + * in order to calibrate for the internal PHY delays. + * + * Note that the increment value ranges up to ~34 bits, and the clock + * frequency is ~29 bits, so multiplying them together should fit within the + * 64 bit arithmetic. + */ +static int ice_phy_cfg_parpcs_e822(struct ice_hw *hw, u8 port) +{ + u64 cur_freq, clk_incval, tu_per_sec, phy_tus; + enum ice_ptp_link_spd link_spd; + enum ice_ptp_fec_mode fec_mode; + int err; + + err = ice_phy_get_speed_and_fec_e822(hw, port, &link_spd, &fec_mode); + if (err) + return err; + + cur_freq = ice_e822_pll_freq(ice_e822_time_ref(hw)); + clk_incval = ice_ptp_read_src_incval(hw); + + /* Calculate TUs per cycle of the PHC clock */ + tu_per_sec = cur_freq * clk_incval; + + /* For each PHY conversion register, look up the appropriate link + * speed frequency and determine the TUs per that clock's cycle time. + * Split this into a high and low value and then program the + * appropriate register. If that link speed does not use the + * associated register, write zeros to clear it instead. + */ + + /* P_REG_PAR_TX_TUS */ + if (e822_vernier[link_spd].tx_par_clk) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].tx_par_clk); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_PAR_TX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_PAR_RX_TUS */ + if (e822_vernier[link_spd].rx_par_clk) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].rx_par_clk); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_PAR_RX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_PCS_TX_TUS */ + if (e822_vernier[link_spd].tx_pcs_clk) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].tx_pcs_clk); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_PCS_TX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_PCS_RX_TUS */ + if (e822_vernier[link_spd].rx_pcs_clk) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].rx_pcs_clk); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_PCS_RX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_DESK_PAR_TX_TUS */ + if (e822_vernier[link_spd].tx_desk_rsgb_par) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].tx_desk_rsgb_par); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_DESK_PAR_TX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_DESK_PAR_RX_TUS */ + if (e822_vernier[link_spd].rx_desk_rsgb_par) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].rx_desk_rsgb_par); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_DESK_PAR_RX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_DESK_PCS_TX_TUS */ + if (e822_vernier[link_spd].tx_desk_rsgb_pcs) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].tx_desk_rsgb_pcs); + else + phy_tus = 0; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_DESK_PCS_TX_TUS_L, + phy_tus); + if (err) + return err; + + /* P_REG_DESK_PCS_RX_TUS */ + if (e822_vernier[link_spd].rx_desk_rsgb_pcs) + phy_tus = div_u64(tu_per_sec, + e822_vernier[link_spd].rx_desk_rsgb_pcs); + else + phy_tus = 0; + + return ice_write_40b_phy_reg_e822(hw, port, P_REG_DESK_PCS_RX_TUS_L, + phy_tus); +} + +/** + * ice_calc_fixed_tx_offset_e822 - Calculated Fixed Tx offset for a port + * @hw: pointer to the HW struct + * @link_spd: the Link speed to calculate for + * + * Calculate the fixed offset due to known static latency data. + */ +static u64 +ice_calc_fixed_tx_offset_e822(struct ice_hw *hw, enum ice_ptp_link_spd link_spd) +{ + u64 cur_freq, clk_incval, tu_per_sec, fixed_offset; + + cur_freq = ice_e822_pll_freq(ice_e822_time_ref(hw)); + clk_incval = ice_ptp_read_src_incval(hw); + + /* Calculate TUs per second */ + tu_per_sec = cur_freq * clk_incval; + + /* Calculate number of TUs to add for the fixed Tx latency. Since the + * latency measurement is in 1/100th of a nanosecond, we need to + * multiply by tu_per_sec and then divide by 1e11. This calculation + * overflows 64 bit integer arithmetic, so break it up into two + * divisions by 1e4 first then by 1e7. + */ + fixed_offset = div_u64(tu_per_sec, 10000); + fixed_offset *= e822_vernier[link_spd].tx_fixed_delay; + fixed_offset = div_u64(fixed_offset, 10000000); + + return fixed_offset; +} + +/** + * ice_phy_cfg_tx_offset_e822 - Configure total Tx timestamp offset + * @hw: pointer to the HW struct + * @port: the PHY port to configure + * + * Program the P_REG_TOTAL_TX_OFFSET register with the total number of TUs to + * adjust Tx timestamps by. This is calculated by combining some known static + * latency along with the Vernier offset computations done by hardware. + * + * This function must be called only after the offset registers are valid, + * i.e. after the Vernier calibration wait has passed, to ensure that the PHY + * has measured the offset. + * + * To avoid overflow, when calculating the offset based on the known static + * latency values, we use measurements in 1/100th of a nanosecond, and divide + * the TUs per second up front. This avoids overflow while allowing + * calculation of the adjustment using integer arithmetic. + */ +static int ice_phy_cfg_tx_offset_e822(struct ice_hw *hw, u8 port) +{ + enum ice_ptp_link_spd link_spd; + enum ice_ptp_fec_mode fec_mode; + u64 total_offset, val; + int err; + + err = ice_phy_get_speed_and_fec_e822(hw, port, &link_spd, &fec_mode); + if (err) + return err; + + total_offset = ice_calc_fixed_tx_offset_e822(hw, link_spd); + + /* Read the first Vernier offset from the PHY register and add it to + * the total offset. + */ + if (link_spd == ICE_PTP_LNK_SPD_1G || + link_spd == ICE_PTP_LNK_SPD_10G || + link_spd == ICE_PTP_LNK_SPD_25G || + link_spd == ICE_PTP_LNK_SPD_25G_RS || + link_spd == ICE_PTP_LNK_SPD_40G || + link_spd == ICE_PTP_LNK_SPD_50G) { + err = ice_read_64b_phy_reg_e822(hw, port, + P_REG_PAR_PCS_TX_OFFSET_L, + &val); + if (err) + return err; + + total_offset += val; + } + + /* For Tx, we only need to use the second Vernier offset for + * multi-lane link speeds with RS-FEC. The lanes will always be + * aligned. + */ + if (link_spd == ICE_PTP_LNK_SPD_50G_RS || + link_spd == ICE_PTP_LNK_SPD_100G_RS) { + err = ice_read_64b_phy_reg_e822(hw, port, + P_REG_PAR_TX_TIME_L, + &val); + if (err) + return err; + + total_offset += val; + } + + /* Now that the total offset has been calculated, program it to the + * PHY and indicate that the Tx offset is ready. After this, + * timestamps will be enabled. + */ + err = ice_write_64b_phy_reg_e822(hw, port, P_REG_TOTAL_TX_OFFSET_L, + total_offset); + if (err) + return err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_TX_OR, 1); + if (err) + return err; + + return 0; +} + +/** + * ice_phy_cfg_fixed_tx_offset_e822 - Configure Tx offset for bypass mode + * @hw: pointer to the HW struct + * @port: the PHY port to configure + * + * Calculate and program the fixed Tx offset, and indicate that the offset is + * ready. This can be used when operating in bypass mode. + */ +static int +ice_phy_cfg_fixed_tx_offset_e822(struct ice_hw *hw, u8 port) +{ + enum ice_ptp_link_spd link_spd; + enum ice_ptp_fec_mode fec_mode; + u64 total_offset; + int err; + + err = ice_phy_get_speed_and_fec_e822(hw, port, &link_spd, &fec_mode); + if (err) + return err; + + total_offset = ice_calc_fixed_tx_offset_e822(hw, link_spd); + + /* Program the fixed Tx offset into the P_REG_TOTAL_TX_OFFSET_L + * register, then indicate that the Tx offset is ready. After this, + * timestamps will be enabled. + * + * Note that this skips including the more precise offsets generated + * by the Vernier calibration. + */ + err = ice_write_64b_phy_reg_e822(hw, port, P_REG_TOTAL_TX_OFFSET_L, + total_offset); + if (err) + return err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_TX_OR, 1); + if (err) + return err; + + return 0; +} + +/** + * ice_phy_calc_pmd_adj_e822 - Calculate PMD adjustment for Rx + * @hw: pointer to the HW struct + * @port: the PHY port to adjust for + * @link_spd: the current link speed of the PHY + * @fec_mode: the current FEC mode of the PHY + * @pmd_adj: on return, the amount to adjust the Rx total offset by + * + * Calculates the adjustment to Rx timestamps due to PMD alignment in the PHY. + * This varies by link speed and FEC mode. The value calculated accounts for + * various delays caused when receiving a packet. + */ +static int +ice_phy_calc_pmd_adj_e822(struct ice_hw *hw, u8 port, + enum ice_ptp_link_spd link_spd, + enum ice_ptp_fec_mode fec_mode, u64 *pmd_adj) +{ + u64 cur_freq, clk_incval, tu_per_sec, mult, adj; + u8 pmd_align; + u32 val; + int err; + + err = ice_read_phy_reg_e822(hw, port, P_REG_PMD_ALIGNMENT, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read PMD alignment, err %d\n", + err); + return err; + } + + pmd_align = (u8)val; + + cur_freq = ice_e822_pll_freq(ice_e822_time_ref(hw)); + clk_incval = ice_ptp_read_src_incval(hw); + + /* Calculate TUs per second */ + tu_per_sec = cur_freq * clk_incval; + + /* The PMD alignment adjustment measurement depends on the link speed, + * and whether FEC is enabled. For each link speed, the alignment + * adjustment is calculated by dividing a value by the length of + * a Time Unit in nanoseconds. + * + * 1G: align == 4 ? 10 * 0.8 : (align + 6 % 10) * 0.8 + * 10G: align == 65 ? 0 : (align * 0.1 * 32/33) + * 10G w/FEC: align * 0.1 * 32/33 + * 25G: align == 65 ? 0 : (align * 0.4 * 32/33) + * 25G w/FEC: align * 0.4 * 32/33 + * 40G: align == 65 ? 0 : (align * 0.1 * 32/33) + * 40G w/FEC: align * 0.1 * 32/33 + * 50G: align == 65 ? 0 : (align * 0.4 * 32/33) + * 50G w/FEC: align * 0.8 * 32/33 + * + * For RS-FEC, if align is < 17 then we must also add 1.6 * 32/33. + * + * To allow for calculating this value using integer arithmetic, we + * instead start with the number of TUs per second, (inverse of the + * length of a Time Unit in nanoseconds), multiply by a value based + * on the PMD alignment register, and then divide by the right value + * calculated based on the table above. To avoid integer overflow this + * division is broken up into a step of dividing by 125 first. + */ + if (link_spd == ICE_PTP_LNK_SPD_1G) { + if (pmd_align == 4) + mult = 10; + else + mult = (pmd_align + 6) % 10; + } else if (link_spd == ICE_PTP_LNK_SPD_10G || + link_spd == ICE_PTP_LNK_SPD_25G || + link_spd == ICE_PTP_LNK_SPD_40G || + link_spd == ICE_PTP_LNK_SPD_50G) { + /* If Clause 74 FEC, always calculate PMD adjust */ + if (pmd_align != 65 || fec_mode == ICE_PTP_FEC_MODE_CLAUSE74) + mult = pmd_align; + else + mult = 0; + } else if (link_spd == ICE_PTP_LNK_SPD_25G_RS || + link_spd == ICE_PTP_LNK_SPD_50G_RS || + link_spd == ICE_PTP_LNK_SPD_100G_RS) { + if (pmd_align < 17) + mult = pmd_align + 40; + else + mult = pmd_align; + } else { + ice_debug(hw, ICE_DBG_PTP, "Unknown link speed %d, skipping PMD adjustment\n", + link_spd); + mult = 0; + } + + /* In some cases, there's no need to adjust for the PMD alignment */ + if (!mult) { + *pmd_adj = 0; + return 0; + } + + /* Calculate the adjustment by multiplying TUs per second by the + * appropriate multiplier and divisor. To avoid overflow, we first + * divide by 125, and then handle remaining divisor based on the link + * speed pmd_adj_divisor value. + */ + adj = div_u64(tu_per_sec, 125); + adj *= mult; + adj = div_u64(adj, e822_vernier[link_spd].pmd_adj_divisor); + + /* Finally, for 25G-RS and 50G-RS, a further adjustment for the Rx + * cycle count is necessary. + */ + if (link_spd == ICE_PTP_LNK_SPD_25G_RS) { + u64 cycle_adj; + u8 rx_cycle; + + err = ice_read_phy_reg_e822(hw, port, P_REG_RX_40_TO_160_CNT, + &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read 25G-RS Rx cycle count, err %d\n", + err); + return err; + } + + rx_cycle = val & P_REG_RX_40_TO_160_CNT_RXCYC_M; + if (rx_cycle) { + mult = (4 - rx_cycle) * 40; + + cycle_adj = div_u64(tu_per_sec, 125); + cycle_adj *= mult; + cycle_adj = div_u64(cycle_adj, e822_vernier[link_spd].pmd_adj_divisor); + + adj += cycle_adj; + } + } else if (link_spd == ICE_PTP_LNK_SPD_50G_RS) { + u64 cycle_adj; + u8 rx_cycle; + + err = ice_read_phy_reg_e822(hw, port, P_REG_RX_80_TO_160_CNT, + &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read 50G-RS Rx cycle count, err %d\n", + err); + return err; + } + + rx_cycle = val & P_REG_RX_80_TO_160_CNT_RXCYC_M; + if (rx_cycle) { + mult = rx_cycle * 40; + + cycle_adj = div_u64(tu_per_sec, 125); + cycle_adj *= mult; + cycle_adj = div_u64(cycle_adj, e822_vernier[link_spd].pmd_adj_divisor); + + adj += cycle_adj; + } + } + + /* Return the calculated adjustment */ + *pmd_adj = adj; + + return 0; +} + +/** + * ice_calc_fixed_rx_offset_e822 - Calculated the fixed Rx offset for a port + * @hw: pointer to HW struct + * @link_spd: The Link speed to calculate for + * + * Determine the fixed Rx latency for a given link speed. + */ +static u64 +ice_calc_fixed_rx_offset_e822(struct ice_hw *hw, enum ice_ptp_link_spd link_spd) +{ + u64 cur_freq, clk_incval, tu_per_sec, fixed_offset; + + cur_freq = ice_e822_pll_freq(ice_e822_time_ref(hw)); + clk_incval = ice_ptp_read_src_incval(hw); + + /* Calculate TUs per second */ + tu_per_sec = cur_freq * clk_incval; + + /* Calculate number of TUs to add for the fixed Rx latency. Since the + * latency measurement is in 1/100th of a nanosecond, we need to + * multiply by tu_per_sec and then divide by 1e11. This calculation + * overflows 64 bit integer arithmetic, so break it up into two + * divisions by 1e4 first then by 1e7. + */ + fixed_offset = div_u64(tu_per_sec, 10000); + fixed_offset *= e822_vernier[link_spd].rx_fixed_delay; + fixed_offset = div_u64(fixed_offset, 10000000); + + return fixed_offset; +} + +/** + * ice_phy_cfg_rx_offset_e822 - Configure total Rx timestamp offset + * @hw: pointer to the HW struct + * @port: the PHY port to configure + * + * Program the P_REG_TOTAL_RX_OFFSET register with the number of Time Units to + * adjust Rx timestamps by. This combines calculations from the Vernier offset + * measurements taken in hardware with some data about known fixed delay as + * well as adjusting for multi-lane alignment delay. + * + * This function must be called only after the offset registers are valid, + * i.e. after the Vernier calibration wait has passed, to ensure that the PHY + * has measured the offset. + * + * To avoid overflow, when calculating the offset based on the known static + * latency values, we use measurements in 1/100th of a nanosecond, and divide + * the TUs per second up front. This avoids overflow while allowing + * calculation of the adjustment using integer arithmetic. + */ +static int ice_phy_cfg_rx_offset_e822(struct ice_hw *hw, u8 port) +{ + enum ice_ptp_link_spd link_spd; + enum ice_ptp_fec_mode fec_mode; + u64 total_offset, pmd, val; + int err; + + err = ice_phy_get_speed_and_fec_e822(hw, port, &link_spd, &fec_mode); + if (err) + return err; + + total_offset = ice_calc_fixed_rx_offset_e822(hw, link_spd); + + /* Read the first Vernier offset from the PHY register and add it to + * the total offset. + */ + err = ice_read_64b_phy_reg_e822(hw, port, + P_REG_PAR_PCS_RX_OFFSET_L, + &val); + if (err) + return err; + + total_offset += val; + + /* For Rx, all multi-lane link speeds include a second Vernier + * calibration, because the lanes might not be aligned. + */ + if (link_spd == ICE_PTP_LNK_SPD_40G || + link_spd == ICE_PTP_LNK_SPD_50G || + link_spd == ICE_PTP_LNK_SPD_50G_RS || + link_spd == ICE_PTP_LNK_SPD_100G_RS) { + err = ice_read_64b_phy_reg_e822(hw, port, + P_REG_PAR_RX_TIME_L, + &val); + if (err) + return err; + + total_offset += val; + } + + /* In addition, Rx must account for the PMD alignment */ + err = ice_phy_calc_pmd_adj_e822(hw, port, link_spd, fec_mode, &pmd); + if (err) + return err; + + /* For RS-FEC, this adjustment adds delay, but for other modes, it + * subtracts delay. + */ + if (fec_mode == ICE_PTP_FEC_MODE_RS_FEC) + total_offset += pmd; + else + total_offset -= pmd; + + /* Now that the total offset has been calculated, program it to the + * PHY and indicate that the Rx offset is ready. After this, + * timestamps will be enabled. + */ + err = ice_write_64b_phy_reg_e822(hw, port, P_REG_TOTAL_RX_OFFSET_L, + total_offset); + if (err) + return err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_RX_OR, 1); + if (err) + return err; + + return 0; +} + +/** + * ice_phy_cfg_fixed_rx_offset_e822 - Configure fixed Rx offset for bypass mode + * @hw: pointer to the HW struct + * @port: the PHY port to configure + * + * Calculate and program the fixed Rx offset, and indicate that the offset is + * ready. This can be used when operating in bypass mode. + */ +static int +ice_phy_cfg_fixed_rx_offset_e822(struct ice_hw *hw, u8 port) +{ + enum ice_ptp_link_spd link_spd; + enum ice_ptp_fec_mode fec_mode; + u64 total_offset; + int err; + + err = ice_phy_get_speed_and_fec_e822(hw, port, &link_spd, &fec_mode); + if (err) + return err; + + total_offset = ice_calc_fixed_rx_offset_e822(hw, link_spd); + + /* Program the fixed Rx offset into the P_REG_TOTAL_RX_OFFSET_L + * register, then indicate that the Rx offset is ready. After this, + * timestamps will be enabled. + * + * Note that this skips including the more precise offsets generated + * by Vernier calibration. + */ + err = ice_write_64b_phy_reg_e822(hw, port, P_REG_TOTAL_RX_OFFSET_L, + total_offset); + if (err) + return err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_RX_OR, 1); + if (err) + return err; + + return 0; +} + +/** + * ice_read_phy_and_phc_time_e822 - Simultaneously capture PHC and PHY time + * @hw: pointer to the HW struct + * @port: the PHY port to read + * @phy_time: on return, the 64bit PHY timer value + * @phc_time: on return, the lower 64bits of PHC time + * + * Issue a READ_TIME timer command to simultaneously capture the PHY and PHC + * timer values. + */ +static int +ice_read_phy_and_phc_time_e822(struct ice_hw *hw, u8 port, u64 *phy_time, + u64 *phc_time) +{ + u64 tx_time, rx_time; + u32 zo, lo; + u8 tmr_idx; + int err; + + tmr_idx = ice_get_ptp_src_clock_index(hw); + + /* Prepare the PHC timer for a READ_TIME capture command */ + ice_ptp_src_cmd(hw, READ_TIME); + + /* Prepare the PHY timer for a READ_TIME capture command */ + err = ice_ptp_one_port_cmd(hw, port, READ_TIME); + if (err) + return err; + + /* Issue the sync to start the READ_TIME capture */ + ice_ptp_exec_tmr_cmd(hw); + + /* Read the captured PHC time from the shadow time registers */ + zo = rd32(hw, GLTSYN_SHTIME_0(tmr_idx)); + lo = rd32(hw, GLTSYN_SHTIME_L(tmr_idx)); + *phc_time = (u64)lo << 32 | zo; + + /* Read the captured PHY time from the PHY shadow registers */ + err = ice_ptp_read_port_capture(hw, port, &tx_time, &rx_time); + if (err) + return err; + + /* If the PHY Tx and Rx timers don't match, log a warning message. + * Note that this should not happen in normal circumstances since the + * driver always programs them together. + */ + if (tx_time != rx_time) + dev_warn(ice_hw_to_dev(hw), + "PHY port %u Tx and Rx timers do not match, tx_time 0x%016llX, rx_time 0x%016llX\n", + port, (unsigned long long)tx_time, + (unsigned long long)rx_time); + + *phy_time = tx_time; + + return 0; +} + +/** + * ice_sync_phy_timer_e822 - Synchronize the PHY timer with PHC timer + * @hw: pointer to the HW struct + * @port: the PHY port to synchronize + * + * Perform an adjustment to ensure that the PHY and PHC timers are in sync. + * This is done by issuing a READ_TIME command which triggers a simultaneous + * read of the PHY timer and PHC timer. Then we use the difference to + * calculate an appropriate 2s complement addition to add to the PHY timer in + * order to ensure it reads the same value as the primary PHC timer. + */ +static int ice_sync_phy_timer_e822(struct ice_hw *hw, u8 port) +{ + u64 phc_time, phy_time, difference; + int err; + + if (!ice_ptp_lock(hw)) { + ice_debug(hw, ICE_DBG_PTP, "Failed to acquire PTP semaphore\n"); + return -EBUSY; + } + + err = ice_read_phy_and_phc_time_e822(hw, port, &phy_time, &phc_time); + if (err) + goto err_unlock; + + /* Calculate the amount required to add to the port time in order for + * it to match the PHC time. + * + * Note that the port adjustment is done using 2s complement + * arithmetic. This is convenient since it means that we can simply + * calculate the difference between the PHC time and the port time, + * and it will be interpreted correctly. + */ + difference = phc_time - phy_time; + + err = ice_ptp_prep_port_adj_e822(hw, port, (s64)difference); + if (err) + goto err_unlock; + + err = ice_ptp_one_port_cmd(hw, port, ADJ_TIME); + if (err) + goto err_unlock; + + /* Issue the sync to activate the time adjustment */ + ice_ptp_exec_tmr_cmd(hw); + + /* Re-capture the timer values to flush the command registers and + * verify that the time was properly adjusted. + */ + err = ice_read_phy_and_phc_time_e822(hw, port, &phy_time, &phc_time); + if (err) + goto err_unlock; + + dev_info(ice_hw_to_dev(hw), + "Port %u PHY time synced to PHC: 0x%016llX, 0x%016llX\n", + port, (unsigned long long)phy_time, + (unsigned long long)phc_time); + + ice_ptp_unlock(hw); + + return 0; + +err_unlock: + ice_ptp_unlock(hw); + return err; +} + +/** + * ice_stop_phy_timer_e822 - Stop the PHY clock timer + * @hw: pointer to the HW struct + * @port: the PHY port to stop + * @soft_reset: if true, hold the SOFT_RESET bit of P_REG_PS + * + * Stop the clock of a PHY port. This must be done as part of the flow to + * re-calibrate Tx and Rx timestamping offsets whenever the clock time is + * initialized or when link speed changes. + */ +int +ice_stop_phy_timer_e822(struct ice_hw *hw, u8 port, bool soft_reset) +{ + int err; + u32 val; + + err = ice_write_phy_reg_e822(hw, port, P_REG_TX_OR, 0); + if (err) + return err; + + err = ice_write_phy_reg_e822(hw, port, P_REG_RX_OR, 0); + if (err) + return err; + + err = ice_read_phy_reg_e822(hw, port, P_REG_PS, &val); + if (err) + return err; + + val &= ~P_REG_PS_START_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + val &= ~P_REG_PS_ENA_CLK_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + if (soft_reset) { + val |= P_REG_PS_SFT_RESET_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + } + + ice_debug(hw, ICE_DBG_PTP, "Disabled clock on PHY port %u\n", port); + + return 0; +} + +/** + * ice_start_phy_timer_e822 - Start the PHY clock timer + * @hw: pointer to the HW struct + * @port: the PHY port to start + * @bypass: if true, start the PHY in bypass mode + * + * Start the clock of a PHY port. This must be done as part of the flow to + * re-calibrate Tx and Rx timestamping offsets whenever the clock time is + * initialized or when link speed changes. + * + * Bypass mode enables timestamps immediately without waiting for Vernier + * calibration to complete. Hardware will still continue taking Vernier + * measurements on Tx or Rx of packets, but they will not be applied to + * timestamps. Use ice_phy_exit_bypass_e822 to exit bypass mode once hardware + * has completed offset calculation. + */ +int +ice_start_phy_timer_e822(struct ice_hw *hw, u8 port, bool bypass) +{ + u32 lo, hi, val; + u64 incval; + u8 tmr_idx; + int err; + + tmr_idx = ice_get_ptp_src_clock_index(hw); + + err = ice_stop_phy_timer_e822(hw, port, false); + if (err) + return err; + + ice_phy_cfg_lane_e822(hw, port); + + err = ice_phy_cfg_uix_e822(hw, port); + if (err) + return err; + + err = ice_phy_cfg_parpcs_e822(hw, port); + if (err) + return err; + + lo = rd32(hw, GLTSYN_INCVAL_L(tmr_idx)); + hi = rd32(hw, GLTSYN_INCVAL_H(tmr_idx)); + incval = (u64)hi << 32 | lo; + + err = ice_write_40b_phy_reg_e822(hw, port, P_REG_TIMETUS_L, incval); + if (err) + return err; + + err = ice_ptp_one_port_cmd(hw, port, INIT_INCVAL); + if (err) + return err; + + ice_ptp_exec_tmr_cmd(hw); + + err = ice_read_phy_reg_e822(hw, port, P_REG_PS, &val); + if (err) + return err; + + val |= P_REG_PS_SFT_RESET_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + val |= P_REG_PS_START_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + val &= ~P_REG_PS_SFT_RESET_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + err = ice_ptp_one_port_cmd(hw, port, INIT_INCVAL); + if (err) + return err; + + ice_ptp_exec_tmr_cmd(hw); + + val |= P_REG_PS_ENA_CLK_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + val |= P_REG_PS_LOAD_OFFSET_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + ice_ptp_exec_tmr_cmd(hw); + + err = ice_sync_phy_timer_e822(hw, port); + if (err) + return err; + + if (bypass) { + val |= P_REG_PS_BYPASS_MODE_M; + /* Enter BYPASS mode, enabling timestamps immediately. */ + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) + return err; + + /* Program the fixed Tx offset */ + err = ice_phy_cfg_fixed_tx_offset_e822(hw, port); + if (err) + return err; + + /* Program the fixed Rx offset */ + err = ice_phy_cfg_fixed_rx_offset_e822(hw, port); + if (err) + return err; + } + + ice_debug(hw, ICE_DBG_PTP, "Enabled clock on PHY port %u\n", port); + + return 0; +} + +/** + * ice_phy_exit_bypass_e822 - Exit bypass mode, after vernier calculations + * @hw: pointer to the HW struct + * @port: the PHY port to configure + * + * After hardware finishes vernier calculations for the Tx and Rx offset, this + * function can be used to exit bypass mode by updating the total Tx and Rx + * offsets, and then disabling bypass. This will enable hardware to include + * the more precise offset calibrations, increasing precision of the generated + * timestamps. + * + * This cannot be done until hardware has measured the offsets, which requires + * waiting until at least one packet has been sent and received by the device. + */ +int ice_phy_exit_bypass_e822(struct ice_hw *hw, u8 port) +{ + int err; + u32 val; + + err = ice_read_phy_reg_e822(hw, port, P_REG_TX_OV_STATUS, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read TX_OV_STATUS for port %u, err %d\n", + port, err); + return err; + } + + if (!(val & P_REG_TX_OV_STATUS_OV_M)) { + ice_debug(hw, ICE_DBG_PTP, "Tx offset is not yet valid for port %u\n", + port); + return -EBUSY; + } + + err = ice_read_phy_reg_e822(hw, port, P_REG_RX_OV_STATUS, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read RX_OV_STATUS for port %u, err %d\n", + port, err); + return err; + } + + if (!(val & P_REG_TX_OV_STATUS_OV_M)) { + ice_debug(hw, ICE_DBG_PTP, "Rx offset is not yet valid for port %u\n", + port); + return -EBUSY; + } + + err = ice_phy_cfg_tx_offset_e822(hw, port); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to program total Tx offset for port %u, err %d\n", + port, err); + return err; + } + + err = ice_phy_cfg_rx_offset_e822(hw, port); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to program total Rx offset for port %u, err %d\n", + port, err); + return err; + } + + /* Exit bypass mode now that the offset has been updated */ + err = ice_read_phy_reg_e822(hw, port, P_REG_PS, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read P_REG_PS for port %u, err %d\n", + port, err); + return err; + } + + if (!(val & P_REG_PS_BYPASS_MODE_M)) + ice_debug(hw, ICE_DBG_PTP, "Port %u not in bypass mode\n", + port); + + val &= ~P_REG_PS_BYPASS_MODE_M; + err = ice_write_phy_reg_e822(hw, port, P_REG_PS, val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to disable bypass for port %u, err %d\n", + port, err); + return err; + } + + dev_info(ice_hw_to_dev(hw), "Exiting bypass mode on PHY port %u\n", + port); + + return 0; +} + /* E810 functions * * The following functions operate on the E810 series devices which use @@ -68,18 +2538,18 @@ u8 ice_get_ptp_src_clock_index(struct ice_hw *hw) static int ice_read_phy_reg_e810(struct ice_hw *hw, u32 addr, u32 *val) { struct ice_sbq_msg_input msg = {0}; - int status; + int err; msg.msg_addr_low = lower_16_bits(addr); msg.msg_addr_high = upper_16_bits(addr); msg.opcode = ice_sbq_msg_rd; msg.dest_dev = rmn_0; - status = ice_sbq_rw_reg(hw, &msg); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, status %d\n", - status); - return status; + err = ice_sbq_rw_reg(hw, &msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n", + err); + return err; } *val = msg.data; @@ -98,7 +2568,7 @@ static int ice_read_phy_reg_e810(struct ice_hw *hw, u32 addr, u32 *val) static int ice_write_phy_reg_e810(struct ice_hw *hw, u32 addr, u32 val) { struct ice_sbq_msg_input msg = {0}; - int status; + int err; msg.msg_addr_low = lower_16_bits(addr); msg.msg_addr_high = upper_16_bits(addr); @@ -106,11 +2576,11 @@ static int ice_write_phy_reg_e810(struct ice_hw *hw, u32 addr, u32 val) msg.dest_dev = rmn_0; msg.data = val; - status = ice_sbq_rw_reg(hw, &msg); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, status %d\n", - status); - return status; + err = ice_sbq_rw_reg(hw, &msg); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n", + err); + return err; } return 0; @@ -130,23 +2600,23 @@ static int ice_read_phy_tstamp_e810(struct ice_hw *hw, u8 lport, u8 idx, u64 *tstamp) { u32 lo_addr, hi_addr, lo, hi; - int status; + int err; lo_addr = TS_EXT(LOW_TX_MEMORY_BANK_START, lport, idx); hi_addr = TS_EXT(HIGH_TX_MEMORY_BANK_START, lport, idx); - status = ice_read_phy_reg_e810(hw, lo_addr, &lo); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to read low PTP timestamp register, status %d\n", - status); - return status; + err = ice_read_phy_reg_e810(hw, lo_addr, &lo); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read low PTP timestamp register, err %d\n", + err); + return err; } - status = ice_read_phy_reg_e810(hw, hi_addr, &hi); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to read high PTP timestamp register, status %d\n", - status); - return status; + err = ice_read_phy_reg_e810(hw, hi_addr, &hi); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read high PTP timestamp register, err %d\n", + err); + return err; } /* For E810 devices, the timestamp is reported with the lower 32 bits @@ -169,23 +2639,23 @@ ice_read_phy_tstamp_e810(struct ice_hw *hw, u8 lport, u8 idx, u64 *tstamp) static int ice_clear_phy_tstamp_e810(struct ice_hw *hw, u8 lport, u8 idx) { u32 lo_addr, hi_addr; - int status; + int err; lo_addr = TS_EXT(LOW_TX_MEMORY_BANK_START, lport, idx); hi_addr = TS_EXT(HIGH_TX_MEMORY_BANK_START, lport, idx); - status = ice_write_phy_reg_e810(hw, lo_addr, 0); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to clear low PTP timestamp register, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, lo_addr, 0); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to clear low PTP timestamp register, err %d\n", + err); + return err; } - status = ice_write_phy_reg_e810(hw, hi_addr, 0); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to clear high PTP timestamp register, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, hi_addr, 0); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to clear high PTP timestamp register, err %d\n", + err); + return err; } return 0; @@ -200,17 +2670,32 @@ static int ice_clear_phy_tstamp_e810(struct ice_hw *hw, u8 lport, u8 idx) */ int ice_ptp_init_phy_e810(struct ice_hw *hw) { - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_ENA(tmr_idx), - GLTSYN_ENA_TSYN_ENA_M); - if (status) + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_ENA(tmr_idx), + GLTSYN_ENA_TSYN_ENA_M); + if (err) ice_debug(hw, ICE_DBG_PTP, "PTP failed in ena_phy_time_syn %d\n", - status); + err); - return status; + return err; +} + +/** + * ice_ptp_init_phc_e810 - Perform E810 specific PHC initialization + * @hw: pointer to HW struct + * + * Perform E810-specific PTP hardware clock initialization steps. + */ +static int ice_ptp_init_phc_e810(struct ice_hw *hw) +{ + /* Ensure synchronization delay is zero */ + wr32(hw, GLTSYN_SYNC_DLAY, 0); + + /* Initialize the PHY */ + return ice_ptp_init_phy_e810(hw); } /** @@ -227,22 +2712,22 @@ int ice_ptp_init_phy_e810(struct ice_hw *hw) */ static int ice_ptp_prep_phy_time_e810(struct ice_hw *hw, u32 time) { - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHTIME_0(tmr_idx), 0); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write SHTIME_0, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHTIME_0(tmr_idx), 0); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write SHTIME_0, err %d\n", + err); + return err; } - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHTIME_L(tmr_idx), time); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write SHTIME_L, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHTIME_L(tmr_idx), time); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write SHTIME_L, err %d\n", + err); + return err; } return 0; @@ -263,26 +2748,26 @@ static int ice_ptp_prep_phy_time_e810(struct ice_hw *hw, u32 time) */ static int ice_ptp_prep_phy_adj_e810(struct ice_hw *hw, s32 adj) { - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; /* Adjustments are represented as signed 2's complement values in * nanoseconds. Sub-nanosecond adjustment is not supported. */ - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_L(tmr_idx), 0); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write adj to PHY SHADJ_L, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_L(tmr_idx), 0); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write adj to PHY SHADJ_L, err %d\n", + err); + return err; } - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_H(tmr_idx), adj); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write adj to PHY SHADJ_H, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_H(tmr_idx), adj); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write adj to PHY SHADJ_H, err %d\n", + err); + return err; } return 0; @@ -300,25 +2785,25 @@ static int ice_ptp_prep_phy_adj_e810(struct ice_hw *hw, s32 adj) static int ice_ptp_prep_phy_incval_e810(struct ice_hw *hw, u64 incval) { u32 high, low; - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; low = lower_32_bits(incval); high = upper_32_bits(incval); - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_L(tmr_idx), low); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write incval to PHY SHADJ_L, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_L(tmr_idx), low); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write incval to PHY SHADJ_L, err %d\n", + err); + return err; } - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_H(tmr_idx), high); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write incval PHY SHADJ_H, status %d\n", - status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_SHADJ_H(tmr_idx), high); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write incval PHY SHADJ_H, err %d\n", + err); + return err; } return 0; @@ -335,7 +2820,7 @@ static int ice_ptp_prep_phy_incval_e810(struct ice_hw *hw, u64 incval) static int ice_ptp_port_cmd_e810(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) { u32 cmd_val, val; - int status; + int err; switch (cmd) { case INIT_TIME: @@ -356,20 +2841,20 @@ static int ice_ptp_port_cmd_e810(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) } /* Read, modify, write */ - status = ice_read_phy_reg_e810(hw, ETH_GLTSYN_CMD, &val); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to read GLTSYN_CMD, status %d\n", status); - return status; + err = ice_read_phy_reg_e810(hw, ETH_GLTSYN_CMD, &val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to read GLTSYN_CMD, err %d\n", err); + return err; } /* Modify necessary bits only and perform write */ val &= ~TS_CMD_MASK_E810; val |= cmd_val; - status = ice_write_phy_reg_e810(hw, ETH_GLTSYN_CMD, val); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to write back GLTSYN_CMD, status %d\n", status); - return status; + err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_CMD, val); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to write back GLTSYN_CMD, err %d\n", err); + return err; } return 0; @@ -377,12 +2862,9 @@ static int ice_ptp_port_cmd_e810(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) /* Device agnostic functions * - * The following functions implement useful behavior to hide the differences - * between E810 and other devices. They call the device-specific - * implementations where necessary. - * - * Currently, the driver only supports E810, but future work will enable - * support for E822-based devices. + * The following functions implement shared behavior common to both E822 and + * E810 devices, possibly calling a device specific implementation where + * necessary. */ /** @@ -433,42 +2915,6 @@ void ice_ptp_unlock(struct ice_hw *hw) } /** - * ice_ptp_src_cmd - Prepare source timer for a timer command - * @hw: pointer to HW structure - * @cmd: Timer command - * - * Prepare the source timer for an upcoming timer sync command. - */ -static void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) -{ - u32 cmd_val; - u8 tmr_idx; - - tmr_idx = ice_get_ptp_src_clock_index(hw); - cmd_val = tmr_idx << SEL_CPK_SRC; - - switch (cmd) { - case INIT_TIME: - cmd_val |= GLTSYN_CMD_INIT_TIME; - break; - case INIT_INCVAL: - cmd_val |= GLTSYN_CMD_INIT_INCVAL; - break; - case ADJ_TIME: - cmd_val |= GLTSYN_CMD_ADJ_TIME; - break; - case ADJ_TIME_AT_TIME: - cmd_val |= GLTSYN_CMD_ADJ_INIT_TIME; - break; - case READ_TIME: - cmd_val |= GLTSYN_CMD_READ_TIME; - break; - } - - wr32(hw, GLTSYN_CMD, cmd_val); -} - -/** * ice_ptp_tmr_cmd - Prepare and trigger a timer sync command * @hw: pointer to HW struct * @cmd: the command to issue @@ -480,23 +2926,26 @@ static void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) */ static int ice_ptp_tmr_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) { - int status; + int err; /* First, prepare the source timer */ ice_ptp_src_cmd(hw, cmd); /* Next, prepare the ports */ - status = ice_ptp_port_cmd_e810(hw, cmd); - if (status) { - ice_debug(hw, ICE_DBG_PTP, "Failed to prepare PHY ports for timer command %u, status %d\n", - cmd, status); - return status; + if (ice_is_e810(hw)) + err = ice_ptp_port_cmd_e810(hw, cmd); + else + err = ice_ptp_port_cmd_e822(hw, cmd); + if (err) { + ice_debug(hw, ICE_DBG_PTP, "Failed to prepare PHY ports for timer command %u, err %d\n", + cmd, err); + return err; } - /* Write the sync command register to drive both source and PHY timer commands - * synchronously + /* Write the sync command register to drive both source and PHY timer + * commands synchronously */ - wr32(hw, GLTSYN_CMD_SYNC, SYNC_EXEC_CMD); + ice_ptp_exec_tmr_cmd(hw); return 0; } @@ -516,8 +2965,8 @@ static int ice_ptp_tmr_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd) */ int ice_ptp_init_time(struct ice_hw *hw, u64 time) { - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; @@ -528,9 +2977,12 @@ int ice_ptp_init_time(struct ice_hw *hw, u64 time) /* PHY timers */ /* Fill Rx and Tx ports and send msg to PHY */ - status = ice_ptp_prep_phy_time_e810(hw, time & 0xFFFFFFFF); - if (status) - return status; + if (ice_is_e810(hw)) + err = ice_ptp_prep_phy_time_e810(hw, time & 0xFFFFFFFF); + else + err = ice_ptp_prep_phy_time_e822(hw, time & 0xFFFFFFFF); + if (err) + return err; return ice_ptp_tmr_cmd(hw, INIT_TIME); } @@ -551,8 +3003,8 @@ int ice_ptp_init_time(struct ice_hw *hw, u64 time) */ int ice_ptp_write_incval(struct ice_hw *hw, u64 incval) { - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; @@ -560,9 +3012,12 @@ int ice_ptp_write_incval(struct ice_hw *hw, u64 incval) wr32(hw, GLTSYN_SHADJ_L(tmr_idx), lower_32_bits(incval)); wr32(hw, GLTSYN_SHADJ_H(tmr_idx), upper_32_bits(incval)); - status = ice_ptp_prep_phy_incval_e810(hw, incval); - if (status) - return status; + if (ice_is_e810(hw)) + err = ice_ptp_prep_phy_incval_e810(hw, incval); + else + err = ice_ptp_prep_phy_incval_e822(hw, incval); + if (err) + return err; return ice_ptp_tmr_cmd(hw, INIT_INCVAL); } @@ -576,16 +3031,16 @@ int ice_ptp_write_incval(struct ice_hw *hw, u64 incval) */ int ice_ptp_write_incval_locked(struct ice_hw *hw, u64 incval) { - int status; + int err; if (!ice_ptp_lock(hw)) return -EBUSY; - status = ice_ptp_write_incval(hw, incval); + err = ice_ptp_write_incval(hw, incval); ice_ptp_unlock(hw); - return status; + return err; } /** @@ -603,8 +3058,8 @@ int ice_ptp_write_incval_locked(struct ice_hw *hw, u64 incval) */ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj) { - int status; u8 tmr_idx; + int err; tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; @@ -616,9 +3071,12 @@ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj) wr32(hw, GLTSYN_SHADJ_L(tmr_idx), 0); wr32(hw, GLTSYN_SHADJ_H(tmr_idx), adj); - status = ice_ptp_prep_phy_adj_e810(hw, adj); - if (status) - return status; + if (ice_is_e810(hw)) + err = ice_ptp_prep_phy_adj_e810(hw, adj); + else + err = ice_ptp_prep_phy_adj_e822(hw, adj); + if (err) + return err; return ice_ptp_tmr_cmd(hw, ADJ_TIME); } @@ -630,11 +3088,16 @@ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj) * @idx: the timestamp index to read * @tstamp: on return, the 40bit timestamp value * - * Read a 40bit timestamp value out of the timestamp block. + * Read a 40bit timestamp value out of the timestamp block. For E822 devices, + * the block is the quad to read from. For E810 devices, the block is the + * logical port to read from. */ int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp) { - return ice_read_phy_tstamp_e810(hw, block, idx, tstamp); + if (ice_is_e810(hw)) + return ice_read_phy_tstamp_e810(hw, block, idx, tstamp); + else + return ice_read_phy_tstamp_e822(hw, block, idx, tstamp); } /** @@ -643,11 +3106,16 @@ int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp) * @block: the block to read from * @idx: the timestamp index to reset * - * Clear a timestamp, resetting its valid bit, from the timestamp block. + * Clear a timestamp, resetting its valid bit, from the timestamp block. For + * E822 devices, the block is the quad to clear from. For E810 devices, the + * block is the logical port to clear from. */ int ice_clear_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx) { - return ice_clear_phy_tstamp_e810(hw, block, idx); + if (ice_is_e810(hw)) + return ice_clear_phy_tstamp_e810(hw, block, idx); + else + return ice_clear_phy_tstamp_e822(hw, block, idx); } /* E810T SMA functions @@ -800,3 +3268,25 @@ bool ice_is_pca9575_present(struct ice_hw *hw) return !status && handle; } + +/** + * ice_ptp_init_phc - Initialize PTP hardware clock + * @hw: pointer to the HW struct + * + * Perform the steps required to initialize the PTP hardware clock. + */ +int ice_ptp_init_phc(struct ice_hw *hw) +{ + u8 src_idx = hw->func_caps.ts_func_info.tmr_index_owned; + + /* Enable source clocks */ + wr32(hw, GLTSYN_ENA(src_idx), GLTSYN_ENA_TSYN_ENA_M); + + /* Clear event err indications for auxiliary pins */ + (void)rd32(hw, GLTSYN_STAT(src_idx)); + + if (ice_is_e810(hw)) + return ice_ptp_init_phc_e810(hw); + else + return ice_ptp_init_phc_e822(hw); +} diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h index b2984b5c22c1..519e75462e67 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h +++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h @@ -12,6 +12,112 @@ enum ice_ptp_tmr_cmd { READ_TIME }; +enum ice_ptp_serdes { + ICE_PTP_SERDES_1G, + ICE_PTP_SERDES_10G, + ICE_PTP_SERDES_25G, + ICE_PTP_SERDES_40G, + ICE_PTP_SERDES_50G, + ICE_PTP_SERDES_100G +}; + +enum ice_ptp_link_spd { + ICE_PTP_LNK_SPD_1G, + ICE_PTP_LNK_SPD_10G, + ICE_PTP_LNK_SPD_25G, + ICE_PTP_LNK_SPD_25G_RS, + ICE_PTP_LNK_SPD_40G, + ICE_PTP_LNK_SPD_50G, + ICE_PTP_LNK_SPD_50G_RS, + ICE_PTP_LNK_SPD_100G_RS, + NUM_ICE_PTP_LNK_SPD /* Must be last */ +}; + +enum ice_ptp_fec_mode { + ICE_PTP_FEC_MODE_NONE, + ICE_PTP_FEC_MODE_CLAUSE74, + ICE_PTP_FEC_MODE_RS_FEC +}; + +/** + * struct ice_time_ref_info_e822 + * @pll_freq: Frequency of PLL that drives timer ticks in Hz + * @nominal_incval: increment to generate nanoseconds in GLTSYN_TIME_L + * @pps_delay: propagation delay of the PPS output signal + * + * Characteristic information for the various TIME_REF sources possible in the + * E822 devices + */ +struct ice_time_ref_info_e822 { + u64 pll_freq; + u64 nominal_incval; + u8 pps_delay; +}; + +/** + * struct ice_vernier_info_e822 + * @tx_par_clk: Frequency used to calculate P_REG_PAR_TX_TUS + * @rx_par_clk: Frequency used to calculate P_REG_PAR_RX_TUS + * @tx_pcs_clk: Frequency used to calculate P_REG_PCS_TX_TUS + * @rx_pcs_clk: Frequency used to calculate P_REG_PCS_RX_TUS + * @tx_desk_rsgb_par: Frequency used to calculate P_REG_DESK_PAR_TX_TUS + * @rx_desk_rsgb_par: Frequency used to calculate P_REG_DESK_PAR_RX_TUS + * @tx_desk_rsgb_pcs: Frequency used to calculate P_REG_DESK_PCS_TX_TUS + * @rx_desk_rsgb_pcs: Frequency used to calculate P_REG_DESK_PCS_RX_TUS + * @tx_fixed_delay: Fixed Tx latency measured in 1/100th nanoseconds + * @pmd_adj_divisor: Divisor used to calculate PDM alignment adjustment + * @rx_fixed_delay: Fixed Rx latency measured in 1/100th nanoseconds + * + * Table of constants used during as part of the Vernier calibration of the Tx + * and Rx timestamps. This includes frequency values used to compute TUs per + * PAR/PCS clock cycle, and static delay values measured during hardware + * design. + * + * Note that some values are not used for all link speeds, and the + * P_REG_DESK_PAR* registers may represent different clock markers at + * different link speeds, either the deskew marker for multi-lane link speeds + * or the Reed Solomon gearbox marker for RS-FEC. + */ +struct ice_vernier_info_e822 { + u32 tx_par_clk; + u32 rx_par_clk; + u32 tx_pcs_clk; + u32 rx_pcs_clk; + u32 tx_desk_rsgb_par; + u32 rx_desk_rsgb_par; + u32 tx_desk_rsgb_pcs; + u32 rx_desk_rsgb_pcs; + u32 tx_fixed_delay; + u32 pmd_adj_divisor; + u32 rx_fixed_delay; +}; + +/** + * struct ice_cgu_pll_params_e822 + * @refclk_pre_div: Reference clock pre-divisor + * @feedback_div: Feedback divisor + * @frac_n_div: Fractional divisor + * @post_pll_div: Post PLL divisor + * + * Clock Generation Unit parameters used to program the PLL based on the + * selected TIME_REF frequency. + */ +struct ice_cgu_pll_params_e822 { + u32 refclk_pre_div; + u32 feedback_div; + u32 frac_n_div; + u32 post_pll_div; +}; + +extern const struct +ice_cgu_pll_params_e822 e822_cgu_params[NUM_ICE_TIME_REF_FREQ]; + +/* Table of constants related to possible TIME_REF sources */ +extern const struct ice_time_ref_info_e822 e822_time_ref[NUM_ICE_TIME_REF_FREQ]; + +/* Table of constants for Vernier calibration on E822 */ +extern const struct ice_vernier_info_e822 e822_vernier[NUM_ICE_PTP_LNK_SPD]; + /* Increment value to generate nanoseconds in the GLTSYN_TIME_L register for * the E810 devices. Based off of a PLL with an 812.5 MHz frequency. */ @@ -27,6 +133,59 @@ int ice_ptp_write_incval_locked(struct ice_hw *hw, u64 incval); int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj); int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp); int ice_clear_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx); +int ice_ptp_init_phc(struct ice_hw *hw); + +/* E822 family functions */ +int ice_read_phy_reg_e822(struct ice_hw *hw, u8 port, u16 offset, u32 *val); +int ice_write_phy_reg_e822(struct ice_hw *hw, u8 port, u16 offset, u32 val); +int ice_read_quad_reg_e822(struct ice_hw *hw, u8 quad, u16 offset, u32 *val); +int ice_write_quad_reg_e822(struct ice_hw *hw, u8 quad, u16 offset, u32 val); +int ice_ptp_prep_port_adj_e822(struct ice_hw *hw, u8 port, s64 time); + +/** + * ice_e822_time_ref - Get the current TIME_REF from capabilities + * @hw: pointer to the HW structure + * + * Returns the current TIME_REF from the capabilities structure. + */ +static inline enum ice_time_ref_freq ice_e822_time_ref(struct ice_hw *hw) +{ + return hw->func_caps.ts_func_info.time_ref; +} + +/** + * ice_set_e822_time_ref - Set new TIME_REF + * @hw: pointer to the HW structure + * @time_ref: new TIME_REF to set + * + * Update the TIME_REF in the capabilities structure in response to some + * change, such as an update to the CGU registers. + */ +static inline void +ice_set_e822_time_ref(struct ice_hw *hw, enum ice_time_ref_freq time_ref) +{ + hw->func_caps.ts_func_info.time_ref = time_ref; +} + +static inline u64 ice_e822_pll_freq(enum ice_time_ref_freq time_ref) +{ + return e822_time_ref[time_ref].pll_freq; +} + +static inline u64 ice_e822_nominal_incval(enum ice_time_ref_freq time_ref) +{ + return e822_time_ref[time_ref].nominal_incval; +} + +static inline u64 ice_e822_pps_delay(enum ice_time_ref_freq time_ref) +{ + return e822_time_ref[time_ref].pps_delay; +} + +/* E822 Vernier calibration functions */ +int ice_stop_phy_timer_e822(struct ice_hw *hw, u8 port, bool soft_reset); +int ice_start_phy_timer_e822(struct ice_hw *hw, u8 port, bool bypass); +int ice_phy_exit_bypass_e822(struct ice_hw *hw, u8 port); /* E810 family functions */ int ice_ptp_init_phy_e810(struct ice_hw *hw); @@ -36,19 +195,194 @@ bool ice_is_pca9575_present(struct ice_hw *hw); #define PFTSYN_SEM_BYTES 4 +#define ICE_PTP_CLOCK_INDEX_0 0x00 +#define ICE_PTP_CLOCK_INDEX_1 0x01 + /* PHY timer commands */ #define SEL_CPK_SRC 8 +#define SEL_PHY_SRC 3 /* Time Sync command Definitions */ #define GLTSYN_CMD_INIT_TIME BIT(0) #define GLTSYN_CMD_INIT_INCVAL BIT(1) +#define GLTSYN_CMD_INIT_TIME_INCVAL (BIT(0) | BIT(1)) #define GLTSYN_CMD_ADJ_TIME BIT(2) #define GLTSYN_CMD_ADJ_INIT_TIME (BIT(2) | BIT(3)) #define GLTSYN_CMD_READ_TIME BIT(7) +/* PHY port Time Sync command definitions */ +#define PHY_CMD_INIT_TIME BIT(0) +#define PHY_CMD_INIT_INCVAL BIT(1) +#define PHY_CMD_ADJ_TIME (BIT(0) | BIT(1)) +#define PHY_CMD_ADJ_TIME_AT_TIME (BIT(0) | BIT(2)) +#define PHY_CMD_READ_TIME (BIT(0) | BIT(1) | BIT(2)) + #define TS_CMD_MASK_E810 0xFF +#define TS_CMD_MASK 0xF #define SYNC_EXEC_CMD 0x3 +/* Macros to derive port low and high addresses on both quads */ +#define P_Q0_L(a, p) ((((a) + (0x2000 * (p)))) & 0xFFFF) +#define P_Q0_H(a, p) ((((a) + (0x2000 * (p)))) >> 16) +#define P_Q1_L(a, p) ((((a) - (0x2000 * ((p) - ICE_PORTS_PER_QUAD)))) & 0xFFFF) +#define P_Q1_H(a, p) ((((a) - (0x2000 * ((p) - ICE_PORTS_PER_QUAD)))) >> 16) + +/* PHY QUAD register base addresses */ +#define Q_0_BASE 0x94000 +#define Q_1_BASE 0x114000 + +/* Timestamp memory reset registers */ +#define Q_REG_TS_CTRL 0x618 +#define Q_REG_TS_CTRL_S 0 +#define Q_REG_TS_CTRL_M BIT(0) + +/* Timestamp availability status registers */ +#define Q_REG_TX_MEMORY_STATUS_L 0xCF0 +#define Q_REG_TX_MEMORY_STATUS_U 0xCF4 + +/* Tx FIFO status registers */ +#define Q_REG_FIFO23_STATUS 0xCF8 +#define Q_REG_FIFO01_STATUS 0xCFC +#define Q_REG_FIFO02_S 0 +#define Q_REG_FIFO02_M ICE_M(0x3FF, 0) +#define Q_REG_FIFO13_S 10 +#define Q_REG_FIFO13_M ICE_M(0x3FF, 10) + +/* Interrupt control Config registers */ +#define Q_REG_TX_MEM_GBL_CFG 0xC08 +#define Q_REG_TX_MEM_GBL_CFG_LANE_TYPE_S 0 +#define Q_REG_TX_MEM_GBL_CFG_LANE_TYPE_M BIT(0) +#define Q_REG_TX_MEM_GBL_CFG_TX_TYPE_S 1 +#define Q_REG_TX_MEM_GBL_CFG_TX_TYPE_M ICE_M(0xFF, 1) +#define Q_REG_TX_MEM_GBL_CFG_INTR_THR_S 9 +#define Q_REG_TX_MEM_GBL_CFG_INTR_THR_M ICE_M(0x3F, 9) +#define Q_REG_TX_MEM_GBL_CFG_INTR_ENA_S 15 +#define Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M BIT(15) + +/* Tx Timestamp data registers */ +#define Q_REG_TX_MEMORY_BANK_START 0xA00 + +/* PHY port register base addresses */ +#define P_0_BASE 0x80000 +#define P_4_BASE 0x106000 + +/* Timestamp init registers */ +#define P_REG_RX_TIMER_INC_PRE_L 0x46C +#define P_REG_RX_TIMER_INC_PRE_U 0x470 +#define P_REG_TX_TIMER_INC_PRE_L 0x44C +#define P_REG_TX_TIMER_INC_PRE_U 0x450 + +/* Timestamp match and adjust target registers */ +#define P_REG_RX_TIMER_CNT_ADJ_L 0x474 +#define P_REG_RX_TIMER_CNT_ADJ_U 0x478 +#define P_REG_TX_TIMER_CNT_ADJ_L 0x454 +#define P_REG_TX_TIMER_CNT_ADJ_U 0x458 + +/* Timestamp capture registers */ +#define P_REG_RX_CAPTURE_L 0x4D8 +#define P_REG_RX_CAPTURE_U 0x4DC +#define P_REG_TX_CAPTURE_L 0x4B4 +#define P_REG_TX_CAPTURE_U 0x4B8 + +/* Timestamp PHY incval registers */ +#define P_REG_TIMETUS_L 0x410 +#define P_REG_TIMETUS_U 0x414 + +#define P_REG_40B_LOW_M 0xFF +#define P_REG_40B_HIGH_S 8 + +/* PHY window length registers */ +#define P_REG_WL 0x40C + +#define PTP_VERNIER_WL 0x111ed + +/* PHY start registers */ +#define P_REG_PS 0x408 +#define P_REG_PS_START_S 0 +#define P_REG_PS_START_M BIT(0) +#define P_REG_PS_BYPASS_MODE_S 1 +#define P_REG_PS_BYPASS_MODE_M BIT(1) +#define P_REG_PS_ENA_CLK_S 2 +#define P_REG_PS_ENA_CLK_M BIT(2) +#define P_REG_PS_LOAD_OFFSET_S 3 +#define P_REG_PS_LOAD_OFFSET_M BIT(3) +#define P_REG_PS_SFT_RESET_S 11 +#define P_REG_PS_SFT_RESET_M BIT(11) + +/* PHY offset valid registers */ +#define P_REG_TX_OV_STATUS 0x4D4 +#define P_REG_TX_OV_STATUS_OV_S 0 +#define P_REG_TX_OV_STATUS_OV_M BIT(0) +#define P_REG_RX_OV_STATUS 0x4F8 +#define P_REG_RX_OV_STATUS_OV_S 0 +#define P_REG_RX_OV_STATUS_OV_M BIT(0) + +/* PHY offset ready registers */ +#define P_REG_TX_OR 0x45C +#define P_REG_RX_OR 0x47C + +/* PHY total offset registers */ +#define P_REG_TOTAL_RX_OFFSET_L 0x460 +#define P_REG_TOTAL_RX_OFFSET_U 0x464 +#define P_REG_TOTAL_TX_OFFSET_L 0x440 +#define P_REG_TOTAL_TX_OFFSET_U 0x444 + +/* Timestamp PAR/PCS registers */ +#define P_REG_UIX66_10G_40G_L 0x480 +#define P_REG_UIX66_10G_40G_U 0x484 +#define P_REG_UIX66_25G_100G_L 0x488 +#define P_REG_UIX66_25G_100G_U 0x48C +#define P_REG_DESK_PAR_RX_TUS_L 0x490 +#define P_REG_DESK_PAR_RX_TUS_U 0x494 +#define P_REG_DESK_PAR_TX_TUS_L 0x498 +#define P_REG_DESK_PAR_TX_TUS_U 0x49C +#define P_REG_DESK_PCS_RX_TUS_L 0x4A0 +#define P_REG_DESK_PCS_RX_TUS_U 0x4A4 +#define P_REG_DESK_PCS_TX_TUS_L 0x4A8 +#define P_REG_DESK_PCS_TX_TUS_U 0x4AC +#define P_REG_PAR_RX_TUS_L 0x420 +#define P_REG_PAR_RX_TUS_U 0x424 +#define P_REG_PAR_TX_TUS_L 0x428 +#define P_REG_PAR_TX_TUS_U 0x42C +#define P_REG_PCS_RX_TUS_L 0x430 +#define P_REG_PCS_RX_TUS_U 0x434 +#define P_REG_PCS_TX_TUS_L 0x438 +#define P_REG_PCS_TX_TUS_U 0x43C +#define P_REG_PAR_RX_TIME_L 0x4F0 +#define P_REG_PAR_RX_TIME_U 0x4F4 +#define P_REG_PAR_TX_TIME_L 0x4CC +#define P_REG_PAR_TX_TIME_U 0x4D0 +#define P_REG_PAR_PCS_RX_OFFSET_L 0x4E8 +#define P_REG_PAR_PCS_RX_OFFSET_U 0x4EC +#define P_REG_PAR_PCS_TX_OFFSET_L 0x4C4 +#define P_REG_PAR_PCS_TX_OFFSET_U 0x4C8 +#define P_REG_LINK_SPEED 0x4FC +#define P_REG_LINK_SPEED_SERDES_S 0 +#define P_REG_LINK_SPEED_SERDES_M ICE_M(0x7, 0) +#define P_REG_LINK_SPEED_FEC_MODE_S 3 +#define P_REG_LINK_SPEED_FEC_MODE_M ICE_M(0x3, 3) +#define P_REG_LINK_SPEED_FEC_MODE(reg) \ + (((reg) & P_REG_LINK_SPEED_FEC_MODE_M) >> \ + P_REG_LINK_SPEED_FEC_MODE_S) + +/* PHY timestamp related registers */ +#define P_REG_PMD_ALIGNMENT 0x0FC +#define P_REG_RX_80_TO_160_CNT 0x6FC +#define P_REG_RX_80_TO_160_CNT_RXCYC_S 0 +#define P_REG_RX_80_TO_160_CNT_RXCYC_M BIT(0) +#define P_REG_RX_40_TO_160_CNT 0x8FC +#define P_REG_RX_40_TO_160_CNT_RXCYC_S 0 +#define P_REG_RX_40_TO_160_CNT_RXCYC_M ICE_M(0x3, 0) + +/* Rx FIFO status registers */ +#define P_REG_RX_OV_FS 0x4F8 +#define P_REG_RX_OV_FS_FIFO_STATUS_S 2 +#define P_REG_RX_OV_FS_FIFO_STATUS_M ICE_M(0x3FF, 2) + +/* Timestamp command registers */ +#define P_REG_TX_TMR_CMD 0x448 +#define P_REG_RX_TMR_CMD 0x468 + /* E810 timesync enable register */ #define ETH_GLTSYN_ENA(_i) (0x03000348 + ((_i) * 4)) @@ -68,9 +402,20 @@ bool ice_is_pca9575_present(struct ice_hw *hw); /* Timestamp block macros */ #define TS_LOW_M 0xFFFFFFFF +#define TS_HIGH_M 0xFF #define TS_HIGH_S 32 +#define TS_PHY_LOW_M 0xFF +#define TS_PHY_HIGH_M 0xFFFFFFFF +#define TS_PHY_HIGH_S 8 + #define BYTES_PER_IDX_ADDR_L_U 8 +#define BYTES_PER_IDX_ADDR_L 4 + +/* Internal PHY timestamp address */ +#define TS_L(a, idx) ((a) + ((idx) * BYTES_PER_IDX_ADDR_L_U)) +#define TS_H(a, idx) ((a) + ((idx) * BYTES_PER_IDX_ADDR_L_U + \ + BYTES_PER_IDX_ADDR_L)) /* External PHY timestamp address */ #define TS_EXT(a, port, idx) ((a) + (0x1000 * (port)) + \ diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 58b1907e3ff1..caf0a02b25f5 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -298,9 +298,30 @@ struct ice_hw_common_caps { #define ICE_TS_TMR_IDX_ASSOC_S 24 #define ICE_TS_TMR_IDX_ASSOC_M BIT(24) +/* TIME_REF clock rate specification */ +enum ice_time_ref_freq { + ICE_TIME_REF_FREQ_25_000 = 0, + ICE_TIME_REF_FREQ_122_880 = 1, + ICE_TIME_REF_FREQ_125_000 = 2, + ICE_TIME_REF_FREQ_153_600 = 3, + ICE_TIME_REF_FREQ_156_250 = 4, + ICE_TIME_REF_FREQ_245_760 = 5, + + NUM_ICE_TIME_REF_FREQ +}; + +/* Clock source specification */ +enum ice_clk_src { + ICE_CLK_SRC_TCX0 = 0, /* Temperature compensated oscillator */ + ICE_CLK_SRC_TIME_REF = 1, /* Use TIME_REF reference clock */ + + NUM_ICE_CLK_SRC +}; + struct ice_ts_func_info { /* Function specific info */ - u32 clk_freq; + enum ice_time_ref_freq time_ref; + u8 clk_freq; u8 clk_src; u8 tmr_index_assoc; u8 ena; diff --git a/drivers/net/ethernet/intel/igb/e1000_i210.c b/drivers/net/ethernet/intel/igb/e1000_i210.c index 9265901455cd..b9b9d35494d2 100644 --- a/drivers/net/ethernet/intel/igb/e1000_i210.c +++ b/drivers/net/ethernet/intel/igb/e1000_i210.c @@ -792,7 +792,6 @@ s32 igb_write_xmdio_reg(struct e1000_hw *hw, u16 addr, u8 dev_addr, u16 data) **/ s32 igb_init_nvm_params_i210(struct e1000_hw *hw) { - s32 ret_val = 0; struct e1000_nvm_info *nvm = &hw->nvm; nvm->ops.acquire = igb_acquire_nvm_i210; @@ -813,7 +812,7 @@ s32 igb_init_nvm_params_i210(struct e1000_hw *hw) nvm->ops.validate = NULL; nvm->ops.update = NULL; } - return ret_val; + return 0; } /** diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c index 4d988da68394..b78407289741 100644 --- a/drivers/net/ethernet/intel/igbvf/netdev.c +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -1520,7 +1520,7 @@ static void igbvf_reset(struct igbvf_adapter *adapter) /* Allow time for pending master requests to run */ if (mac->ops.reset_hw(hw)) - dev_warn(&adapter->pdev->dev, "PF still resetting\n"); + dev_info(&adapter->pdev->dev, "PF still resetting\n"); mac->ops.init_hw(hw); diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h index c7fe61509d5b..5c66b97c0cfa 100644 --- a/drivers/net/ethernet/intel/igc/igc_defines.h +++ b/drivers/net/ethernet/intel/igc/igc_defines.h @@ -85,9 +85,6 @@ #define IGC_WUFC_EXT_FILTER_MASK GENMASK(31, 8) -/* Physical Func Reset Done Indication */ -#define IGC_CTRL_EXT_LINK_MODE_MASK 0x00C00000 - /* Loop limit on how long we wait for auto-negotiation to complete */ #define COPPER_LINK_UP_LIMIT 10 #define PHY_AUTO_NEG_LIMIT 45 @@ -584,7 +581,6 @@ #define IGC_GEN_POLL_TIMEOUT 1920 /* PHY Control Register */ -#define MII_CR_FULL_DUPLEX 0x0100 /* FDX =1, half duplex =0 */ #define MII_CR_RESTART_AUTO_NEG 0x0200 /* Restart auto negotiation */ #define MII_CR_POWER_DOWN 0x0800 /* Power down */ #define MII_CR_AUTO_NEG_EN 0x1000 /* Auto Neg Enable */ @@ -605,9 +601,6 @@ #define PHY_1000T_CTRL 0x09 /* 1000Base-T Control Reg */ #define PHY_1000T_STATUS 0x0A /* 1000Base-T Status Reg */ -/* Bit definitions for valid PHY IDs. I = Integrated E = External */ -#define I225_I_PHY_ID 0x67C9DC00 - /* MDI Control */ #define IGC_MDIC_DATA_MASK 0x0000FFFF #define IGC_MDIC_REG_MASK 0x001F0000 diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h index 587db7483f25..b1e72ec5f131 100644 --- a/drivers/net/ethernet/intel/igc/igc_hw.h +++ b/drivers/net/ethernet/intel/igc/igc_hw.h @@ -55,7 +55,6 @@ enum igc_mac_type { enum igc_phy_type { igc_phy_unknown = 0, - igc_phy_none, igc_phy_i225, }; @@ -68,8 +67,6 @@ enum igc_media_type { enum igc_nvm_type { igc_nvm_unknown = 0, igc_nvm_eeprom_spi, - igc_nvm_flash_hw, - igc_nvm_invm, }; struct igc_info { diff --git a/drivers/net/ethernet/intel/igc/igc_i225.c b/drivers/net/ethernet/intel/igc/igc_i225.c index b6807e16eea9..66ea566488d1 100644 --- a/drivers/net/ethernet/intel/igc/igc_i225.c +++ b/drivers/net/ethernet/intel/igc/igc_i225.c @@ -473,13 +473,11 @@ s32 igc_init_nvm_params_i225(struct igc_hw *hw) /* NVM Function Pointers */ if (igc_get_flash_presence_i225(hw)) { - hw->nvm.type = igc_nvm_flash_hw; nvm->ops.read = igc_read_nvm_srrd_i225; nvm->ops.write = igc_write_nvm_srwr_i225; nvm->ops.validate = igc_validate_nvm_checksum_i225; nvm->ops.update = igc_update_nvm_checksum_i225; } else { - hw->nvm.type = igc_nvm_invm; nvm->ops.read = igc_read_nvm_eerd; nvm->ops.write = NULL; nvm->ops.validate = NULL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index 1c98652b244a..d1093bb2d436 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -546,6 +546,13 @@ static int mlx5_devlink_enable_remote_dev_reset_get(struct devlink *devlink, u32 return 0; } +static int mlx5_devlink_eq_depth_validate(struct devlink *devlink, u32 id, + union devlink_param_value val, + struct netlink_ext_ack *extack) +{ + return (val.vu16 >= 64 && val.vu16 <= 4096) ? 0 : -EINVAL; +} + static const struct devlink_param mlx5_devlink_params[] = { DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_FLOW_STEERING_MODE, "flow_steering_mode", DEVLINK_PARAM_TYPE_STRING, @@ -570,6 +577,10 @@ static const struct devlink_param mlx5_devlink_params[] = { DEVLINK_PARAM_GENERIC(ENABLE_REMOTE_DEV_RESET, BIT(DEVLINK_PARAM_CMODE_RUNTIME), mlx5_devlink_enable_remote_dev_reset_get, mlx5_devlink_enable_remote_dev_reset_set, NULL), + DEVLINK_PARAM_GENERIC(IO_EQ_SIZE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), + NULL, NULL, mlx5_devlink_eq_depth_validate), + DEVLINK_PARAM_GENERIC(EVENT_EQ_SIZE, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), + NULL, NULL, mlx5_devlink_eq_depth_validate), }; static void mlx5_devlink_set_params_init_values(struct devlink *devlink) @@ -608,6 +619,16 @@ static void mlx5_devlink_set_params_init_values(struct devlink *devlink) value); } #endif + + value.vu32 = MLX5_COMP_EQ_SIZE; + devlink_param_driverinit_value_set(devlink, + DEVLINK_PARAM_GENERIC_ID_IO_EQ_SIZE, + value); + + value.vu32 = MLX5_NUM_ASYNC_EQE; + devlink_param_driverinit_value_set(devlink, + DEVLINK_PARAM_GENERIC_ID_EVENT_EQ_SIZE, + value); } static const struct devlink_param enable_eth_param = @@ -752,6 +773,66 @@ static void mlx5_devlink_auxdev_params_unregister(struct devlink *devlink) mlx5_devlink_eth_param_unregister(devlink); } +static int mlx5_devlink_max_uc_list_validate(struct devlink *devlink, u32 id, + union devlink_param_value val, + struct netlink_ext_ack *extack) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + + if (val.vu32 == 0) { + NL_SET_ERR_MSG_MOD(extack, "max_macs value must be greater than 0"); + return -EINVAL; + } + + if (!is_power_of_2(val.vu32)) { + NL_SET_ERR_MSG_MOD(extack, "Only power of 2 values are supported for max_macs"); + return -EINVAL; + } + + if (ilog2(val.vu32) > + MLX5_CAP_GEN_MAX(dev, log_max_current_uc_list)) { + NL_SET_ERR_MSG_MOD(extack, "max_macs value is out of the supported range"); + return -EINVAL; + } + + return 0; +} + +static const struct devlink_param max_uc_list_param = + DEVLINK_PARAM_GENERIC(MAX_MACS, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), + NULL, NULL, mlx5_devlink_max_uc_list_validate); + +static int mlx5_devlink_max_uc_list_param_register(struct devlink *devlink) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + union devlink_param_value value; + int err; + + if (!MLX5_CAP_GEN_MAX(dev, log_max_current_uc_list_wr_supported)) + return 0; + + err = devlink_param_register(devlink, &max_uc_list_param); + if (err) + return err; + + value.vu32 = 1 << MLX5_CAP_GEN(dev, log_max_current_uc_list); + devlink_param_driverinit_value_set(devlink, + DEVLINK_PARAM_GENERIC_ID_MAX_MACS, + value); + return 0; +} + +static void +mlx5_devlink_max_uc_list_param_unregister(struct devlink *devlink) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + + if (!MLX5_CAP_GEN_MAX(dev, log_max_current_uc_list_wr_supported)) + return; + + devlink_param_unregister(devlink, &max_uc_list_param); +} + #define MLX5_TRAP_DROP(_id, _group_id) \ DEVLINK_TRAP_GENERIC(DROP, DROP, _id, \ DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ @@ -811,6 +892,10 @@ int mlx5_devlink_register(struct devlink *devlink) if (err) goto auxdev_reg_err; + err = mlx5_devlink_max_uc_list_param_register(devlink); + if (err) + goto max_uc_list_err; + err = mlx5_devlink_traps_register(devlink); if (err) goto traps_reg_err; @@ -821,6 +906,8 @@ int mlx5_devlink_register(struct devlink *devlink) return 0; traps_reg_err: + mlx5_devlink_max_uc_list_param_unregister(devlink); +max_uc_list_err: mlx5_devlink_auxdev_params_unregister(devlink); auxdev_reg_err: devlink_params_unregister(devlink, mlx5_devlink_params, @@ -831,6 +918,7 @@ auxdev_reg_err: void mlx5_devlink_unregister(struct devlink *devlink) { mlx5_devlink_traps_unregister(devlink); + mlx5_devlink_max_uc_list_param_unregister(devlink); mlx5_devlink_auxdev_params_unregister(devlink); devlink_params_unregister(devlink, mlx5_devlink_params, ARRAY_SIZE(mlx5_devlink_params)); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index e77c4159713f..33679467c63a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -145,7 +145,6 @@ struct page_pool; #define MLX5E_MIN_NUM_CHANNELS 0x1 #define MLX5E_MAX_NUM_CHANNELS (MLX5E_INDIR_RQT_SIZE / 2) -#define MLX5E_MAX_NUM_SQS (MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC) #define MLX5E_TX_CQ_POLL_BUDGET 128 #define MLX5E_TX_XSK_POLL_BUDGET 64 #define MLX5E_SQ_RECOVER_MIN_INTERVAL 500 /* msecs */ @@ -875,10 +874,8 @@ struct mlx5e_trap; struct mlx5e_priv { /* priv data path fields - start */ - /* +1 for port ptp ts */ - struct mlx5e_txqsq *txq2sq[(MLX5E_MAX_NUM_CHANNELS + 1) * MLX5E_MAX_NUM_TC + - MLX5E_QOS_MAX_LEAF_NODES]; - int channel_tc2realtxq[MLX5E_MAX_NUM_CHANNELS][MLX5E_MAX_NUM_TC]; + struct mlx5e_txqsq **txq2sq; + int **channel_tc2realtxq; int port_ptp_tc2realtxq[MLX5E_MAX_NUM_TC]; #ifdef CONFIG_MLX5_CORE_EN_DCB struct mlx5e_dcbx_dp dcbx_dp; @@ -893,7 +890,7 @@ struct mlx5e_priv { struct mlx5e_channels channels; u32 tisn[MLX5_MAX_PORTS][MLX5E_MAX_NUM_TC]; struct mlx5e_rx_res *rx_res; - u32 tx_rates[MLX5E_MAX_NUM_SQS]; + u32 *tx_rates; struct mlx5e_flow_steering fs; @@ -909,7 +906,7 @@ struct mlx5e_priv { struct net_device *netdev; struct mlx5e_trap *en_trap; struct mlx5e_stats stats; - struct mlx5e_channel_stats channel_stats[MLX5E_MAX_NUM_CHANNELS]; + struct mlx5e_channel_stats **channel_stats; struct mlx5e_channel_stats trap_stats; struct mlx5e_ptp_stats ptp_stats; u16 stats_nch; @@ -956,6 +953,12 @@ struct mlx5e_rx_handlers { extern const struct mlx5e_rx_handlers mlx5e_rx_handlers_nic; +enum mlx5e_profile_feature { + MLX5E_PROFILE_FEATURE_PTP_RX, + MLX5E_PROFILE_FEATURE_PTP_TX, + MLX5E_PROFILE_FEATURE_QOS_HTB, +}; + struct mlx5e_profile { int (*init)(struct mlx5_core_dev *mdev, struct net_device *netdev); @@ -969,14 +972,18 @@ struct mlx5e_profile { int (*update_rx)(struct mlx5e_priv *priv); void (*update_stats)(struct mlx5e_priv *priv); void (*update_carrier)(struct mlx5e_priv *priv); + int (*max_nch_limit)(struct mlx5_core_dev *mdev); unsigned int (*stats_grps_num)(struct mlx5e_priv *priv); mlx5e_stats_grp_t *stats_grps; const struct mlx5e_rx_handlers *rx_handlers; int max_tc; u8 rq_groups; - bool rx_ptp_support; + u32 features; }; +#define mlx5e_profile_feature_cap(profile, feature) \ + ((profile)->features & (MLX5E_PROFILE_FEATURE_## feature)) + void mlx5e_build_ptys2ethtool_map(void); bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev); @@ -1188,8 +1195,7 @@ int mlx5e_priv_init(struct mlx5e_priv *priv, struct mlx5_core_dev *mdev); void mlx5e_priv_cleanup(struct mlx5e_priv *priv); struct net_device * -mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *profile, - unsigned int txqs, unsigned int rxqs); +mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *profile); int mlx5e_attach_netdev(struct mlx5e_priv *priv); void mlx5e_detach_netdev(struct mlx5e_priv *priv); void mlx5e_destroy_netdev(struct mlx5e_priv *priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/hv_vhca_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en/hv_vhca_stats.c index d290d7276b8d..074ffa4fa5af 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/hv_vhca_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/hv_vhca_stats.c @@ -20,7 +20,7 @@ mlx5e_hv_vhca_fill_ring_stats(struct mlx5e_priv *priv, int ch, struct mlx5e_channel_stats *stats; int tc; - stats = &priv->channel_stats[ch]; + stats = priv->channel_stats[ch]; data->rx_packets = stats->rq.packets; data->rx_bytes = stats->rq.bytes; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index 18d542b1c5cb..82baafd3c00c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -768,7 +768,7 @@ int mlx5e_ptp_alloc_rx_fs(struct mlx5e_priv *priv) { struct mlx5e_ptp_fs *ptp_fs; - if (!priv->profile->rx_ptp_support) + if (!mlx5e_profile_feature_cap(priv->profile, PTP_RX)) return 0; ptp_fs = kzalloc(sizeof(*ptp_fs), GFP_KERNEL); @@ -783,7 +783,7 @@ void mlx5e_ptp_free_rx_fs(struct mlx5e_priv *priv) { struct mlx5e_ptp_fs *ptp_fs = priv->fs.ptp_fs; - if (!priv->profile->rx_ptp_support) + if (!mlx5e_profile_feature_cap(priv->profile, PTP_RX)) return; mlx5e_ptp_rx_unset_fs(priv); @@ -794,7 +794,7 @@ int mlx5e_ptp_rx_manage_fs(struct mlx5e_priv *priv, bool set) { struct mlx5e_ptp *c = priv->channels.ptp; - if (!priv->profile->rx_ptp_support) + if (!mlx5e_profile_feature_cap(priv->profile, PTP_RX)) return 0; if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c index fcb0892c08a9..0991345c4ae5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c @@ -517,6 +517,9 @@ int mlx5e_rep_indr_setup_cb(struct net_device *netdev, struct Qdisc *sch, void * void *data, void (*cleanup)(struct flow_block_cb *block_cb)) { + if (!netdev) + return -EOPNOTSUPP; + switch (type) { case TC_SETUP_BLOCK: return mlx5e_rep_indr_setup_block(netdev, sch, cb_priv, type_data, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c index 0015a81eb9a1..24c32f73040a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c @@ -37,7 +37,6 @@ struct mlx5e_rx_res { /* API for rx_res_rss_* */ static int mlx5e_rx_res_rss_init_def(struct mlx5e_rx_res *res, - const struct mlx5e_packet_merge_param *init_pkt_merge_param, unsigned int init_nch) { bool inner_ft_support = res->features & MLX5E_RX_RES_FEATURE_INNER_FT; @@ -52,7 +51,7 @@ static int mlx5e_rx_res_rss_init_def(struct mlx5e_rx_res *res, return -ENOMEM; err = mlx5e_rss_init(rss, res->mdev, inner_ft_support, res->drop_rqn, - init_pkt_merge_param); + &res->pkt_merge_param); if (err) goto err_rss_free; @@ -277,8 +276,7 @@ struct mlx5e_rx_res *mlx5e_rx_res_alloc(void) return kvzalloc(sizeof(struct mlx5e_rx_res), GFP_KERNEL); } -static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res, - const struct mlx5e_packet_merge_param *init_pkt_merge_param) +static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res) { bool inner_ft_support = res->features & MLX5E_RX_RES_FEATURE_INNER_FT; struct mlx5e_tir_builder *builder; @@ -309,7 +307,7 @@ static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res, mlx5e_tir_builder_build_rqt(builder, res->mdev->mlx5e_res.hw_objs.td.tdn, mlx5e_rqt_get_rqtn(&res->channels[ix].direct_rqt), inner_ft_support); - mlx5e_tir_builder_build_packet_merge(builder, init_pkt_merge_param); + mlx5e_tir_builder_build_packet_merge(builder, &res->pkt_merge_param); mlx5e_tir_builder_build_direct(builder); err = mlx5e_tir_init(&res->channels[ix].direct_tir, builder, res->mdev, true); @@ -339,7 +337,7 @@ static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res, mlx5e_tir_builder_build_rqt(builder, res->mdev->mlx5e_res.hw_objs.td.tdn, mlx5e_rqt_get_rqtn(&res->channels[ix].xsk_rqt), inner_ft_support); - mlx5e_tir_builder_build_packet_merge(builder, init_pkt_merge_param); + mlx5e_tir_builder_build_packet_merge(builder, &res->pkt_merge_param); mlx5e_tir_builder_build_direct(builder); err = mlx5e_tir_init(&res->channels[ix].xsk_tir, builder, res->mdev, true); @@ -454,11 +452,11 @@ int mlx5e_rx_res_init(struct mlx5e_rx_res *res, struct mlx5_core_dev *mdev, res->pkt_merge_param = *init_pkt_merge_param; init_rwsem(&res->pkt_merge_param_sem); - err = mlx5e_rx_res_rss_init_def(res, init_pkt_merge_param, init_nch); + err = mlx5e_rx_res_rss_init_def(res, init_nch); if (err) goto err_out; - err = mlx5e_rx_res_channels_init(res, init_pkt_merge_param); + err = mlx5e_rx_res_channels_init(res); if (err) goto err_rss_destroy; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c index 538bc2419bd8..5f2b67b9c189 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c @@ -67,7 +67,7 @@ static int mlx5e_init_xsk_rq(struct mlx5e_channel *c, rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->xdpsq = &c->rq_xdpsq; rq->xsk_pool = pool; - rq->stats = &c->priv->channel_stats[c->ix].xskrq; + rq->stats = &c->priv->channel_stats[c->ix]->xskrq; rq->ptp_cyc2time = mlx5_rq_ts_translator(mdev); rq_xdp_ix = c->ix + params->num_channels * MLX5E_RQ_GROUP_XSK; err = mlx5e_rq_set_handlers(rq, params, xsk); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c index 15711814d2d2..96064a2033f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c @@ -611,7 +611,7 @@ int mlx5e_ktls_add_rx(struct net_device *netdev, struct sock *sk, priv_rx->rxq = rxq; priv_rx->sk = sk; - priv_rx->rq_stats = &priv->channel_stats[rxq].rq; + priv_rx->rq_stats = &priv->channel_stats[rxq]->rq; priv_rx->sw_stats = &priv->tls->sw_stats; mlx5e_set_ktls_rx_priv_ctx(tls_ctx, priv_rx); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index fe5d82fa6e92..49cca6bd49a1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -556,7 +556,7 @@ static struct mlx5_flow_handle *arfs_add_rule(struct mlx5e_priv *priv, rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1); if (IS_ERR(rule)) { err = PTR_ERR(rule); - priv->channel_stats[arfs_rule->rxq].rq.arfs_err++; + priv->channel_stats[arfs_rule->rxq]->rq.arfs_err++; mlx5e_dbg(HW, priv, "%s: add rule(filter id=%d, rq idx=%d, ip proto=0x%x) failed,err=%d\n", __func__, arfs_rule->filter_id, arfs_rule->rxq, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index c8757c5a812b..536fcb2c5e90 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -1934,7 +1934,7 @@ int mlx5e_modify_rx_cqe_compression_locked(struct mlx5e_priv *priv, bool new_val if (curr_val == new_val) return 0; - if (new_val && !priv->profile->rx_ptp_support && rx_filter) { + if (new_val && !mlx5e_profile_feature_cap(priv->profile, PTP_RX) && rx_filter) { netdev_err(priv->netdev, "Profile doesn't support enabling of CQE compression while hardware time-stamping is enabled.\n"); return -EINVAL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 496977e7406e..1a47108805fd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -479,7 +479,7 @@ static int mlx5e_init_rxq_rq(struct mlx5e_channel *c, struct mlx5e_params *param rq->mdev = mdev; rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->xdpsq = &c->rq_xdpsq; - rq->stats = &c->priv->channel_stats[c->ix].rq; + rq->stats = &c->priv->channel_stats[c->ix]->rq; rq->ptp_cyc2time = mlx5_rq_ts_translator(mdev); err = mlx5e_rq_set_handlers(rq, params, NULL); if (err) @@ -1161,10 +1161,10 @@ static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c, sq->xsk_pool = xsk_pool; sq->stats = sq->xsk_pool ? - &c->priv->channel_stats[c->ix].xsksq : + &c->priv->channel_stats[c->ix]->xsksq : is_redirect ? - &c->priv->channel_stats[c->ix].xdpsq : - &c->priv->channel_stats[c->ix].rq_xdpsq; + &c->priv->channel_stats[c->ix]->xdpsq : + &c->priv->channel_stats[c->ix]->rq_xdpsq; param->wq.db_numa_node = cpu_to_node(c->cpu); err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, wq, &sq->wq_ctrl); @@ -1928,7 +1928,7 @@ static int mlx5e_open_sqs(struct mlx5e_channel *c, err = mlx5e_open_txqsq(c, c->priv->tisn[c->lag_port][tc], txq_ix, params, &cparam->txq_sq, &c->sq[tc], tc, qos_queue_group_id, - &c->priv->channel_stats[c->ix].sq[tc]); + &c->priv->channel_stats[c->ix]->sq[tc]); if (err) goto err_close_sqs; } @@ -2176,6 +2176,30 @@ static u8 mlx5e_enumerate_lag_port(struct mlx5_core_dev *mdev, int ix) return (ix + port_aff_bias) % mlx5e_get_num_lag_ports(mdev); } +static int mlx5e_channel_stats_alloc(struct mlx5e_priv *priv, int ix, int cpu) +{ + if (ix > priv->stats_nch) { + netdev_warn(priv->netdev, "Unexpected channel stats index %d > %d\n", ix, + priv->stats_nch); + return -EINVAL; + } + + if (priv->channel_stats[ix]) + return 0; + + /* Asymmetric dynamic memory allocation. + * Freed in mlx5e_priv_arrays_free, not on channel closure. + */ + mlx5e_dbg(DRV, priv, "Creating channel stats %d\n", ix); + priv->channel_stats[ix] = kvzalloc_node(sizeof(**priv->channel_stats), + GFP_KERNEL, cpu_to_node(cpu)); + if (!priv->channel_stats[ix]) + return -ENOMEM; + priv->stats_nch++; + + return 0; +} + static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_params *params, struct mlx5e_channel_param *cparam, @@ -2193,6 +2217,10 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, if (err) return err; + err = mlx5e_channel_stats_alloc(priv, ix, cpu); + if (err) + return err; + c = kvzalloc_node(sizeof(*c), GFP_KERNEL, cpu_to_node(cpu)); if (!c) return -ENOMEM; @@ -2207,7 +2235,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.hw_objs.mkey); c->num_tc = mlx5e_get_dcb_num_tc(params); c->xdp = !!params->xdp_prog; - c->stats = &priv->channel_stats[ix].ch; + c->stats = &priv->channel_stats[ix]->ch; c->aff_mask = irq_get_effective_affinity_mask(irq); c->lag_port = mlx5e_enumerate_lag_port(priv->mdev, ix); @@ -3371,7 +3399,7 @@ void mlx5e_fold_sw_stats64(struct mlx5e_priv *priv, struct rtnl_link_stats64 *s) int i; for (i = 0; i < priv->stats_nch; i++) { - struct mlx5e_channel_stats *channel_stats = &priv->channel_stats[i]; + struct mlx5e_channel_stats *channel_stats = priv->channel_stats[i]; struct mlx5e_rq_stats *xskrq_stats = &channel_stats->xskrq; struct mlx5e_rq_stats *rq_stats = &channel_stats->rq; int j; @@ -4038,7 +4066,7 @@ int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) goto err_unlock; } - if (!priv->profile->rx_ptp_support) + if (!mlx5e_profile_feature_cap(priv->profile, PTP_RX)) err = mlx5e_hwstamp_config_no_ptp_rx(priv, config.rx_filter != HWTSTAMP_FILTER_NONE); else @@ -5093,9 +5121,23 @@ static const struct mlx5e_profile mlx5e_nic_profile = { .rq_groups = MLX5E_NUM_RQ_GROUPS(XSK), .stats_grps = mlx5e_nic_stats_grps, .stats_grps_num = mlx5e_nic_stats_grps_num, - .rx_ptp_support = true, + .features = BIT(MLX5E_PROFILE_FEATURE_PTP_RX) | + BIT(MLX5E_PROFILE_FEATURE_PTP_TX) | + BIT(MLX5E_PROFILE_FEATURE_QOS_HTB), }; +static int mlx5e_profile_max_num_channels(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) +{ + int nch; + + nch = mlx5e_get_max_num_channels(mdev); + + if (profile->max_nch_limit) + nch = min_t(int, nch, profile->max_nch_limit(mdev)); + return nch; +} + static unsigned int mlx5e_calc_max_nch(struct mlx5_core_dev *mdev, struct net_device *netdev, const struct mlx5e_profile *profile) @@ -5104,7 +5146,7 @@ mlx5e_calc_max_nch(struct mlx5_core_dev *mdev, struct net_device *netdev, unsigned int max_nch, tmp; /* core resources */ - max_nch = mlx5e_get_max_num_channels(mdev); + max_nch = mlx5e_profile_max_num_channels(mdev, profile); /* netdev rx queues */ tmp = netdev->num_rx_queues / max_t(u8, profile->rq_groups, 1); @@ -5128,12 +5170,17 @@ int mlx5e_priv_init(struct mlx5e_priv *priv, struct net_device *netdev, struct mlx5_core_dev *mdev) { + int nch, num_txqs, node, i; + + num_txqs = netdev->num_tx_queues; + nch = mlx5e_calc_max_nch(mdev, netdev, profile); + node = dev_to_node(mlx5_core_dma_dev(mdev)); + /* priv init */ priv->mdev = mdev; priv->netdev = netdev; priv->msglevel = MLX5E_MSG_LEVEL; - priv->max_nch = mlx5e_calc_max_nch(mdev, netdev, profile); - priv->stats_nch = priv->max_nch; + priv->max_nch = nch; priv->max_opened_tc = 1; if (!alloc_cpumask_var(&priv->scratchpad.cpumask, GFP_KERNEL)) @@ -5150,11 +5197,46 @@ int mlx5e_priv_init(struct mlx5e_priv *priv, if (!priv->wq) goto err_free_cpumask; + priv->txq2sq = kcalloc_node(num_txqs, sizeof(*priv->txq2sq), GFP_KERNEL, node); + if (!priv->txq2sq) + goto err_destroy_workqueue; + + priv->tx_rates = kcalloc_node(num_txqs, sizeof(*priv->tx_rates), GFP_KERNEL, node); + if (!priv->tx_rates) + goto err_free_txq2sq; + + priv->channel_tc2realtxq = + kcalloc_node(nch, sizeof(*priv->channel_tc2realtxq), GFP_KERNEL, node); + if (!priv->channel_tc2realtxq) + goto err_free_tx_rates; + + for (i = 0; i < nch; i++) { + priv->channel_tc2realtxq[i] = + kcalloc_node(profile->max_tc, sizeof(**priv->channel_tc2realtxq), + GFP_KERNEL, node); + if (!priv->channel_tc2realtxq[i]) + goto err_free_channel_tc2realtxq; + } + + priv->channel_stats = + kcalloc_node(nch, sizeof(*priv->channel_stats), GFP_KERNEL, node); + if (!priv->channel_stats) + goto err_free_channel_tc2realtxq; + return 0; +err_free_channel_tc2realtxq: + while (--i >= 0) + kfree(priv->channel_tc2realtxq[i]); + kfree(priv->channel_tc2realtxq); +err_free_tx_rates: + kfree(priv->tx_rates); +err_free_txq2sq: + kfree(priv->txq2sq); +err_destroy_workqueue: + destroy_workqueue(priv->wq); err_free_cpumask: free_cpumask_var(priv->scratchpad.cpumask); - return -ENOMEM; } @@ -5166,6 +5248,14 @@ void mlx5e_priv_cleanup(struct mlx5e_priv *priv) if (!priv->mdev) return; + for (i = 0; i < priv->stats_nch; i++) + kvfree(priv->channel_stats[i]); + kfree(priv->channel_stats); + for (i = 0; i < priv->max_nch; i++) + kfree(priv->channel_tc2realtxq[i]); + kfree(priv->channel_tc2realtxq); + kfree(priv->tx_rates); + kfree(priv->txq2sq); destroy_workqueue(priv->wq); free_cpumask_var(priv->scratchpad.cpumask); @@ -5181,13 +5271,44 @@ void mlx5e_priv_cleanup(struct mlx5e_priv *priv) memset(priv, 0, sizeof(*priv)); } +static unsigned int mlx5e_get_max_num_txqs(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) +{ + unsigned int nch, ptp_txqs, qos_txqs; + + nch = mlx5e_profile_max_num_channels(mdev, profile); + + ptp_txqs = MLX5_CAP_GEN(mdev, ts_cqe_to_dest_cqn) && + mlx5e_profile_feature_cap(profile, PTP_TX) ? + profile->max_tc : 0; + + qos_txqs = mlx5_qos_is_supported(mdev) && + mlx5e_profile_feature_cap(profile, QOS_HTB) ? + mlx5e_qos_max_leaf_nodes(mdev) : 0; + + return nch * profile->max_tc + ptp_txqs + qos_txqs; +} + +static unsigned int mlx5e_get_max_num_rxqs(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) +{ + unsigned int nch; + + nch = mlx5e_profile_max_num_channels(mdev, profile); + + return nch * profile->rq_groups; +} + struct net_device * -mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *profile, - unsigned int txqs, unsigned int rxqs) +mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *profile) { struct net_device *netdev; + unsigned int txqs, rxqs; int err; + txqs = mlx5e_get_max_num_txqs(mdev, profile); + rxqs = mlx5e_get_max_num_rxqs(mdev, profile); + netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), txqs, rxqs); if (!netdev) { mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n"); @@ -5432,22 +5553,10 @@ static int mlx5e_probe(struct auxiliary_device *adev, struct mlx5_core_dev *mdev = edev->mdev; struct net_device *netdev; pm_message_t state = {}; - unsigned int txqs, rxqs, ptp_txqs = 0; struct mlx5e_priv *priv; - int qos_sqs = 0; int err; - int nch; - - if (MLX5_CAP_GEN(mdev, ts_cqe_to_dest_cqn)) - ptp_txqs = profile->max_tc; - if (mlx5_qos_is_supported(mdev)) - qos_sqs = mlx5e_qos_max_leaf_nodes(mdev); - - nch = mlx5e_get_max_num_channels(mdev); - txqs = nch * profile->max_tc + ptp_txqs + qos_sqs; - rxqs = nch * profile->rq_groups; - netdev = mlx5e_create_netdev(mdev, profile, txqs, rxqs); + netdev = mlx5e_create_netdev(mdev, profile); if (!netdev) { mlx5_core_err(mdev, "mlx5e_create_netdev failed\n"); return -ENOMEM; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 6e0f88ea3701..8c0f4cfbe471 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -591,6 +591,12 @@ bool mlx5e_eswitch_vf_rep(const struct net_device *netdev) return netdev->netdev_ops == &mlx5e_netdev_ops_rep; } +static int mlx5e_rep_max_nch_limit(struct mlx5_core_dev *mdev) +{ + return (1 << MLX5_CAP_GEN(mdev, log_max_tir)) / + mlx5_eswitch_get_total_vports(mdev); +} + static void mlx5e_build_rep_params(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); @@ -1113,7 +1119,7 @@ static const struct mlx5e_profile mlx5e_rep_profile = { .rq_groups = MLX5E_NUM_RQ_GROUPS(REGULAR), .stats_grps = mlx5e_rep_stats_grps, .stats_grps_num = mlx5e_rep_stats_grps_num, - .rx_ptp_support = false, + .max_nch_limit = mlx5e_rep_max_nch_limit, }; static const struct mlx5e_profile mlx5e_uplink_rep_profile = { @@ -1134,7 +1140,6 @@ static const struct mlx5e_profile mlx5e_uplink_rep_profile = { .rq_groups = MLX5E_NUM_RQ_GROUPS(XSK), .stats_grps = mlx5e_ul_rep_stats_grps, .stats_grps_num = mlx5e_ul_rep_stats_grps_num, - .rx_ptp_support = false, }; /* e-Switch vport representors */ @@ -1185,14 +1190,10 @@ mlx5e_vport_vf_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) struct devlink_port *dl_port; struct net_device *netdev; struct mlx5e_priv *priv; - unsigned int txqs, rxqs; - int nch, err; + int err; profile = &mlx5e_rep_profile; - nch = mlx5e_get_max_num_channels(dev); - txqs = nch * profile->max_tc; - rxqs = nch * profile->rq_groups; - netdev = mlx5e_create_netdev(dev, profile, txqs, rxqs); + netdev = mlx5e_create_netdev(dev, profile); if (!netdev) { mlx5_core_warn(dev, "Failed to create representor netdev for vport %d\n", diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 7e05d7592bce..f09b57c31ed7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -2189,7 +2189,7 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, priv = mlx5i_epriv(netdev); tstamp = &priv->tstamp; - stats = &priv->channel_stats[rq->ix].rq; + stats = rq->stats; flags_rqpn = be32_to_cpu(cqe->flags_rqpn); g = (flags_rqpn >> 28) & 3; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index 3c91a11e27ad..73fcd9fb17dd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -463,7 +463,7 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(sw) for (i = 0; i < priv->stats_nch; i++) { struct mlx5e_channel_stats *channel_stats = - &priv->channel_stats[i]; + priv->channel_stats[i]; int j; mlx5e_stats_grp_sw_update_stats_rq_stats(s, &channel_stats->rq); @@ -2197,21 +2197,21 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(channels) for (i = 0; i < max_nch; i++) for (j = 0; j < NUM_CH_STATS; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].ch, + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->ch, ch_stats_desc, j); for (i = 0; i < max_nch; i++) { for (j = 0; j < NUM_RQ_STATS; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].rq, + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->rq, rq_stats_desc, j); for (j = 0; j < NUM_XSKRQ_STATS * is_xsk; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].xskrq, + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->xskrq, xskrq_stats_desc, j); for (j = 0; j < NUM_RQ_XDPSQ_STATS; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].rq_xdpsq, + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->rq_xdpsq, rq_xdpsq_stats_desc, j); } @@ -2219,17 +2219,17 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(channels) for (i = 0; i < max_nch; i++) for (j = 0; j < NUM_SQ_STATS; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].sq[tc], + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->sq[tc], sq_stats_desc, j); for (i = 0; i < max_nch; i++) { for (j = 0; j < NUM_XSKSQ_STATS * is_xsk; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].xsksq, + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->xsksq, xsksq_stats_desc, j); for (j = 0; j < NUM_XDPSQ_STATS; j++) data[idx++] = - MLX5E_READ_CTR64_CPU(&priv->channel_stats[i].xdpsq, + MLX5E_READ_CTR64_CPU(&priv->channel_stats[i]->xdpsq, xdpsq_stats_desc, j); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 792e0d6aa861..b695aad71ee1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -19,6 +19,7 @@ #include "lib/clock.h" #include "diag/fw_tracer.h" #include "mlx5_irq.h" +#include "devlink.h" enum { MLX5_EQE_OWNER_INIT_VAL = 0x1, @@ -622,6 +623,20 @@ static void cleanup_async_eq(struct mlx5_core_dev *dev, name, err); } +static u16 async_eq_depth_devlink_param_get(struct mlx5_core_dev *dev) +{ + struct devlink *devlink = priv_to_devlink(dev); + union devlink_param_value val; + int err; + + err = devlink_param_driverinit_value_get(devlink, + DEVLINK_PARAM_GENERIC_ID_EVENT_EQ_SIZE, + &val); + if (!err) + return val.vu32; + mlx5_core_dbg(dev, "Failed to get param. using default. err = %d\n", err); + return MLX5_NUM_ASYNC_EQE; +} static int create_async_eqs(struct mlx5_core_dev *dev) { struct mlx5_eq_table *table = dev->priv.eq_table; @@ -646,7 +661,7 @@ static int create_async_eqs(struct mlx5_core_dev *dev) param = (struct mlx5_eq_param) { .irq_index = MLX5_IRQ_EQ_CTRL, - .nent = MLX5_NUM_ASYNC_EQE, + .nent = async_eq_depth_devlink_param_get(dev), }; gather_async_events_mask(dev, param.mask); @@ -796,6 +811,21 @@ static void destroy_comp_eqs(struct mlx5_core_dev *dev) } } +static u16 comp_eq_depth_devlink_param_get(struct mlx5_core_dev *dev) +{ + struct devlink *devlink = priv_to_devlink(dev); + union devlink_param_value val; + int err; + + err = devlink_param_driverinit_value_get(devlink, + DEVLINK_PARAM_GENERIC_ID_IO_EQ_SIZE, + &val); + if (!err) + return val.vu32; + mlx5_core_dbg(dev, "Failed to get param. using default. err = %d\n", err); + return MLX5_COMP_EQ_SIZE; +} + static int create_comp_eqs(struct mlx5_core_dev *dev) { struct mlx5_eq_table *table = dev->priv.eq_table; @@ -807,7 +837,7 @@ static int create_comp_eqs(struct mlx5_core_dev *dev) INIT_LIST_HEAD(&table->comp_eqs_list); ncomp_eqs = table->num_comp_eqs; - nent = MLX5_COMP_EQ_SIZE; + nent = comp_eq_depth_devlink_param_get(dev); for (i = 0; i < ncomp_eqs; i++) { struct mlx5_eq_param param = {}; int vecidx = i; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 513f741d16c7..ead5e8acc8be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -343,9 +343,6 @@ void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata); int mlx5_esw_qos_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32 rate_mbps); -bool mlx5_esw_vport_match_metadata_supported(const struct mlx5_eswitch *esw); -int mlx5_esw_offloads_vport_metadata_set(struct mlx5_eswitch *esw, bool enable); - /* E-Switch API */ int mlx5_eswitch_init(struct mlx5_core_dev *dev); void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 051b20ec7bdb..0a99a020a3b2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -117,7 +117,7 @@ static void mlx5i_grp_sw_update_stats(struct mlx5e_priv *priv) struct mlx5e_channel_stats *channel_stats; struct mlx5e_rq_stats *rq_stats; - channel_stats = &priv->channel_stats[i]; + channel_stats = priv->channel_stats[i]; rq_stats = &channel_stats->rq; s.rx_packets += rq_stats->packets; @@ -449,7 +449,6 @@ static const struct mlx5e_profile mlx5i_nic_profile = { .rq_groups = MLX5E_NUM_RQ_GROUPS(REGULAR), .stats_grps = mlx5i_stats_grps, .stats_grps_num = mlx5i_stats_grps_num, - .rx_ptp_support = false, }; /* mlx5i netdev NDos */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c index 5308f23702bc..0b86e78dbc0e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c @@ -350,7 +350,6 @@ static const struct mlx5e_profile mlx5i_pkey_nic_profile = { .rx_handlers = &mlx5i_rx_handlers, .max_tc = MLX5I_MAX_NUM_TC, .rq_groups = MLX5E_NUM_RQ_GROUPS(REGULAR), - .rx_ptp_support = false, }; const struct mlx5e_profile *mlx5i_pkey_get_profile(void) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index d97c9e86d7b3..b1a82226623c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -484,10 +484,26 @@ static int handle_hca_cap_odp(struct mlx5_core_dev *dev, void *set_ctx) return set_caps(dev, set_ctx, MLX5_SET_HCA_CAP_OP_MOD_ODP); } +static int max_uc_list_get_devlink_param(struct mlx5_core_dev *dev) +{ + struct devlink *devlink = priv_to_devlink(dev); + union devlink_param_value val; + int err; + + err = devlink_param_driverinit_value_get(devlink, + DEVLINK_PARAM_GENERIC_ID_MAX_MACS, + &val); + if (!err) + return val.vu32; + mlx5_core_dbg(dev, "Failed to get param. err = %d\n", err); + return err; +} + static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx) { struct mlx5_profile *prof = &dev->profile; void *set_hca_cap; + int max_uc_list; int err; err = mlx5_core_get_caps(dev, MLX5_CAP_GENERAL); @@ -561,6 +577,11 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx) if (MLX5_CAP_GEN(dev, roce_rw_supported)) MLX5_SET(cmd_hca_cap, set_hca_cap, roce, mlx5_is_roce_init_enabled(dev)); + max_uc_list = max_uc_list_get_devlink_param(dev); + if (max_uc_list > 0) + MLX5_SET(cmd_hca_cap, set_hca_cap, log_max_current_uc_list, + ilog2(max_uc_list)); + return set_caps(dev, set_ctx, MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 7e89d5980d5e..866b9357939b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -1708,6 +1708,124 @@ static void mlxsw_core_health_listener_func(const struct mlxsw_reg_info *reg, static const struct mlxsw_listener mlxsw_core_health_listener = MLXSW_EVENTL(mlxsw_core_health_listener_func, MFDE, MFDE); +static int +mlxsw_core_health_fw_fatal_dump_fatal_cause(const char *mfde_pl, + struct devlink_fmsg *fmsg) +{ + u32 val, tile_v; + int err; + + val = mlxsw_reg_mfde_fatal_cause_id_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "cause_id", val); + if (err) + return err; + tile_v = mlxsw_reg_mfde_fatal_cause_tile_v_get(mfde_pl); + if (tile_v) { + val = mlxsw_reg_mfde_fatal_cause_tile_index_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "tile_index", val); + if (err) + return err; + } + + return 0; +} + +static int +mlxsw_core_health_fw_fatal_dump_fw_assert(const char *mfde_pl, + struct devlink_fmsg *fmsg) +{ + u32 val, tile_v; + int err; + + val = mlxsw_reg_mfde_fw_assert_var0_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "var0", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_var1_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "var1", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_var2_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "var2", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_var3_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "var3", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_var4_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "var4", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_existptr_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "existptr", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_callra_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "callra", val); + if (err) + return err; + val = mlxsw_reg_mfde_fw_assert_oe_get(mfde_pl); + err = devlink_fmsg_bool_pair_put(fmsg, "old_event", val); + if (err) + return err; + tile_v = mlxsw_reg_mfde_fw_assert_tile_v_get(mfde_pl); + if (tile_v) { + val = mlxsw_reg_mfde_fw_assert_tile_index_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "tile_index", val); + if (err) + return err; + } + val = mlxsw_reg_mfde_fw_assert_ext_synd_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "ext_synd", val); + if (err) + return err; + + return 0; +} + +static int +mlxsw_core_health_fw_fatal_dump_kvd_im_stop(const char *mfde_pl, + struct devlink_fmsg *fmsg) +{ + u32 val; + int err; + + val = mlxsw_reg_mfde_kvd_im_stop_oe_get(mfde_pl); + err = devlink_fmsg_bool_pair_put(fmsg, "old_event", val); + if (err) + return err; + val = mlxsw_reg_mfde_kvd_im_stop_pipes_mask_get(mfde_pl); + return devlink_fmsg_u32_pair_put(fmsg, "pipes_mask", val); +} + +static int +mlxsw_core_health_fw_fatal_dump_crspace_to(const char *mfde_pl, + struct devlink_fmsg *fmsg) +{ + u32 val; + int err; + + val = mlxsw_reg_mfde_crspace_to_log_address_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "log_address", val); + if (err) + return err; + val = mlxsw_reg_mfde_crspace_to_oe_get(mfde_pl); + err = devlink_fmsg_bool_pair_put(fmsg, "old_event", val); + if (err) + return err; + val = mlxsw_reg_mfde_crspace_to_log_id_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "log_irisc_id", val); + if (err) + return err; + val = mlxsw_reg_mfde_crspace_to_log_ip_get(mfde_pl); + err = devlink_fmsg_u64_pair_put(fmsg, "log_ip", val); + if (err) + return err; + + return 0; +} + static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *reporter, struct devlink_fmsg *fmsg, void *priv_ctx, struct netlink_ext_ack *extack) @@ -1741,6 +1859,46 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor case MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP: val_str = "KVD insertion machine stopped"; break; + case MLXSW_REG_MFDE_EVENT_ID_TEST: + val_str = "Test"; + break; + case MLXSW_REG_MFDE_EVENT_ID_FW_ASSERT: + val_str = "FW assert"; + break; + case MLXSW_REG_MFDE_EVENT_ID_FATAL_CAUSE: + val_str = "Fatal cause"; + break; + default: + val_str = NULL; + } + if (val_str) { + err = devlink_fmsg_string_pair_put(fmsg, "desc", val_str); + if (err) + return err; + } + + err = devlink_fmsg_arr_pair_nest_end(fmsg); + if (err) + return err; + + err = devlink_fmsg_arr_pair_nest_start(fmsg, "severity"); + if (err) + return err; + + val = mlxsw_reg_mfde_severity_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "id", val); + if (err) + return err; + switch (val) { + case MLXSW_REG_MFDE_SEVERITY_FATL: + val_str = "Fatal"; + break; + case MLXSW_REG_MFDE_SEVERITY_NRML: + val_str = "Normal"; + break; + case MLXSW_REG_MFDE_SEVERITY_INTR: + val_str = "Debug"; + break; default: val_str = NULL; } @@ -1749,6 +1907,7 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor if (err) return err; } + err = devlink_fmsg_arr_pair_nest_end(fmsg); if (err) return err; @@ -1800,24 +1959,18 @@ static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *repor if (err) return err; - if (event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO) { - val = mlxsw_reg_mfde_log_address_get(mfde_pl); - err = devlink_fmsg_u32_pair_put(fmsg, "log_address", val); - if (err) - return err; - val = mlxsw_reg_mfde_log_id_get(mfde_pl); - err = devlink_fmsg_u8_pair_put(fmsg, "log_irisc_id", val); - if (err) - return err; - val = mlxsw_reg_mfde_log_ip_get(mfde_pl); - err = devlink_fmsg_u64_pair_put(fmsg, "log_ip", val); - if (err) - return err; - } else if (event_id == MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP) { - val = mlxsw_reg_mfde_pipes_mask_get(mfde_pl); - err = devlink_fmsg_u32_pair_put(fmsg, "pipes_mask", val); - if (err) - return err; + switch (event_id) { + case MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO: + return mlxsw_core_health_fw_fatal_dump_crspace_to(mfde_pl, + fmsg); + case MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP: + return mlxsw_core_health_fw_fatal_dump_kvd_im_stop(mfde_pl, + fmsg); + case MLXSW_REG_MFDE_EVENT_ID_FW_ASSERT: + return mlxsw_core_health_fw_fatal_dump_fw_assert(mfde_pl, fmsg); + case MLXSW_REG_MFDE_EVENT_ID_FATAL_CAUSE: + return mlxsw_core_health_fw_fatal_dump_fatal_cause(mfde_pl, + fmsg); } return 0; diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index f748b537bdab..c97d2c744725 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -11318,7 +11318,7 @@ mlxsw_reg_mgpir_unpack(char *payload, u8 *num_of_devices, * ----------------------------------- */ #define MLXSW_REG_MFDE_ID 0x9200 -#define MLXSW_REG_MFDE_LEN 0x18 +#define MLXSW_REG_MFDE_LEN 0x30 MLXSW_REG_DEFINE(mfde, MLXSW_REG_MFDE_ID, MLXSW_REG_MFDE_LEN); @@ -11328,10 +11328,32 @@ MLXSW_REG_DEFINE(mfde, MLXSW_REG_MFDE_ID, MLXSW_REG_MFDE_LEN); */ MLXSW_ITEM32(reg, mfde, irisc_id, 0x00, 24, 8); +enum mlxsw_reg_mfde_severity { + /* Unrecoverable switch behavior */ + MLXSW_REG_MFDE_SEVERITY_FATL = 2, + /* Unexpected state with possible systemic failure */ + MLXSW_REG_MFDE_SEVERITY_NRML = 3, + /* Unexpected state without systemic failure */ + MLXSW_REG_MFDE_SEVERITY_INTR = 5, +}; + +/* reg_mfde_severity + * The severity of the event. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, severity, 0x00, 16, 8); + enum mlxsw_reg_mfde_event_id { + /* CRspace timeout */ MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO = 1, /* KVD insertion machine stopped */ MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP, + /* Triggered by MFGD.trigger_test */ + MLXSW_REG_MFDE_EVENT_ID_TEST, + /* Triggered when firmware hits an assert */ + MLXSW_REG_MFDE_EVENT_ID_FW_ASSERT, + /* Fatal error interrupt from hardware */ + MLXSW_REG_MFDE_EVENT_ID_FATAL_CAUSE, }; /* reg_mfde_event_id @@ -11372,32 +11394,110 @@ MLXSW_ITEM32(reg, mfde, command_type, 0x04, 24, 2); */ MLXSW_ITEM32(reg, mfde, reg_attr_id, 0x04, 0, 16); -/* reg_mfde_log_address +/* reg_mfde_crspace_to_log_address * crspace address accessed, which resulted in timeout. - * Valid in case event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO * Access: RO */ -MLXSW_ITEM32(reg, mfde, log_address, 0x10, 0, 32); +MLXSW_ITEM32(reg, mfde, crspace_to_log_address, 0x10, 0, 32); -/* reg_mfde_log_id +/* reg_mfde_crspace_to_oe + * 0 - New event + * 1 - Old event, occurred before MFGD activation. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, crspace_to_oe, 0x14, 24, 1); + +/* reg_mfde_crspace_to_log_id * Which irisc triggered the timeout. - * Valid in case event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO * Access: RO */ -MLXSW_ITEM32(reg, mfde, log_id, 0x14, 0, 4); +MLXSW_ITEM32(reg, mfde, crspace_to_log_id, 0x14, 0, 4); -/* reg_mfde_log_ip +/* reg_mfde_crspace_to_log_ip * IP (instruction pointer) that triggered the timeout. - * Valid in case event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO * Access: RO */ -MLXSW_ITEM64(reg, mfde, log_ip, 0x18, 0, 64); +MLXSW_ITEM64(reg, mfde, crspace_to_log_ip, 0x18, 0, 64); + +/* reg_mfde_kvd_im_stop_oe + * 0 - New event + * 1 - Old event, occurred before MFGD activation. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, kvd_im_stop_oe, 0x10, 24, 1); -/* reg_mfde_pipes_mask +/* reg_mfde_kvd_im_stop_pipes_mask * Bit per kvh pipe. * Access: RO */ -MLXSW_ITEM32(reg, mfde, pipes_mask, 0x10, 0, 16); +MLXSW_ITEM32(reg, mfde, kvd_im_stop_pipes_mask, 0x10, 0, 16); + +/* reg_mfde_fw_assert_var0-4 + * Variables passed to assert. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_var0, 0x10, 0, 32); +MLXSW_ITEM32(reg, mfde, fw_assert_var1, 0x14, 0, 32); +MLXSW_ITEM32(reg, mfde, fw_assert_var2, 0x18, 0, 32); +MLXSW_ITEM32(reg, mfde, fw_assert_var3, 0x1C, 0, 32); +MLXSW_ITEM32(reg, mfde, fw_assert_var4, 0x20, 0, 32); + +/* reg_mfde_fw_assert_existptr + * The instruction pointer when assert was triggered. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_existptr, 0x24, 0, 32); + +/* reg_mfde_fw_assert_callra + * The return address after triggering assert. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_callra, 0x28, 0, 32); + +/* reg_mfde_fw_assert_oe + * 0 - New event + * 1 - Old event, occurred before MFGD activation. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_oe, 0x2C, 24, 1); + +/* reg_mfde_fw_assert_tile_v + * 0: The assert was from main + * 1: The assert was from a tile + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_tile_v, 0x2C, 23, 1); + +/* reg_mfde_fw_assert_tile_index + * When tile_v=1, the tile_index that caused the assert. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_tile_index, 0x2C, 16, 6); + +/* reg_mfde_fw_assert_ext_synd + * A generated one-to-one identifier which is specific per-assert. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fw_assert_ext_synd, 0x2C, 0, 16); + +/* reg_mfde_fatal_cause_id + * HW interrupt cause id. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fatal_cause_id, 0x10, 0, 18); + +/* reg_mfde_fatal_cause_tile_v + * 0: The assert was from main + * 1: The assert was from a tile + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fatal_cause_tile_v, 0x14, 23, 1); + +/* reg_mfde_fatal_cause_tile_index + * When tile_v=1, the tile_index that caused the assert. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, fatal_cause_tile_index, 0x14, 16, 6); /* TNGCR - Tunneling NVE General Configuration Register * ---------------------------------------------------- diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c index be3791ca6069..bb417db773b9 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c @@ -203,7 +203,7 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, */ burst = roundup_pow_of_two(act->police.burst); err = mlxsw_sp_acl_rulei_act_police(mlxsw_sp, rulei, - act->police.index, + act->hw_index, act->police.rate_bytes_ps, burst, extack); if (err) @@ -508,7 +508,8 @@ static int mlxsw_sp_flower_parse(struct mlxsw_sp *mlxsw_sp, struct flow_match_vlan match; flow_rule_match_vlan(rule, &match); - if (mlxsw_sp_flow_block_is_egress_bound(block)) { + if (mlxsw_sp_flow_block_is_egress_bound(block) && + match.mask->vlan_id) { NL_SET_ERR_MSG_MOD(f->common.extack, "vlan_id key is not supported on egress"); return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/microchip/lan966x/Kconfig b/drivers/net/ethernet/microchip/lan966x/Kconfig index 2860a8c9923d..ac273f84b69e 100644 --- a/drivers/net/ethernet/microchip/lan966x/Kconfig +++ b/drivers/net/ethernet/microchip/lan966x/Kconfig @@ -2,6 +2,7 @@ config LAN966X_SWITCH tristate "Lan966x switch driver" depends on HAS_IOMEM depends on OF + depends on NET_SWITCHDEV select PHYLINK select PACKING help diff --git a/drivers/net/ethernet/microchip/lan966x/Makefile b/drivers/net/ethernet/microchip/lan966x/Makefile index 2989ba528236..ec1a1fa8b0d5 100644 --- a/drivers/net/ethernet/microchip/lan966x/Makefile +++ b/drivers/net/ethernet/microchip/lan966x/Makefile @@ -6,4 +6,5 @@ obj-$(CONFIG_LAN966X_SWITCH) += lan966x-switch.o lan966x-switch-objs := lan966x_main.o lan966x_phylink.o lan966x_port.o \ - lan966x_mac.o lan966x_ethtool.o + lan966x_mac.o lan966x_ethtool.o lan966x_switchdev.o \ + lan966x_vlan.o lan966x_fdb.o diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_fdb.c b/drivers/net/ethernet/microchip/lan966x/lan966x_fdb.c new file mode 100644 index 000000000000..da5ca7188679 --- /dev/null +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_fdb.c @@ -0,0 +1,244 @@ +// SPDX-License-Identifier: GPL-2.0+ + +#include <net/switchdev.h> + +#include "lan966x_main.h" + +struct lan966x_fdb_event_work { + struct work_struct work; + struct switchdev_notifier_fdb_info fdb_info; + struct net_device *dev; + struct lan966x *lan966x; + unsigned long event; +}; + +struct lan966x_fdb_entry { + struct list_head list; + unsigned char mac[ETH_ALEN] __aligned(2); + u16 vid; + u32 references; +}; + +static struct lan966x_fdb_entry * +lan966x_fdb_find_entry(struct lan966x *lan966x, + struct switchdev_notifier_fdb_info *fdb_info) +{ + struct lan966x_fdb_entry *fdb_entry; + + list_for_each_entry(fdb_entry, &lan966x->fdb_entries, list) { + if (fdb_entry->vid == fdb_info->vid && + ether_addr_equal(fdb_entry->mac, fdb_info->addr)) + return fdb_entry; + } + + return NULL; +} + +static void lan966x_fdb_add_entry(struct lan966x *lan966x, + struct switchdev_notifier_fdb_info *fdb_info) +{ + struct lan966x_fdb_entry *fdb_entry; + + fdb_entry = lan966x_fdb_find_entry(lan966x, fdb_info); + if (fdb_entry) { + fdb_entry->references++; + return; + } + + fdb_entry = kzalloc(sizeof(*fdb_entry), GFP_KERNEL); + if (!fdb_entry) + return; + + ether_addr_copy(fdb_entry->mac, fdb_info->addr); + fdb_entry->vid = fdb_info->vid; + fdb_entry->references = 1; + list_add_tail(&fdb_entry->list, &lan966x->fdb_entries); +} + +static bool lan966x_fdb_del_entry(struct lan966x *lan966x, + struct switchdev_notifier_fdb_info *fdb_info) +{ + struct lan966x_fdb_entry *fdb_entry, *tmp; + + list_for_each_entry_safe(fdb_entry, tmp, &lan966x->fdb_entries, + list) { + if (fdb_entry->vid == fdb_info->vid && + ether_addr_equal(fdb_entry->mac, fdb_info->addr)) { + fdb_entry->references--; + if (!fdb_entry->references) { + list_del(&fdb_entry->list); + kfree(fdb_entry); + return true; + } + break; + } + } + + return false; +} + +void lan966x_fdb_write_entries(struct lan966x *lan966x, u16 vid) +{ + struct lan966x_fdb_entry *fdb_entry; + + list_for_each_entry(fdb_entry, &lan966x->fdb_entries, list) { + if (fdb_entry->vid != vid) + continue; + + lan966x_mac_cpu_learn(lan966x, fdb_entry->mac, fdb_entry->vid); + } +} + +void lan966x_fdb_erase_entries(struct lan966x *lan966x, u16 vid) +{ + struct lan966x_fdb_entry *fdb_entry; + + list_for_each_entry(fdb_entry, &lan966x->fdb_entries, list) { + if (fdb_entry->vid != vid) + continue; + + lan966x_mac_cpu_forget(lan966x, fdb_entry->mac, fdb_entry->vid); + } +} + +static void lan966x_fdb_purge_entries(struct lan966x *lan966x) +{ + struct lan966x_fdb_entry *fdb_entry, *tmp; + + list_for_each_entry_safe(fdb_entry, tmp, &lan966x->fdb_entries, list) { + list_del(&fdb_entry->list); + kfree(fdb_entry); + } +} + +int lan966x_fdb_init(struct lan966x *lan966x) +{ + INIT_LIST_HEAD(&lan966x->fdb_entries); + lan966x->fdb_work = alloc_ordered_workqueue("lan966x_order", 0); + if (!lan966x->fdb_work) + return -ENOMEM; + + return 0; +} + +void lan966x_fdb_deinit(struct lan966x *lan966x) +{ + destroy_workqueue(lan966x->fdb_work); + lan966x_fdb_purge_entries(lan966x); +} + +static void lan966x_fdb_event_work(struct work_struct *work) +{ + struct lan966x_fdb_event_work *fdb_work = + container_of(work, struct lan966x_fdb_event_work, work); + struct switchdev_notifier_fdb_info *fdb_info; + struct net_device *dev = fdb_work->dev; + struct lan966x_port *port; + struct lan966x *lan966x; + int ret; + + fdb_info = &fdb_work->fdb_info; + lan966x = fdb_work->lan966x; + + if (lan966x_netdevice_check(dev)) { + port = netdev_priv(dev); + + switch (fdb_work->event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: + if (!fdb_info->added_by_user) + break; + lan966x_mac_add_entry(lan966x, port, fdb_info->addr, + fdb_info->vid); + break; + case SWITCHDEV_FDB_DEL_TO_DEVICE: + if (!fdb_info->added_by_user) + break; + lan966x_mac_del_entry(lan966x, fdb_info->addr, + fdb_info->vid); + break; + } + } else { + if (!netif_is_bridge_master(dev)) + goto out; + + /* In case the bridge is called */ + switch (fdb_work->event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: + /* If there is no front port in this vlan, there is no + * point to copy the frame to CPU because it would be + * just dropped at later point. So add it only if + * there is a port but it is required to store the fdb + * entry for later point when a port actually gets in + * the vlan. + */ + lan966x_fdb_add_entry(lan966x, fdb_info); + if (!lan966x_vlan_cpu_member_cpu_vlan_mask(lan966x, + fdb_info->vid)) + break; + + lan966x_mac_cpu_learn(lan966x, fdb_info->addr, + fdb_info->vid); + break; + case SWITCHDEV_FDB_DEL_TO_DEVICE: + ret = lan966x_fdb_del_entry(lan966x, fdb_info); + if (!lan966x_vlan_cpu_member_cpu_vlan_mask(lan966x, + fdb_info->vid)) + break; + + if (ret) + lan966x_mac_cpu_forget(lan966x, fdb_info->addr, + fdb_info->vid); + break; + } + } + +out: + kfree(fdb_work->fdb_info.addr); + kfree(fdb_work); + dev_put(dev); +} + +int lan966x_handle_fdb(struct net_device *dev, + struct net_device *orig_dev, + unsigned long event, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info) +{ + struct lan966x_port *port = netdev_priv(dev); + struct lan966x *lan966x = port->lan966x; + struct lan966x_fdb_event_work *fdb_work; + + if (ctx && ctx != port) + return 0; + + switch (event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: + case SWITCHDEV_FDB_DEL_TO_DEVICE: + if (lan966x_netdevice_check(orig_dev) && + !fdb_info->added_by_user) + break; + + fdb_work = kzalloc(sizeof(*fdb_work), GFP_ATOMIC); + if (!fdb_work) + return -ENOMEM; + + fdb_work->dev = orig_dev; + fdb_work->lan966x = lan966x; + fdb_work->event = event; + INIT_WORK(&fdb_work->work, lan966x_fdb_event_work); + memcpy(&fdb_work->fdb_info, fdb_info, sizeof(fdb_work->fdb_info)); + fdb_work->fdb_info.addr = kzalloc(ETH_ALEN, GFP_ATOMIC); + if (!fdb_work->fdb_info.addr) + goto err_addr_alloc; + + ether_addr_copy((u8 *)fdb_work->fdb_info.addr, fdb_info->addr); + dev_hold(orig_dev); + + queue_work(lan966x->fdb_work, &fdb_work->work); + break; + } + + return 0; +err_addr_alloc: + kfree(fdb_work); + return -ENOMEM; +} diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_mac.c b/drivers/net/ethernet/microchip/lan966x/lan966x_mac.c index f6878b9f57ef..efadb8d326cc 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_mac.c +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_mac.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0+ +#include <net/switchdev.h> #include "lan966x_main.h" #define LAN966X_MAC_COLUMNS 4 @@ -13,6 +14,23 @@ #define MACACCESS_CMD_WRITE 7 #define MACACCESS_CMD_SYNC_GET_NEXT 8 +#define LAN966X_MAC_INVALID_ROW -1 + +struct lan966x_mac_entry { + struct list_head list; + unsigned char mac[ETH_ALEN] __aligned(2); + u16 vid; + u16 port_index; + int row; +}; + +struct lan966x_mac_raw_entry { + u32 mach; + u32 macl; + u32 maca; + bool processed; +}; + static int lan966x_mac_get_status(struct lan966x *lan966x) { return lan_rd(lan966x, ANA_MACACCESS); @@ -93,9 +111,333 @@ int lan966x_mac_cpu_forget(struct lan966x *lan966x, const char *addr, u16 vid) return lan966x_mac_forget(lan966x, addr, vid, ENTRYTYPE_LOCKED); } +void lan966x_mac_set_ageing(struct lan966x *lan966x, + u32 ageing) +{ + lan_rmw(ANA_AUTOAGE_AGE_PERIOD_SET(ageing / 2), + ANA_AUTOAGE_AGE_PERIOD, + lan966x, ANA_AUTOAGE); +} + void lan966x_mac_init(struct lan966x *lan966x) { /* Clear the MAC table */ lan_wr(MACACCESS_CMD_INIT, lan966x, ANA_MACACCESS); lan966x_mac_wait_for_completion(lan966x); + + spin_lock_init(&lan966x->mac_lock); + INIT_LIST_HEAD(&lan966x->mac_entries); +} + +static struct lan966x_mac_entry *lan966x_mac_alloc_entry(const unsigned char *mac, + u16 vid, u16 port_index) +{ + struct lan966x_mac_entry *mac_entry; + + mac_entry = kzalloc(sizeof(*mac_entry), GFP_KERNEL); + if (!mac_entry) + return NULL; + + memcpy(mac_entry->mac, mac, ETH_ALEN); + mac_entry->vid = vid; + mac_entry->port_index = port_index; + mac_entry->row = LAN966X_MAC_INVALID_ROW; + return mac_entry; +} + +static struct lan966x_mac_entry *lan966x_mac_find_entry(struct lan966x *lan966x, + const unsigned char *mac, + u16 vid, u16 port_index) +{ + struct lan966x_mac_entry *res = NULL; + struct lan966x_mac_entry *mac_entry; + + spin_lock(&lan966x->mac_lock); + list_for_each_entry(mac_entry, &lan966x->mac_entries, list) { + if (mac_entry->vid == vid && + ether_addr_equal(mac, mac_entry->mac) && + mac_entry->port_index == port_index) { + res = mac_entry; + break; + } + } + spin_unlock(&lan966x->mac_lock); + + return res; +} + +static int lan966x_mac_lookup(struct lan966x *lan966x, + const unsigned char mac[ETH_ALEN], + unsigned int vid, enum macaccess_entry_type type) +{ + int ret; + + lan966x_mac_select(lan966x, mac, vid); + + /* Issue a read command */ + lan_wr(ANA_MACACCESS_ENTRYTYPE_SET(type) | + ANA_MACACCESS_VALID_SET(1) | + ANA_MACACCESS_MAC_TABLE_CMD_SET(MACACCESS_CMD_READ), + lan966x, ANA_MACACCESS); + + ret = lan966x_mac_wait_for_completion(lan966x); + if (ret) + return ret; + + return ANA_MACACCESS_VALID_GET(lan_rd(lan966x, ANA_MACACCESS)); +} + +static void lan966x_fdb_call_notifiers(enum switchdev_notifier_type type, + const char *mac, u16 vid, + struct net_device *dev) +{ + struct switchdev_notifier_fdb_info info = { 0 }; + + info.addr = mac; + info.vid = vid; + info.offloaded = true; + call_switchdev_notifiers(type, dev, &info.info, NULL); +} + +int lan966x_mac_add_entry(struct lan966x *lan966x, struct lan966x_port *port, + const unsigned char *addr, u16 vid) +{ + struct lan966x_mac_entry *mac_entry; + + if (lan966x_mac_lookup(lan966x, addr, vid, ENTRYTYPE_NORMAL)) + return 0; + + /* In case the entry already exists, don't add it again to SW, + * just update HW, but we need to look in the actual HW because + * it is possible for an entry to be learn by HW and before we + * get the interrupt the frame will reach CPU and the CPU will + * add the entry but without the extern_learn flag. + */ + mac_entry = lan966x_mac_find_entry(lan966x, addr, vid, port->chip_port); + if (mac_entry) + return lan966x_mac_learn(lan966x, port->chip_port, + addr, vid, ENTRYTYPE_LOCKED); + + mac_entry = lan966x_mac_alloc_entry(addr, vid, port->chip_port); + if (!mac_entry) + return -ENOMEM; + + spin_lock(&lan966x->mac_lock); + list_add_tail(&mac_entry->list, &lan966x->mac_entries); + spin_unlock(&lan966x->mac_lock); + + lan966x_mac_learn(lan966x, port->chip_port, addr, vid, ENTRYTYPE_LOCKED); + lan966x_fdb_call_notifiers(SWITCHDEV_FDB_OFFLOADED, addr, vid, port->dev); + + return 0; +} + +int lan966x_mac_del_entry(struct lan966x *lan966x, const unsigned char *addr, + u16 vid) +{ + struct lan966x_mac_entry *mac_entry, *tmp; + + spin_lock(&lan966x->mac_lock); + list_for_each_entry_safe(mac_entry, tmp, &lan966x->mac_entries, + list) { + if (mac_entry->vid == vid && + ether_addr_equal(addr, mac_entry->mac)) { + lan966x_mac_forget(lan966x, mac_entry->mac, mac_entry->vid, + ENTRYTYPE_LOCKED); + + list_del(&mac_entry->list); + kfree(mac_entry); + } + } + spin_unlock(&lan966x->mac_lock); + + return 0; +} + +void lan966x_mac_purge_entries(struct lan966x *lan966x) +{ + struct lan966x_mac_entry *mac_entry, *tmp; + + spin_lock(&lan966x->mac_lock); + list_for_each_entry_safe(mac_entry, tmp, &lan966x->mac_entries, + list) { + lan966x_mac_forget(lan966x, mac_entry->mac, mac_entry->vid, + ENTRYTYPE_LOCKED); + + list_del(&mac_entry->list); + kfree(mac_entry); + } + spin_unlock(&lan966x->mac_lock); +} + +static void lan966x_mac_notifiers(enum switchdev_notifier_type type, + unsigned char *mac, u32 vid, + struct net_device *dev) +{ + rtnl_lock(); + lan966x_fdb_call_notifiers(type, mac, vid, dev); + rtnl_unlock(); +} + +static void lan966x_mac_process_raw_entry(struct lan966x_mac_raw_entry *raw_entry, + u8 *mac, u16 *vid, u32 *dest_idx) +{ + mac[0] = (raw_entry->mach >> 8) & 0xff; + mac[1] = (raw_entry->mach >> 0) & 0xff; + mac[2] = (raw_entry->macl >> 24) & 0xff; + mac[3] = (raw_entry->macl >> 16) & 0xff; + mac[4] = (raw_entry->macl >> 8) & 0xff; + mac[5] = (raw_entry->macl >> 0) & 0xff; + + *vid = (raw_entry->mach >> 16) & 0xfff; + *dest_idx = ANA_MACACCESS_DEST_IDX_GET(raw_entry->maca); +} + +static void lan966x_mac_irq_process(struct lan966x *lan966x, u32 row, + struct lan966x_mac_raw_entry *raw_entries) +{ + struct lan966x_mac_entry *mac_entry, *tmp; + unsigned char mac[ETH_ALEN] __aligned(2); + u32 dest_idx; + u32 column; + u16 vid; + + spin_lock(&lan966x->mac_lock); + list_for_each_entry_safe(mac_entry, tmp, &lan966x->mac_entries, list) { + bool found = false; + + if (mac_entry->row != row) + continue; + + for (column = 0; column < LAN966X_MAC_COLUMNS; ++column) { + /* All the valid entries are at the start of the row, + * so when get one invalid entry it can just skip the + * rest of the columns + */ + if (!ANA_MACACCESS_VALID_GET(raw_entries[column].maca)) + break; + + lan966x_mac_process_raw_entry(&raw_entries[column], + mac, &vid, &dest_idx); + WARN_ON(dest_idx > lan966x->num_phys_ports); + + /* If the entry in SW is found, then there is nothing + * to do + */ + if (mac_entry->vid == vid && + ether_addr_equal(mac_entry->mac, mac) && + mac_entry->port_index == dest_idx) { + raw_entries[column].processed = true; + found = true; + break; + } + } + + if (!found) { + /* Notify the bridge that the entry doesn't exist + * anymore in the HW and remove the entry from the SW + * list + */ + lan966x_mac_notifiers(SWITCHDEV_FDB_DEL_TO_BRIDGE, + mac_entry->mac, mac_entry->vid, + lan966x->ports[mac_entry->port_index]->dev); + + list_del(&mac_entry->list); + kfree(mac_entry); + } + } + spin_unlock(&lan966x->mac_lock); + + /* Now go to the list of columns and see if any entry was not in the SW + * list, then that means that the entry is new so it needs to notify the + * bridge. + */ + for (column = 0; column < LAN966X_MAC_COLUMNS; ++column) { + /* All the valid entries are at the start of the row, so when + * get one invalid entry it can just skip the rest of the columns + */ + if (!ANA_MACACCESS_VALID_GET(raw_entries[column].maca)) + break; + + /* If the entry already exists then don't do anything */ + if (raw_entries[column].processed) + continue; + + lan966x_mac_process_raw_entry(&raw_entries[column], + mac, &vid, &dest_idx); + WARN_ON(dest_idx > lan966x->num_phys_ports); + + mac_entry = lan966x_mac_alloc_entry(mac, vid, dest_idx); + if (!mac_entry) + return; + + mac_entry->row = row; + + spin_lock(&lan966x->mac_lock); + list_add_tail(&mac_entry->list, &lan966x->mac_entries); + spin_unlock(&lan966x->mac_lock); + + lan966x_mac_notifiers(SWITCHDEV_FDB_ADD_TO_BRIDGE, + mac, vid, lan966x->ports[dest_idx]->dev); + } +} + +irqreturn_t lan966x_mac_irq_handler(struct lan966x *lan966x) +{ + struct lan966x_mac_raw_entry entry[LAN966X_MAC_COLUMNS] = { 0 }; + u32 index, column; + bool stop = true; + u32 val; + + /* Start the scan from 0, 0 */ + lan_wr(ANA_MACTINDX_M_INDEX_SET(0) | + ANA_MACTINDX_BUCKET_SET(0), + lan966x, ANA_MACTINDX); + + while (1) { + lan_rmw(ANA_MACACCESS_MAC_TABLE_CMD_SET(MACACCESS_CMD_SYNC_GET_NEXT), + ANA_MACACCESS_MAC_TABLE_CMD, + lan966x, ANA_MACACCESS); + lan966x_mac_wait_for_completion(lan966x); + + val = lan_rd(lan966x, ANA_MACTINDX); + index = ANA_MACTINDX_M_INDEX_GET(val); + column = ANA_MACTINDX_BUCKET_GET(val); + + /* The SYNC-GET-NEXT returns all the entries(4) in a row in + * which is suffered a change. By change it means that new entry + * was added or an entry was removed because of ageing. + * It would return all the columns for that row. And after that + * it would return the next row The stop conditions of the + * SYNC-GET-NEXT is when it reaches 'directly' to row 0 + * column 3. So if SYNC-GET-NEXT returns row 0 and column 0 + * then it is required to continue to read more even if it + * reaches row 0 and column 3. + */ + if (index == 0 && column == 0) + stop = false; + + if (column == LAN966X_MAC_COLUMNS - 1 && + index == 0 && stop) + break; + + entry[column].mach = lan_rd(lan966x, ANA_MACHDATA); + entry[column].macl = lan_rd(lan966x, ANA_MACLDATA); + entry[column].maca = lan_rd(lan966x, ANA_MACACCESS); + + /* Once all the columns are read process them */ + if (column == LAN966X_MAC_COLUMNS - 1) { + lan966x_mac_irq_process(lan966x, index, entry); + /* A row was processed so it is safe to assume that the + * next row/column can be the stop condition + */ + stop = true; + } + } + + lan_rmw(ANA_ANAINTR_INTR_SET(0), + ANA_ANAINTR_INTR, + lan966x, ANA_ANAINTR); + + return IRQ_HANDLED; } diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c index 101c1f005baf..16f4d8737d7b 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c @@ -108,12 +108,12 @@ static int lan966x_port_set_mac_address(struct net_device *dev, void *p) int ret; /* Learn the new net device MAC address in the mac table. */ - ret = lan966x_mac_cpu_learn(lan966x, addr->sa_data, port->pvid); + ret = lan966x_mac_cpu_learn(lan966x, addr->sa_data, HOST_PVID); if (ret) return ret; /* Then forget the previous one. */ - ret = lan966x_mac_cpu_forget(lan966x, dev->dev_addr, port->pvid); + ret = lan966x_mac_cpu_forget(lan966x, dev->dev_addr, HOST_PVID); if (ret) return ret; @@ -283,6 +283,12 @@ static void lan966x_ifh_set_ipv(void *ifh, u64 bypass) IFH_POS_IPV, IFH_LEN * 4, PACK, 0); } +static void lan966x_ifh_set_vid(void *ifh, u64 vid) +{ + packing(ifh, &vid, IFH_POS_TCI + IFH_WID_TCI - 1, + IFH_POS_TCI, IFH_LEN * 4, PACK, 0); +} + static int lan966x_port_xmit(struct sk_buff *skb, struct net_device *dev) { struct lan966x_port *port = netdev_priv(dev); @@ -294,32 +300,11 @@ static int lan966x_port_xmit(struct sk_buff *skb, struct net_device *dev) lan966x_ifh_set_port(ifh, BIT_ULL(port->chip_port)); lan966x_ifh_set_qos_class(ifh, skb->priority >= 7 ? 0x7 : skb->priority); lan966x_ifh_set_ipv(ifh, skb->priority >= 7 ? 0x7 : skb->priority); + lan966x_ifh_set_vid(ifh, skb_vlan_tag_get(skb)); return lan966x_port_ifh_xmit(skb, ifh, dev); } -static void lan966x_set_promisc(struct lan966x_port *port, bool enable) -{ - struct lan966x *lan966x = port->lan966x; - - lan_rmw(ANA_CPU_FWD_CFG_SRC_COPY_ENA_SET(enable), - ANA_CPU_FWD_CFG_SRC_COPY_ENA, - lan966x, ANA_CPU_FWD_CFG(port->chip_port)); -} - -static void lan966x_port_change_rx_flags(struct net_device *dev, int flags) -{ - struct lan966x_port *port = netdev_priv(dev); - - if (!(flags & IFF_PROMISC)) - return; - - if (dev->flags & IFF_PROMISC) - lan966x_set_promisc(port, true); - else - lan966x_set_promisc(port, false); -} - static int lan966x_port_change_mtu(struct net_device *dev, int new_mtu) { struct lan966x_port *port = netdev_priv(dev); @@ -369,7 +354,6 @@ static const struct net_device_ops lan966x_port_netdev_ops = { .ndo_open = lan966x_port_open, .ndo_stop = lan966x_port_stop, .ndo_start_xmit = lan966x_port_xmit, - .ndo_change_rx_flags = lan966x_port_change_rx_flags, .ndo_change_mtu = lan966x_port_change_mtu, .ndo_set_rx_mode = lan966x_port_set_rx_mode, .ndo_get_phys_port_name = lan966x_port_get_phys_port_name, @@ -378,6 +362,11 @@ static const struct net_device_ops lan966x_port_netdev_ops = { .ndo_get_port_parent_id = lan966x_port_get_parent_id, }; +bool lan966x_netdevice_check(const struct net_device *dev) +{ + return dev->netdev_ops == &lan966x_port_netdev_ops; +} + static int lan966x_port_xtr_status(struct lan966x *lan966x, u8 grp) { return lan_rd(lan966x, QS_XTR_RD(grp)); @@ -514,6 +503,9 @@ static irqreturn_t lan966x_xtr_irq_handler(int irq, void *args) skb->protocol = eth_type_trans(skb, dev); + if (lan966x->bridge_mask & BIT(src_port)) + skb->offload_fwd_mark = 1; + netif_rx_ni(skb); dev->stats.rx_bytes += len; dev->stats.rx_packets++; @@ -527,6 +519,13 @@ recover: return IRQ_HANDLED; } +static irqreturn_t lan966x_ana_irq_handler(int irq, void *args) +{ + struct lan966x *lan966x = args; + + return lan966x_mac_irq_handler(lan966x); +} + static void lan966x_cleanup_ports(struct lan966x *lan966x) { struct lan966x_port *port; @@ -554,6 +553,11 @@ static void lan966x_cleanup_ports(struct lan966x *lan966x) disable_irq(lan966x->xtr_irq); lan966x->xtr_irq = -ENXIO; + + if (lan966x->ana_irq) { + disable_irq(lan966x->ana_irq); + lan966x->ana_irq = -ENXIO; + } } static int lan966x_probe_port(struct lan966x *lan966x, u32 p, @@ -578,13 +582,14 @@ static int lan966x_probe_port(struct lan966x *lan966x, u32 p, port->dev = dev; port->lan966x = lan966x; port->chip_port = p; - port->pvid = PORT_PVID; lan966x->ports[p] = port; dev->max_mtu = ETH_MAX_MTU; dev->netdev_ops = &lan966x_port_netdev_ops; dev->ethtool_ops = &lan966x_ethtool_ops; + dev->features |= NETIF_F_HW_VLAN_CTAG_TX | + NETIF_F_HW_VLAN_STAG_TX; dev->needed_headroom = IFH_LEN * sizeof(u32); eth_hw_addr_gen(dev, lan966x->base_mac, p + 1); @@ -631,6 +636,10 @@ static int lan966x_probe_port(struct lan966x *lan966x, u32 p, return err; } + lan966x_vlan_port_set_vlan_aware(port, 0); + lan966x_vlan_port_set_vid(port, HOST_PVID, false, false); + lan966x_vlan_port_apply(port); + return 0; } @@ -641,6 +650,8 @@ static void lan966x_init(struct lan966x *lan966x) /* MAC table initialization */ lan966x_mac_init(lan966x); + lan966x_vlan_init(lan966x); + /* Flush queues */ lan_wr(lan_rd(lan966x, QS_XTR_FLUSH) | GENMASK(1, 0), @@ -704,8 +715,10 @@ static void lan966x_init(struct lan966x *lan966x) /* There are 8 priorities */ for (i = 0; i < 8; ++i) lan_rmw(ANA_FLOODING_FLD_MULTICAST_SET(PGID_MC) | + ANA_FLOODING_FLD_UNICAST_SET(PGID_UC) | ANA_FLOODING_FLD_BROADCAST_SET(PGID_BC), ANA_FLOODING_FLD_MULTICAST | + ANA_FLOODING_FLD_UNICAST | ANA_FLOODING_FLD_BROADCAST, lan966x, ANA_FLOODING(i)); @@ -757,6 +770,11 @@ static void lan966x_init(struct lan966x *lan966x) ANA_PGID_PGID, lan966x, ANA_PGID(PGID_MCIPV4)); + /* Unicast to all other ports */ + lan_rmw(GENMASK(lan966x->num_phys_ports - 1, 0), + ANA_PGID_PGID, + lan966x, ANA_PGID(PGID_UC)); + /* Broadcast to the CPU port and to other ports */ lan_rmw(ANA_PGID_PGID_SET(BIT(CPU_PORT) | GENMASK(lan966x->num_phys_ports - 1, 0)), ANA_PGID_PGID, @@ -870,6 +888,15 @@ static int lan966x_probe(struct platform_device *pdev) return -ENODEV; } + lan966x->ana_irq = platform_get_irq_byname(pdev, "ana"); + if (lan966x->ana_irq) { + err = devm_request_threaded_irq(&pdev->dev, lan966x->ana_irq, NULL, + lan966x_ana_irq_handler, IRQF_ONESHOT, + "ana irq", lan966x); + if (err) + return dev_err_probe(&pdev->dev, err, "Unable to use ana irq"); + } + /* init switch */ lan966x_init(lan966x); lan966x_stats_init(lan966x); @@ -899,6 +926,10 @@ static int lan966x_probe(struct platform_device *pdev) lan966x_port_init(lan966x->ports[p]); } + err = lan966x_fdb_init(lan966x); + if (err) + goto cleanup_ports; + return 0; cleanup_ports: @@ -923,6 +954,9 @@ static int lan966x_remove(struct platform_device *pdev) destroy_workqueue(lan966x->stats_queue); mutex_destroy(&lan966x->stats_lock); + lan966x_mac_purge_entries(lan966x); + lan966x_fdb_deinit(lan966x); + return 0; } @@ -934,7 +968,32 @@ static struct platform_driver lan966x_driver = { .of_match_table = lan966x_match, }, }; -module_platform_driver(lan966x_driver); + +static int __init lan966x_switch_driver_init(void) +{ + int ret; + + lan966x_register_notifier_blocks(); + + ret = platform_driver_register(&lan966x_driver); + if (ret) + goto err; + + return 0; + +err: + lan966x_unregister_notifier_blocks(); + return ret; +} + +static void __exit lan966x_switch_driver_exit(void) +{ + platform_driver_unregister(&lan966x_driver); + lan966x_unregister_notifier_blocks(); +} + +module_init(lan966x_switch_driver_init); +module_exit(lan966x_switch_driver_exit); MODULE_DESCRIPTION("Microchip LAN966X switch driver"); MODULE_AUTHOR("Horatiu Vultur <horatiu.vultur@microchip.com>"); diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h index 7e5a3b6f168d..c399b1256edc 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.h +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.h @@ -4,9 +4,11 @@ #define __LAN966X_MAIN_H__ #include <linux/etherdevice.h> +#include <linux/if_vlan.h> #include <linux/jiffies.h> #include <linux/phy.h> #include <linux/phylink.h> +#include <net/switchdev.h> #include "lan966x_regs.h" #include "lan966x_ifh.h" @@ -22,7 +24,8 @@ #define PGID_SRC 80 #define PGID_ENTRIES 89 -#define PORT_PVID 0 +#define UNAWARE_PVID 0 +#define HOST_PVID 4095 /* Reserved amount for (SRC, PRIO) at index 8*SRC + PRIO */ #define QSYS_Q_RSRV 95 @@ -75,6 +78,16 @@ struct lan966x { u8 base_mac[ETH_ALEN]; + struct net_device *bridge; + u16 bridge_mask; + u16 bridge_fwd_mask; + + struct list_head mac_entries; + spinlock_t mac_lock; /* lock for mac_entries list */ + + u16 vlan_mask[VLAN_N_VID]; + DECLARE_BITMAP(cpu_vlan_mask, VLAN_N_VID); + /* stats */ const struct lan966x_stat_layout *stats_layout; u32 num_stats; @@ -87,6 +100,11 @@ struct lan966x { /* interrupts */ int xtr_irq; + int ana_irq; + + /* worqueue for fdb */ + struct workqueue_struct *fdb_work; + struct list_head fdb_entries; }; struct lan966x_port_config { @@ -105,6 +123,10 @@ struct lan966x_port { u8 chip_port; u16 pvid; + u16 vid; + bool vlan_aware; + + bool learn_ena; struct phylink_config phylink_config; struct phylink_pcs phylink_pcs; @@ -118,6 +140,11 @@ extern const struct phylink_mac_ops lan966x_phylink_mac_ops; extern const struct phylink_pcs_ops lan966x_phylink_pcs_ops; extern const struct ethtool_ops lan966x_ethtool_ops; +bool lan966x_netdevice_check(const struct net_device *dev); + +void lan966x_register_notifier_blocks(void); +void lan966x_unregister_notifier_blocks(void); + void lan966x_stats_get(struct net_device *dev, struct rtnl_link_stats64 *stats); int lan966x_stats_init(struct lan966x *lan966x); @@ -141,6 +168,43 @@ int lan966x_mac_forget(struct lan966x *lan966x, int lan966x_mac_cpu_learn(struct lan966x *lan966x, const char *addr, u16 vid); int lan966x_mac_cpu_forget(struct lan966x *lan966x, const char *addr, u16 vid); void lan966x_mac_init(struct lan966x *lan966x); +void lan966x_mac_set_ageing(struct lan966x *lan966x, + u32 ageing); +int lan966x_mac_del_entry(struct lan966x *lan966x, + const unsigned char *addr, + u16 vid); +int lan966x_mac_add_entry(struct lan966x *lan966x, + struct lan966x_port *port, + const unsigned char *addr, + u16 vid); +void lan966x_mac_purge_entries(struct lan966x *lan966x); +irqreturn_t lan966x_mac_irq_handler(struct lan966x *lan966x); + +void lan966x_vlan_init(struct lan966x *lan966x); +void lan966x_vlan_port_apply(struct lan966x_port *port); +bool lan966x_vlan_cpu_member_cpu_vlan_mask(struct lan966x *lan966x, u16 vid); +void lan966x_vlan_port_set_vlan_aware(struct lan966x_port *port, + bool vlan_aware); +int lan966x_vlan_port_set_vid(struct lan966x_port *port, + u16 vid, + bool pvid, + bool untagged); +void lan966x_vlan_port_add_vlan(struct lan966x_port *port, + u16 vid, + bool pvid, + bool untagged); +void lan966x_vlan_port_del_vlan(struct lan966x_port *port, u16 vid); +void lan966x_vlan_cpu_add_vlan(struct lan966x *lan966x, u16 vid); +void lan966x_vlan_cpu_del_vlan(struct lan966x *lan966x, u16 vid); + +void lan966x_fdb_write_entries(struct lan966x *lan966x, u16 vid); +void lan966x_fdb_erase_entries(struct lan966x *lan966x, u16 vid); +int lan966x_fdb_init(struct lan966x *lan966x); +void lan966x_fdb_deinit(struct lan966x *lan966x); +int lan966x_handle_fdb(struct net_device *dev, + struct net_device *orig_dev, + unsigned long event, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info); static inline void __iomem *lan_addr(void __iomem *base[], int id, int tinst, int tcnt, diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h b/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h index 879dcd807dec..a13c469e139a 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_regs.h @@ -61,6 +61,9 @@ enum lan966x_target { #define ANA_ADVLEARN_VLAN_CHK_GET(x)\ FIELD_GET(ANA_ADVLEARN_VLAN_CHK, x) +/* ANA:ANA:VLANMASK */ +#define ANA_VLANMASK __REG(TARGET_ANA, 0, 1, 29824, 0, 1, 244, 8, 0, 1, 4) + /* ANA:ANA:ANAINTR */ #define ANA_ANAINTR __REG(TARGET_ANA, 0, 1, 29824, 0, 1, 244, 16, 0, 1, 4) @@ -88,6 +91,12 @@ enum lan966x_target { /* ANA:ANA:FLOODING */ #define ANA_FLOODING(r) __REG(TARGET_ANA, 0, 1, 29824, 0, 1, 244, 68, r, 8, 4) +#define ANA_FLOODING_FLD_UNICAST GENMASK(17, 12) +#define ANA_FLOODING_FLD_UNICAST_SET(x)\ + FIELD_PREP(ANA_FLOODING_FLD_UNICAST, x) +#define ANA_FLOODING_FLD_UNICAST_GET(x)\ + FIELD_GET(ANA_FLOODING_FLD_UNICAST, x) + #define ANA_FLOODING_FLD_BROADCAST GENMASK(11, 6) #define ANA_FLOODING_FLD_BROADCAST_SET(x)\ FIELD_PREP(ANA_FLOODING_FLD_BROADCAST, x) @@ -184,6 +193,102 @@ enum lan966x_target { #define ANA_MACACCESS_MAC_TABLE_CMD_GET(x)\ FIELD_GET(ANA_MACACCESS_MAC_TABLE_CMD, x) +/* ANA:ANA_TABLES:MACTINDX */ +#define ANA_MACTINDX __REG(TARGET_ANA, 0, 1, 27520, 0, 1, 128, 52, 0, 1, 4) + +#define ANA_MACTINDX_BUCKET GENMASK(12, 11) +#define ANA_MACTINDX_BUCKET_SET(x)\ + FIELD_PREP(ANA_MACTINDX_BUCKET, x) +#define ANA_MACTINDX_BUCKET_GET(x)\ + FIELD_GET(ANA_MACTINDX_BUCKET, x) + +#define ANA_MACTINDX_M_INDEX GENMASK(10, 0) +#define ANA_MACTINDX_M_INDEX_SET(x)\ + FIELD_PREP(ANA_MACTINDX_M_INDEX, x) +#define ANA_MACTINDX_M_INDEX_GET(x)\ + FIELD_GET(ANA_MACTINDX_M_INDEX, x) + +/* ANA:ANA_TABLES:VLAN_PORT_MASK */ +#define ANA_VLAN_PORT_MASK __REG(TARGET_ANA, 0, 1, 27520, 0, 1, 128, 56, 0, 1, 4) + +#define ANA_VLAN_PORT_MASK_VLAN_PORT_MASK GENMASK(8, 0) +#define ANA_VLAN_PORT_MASK_VLAN_PORT_MASK_SET(x)\ + FIELD_PREP(ANA_VLAN_PORT_MASK_VLAN_PORT_MASK, x) +#define ANA_VLAN_PORT_MASK_VLAN_PORT_MASK_GET(x)\ + FIELD_GET(ANA_VLAN_PORT_MASK_VLAN_PORT_MASK, x) + +/* ANA:ANA_TABLES:VLANACCESS */ +#define ANA_VLANACCESS __REG(TARGET_ANA, 0, 1, 27520, 0, 1, 128, 60, 0, 1, 4) + +#define ANA_VLANACCESS_VLAN_TBL_CMD GENMASK(1, 0) +#define ANA_VLANACCESS_VLAN_TBL_CMD_SET(x)\ + FIELD_PREP(ANA_VLANACCESS_VLAN_TBL_CMD, x) +#define ANA_VLANACCESS_VLAN_TBL_CMD_GET(x)\ + FIELD_GET(ANA_VLANACCESS_VLAN_TBL_CMD, x) + +/* ANA:ANA_TABLES:VLANTIDX */ +#define ANA_VLANTIDX __REG(TARGET_ANA, 0, 1, 27520, 0, 1, 128, 64, 0, 1, 4) + +#define ANA_VLANTIDX_VLAN_PGID_CPU_DIS BIT(18) +#define ANA_VLANTIDX_VLAN_PGID_CPU_DIS_SET(x)\ + FIELD_PREP(ANA_VLANTIDX_VLAN_PGID_CPU_DIS, x) +#define ANA_VLANTIDX_VLAN_PGID_CPU_DIS_GET(x)\ + FIELD_GET(ANA_VLANTIDX_VLAN_PGID_CPU_DIS, x) + +#define ANA_VLANTIDX_V_INDEX GENMASK(11, 0) +#define ANA_VLANTIDX_V_INDEX_SET(x)\ + FIELD_PREP(ANA_VLANTIDX_V_INDEX, x) +#define ANA_VLANTIDX_V_INDEX_GET(x)\ + FIELD_GET(ANA_VLANTIDX_V_INDEX, x) + +/* ANA:PORT:VLAN_CFG */ +#define ANA_VLAN_CFG(g) __REG(TARGET_ANA, 0, 1, 28672, g, 9, 128, 0, 0, 1, 4) + +#define ANA_VLAN_CFG_VLAN_AWARE_ENA BIT(20) +#define ANA_VLAN_CFG_VLAN_AWARE_ENA_SET(x)\ + FIELD_PREP(ANA_VLAN_CFG_VLAN_AWARE_ENA, x) +#define ANA_VLAN_CFG_VLAN_AWARE_ENA_GET(x)\ + FIELD_GET(ANA_VLAN_CFG_VLAN_AWARE_ENA, x) + +#define ANA_VLAN_CFG_VLAN_POP_CNT GENMASK(19, 18) +#define ANA_VLAN_CFG_VLAN_POP_CNT_SET(x)\ + FIELD_PREP(ANA_VLAN_CFG_VLAN_POP_CNT, x) +#define ANA_VLAN_CFG_VLAN_POP_CNT_GET(x)\ + FIELD_GET(ANA_VLAN_CFG_VLAN_POP_CNT, x) + +#define ANA_VLAN_CFG_VLAN_VID GENMASK(11, 0) +#define ANA_VLAN_CFG_VLAN_VID_SET(x)\ + FIELD_PREP(ANA_VLAN_CFG_VLAN_VID, x) +#define ANA_VLAN_CFG_VLAN_VID_GET(x)\ + FIELD_GET(ANA_VLAN_CFG_VLAN_VID, x) + +/* ANA:PORT:DROP_CFG */ +#define ANA_DROP_CFG(g) __REG(TARGET_ANA, 0, 1, 28672, g, 9, 128, 4, 0, 1, 4) + +#define ANA_DROP_CFG_DROP_UNTAGGED_ENA BIT(6) +#define ANA_DROP_CFG_DROP_UNTAGGED_ENA_SET(x)\ + FIELD_PREP(ANA_DROP_CFG_DROP_UNTAGGED_ENA, x) +#define ANA_DROP_CFG_DROP_UNTAGGED_ENA_GET(x)\ + FIELD_GET(ANA_DROP_CFG_DROP_UNTAGGED_ENA, x) + +#define ANA_DROP_CFG_DROP_PRIO_S_TAGGED_ENA BIT(3) +#define ANA_DROP_CFG_DROP_PRIO_S_TAGGED_ENA_SET(x)\ + FIELD_PREP(ANA_DROP_CFG_DROP_PRIO_S_TAGGED_ENA, x) +#define ANA_DROP_CFG_DROP_PRIO_S_TAGGED_ENA_GET(x)\ + FIELD_GET(ANA_DROP_CFG_DROP_PRIO_S_TAGGED_ENA, x) + +#define ANA_DROP_CFG_DROP_PRIO_C_TAGGED_ENA BIT(2) +#define ANA_DROP_CFG_DROP_PRIO_C_TAGGED_ENA_SET(x)\ + FIELD_PREP(ANA_DROP_CFG_DROP_PRIO_C_TAGGED_ENA, x) +#define ANA_DROP_CFG_DROP_PRIO_C_TAGGED_ENA_GET(x)\ + FIELD_GET(ANA_DROP_CFG_DROP_PRIO_C_TAGGED_ENA, x) + +#define ANA_DROP_CFG_DROP_MC_SMAC_ENA BIT(0) +#define ANA_DROP_CFG_DROP_MC_SMAC_ENA_SET(x)\ + FIELD_PREP(ANA_DROP_CFG_DROP_MC_SMAC_ENA, x) +#define ANA_DROP_CFG_DROP_MC_SMAC_ENA_GET(x)\ + FIELD_GET(ANA_DROP_CFG_DROP_MC_SMAC_ENA, x) + /* ANA:PORT:CPU_FWD_CFG */ #define ANA_CPU_FWD_CFG(g) __REG(TARGET_ANA, 0, 1, 28672, g, 9, 128, 96, 0, 1, 4) @@ -589,6 +694,36 @@ enum lan966x_target { /* QSYS:RES_CTRL:RES_CFG */ #define QSYS_RES_CFG(g) __REG(TARGET_QSYS, 0, 1, 32768, g, 1024, 8, 0, 0, 1, 4) +/* REW:PORT:PORT_VLAN_CFG */ +#define REW_PORT_VLAN_CFG(g) __REG(TARGET_REW, 0, 1, 0, g, 10, 128, 0, 0, 1, 4) + +#define REW_PORT_VLAN_CFG_PORT_TPID GENMASK(31, 16) +#define REW_PORT_VLAN_CFG_PORT_TPID_SET(x)\ + FIELD_PREP(REW_PORT_VLAN_CFG_PORT_TPID, x) +#define REW_PORT_VLAN_CFG_PORT_TPID_GET(x)\ + FIELD_GET(REW_PORT_VLAN_CFG_PORT_TPID, x) + +#define REW_PORT_VLAN_CFG_PORT_VID GENMASK(11, 0) +#define REW_PORT_VLAN_CFG_PORT_VID_SET(x)\ + FIELD_PREP(REW_PORT_VLAN_CFG_PORT_VID, x) +#define REW_PORT_VLAN_CFG_PORT_VID_GET(x)\ + FIELD_GET(REW_PORT_VLAN_CFG_PORT_VID, x) + +/* REW:PORT:TAG_CFG */ +#define REW_TAG_CFG(g) __REG(TARGET_REW, 0, 1, 0, g, 10, 128, 4, 0, 1, 4) + +#define REW_TAG_CFG_TAG_CFG GENMASK(8, 7) +#define REW_TAG_CFG_TAG_CFG_SET(x)\ + FIELD_PREP(REW_TAG_CFG_TAG_CFG, x) +#define REW_TAG_CFG_TAG_CFG_GET(x)\ + FIELD_GET(REW_TAG_CFG_TAG_CFG, x) + +#define REW_TAG_CFG_TAG_TPID_CFG GENMASK(6, 5) +#define REW_TAG_CFG_TAG_TPID_CFG_SET(x)\ + FIELD_PREP(REW_TAG_CFG_TAG_TPID_CFG, x) +#define REW_TAG_CFG_TAG_TPID_CFG_GET(x)\ + FIELD_GET(REW_TAG_CFG_TAG_TPID_CFG, x) + /* REW:PORT:PORT_CFG */ #define REW_PORT_CFG(g) __REG(TARGET_REW, 0, 1, 0, g, 10, 128, 8, 0, 1, 4) diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_switchdev.c b/drivers/net/ethernet/microchip/lan966x/lan966x_switchdev.c new file mode 100644 index 000000000000..deb3dd5be67a --- /dev/null +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_switchdev.c @@ -0,0 +1,536 @@ +// SPDX-License-Identifier: GPL-2.0+ + +#include <linux/if_bridge.h> +#include <net/switchdev.h> + +#include "lan966x_main.h" + +static struct notifier_block lan966x_netdevice_nb __read_mostly; +static struct notifier_block lan966x_switchdev_nb __read_mostly; +static struct notifier_block lan966x_switchdev_blocking_nb __read_mostly; + +static void lan966x_port_set_mcast_flood(struct lan966x_port *port, + bool enabled) +{ + u32 val = lan_rd(port->lan966x, ANA_PGID(PGID_MC)); + + val = ANA_PGID_PGID_GET(val); + if (enabled) + val |= BIT(port->chip_port); + else + val &= ~BIT(port->chip_port); + + lan_rmw(ANA_PGID_PGID_SET(val), + ANA_PGID_PGID, + port->lan966x, ANA_PGID(PGID_MC)); +} + +static void lan966x_port_set_ucast_flood(struct lan966x_port *port, + bool enabled) +{ + u32 val = lan_rd(port->lan966x, ANA_PGID(PGID_UC)); + + val = ANA_PGID_PGID_GET(val); + if (enabled) + val |= BIT(port->chip_port); + else + val &= ~BIT(port->chip_port); + + lan_rmw(ANA_PGID_PGID_SET(val), + ANA_PGID_PGID, + port->lan966x, ANA_PGID(PGID_UC)); +} + +static void lan966x_port_set_bcast_flood(struct lan966x_port *port, + bool enabled) +{ + u32 val = lan_rd(port->lan966x, ANA_PGID(PGID_BC)); + + val = ANA_PGID_PGID_GET(val); + if (enabled) + val |= BIT(port->chip_port); + else + val &= ~BIT(port->chip_port); + + lan_rmw(ANA_PGID_PGID_SET(val), + ANA_PGID_PGID, + port->lan966x, ANA_PGID(PGID_BC)); +} + +static void lan966x_port_set_learning(struct lan966x_port *port, bool enabled) +{ + lan_rmw(ANA_PORT_CFG_LEARN_ENA_SET(enabled), + ANA_PORT_CFG_LEARN_ENA, + port->lan966x, ANA_PORT_CFG(port->chip_port)); + + port->learn_ena = enabled; +} + +static void lan966x_port_bridge_flags(struct lan966x_port *port, + struct switchdev_brport_flags flags) +{ + if (flags.mask & BR_MCAST_FLOOD) + lan966x_port_set_mcast_flood(port, + !!(flags.val & BR_MCAST_FLOOD)); + + if (flags.mask & BR_FLOOD) + lan966x_port_set_ucast_flood(port, + !!(flags.val & BR_FLOOD)); + + if (flags.mask & BR_BCAST_FLOOD) + lan966x_port_set_bcast_flood(port, + !!(flags.val & BR_BCAST_FLOOD)); + + if (flags.mask & BR_LEARNING) + lan966x_port_set_learning(port, + !!(flags.val & BR_LEARNING)); +} + +static int lan966x_port_pre_bridge_flags(struct lan966x_port *port, + struct switchdev_brport_flags flags) +{ + if (flags.mask & ~(BR_MCAST_FLOOD | BR_FLOOD | BR_BCAST_FLOOD | + BR_LEARNING)) + return -EINVAL; + + return 0; +} + +static void lan966x_update_fwd_mask(struct lan966x *lan966x) +{ + int i; + + for (i = 0; i < lan966x->num_phys_ports; i++) { + struct lan966x_port *port = lan966x->ports[i]; + unsigned long mask = 0; + + if (port && lan966x->bridge_fwd_mask & BIT(i)) + mask = lan966x->bridge_fwd_mask & ~BIT(i); + + mask |= BIT(CPU_PORT); + + lan_wr(ANA_PGID_PGID_SET(mask), + lan966x, ANA_PGID(PGID_SRC + i)); + } +} + +static void lan966x_port_stp_state_set(struct lan966x_port *port, u8 state) +{ + struct lan966x *lan966x = port->lan966x; + bool learn_ena = false; + + if ((state == BR_STATE_FORWARDING || state == BR_STATE_LEARNING) && + port->learn_ena) + learn_ena = true; + + if (state == BR_STATE_FORWARDING) + lan966x->bridge_fwd_mask |= BIT(port->chip_port); + else + lan966x->bridge_fwd_mask &= ~BIT(port->chip_port); + + lan_rmw(ANA_PORT_CFG_LEARN_ENA_SET(learn_ena), + ANA_PORT_CFG_LEARN_ENA, + lan966x, ANA_PORT_CFG(port->chip_port)); + + lan966x_update_fwd_mask(lan966x); +} + +static void lan966x_port_ageing_set(struct lan966x_port *port, + unsigned long ageing_clock_t) +{ + unsigned long ageing_jiffies = clock_t_to_jiffies(ageing_clock_t); + u32 ageing_time = jiffies_to_msecs(ageing_jiffies) / 1000; + + lan966x_mac_set_ageing(port->lan966x, ageing_time); +} + +static int lan966x_port_attr_set(struct net_device *dev, const void *ctx, + const struct switchdev_attr *attr, + struct netlink_ext_ack *extack) +{ + struct lan966x_port *port = netdev_priv(dev); + int err = 0; + + if (ctx && ctx != port) + return 0; + + switch (attr->id) { + case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS: + lan966x_port_bridge_flags(port, attr->u.brport_flags); + break; + case SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS: + err = lan966x_port_pre_bridge_flags(port, attr->u.brport_flags); + break; + case SWITCHDEV_ATTR_ID_PORT_STP_STATE: + lan966x_port_stp_state_set(port, attr->u.stp_state); + break; + case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME: + lan966x_port_ageing_set(port, attr->u.ageing_time); + break; + case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING: + lan966x_vlan_port_set_vlan_aware(port, attr->u.vlan_filtering); + lan966x_vlan_port_apply(port); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static int lan966x_port_bridge_join(struct lan966x_port *port, + struct net_device *bridge, + struct netlink_ext_ack *extack) +{ + struct switchdev_brport_flags flags = {0}; + struct lan966x *lan966x = port->lan966x; + struct net_device *dev = port->dev; + int err; + + if (!lan966x->bridge_mask) { + lan966x->bridge = bridge; + } else { + if (lan966x->bridge != bridge) { + NL_SET_ERR_MSG_MOD(extack, "Not allow to add port to different bridge"); + return -ENODEV; + } + } + + err = switchdev_bridge_port_offload(dev, dev, port, + &lan966x_switchdev_nb, + &lan966x_switchdev_blocking_nb, + false, extack); + if (err) + return err; + + lan966x->bridge_mask |= BIT(port->chip_port); + + flags.mask = BR_LEARNING | BR_FLOOD | BR_MCAST_FLOOD | BR_BCAST_FLOOD; + flags.val = flags.mask; + lan966x_port_bridge_flags(port, flags); + + return 0; +} + +static void lan966x_port_bridge_leave(struct lan966x_port *port, + struct net_device *bridge) +{ + struct switchdev_brport_flags flags = {0}; + struct lan966x *lan966x = port->lan966x; + + flags.mask = BR_LEARNING | BR_FLOOD | BR_MCAST_FLOOD | BR_BCAST_FLOOD; + flags.val = flags.mask & ~BR_LEARNING; + lan966x_port_bridge_flags(port, flags); + + lan966x->bridge_mask &= ~BIT(port->chip_port); + + if (!lan966x->bridge_mask) + lan966x->bridge = NULL; + + /* Set the port back to host mode */ + lan966x_vlan_port_set_vlan_aware(port, false); + lan966x_vlan_port_set_vid(port, HOST_PVID, false, false); + lan966x_vlan_port_apply(port); +} + +static int lan966x_port_changeupper(struct net_device *dev, + struct netdev_notifier_changeupper_info *info) +{ + struct lan966x_port *port = netdev_priv(dev); + struct netlink_ext_ack *extack; + int err = 0; + + extack = netdev_notifier_info_to_extack(&info->info); + + if (netif_is_bridge_master(info->upper_dev)) { + if (info->linking) + err = lan966x_port_bridge_join(port, info->upper_dev, + extack); + else + lan966x_port_bridge_leave(port, info->upper_dev); + } + + return err; +} + +static int lan966x_port_prechangeupper(struct net_device *dev, + struct netdev_notifier_changeupper_info *info) +{ + struct lan966x_port *port = netdev_priv(dev); + + if (netif_is_bridge_master(info->upper_dev) && !info->linking) + switchdev_bridge_port_unoffload(port->dev, port, + &lan966x_switchdev_nb, + &lan966x_switchdev_blocking_nb); + + return NOTIFY_DONE; +} + +static int lan966x_foreign_bridging_check(struct net_device *bridge, + struct netlink_ext_ack *extack) +{ + struct lan966x *lan966x = NULL; + bool has_foreign = false; + struct net_device *dev; + struct list_head *iter; + + if (!netif_is_bridge_master(bridge)) + return 0; + + netdev_for_each_lower_dev(bridge, dev, iter) { + if (lan966x_netdevice_check(dev)) { + struct lan966x_port *port = netdev_priv(dev); + + if (lan966x) { + /* Bridge already has at least one port of a + * lan966x switch inside it, check that it's + * the same instance of the driver. + */ + if (port->lan966x != lan966x) { + NL_SET_ERR_MSG_MOD(extack, + "Bridging between multiple lan966x switches disallowed"); + return -EINVAL; + } + } else { + /* This is the first lan966x port inside this + * bridge + */ + lan966x = port->lan966x; + } + } else { + has_foreign = true; + } + + if (lan966x && has_foreign) { + NL_SET_ERR_MSG_MOD(extack, + "Bridging lan966x ports with foreign interfaces disallowed"); + return -EINVAL; + } + } + + return 0; +} + +static int lan966x_bridge_check(struct net_device *dev, + struct netdev_notifier_changeupper_info *info) +{ + return lan966x_foreign_bridging_check(info->upper_dev, + info->info.extack); +} + +static int lan966x_netdevice_port_event(struct net_device *dev, + struct notifier_block *nb, + unsigned long event, void *ptr) +{ + int err = 0; + + if (!lan966x_netdevice_check(dev)) { + if (event == NETDEV_CHANGEUPPER) + return lan966x_bridge_check(dev, ptr); + return 0; + } + + switch (event) { + case NETDEV_PRECHANGEUPPER: + err = lan966x_port_prechangeupper(dev, ptr); + break; + case NETDEV_CHANGEUPPER: + err = lan966x_bridge_check(dev, ptr); + if (err) + return err; + + err = lan966x_port_changeupper(dev, ptr); + break; + } + + return err; +} + +static int lan966x_netdevice_event(struct notifier_block *nb, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + int ret; + + ret = lan966x_netdevice_port_event(dev, nb, event, ptr); + + return notifier_from_errno(ret); +} + +static bool lan966x_foreign_dev_check(const struct net_device *dev, + const struct net_device *foreign_dev) +{ + struct lan966x_port *port = netdev_priv(dev); + struct lan966x *lan966x = port->lan966x; + + if (netif_is_bridge_master(foreign_dev)) + if (lan966x->bridge != foreign_dev) + return true; + + return false; +} + +static int lan966x_switchdev_event(struct notifier_block *nb, + unsigned long event, void *ptr) +{ + struct net_device *dev = switchdev_notifier_info_to_dev(ptr); + int err; + + switch (event) { + case SWITCHDEV_PORT_ATTR_SET: + err = switchdev_handle_port_attr_set(dev, ptr, + lan966x_netdevice_check, + lan966x_port_attr_set); + return notifier_from_errno(err); + case SWITCHDEV_FDB_ADD_TO_DEVICE: + case SWITCHDEV_FDB_DEL_TO_DEVICE: + err = switchdev_handle_fdb_event_to_device(dev, event, ptr, + lan966x_netdevice_check, + lan966x_foreign_dev_check, + lan966x_handle_fdb, + NULL); + return notifier_from_errno(err); + } + + return NOTIFY_DONE; +} + +static int lan966x_handle_port_vlan_add(struct lan966x_port *port, + const struct switchdev_obj *obj) +{ + const struct switchdev_obj_port_vlan *v = SWITCHDEV_OBJ_PORT_VLAN(obj); + struct lan966x *lan966x = port->lan966x; + + /* When adding a port to a vlan, we get a callback for the port but + * also for the bridge. When get the callback for the bridge just bail + * out. Then when the bridge is added to the vlan, then we get a + * callback here but in this case the flags has set: + * BRIDGE_VLAN_INFO_BRENTRY. In this case it means that the CPU + * port is added to the vlan, so the broadcast frames and unicast frames + * with dmac of the bridge should be foward to CPU. + */ + if (netif_is_bridge_master(obj->orig_dev) && + !(v->flags & BRIDGE_VLAN_INFO_BRENTRY)) + return 0; + + if (!netif_is_bridge_master(obj->orig_dev)) + lan966x_vlan_port_add_vlan(port, v->vid, + v->flags & BRIDGE_VLAN_INFO_PVID, + v->flags & BRIDGE_VLAN_INFO_UNTAGGED); + else + lan966x_vlan_cpu_add_vlan(lan966x, v->vid); + + return 0; +} + +static int lan966x_handle_port_obj_add(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj, + struct netlink_ext_ack *extack) +{ + struct lan966x_port *port = netdev_priv(dev); + int err; + + if (ctx && ctx != port) + return 0; + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + err = lan966x_handle_port_vlan_add(port, obj); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static int lan966x_handle_port_vlan_del(struct lan966x_port *port, + const struct switchdev_obj *obj) +{ + const struct switchdev_obj_port_vlan *v = SWITCHDEV_OBJ_PORT_VLAN(obj); + struct lan966x *lan966x = port->lan966x; + + if (!netif_is_bridge_master(obj->orig_dev)) + lan966x_vlan_port_del_vlan(port, v->vid); + else + lan966x_vlan_cpu_del_vlan(lan966x, v->vid); + + return 0; +} + +static int lan966x_handle_port_obj_del(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj) +{ + struct lan966x_port *port = netdev_priv(dev); + int err; + + if (ctx && ctx != port) + return 0; + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + err = lan966x_handle_port_vlan_del(port, obj); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static int lan966x_switchdev_blocking_event(struct notifier_block *nb, + unsigned long event, + void *ptr) +{ + struct net_device *dev = switchdev_notifier_info_to_dev(ptr); + int err; + + switch (event) { + case SWITCHDEV_PORT_OBJ_ADD: + err = switchdev_handle_port_obj_add(dev, ptr, + lan966x_netdevice_check, + lan966x_handle_port_obj_add); + return notifier_from_errno(err); + case SWITCHDEV_PORT_OBJ_DEL: + err = switchdev_handle_port_obj_del(dev, ptr, + lan966x_netdevice_check, + lan966x_handle_port_obj_del); + return notifier_from_errno(err); + case SWITCHDEV_PORT_ATTR_SET: + err = switchdev_handle_port_attr_set(dev, ptr, + lan966x_netdevice_check, + lan966x_port_attr_set); + return notifier_from_errno(err); + } + + return NOTIFY_DONE; +} + +static struct notifier_block lan966x_netdevice_nb __read_mostly = { + .notifier_call = lan966x_netdevice_event, +}; + +static struct notifier_block lan966x_switchdev_nb __read_mostly = { + .notifier_call = lan966x_switchdev_event, +}; + +static struct notifier_block lan966x_switchdev_blocking_nb __read_mostly = { + .notifier_call = lan966x_switchdev_blocking_event, +}; + +void lan966x_register_notifier_blocks(void) +{ + register_netdevice_notifier(&lan966x_netdevice_nb); + register_switchdev_notifier(&lan966x_switchdev_nb); + register_switchdev_blocking_notifier(&lan966x_switchdev_blocking_nb); +} + +void lan966x_unregister_notifier_blocks(void) +{ + unregister_switchdev_blocking_notifier(&lan966x_switchdev_blocking_nb); + unregister_switchdev_notifier(&lan966x_switchdev_nb); + unregister_netdevice_notifier(&lan966x_netdevice_nb); +} diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_vlan.c b/drivers/net/ethernet/microchip/lan966x/lan966x_vlan.c new file mode 100644 index 000000000000..057f48ddf22c --- /dev/null +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_vlan.c @@ -0,0 +1,312 @@ +// SPDX-License-Identifier: GPL-2.0+ + +#include "lan966x_main.h" + +#define VLANACCESS_CMD_IDLE 0 +#define VLANACCESS_CMD_READ 1 +#define VLANACCESS_CMD_WRITE 2 +#define VLANACCESS_CMD_INIT 3 + +static int lan966x_vlan_get_status(struct lan966x *lan966x) +{ + return lan_rd(lan966x, ANA_VLANACCESS); +} + +static int lan966x_vlan_wait_for_completion(struct lan966x *lan966x) +{ + u32 val; + + return readx_poll_timeout(lan966x_vlan_get_status, + lan966x, val, + (val & ANA_VLANACCESS_VLAN_TBL_CMD) == + VLANACCESS_CMD_IDLE, + TABLE_UPDATE_SLEEP_US, TABLE_UPDATE_TIMEOUT_US); +} + +static void lan966x_vlan_set_mask(struct lan966x *lan966x, u16 vid) +{ + u16 mask = lan966x->vlan_mask[vid]; + bool cpu_dis; + + cpu_dis = !(mask & BIT(CPU_PORT)); + + /* Set flags and the VID to configure */ + lan_rmw(ANA_VLANTIDX_VLAN_PGID_CPU_DIS_SET(cpu_dis) | + ANA_VLANTIDX_V_INDEX_SET(vid), + ANA_VLANTIDX_VLAN_PGID_CPU_DIS | + ANA_VLANTIDX_V_INDEX, + lan966x, ANA_VLANTIDX); + + /* Set the vlan port members mask */ + lan_rmw(ANA_VLAN_PORT_MASK_VLAN_PORT_MASK_SET(mask), + ANA_VLAN_PORT_MASK_VLAN_PORT_MASK, + lan966x, ANA_VLAN_PORT_MASK); + + /* Issue a write command */ + lan_rmw(ANA_VLANACCESS_VLAN_TBL_CMD_SET(VLANACCESS_CMD_WRITE), + ANA_VLANACCESS_VLAN_TBL_CMD, + lan966x, ANA_VLANACCESS); + + if (lan966x_vlan_wait_for_completion(lan966x)) + dev_err(lan966x->dev, "Vlan set mask failed\n"); +} + +static void lan966x_vlan_port_add_vlan_mask(struct lan966x_port *port, u16 vid) +{ + struct lan966x *lan966x = port->lan966x; + u8 p = port->chip_port; + + lan966x->vlan_mask[vid] |= BIT(p); + lan966x_vlan_set_mask(lan966x, vid); +} + +static void lan966x_vlan_port_del_vlan_mask(struct lan966x_port *port, u16 vid) +{ + struct lan966x *lan966x = port->lan966x; + u8 p = port->chip_port; + + lan966x->vlan_mask[vid] &= ~BIT(p); + lan966x_vlan_set_mask(lan966x, vid); +} + +static bool lan966x_vlan_port_any_vlan_mask(struct lan966x *lan966x, u16 vid) +{ + return !!(lan966x->vlan_mask[vid] & ~BIT(CPU_PORT)); +} + +static void lan966x_vlan_cpu_add_vlan_mask(struct lan966x *lan966x, u16 vid) +{ + lan966x->vlan_mask[vid] |= BIT(CPU_PORT); + lan966x_vlan_set_mask(lan966x, vid); +} + +static void lan966x_vlan_cpu_del_vlan_mask(struct lan966x *lan966x, u16 vid) +{ + lan966x->vlan_mask[vid] &= ~BIT(CPU_PORT); + lan966x_vlan_set_mask(lan966x, vid); +} + +static void lan966x_vlan_cpu_add_cpu_vlan_mask(struct lan966x *lan966x, u16 vid) +{ + __set_bit(vid, lan966x->cpu_vlan_mask); +} + +static void lan966x_vlan_cpu_del_cpu_vlan_mask(struct lan966x *lan966x, u16 vid) +{ + __clear_bit(vid, lan966x->cpu_vlan_mask); +} + +bool lan966x_vlan_cpu_member_cpu_vlan_mask(struct lan966x *lan966x, u16 vid) +{ + return test_bit(vid, lan966x->cpu_vlan_mask); +} + +static u16 lan966x_vlan_port_get_pvid(struct lan966x_port *port) +{ + struct lan966x *lan966x = port->lan966x; + + if (!(lan966x->bridge_mask & BIT(port->chip_port))) + return HOST_PVID; + + return port->vlan_aware ? port->pvid : UNAWARE_PVID; +} + +int lan966x_vlan_port_set_vid(struct lan966x_port *port, u16 vid, + bool pvid, bool untagged) +{ + struct lan966x *lan966x = port->lan966x; + + /* Egress vlan classification */ + if (untagged && port->vid != vid) { + if (port->vid) { + dev_err(lan966x->dev, + "Port already has a native VLAN: %d\n", + port->vid); + return -EBUSY; + } + port->vid = vid; + } + + /* Default ingress vlan classification */ + if (pvid) + port->pvid = vid; + + return 0; +} + +static void lan966x_vlan_port_remove_vid(struct lan966x_port *port, u16 vid) +{ + if (port->pvid == vid) + port->pvid = 0; + + if (port->vid == vid) + port->vid = 0; +} + +void lan966x_vlan_port_set_vlan_aware(struct lan966x_port *port, + bool vlan_aware) +{ + port->vlan_aware = vlan_aware; +} + +void lan966x_vlan_port_apply(struct lan966x_port *port) +{ + struct lan966x *lan966x = port->lan966x; + u16 pvid; + u32 val; + + pvid = lan966x_vlan_port_get_pvid(port); + + /* Ingress clasification (ANA_PORT_VLAN_CFG) */ + /* Default vlan to classify for untagged frames (may be zero) */ + val = ANA_VLAN_CFG_VLAN_VID_SET(pvid); + if (port->vlan_aware) + val |= ANA_VLAN_CFG_VLAN_AWARE_ENA_SET(1) | + ANA_VLAN_CFG_VLAN_POP_CNT_SET(1); + + lan_rmw(val, + ANA_VLAN_CFG_VLAN_VID | ANA_VLAN_CFG_VLAN_AWARE_ENA | + ANA_VLAN_CFG_VLAN_POP_CNT, + lan966x, ANA_VLAN_CFG(port->chip_port)); + + /* Drop frames with multicast source address */ + val = ANA_DROP_CFG_DROP_MC_SMAC_ENA_SET(1); + if (port->vlan_aware && !pvid) + /* If port is vlan-aware and tagged, drop untagged and priority + * tagged frames. + */ + val |= ANA_DROP_CFG_DROP_UNTAGGED_ENA_SET(1) | + ANA_DROP_CFG_DROP_PRIO_S_TAGGED_ENA_SET(1) | + ANA_DROP_CFG_DROP_PRIO_C_TAGGED_ENA_SET(1); + + lan_wr(val, lan966x, ANA_DROP_CFG(port->chip_port)); + + /* Egress configuration (REW_TAG_CFG): VLAN tag type to 8021Q */ + val = REW_TAG_CFG_TAG_TPID_CFG_SET(0); + if (port->vlan_aware) { + if (port->vid) + /* Tag all frames except when VID == DEFAULT_VLAN */ + val |= REW_TAG_CFG_TAG_CFG_SET(1); + else + val |= REW_TAG_CFG_TAG_CFG_SET(3); + } + + /* Update only some bits in the register */ + lan_rmw(val, + REW_TAG_CFG_TAG_TPID_CFG | REW_TAG_CFG_TAG_CFG, + lan966x, REW_TAG_CFG(port->chip_port)); + + /* Set default VLAN and tag type to 8021Q */ + lan_rmw(REW_PORT_VLAN_CFG_PORT_TPID_SET(ETH_P_8021Q) | + REW_PORT_VLAN_CFG_PORT_VID_SET(port->vid), + REW_PORT_VLAN_CFG_PORT_TPID | + REW_PORT_VLAN_CFG_PORT_VID, + lan966x, REW_PORT_VLAN_CFG(port->chip_port)); +} + +void lan966x_vlan_port_add_vlan(struct lan966x_port *port, + u16 vid, + bool pvid, + bool untagged) +{ + struct lan966x *lan966x = port->lan966x; + + /* If the CPU(br) is already part of the vlan then add the fdb + * entries in MAC table to copy the frames to the CPU(br). + * If the CPU(br) is not part of the vlan then it would + * just drop the frames. + */ + if (lan966x_vlan_cpu_member_cpu_vlan_mask(lan966x, vid)) { + lan966x_vlan_cpu_add_vlan_mask(lan966x, vid); + lan966x_fdb_write_entries(lan966x, vid); + } + + lan966x_vlan_port_set_vid(port, vid, pvid, untagged); + lan966x_vlan_port_add_vlan_mask(port, vid); + lan966x_vlan_port_apply(port); +} + +void lan966x_vlan_port_del_vlan(struct lan966x_port *port, u16 vid) +{ + struct lan966x *lan966x = port->lan966x; + + lan966x_vlan_port_remove_vid(port, vid); + lan966x_vlan_port_del_vlan_mask(port, vid); + lan966x_vlan_port_apply(port); + + /* In case there are no other ports in vlan then remove the CPU from + * that vlan but still keep it in the mask because it may be needed + * again then another port gets added in that vlan + */ + if (!lan966x_vlan_port_any_vlan_mask(lan966x, vid)) { + lan966x_vlan_cpu_del_vlan_mask(lan966x, vid); + lan966x_fdb_erase_entries(lan966x, vid); + } +} + +void lan966x_vlan_cpu_add_vlan(struct lan966x *lan966x, u16 vid) +{ + /* Add an entry in the MAC table for the CPU + * Add the CPU part of the vlan only if there is another port in that + * vlan otherwise all the broadcast frames in that vlan will go to CPU + * even if none of the ports are in the vlan and then the CPU will just + * need to discard these frames. It is required to store this + * information so when a front port is added then it would add also the + * CPU port. + */ + if (lan966x_vlan_port_any_vlan_mask(lan966x, vid)) + lan966x_vlan_cpu_add_vlan_mask(lan966x, vid); + + lan966x_vlan_cpu_add_cpu_vlan_mask(lan966x, vid); + lan966x_fdb_write_entries(lan966x, vid); +} + +void lan966x_vlan_cpu_del_vlan(struct lan966x *lan966x, u16 vid) +{ + /* Remove the CPU part of the vlan */ + lan966x_vlan_cpu_del_cpu_vlan_mask(lan966x, vid); + lan966x_vlan_cpu_del_vlan_mask(lan966x, vid); + lan966x_fdb_erase_entries(lan966x, vid); +} + +void lan966x_vlan_init(struct lan966x *lan966x) +{ + u16 port, vid; + + /* Clear VLAN table, by default all ports are members of all VLANS */ + lan_rmw(ANA_VLANACCESS_VLAN_TBL_CMD_SET(VLANACCESS_CMD_INIT), + ANA_VLANACCESS_VLAN_TBL_CMD, + lan966x, ANA_VLANACCESS); + lan966x_vlan_wait_for_completion(lan966x); + + for (vid = 1; vid < VLAN_N_VID; vid++) { + lan966x->vlan_mask[vid] = 0; + lan966x_vlan_set_mask(lan966x, vid); + } + + /* Set all the ports + cpu to be part of HOST_PVID and UNAWARE_PVID */ + lan966x->vlan_mask[HOST_PVID] = + GENMASK(lan966x->num_phys_ports - 1, 0) | BIT(CPU_PORT); + lan966x_vlan_set_mask(lan966x, HOST_PVID); + + lan966x->vlan_mask[UNAWARE_PVID] = + GENMASK(lan966x->num_phys_ports - 1, 0) | BIT(CPU_PORT); + lan966x_vlan_set_mask(lan966x, UNAWARE_PVID); + + lan966x_vlan_cpu_add_cpu_vlan_mask(lan966x, UNAWARE_PVID); + + /* Configure the CPU port to be vlan aware */ + lan_wr(ANA_VLAN_CFG_VLAN_VID_SET(0) | + ANA_VLAN_CFG_VLAN_AWARE_ENA_SET(1) | + ANA_VLAN_CFG_VLAN_POP_CNT_SET(1), + lan966x, ANA_VLAN_CFG(CPU_PORT)); + + /* Set vlan ingress filter mask to all ports */ + lan_wr(GENMASK(lan966x->num_phys_ports, 0), + lan966x, ANA_VLANMASK); + + for (port = 0; port < lan966x->num_phys_ports; port++) { + lan_wr(0, lan966x, REW_PORT_VLAN_CFG(port)); + lan_wr(0, lan966x, REW_TAG_CFG(port)); + } +} diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c b/drivers/net/ethernet/mscc/ocelot_flower.c index 58fce173f95b..beb9379424c0 100644 --- a/drivers/net/ethernet/mscc/ocelot_flower.c +++ b/drivers/net/ethernet/mscc/ocelot_flower.c @@ -303,7 +303,7 @@ static int ocelot_flower_parse_action(struct ocelot *ocelot, int port, } filter->action.police_ena = true; - pol_ix = a->police.index + ocelot->vcap_pol.base; + pol_ix = a->hw_index + ocelot->vcap_pol.base; pol_max = ocelot->vcap_pol.max; if (ocelot->vcap_pol.max2 && pol_ix > pol_max) { diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c index 224089d04d98..f97eff5afd12 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/offload.c +++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c @@ -1867,6 +1867,9 @@ nfp_flower_indr_setup_tc_cb(struct net_device *netdev, struct Qdisc *sch, void * void *data, void (*cleanup)(struct flow_block_cb *block_cb)) { + if (!netdev) + return -EOPNOTSUPP; + if (!nfp_fl_is_netdev_to_offload(netdev)) return -EOPNOTSUPP; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 18a262ef17f6..aca3c3f0f0ad 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -174,11 +174,14 @@ struct stmmac_flow_entry { /* Rx Frame Steering */ enum stmmac_rfs_type { STMMAC_RFS_T_VLAN, + STMMAC_RFS_T_LLDP, + STMMAC_RFS_T_1588, STMMAC_RFS_T_MAX, }; struct stmmac_rfs_entry { unsigned long cookie; + u16 etype; int in_use; int type; int tc; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c index d0a2b289f460..d61766eeac6d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c @@ -237,6 +237,8 @@ static int tc_rfs_init(struct stmmac_priv *priv) int i; priv->rfs_entries_max[STMMAC_RFS_T_VLAN] = 8; + priv->rfs_entries_max[STMMAC_RFS_T_LLDP] = 1; + priv->rfs_entries_max[STMMAC_RFS_T_1588] = 1; for (i = 0; i < STMMAC_RFS_T_MAX; i++) priv->rfs_entries_total += priv->rfs_entries_max[i]; @@ -451,6 +453,8 @@ static int tc_parse_flow_actions(struct stmmac_priv *priv, return 0; } +#define ETHER_TYPE_FULL_MASK cpu_to_be16(~0) + static int tc_add_basic_flow(struct stmmac_priv *priv, struct flow_cls_offload *cls, struct stmmac_flow_entry *entry) @@ -464,6 +468,7 @@ static int tc_add_basic_flow(struct stmmac_priv *priv, return -EINVAL; flow_rule_match_basic(rule, &match); + entry->ip_proto = match.key->ip_proto; return 0; } @@ -724,6 +729,114 @@ static int tc_del_vlan_flow(struct stmmac_priv *priv, return 0; } +static int tc_add_ethtype_flow(struct stmmac_priv *priv, + struct flow_cls_offload *cls) +{ + struct stmmac_rfs_entry *entry = tc_find_rfs(priv, cls, false); + struct flow_rule *rule = flow_cls_offload_flow_rule(cls); + struct flow_dissector *dissector = rule->match.dissector; + int tc = tc_classid_to_hwtc(priv->dev, cls->classid); + struct flow_match_basic match; + + if (!entry) { + entry = tc_find_rfs(priv, cls, true); + if (!entry) + return -ENOENT; + } + + /* Nothing to do here */ + if (!dissector_uses_key(dissector, FLOW_DISSECTOR_KEY_BASIC)) + return -EINVAL; + + if (tc < 0) { + netdev_err(priv->dev, "Invalid traffic class\n"); + return -EINVAL; + } + + flow_rule_match_basic(rule, &match); + + if (match.mask->n_proto) { + u16 etype = ntohs(match.key->n_proto); + + if (match.mask->n_proto != ETHER_TYPE_FULL_MASK) { + netdev_err(priv->dev, "Only full mask is supported for EthType filter"); + return -EINVAL; + } + switch (etype) { + case ETH_P_LLDP: + if (priv->rfs_entries_cnt[STMMAC_RFS_T_LLDP] >= + priv->rfs_entries_max[STMMAC_RFS_T_LLDP]) + return -ENOENT; + + entry->type = STMMAC_RFS_T_LLDP; + priv->rfs_entries_cnt[STMMAC_RFS_T_LLDP]++; + + stmmac_rx_queue_routing(priv, priv->hw, + PACKET_DCBCPQ, tc); + break; + case ETH_P_1588: + if (priv->rfs_entries_cnt[STMMAC_RFS_T_1588] >= + priv->rfs_entries_max[STMMAC_RFS_T_1588]) + return -ENOENT; + + entry->type = STMMAC_RFS_T_1588; + priv->rfs_entries_cnt[STMMAC_RFS_T_1588]++; + + stmmac_rx_queue_routing(priv, priv->hw, + PACKET_PTPQ, tc); + break; + default: + netdev_err(priv->dev, "EthType(0x%x) is not supported", etype); + return -EINVAL; + } + + entry->in_use = true; + entry->cookie = cls->cookie; + entry->tc = tc; + entry->etype = etype; + + return 0; + } + + return -EINVAL; +} + +static int tc_del_ethtype_flow(struct stmmac_priv *priv, + struct flow_cls_offload *cls) +{ + struct stmmac_rfs_entry *entry = tc_find_rfs(priv, cls, false); + + if (!entry || !entry->in_use || + entry->type < STMMAC_RFS_T_LLDP || + entry->type > STMMAC_RFS_T_1588) + return -ENOENT; + + switch (entry->etype) { + case ETH_P_LLDP: + stmmac_rx_queue_routing(priv, priv->hw, + PACKET_DCBCPQ, 0); + priv->rfs_entries_cnt[STMMAC_RFS_T_LLDP]--; + break; + case ETH_P_1588: + stmmac_rx_queue_routing(priv, priv->hw, + PACKET_PTPQ, 0); + priv->rfs_entries_cnt[STMMAC_RFS_T_1588]--; + break; + default: + netdev_err(priv->dev, "EthType(0x%x) is not supported", + entry->etype); + return -EINVAL; + } + + entry->in_use = false; + entry->cookie = 0; + entry->tc = 0; + entry->etype = 0; + entry->type = 0; + + return 0; +} + static int tc_add_flow_cls(struct stmmac_priv *priv, struct flow_cls_offload *cls) { @@ -733,6 +846,10 @@ static int tc_add_flow_cls(struct stmmac_priv *priv, if (!ret) return ret; + ret = tc_add_ethtype_flow(priv, cls); + if (!ret) + return ret; + return tc_add_vlan_flow(priv, cls); } @@ -745,6 +862,10 @@ static int tc_del_flow_cls(struct stmmac_priv *priv, if (!ret) return ret; + ret = tc_del_ethtype_flow(priv, cls); + if (!ret) + return ret; + return tc_del_vlan_flow(priv, cls); } diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 44a24b99c894..c6a97fcca0e6 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -66,6 +66,19 @@ #define KSZ8081_LMD_SHORT_INDICATOR BIT(12) #define KSZ8081_LMD_DELTA_TIME_MASK GENMASK(8, 0) +/* Lan8814 general Interrupt control/status reg in GPHY specific block. */ +#define LAN8814_INTC 0x18 +#define LAN8814_INTS 0x1B + +#define LAN8814_INT_LINK_DOWN BIT(2) +#define LAN8814_INT_LINK_UP BIT(0) +#define LAN8814_INT_LINK (LAN8814_INT_LINK_UP |\ + LAN8814_INT_LINK_DOWN) + +#define LAN8814_INTR_CTRL_REG 0x34 +#define LAN8814_INTR_CTRL_REG_POLARITY BIT(1) +#define LAN8814_INTR_CTRL_REG_INTR_ENABLE BIT(0) + /* PHY Control 1 */ #define MII_KSZPHY_CTRL_1 0x1e #define KSZ8081_CTRL1_MDIX_STAT BIT(4) @@ -1620,6 +1633,58 @@ static int lan8804_config_init(struct phy_device *phydev) return 0; } +static irqreturn_t lan8814_handle_interrupt(struct phy_device *phydev) +{ + int irq_status; + + irq_status = phy_read(phydev, LAN8814_INTS); + if (irq_status < 0) + return IRQ_NONE; + + if (!(irq_status & LAN8814_INT_LINK)) + return IRQ_NONE; + + phy_trigger_machine(phydev); + + return IRQ_HANDLED; +} + +static int lan8814_ack_interrupt(struct phy_device *phydev) +{ + /* bit[12..0] int status, which is a read and clear register. */ + int rc; + + rc = phy_read(phydev, LAN8814_INTS); + + return (rc < 0) ? rc : 0; +} + +static int lan8814_config_intr(struct phy_device *phydev) +{ + int err; + + lanphy_write_page_reg(phydev, 4, LAN8814_INTR_CTRL_REG, + LAN8814_INTR_CTRL_REG_POLARITY | + LAN8814_INTR_CTRL_REG_INTR_ENABLE); + + /* enable / disable interrupts */ + if (phydev->interrupts == PHY_INTERRUPT_ENABLED) { + err = lan8814_ack_interrupt(phydev); + if (err) + return err; + + err = phy_write(phydev, LAN8814_INTC, LAN8814_INT_LINK); + } else { + err = phy_write(phydev, LAN8814_INTC, 0); + if (err) + return err; + + err = lan8814_ack_interrupt(phydev); + } + + return err; +} + static struct phy_driver ksphy_driver[] = { { .phy_id = PHY_ID_KS8737, @@ -1802,6 +1867,8 @@ static struct phy_driver ksphy_driver[] = { .get_stats = kszphy_get_stats, .suspend = genphy_suspend, .resume = kszphy_resume, + .config_intr = lan8814_config_intr, + .handle_interrupt = lan8814_handle_interrupt, }, { .phy_id = PHY_ID_LAN8804, .phy_id_mask = MICREL_PHY_ID_MASK, diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c index 4733fd7fb169..62c453a21e49 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.c +++ b/drivers/net/wireless/ath/ath10k/wmi.c @@ -2613,6 +2613,7 @@ int ath10k_wmi_event_mgmt_rx(struct ath10k *ar, struct sk_buff *skb) if (ieee80211_is_beacon(hdr->frame_control) || ieee80211_is_probe_resp(hdr->frame_control)) { struct ieee80211_mgmt *mgmt = (void *)skb->data; + enum cfg80211_bss_frame_type ftype; u8 *ies; int ies_ch; @@ -2623,9 +2624,14 @@ int ath10k_wmi_event_mgmt_rx(struct ath10k *ar, struct sk_buff *skb) ies = mgmt->u.beacon.variable; + if (ieee80211_is_beacon(mgmt->frame_control)) + ftype = CFG80211_BSS_FTYPE_BEACON; + else + ftype = CFG80211_BSS_FTYPE_PRESP; + ies_ch = cfg80211_get_ies_channel_number(mgmt->u.beacon.variable, skb_tail_pointer(skb) - ies, - sband->band); + sband->band, ftype); if (ies_ch > 0 && ies_ch != channel) { ath10k_dbg(ar, ATH10K_DBG_MGMT, diff --git a/drivers/net/wireless/ath/ath11k/debugfs.c b/drivers/net/wireless/ath/ath11k/debugfs.c index ca68cf63dd27..198ade90b725 100644 --- a/drivers/net/wireless/ath/ath11k/debugfs.c +++ b/drivers/net/wireless/ath/ath11k/debugfs.c @@ -3,6 +3,8 @@ * Copyright (c) 2018-2020 The Linux Foundation. All rights reserved. */ +#include <linux/vmalloc.h> + #include "debugfs.h" #include "core.h" diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c index 1f8a81987187..d6575feca5a2 100644 --- a/drivers/net/wireless/ath/ath11k/reg.c +++ b/drivers/net/wireless/ath/ath11k/reg.c @@ -2,6 +2,8 @@ /* * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved. */ +#include <linux/rtnetlink.h> + #include "core.h" #include "debug.h" diff --git a/drivers/net/wireless/ath/carl9170/tx.c b/drivers/net/wireless/ath/carl9170/tx.c index 88444fe6d1c6..1b76f4434c06 100644 --- a/drivers/net/wireless/ath/carl9170/tx.c +++ b/drivers/net/wireless/ath/carl9170/tx.c @@ -275,12 +275,12 @@ static void carl9170_tx_release(struct kref *ref) if (WARN_ON_ONCE(!ar)) return; - BUILD_BUG_ON( - offsetof(struct ieee80211_tx_info, status.ack_signal) != 20); - - memset(&txinfo->status.ack_signal, 0, - sizeof(struct ieee80211_tx_info) - - offsetof(struct ieee80211_tx_info, status.ack_signal)); + /* + * This does not call ieee80211_tx_info_clear_status() because + * carl9170_tx_fill_rateinfo() has filled the rate information + * before we get to this point. + */ + memset_after(&txinfo->status, 0, rates); if (atomic_read(&ar->tx_total_queued)) ar->tx_schedule = true; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c index 9ef89ff09105..dd58c8f9aa11 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c @@ -608,7 +608,8 @@ static const struct ieee80211_sband_iftype_data iwl_he_capa[] = { .phy_cap_info[9] = IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_COMP_SIGB | IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_NON_COMP_SIGB | - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_RESERVED, + (IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED << + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_POS), .phy_cap_info[10] = IEEE80211_HE_PHY_CAP10_HE_MU_M1RU_MAX_LTF, }, @@ -665,7 +666,8 @@ static const struct ieee80211_sband_iftype_data iwl_he_capa[] = { IEEE80211_HE_PHY_CAP8_HE_ER_SU_PPDU_4XLTF_AND_08_US_GI | IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_242, .phy_cap_info[9] = - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_RESERVED, + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED + << IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_POS, }, /* * Set default Tx/Rx HE MCS NSS Support field. diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index efda37645fd4..65f4fe3ef504 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -2215,24 +2215,24 @@ static void iwl_mvm_cfg_he_sta(struct iwl_mvm *mvm, } flags |= STA_CTXT_HE_PACKET_EXT; - } else if ((sta->he_cap.he_cap_elem.phy_cap_info[9] & - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_MASK) != - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_RESERVED) { + } else if (u8_get_bits(sta->he_cap.he_cap_elem.phy_cap_info[9], + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK) + != IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED) { int low_th = -1; int high_th = -1; /* Take the PPE thresholds from the nominal padding info */ - switch (sta->he_cap.he_cap_elem.phy_cap_info[9] & - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_MASK) { - case IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_0US: + switch (u8_get_bits(sta->he_cap.he_cap_elem.phy_cap_info[9], + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK)) { + case IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_0US: low_th = IWL_HE_PKT_EXT_NONE; high_th = IWL_HE_PKT_EXT_NONE; break; - case IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_8US: + case IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_8US: low_th = IWL_HE_PKT_EXT_BPSK; high_th = IWL_HE_PKT_EXT_NONE; break; - case IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_16US: + case IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US: low_th = IWL_HE_PKT_EXT_NONE; high_th = IWL_HE_PKT_EXT_BPSK; break; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index 58d4d3bf6f86..87630d38dc52 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -5,6 +5,7 @@ * Copyright (C) 2016-2017 Intel Deutschland GmbH */ #include <linux/module.h> +#include <linux/rtnetlink.h> #include <linux/vmalloc.h> #include <net/mac80211.h> diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/vendor-cmd.c b/drivers/net/wireless/intel/iwlwifi/mvm/vendor-cmd.c index f702ad85e609..78450366312b 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/vendor-cmd.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/vendor-cmd.c @@ -4,6 +4,7 @@ */ #include "mvm.h" #include <linux/nl80211-vnd-intel.h> +#include <net/netlink.h> static const struct nla_policy iwl_mvm_vendor_attr_policy[NUM_IWL_MVM_VENDOR_ATTR] = { diff --git a/drivers/net/wireless/intersil/p54/txrx.c b/drivers/net/wireless/intersil/p54/txrx.c index 873fea59894f..8414aa208655 100644 --- a/drivers/net/wireless/intersil/p54/txrx.c +++ b/drivers/net/wireless/intersil/p54/txrx.c @@ -431,11 +431,7 @@ static void p54_rx_frame_sent(struct p54_common *priv, struct sk_buff *skb) * Clear manually, ieee80211_tx_info_clear_status would * clear the counts too and we need them. */ - memset(&info->status.ack_signal, 0, - sizeof(struct ieee80211_tx_info) - - offsetof(struct ieee80211_tx_info, status.ack_signal)); - BUILD_BUG_ON(offsetof(struct ieee80211_tx_info, - status.ack_signal) != 20); + memset_after(&info->status, 0, rates); if (entry_hdr->flags & cpu_to_le16(P54_HDR_FLAG_DATA_ALIGN)) pad = entry_data->align[0]; diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 23219f3747f8..0307a6677907 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -1276,7 +1276,7 @@ static void mac80211_hwsim_tx_frame_nl(struct ieee80211_hw *hw, hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PM); /* If the queue contains MAX_QUEUE skb's drop some */ if (skb_queue_len(&data->pending) >= MAX_QUEUE) { - /* Droping until WARN_QUEUE level */ + /* Dropping until WARN_QUEUE level */ while (skb_queue_len(&data->pending) >= WARN_QUEUE) { ieee80211_free_txskb(hw, skb_dequeue(&data->pending)); data->tx_dropped++; diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c index 4fa8e7ba93e6..d054cdecd5f7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c @@ -853,7 +853,8 @@ mt7915_init_he_caps(struct mt7915_phy *phy, enum nl80211_band band, mt7915_gen_ppe_thresh(he_cap->ppe_thres, nss); } else { he_cap_elem->phy_cap_info[9] |= - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_16US; + u8_encode_bits(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US, + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK); } idx++; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c index 3effe22f69c0..7a8d2596c226 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c @@ -152,7 +152,8 @@ mt7921_init_he_caps(struct mt7921_phy *phy, enum nl80211_band band, mt7921_gen_ppe_thresh(he_cap->ppe_thres, nss); } else { he_cap_elem->phy_cap_info[9] |= - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_16US; + u8_encode_bits(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US, + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK); } if (band == NL80211_BAND_6GHZ) { diff --git a/drivers/net/wireless/mediatek/mt76/testmode.h b/drivers/net/wireless/mediatek/mt76/testmode.h index 5e2792d81576..0590c35c7126 100644 --- a/drivers/net/wireless/mediatek/mt76/testmode.h +++ b/drivers/net/wireless/mediatek/mt76/testmode.h @@ -7,6 +7,8 @@ #define MT76_TM_TIMEOUT 10 +#include <net/netlink.h> + /** * enum mt76_testmode_attr - testmode attributes inside NL80211_ATTR_TESTDATA * diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 00ae86807dc2..a0737eea9f81 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -1,6 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* Copyright(c) 2019-2020 Realtek Corporation */ +#include <linux/ip.h> +#include <linux/udp.h> #include "coex.h" #include "core.h" @@ -2119,7 +2121,8 @@ static void rtw89_init_he_cap(struct rtw89_dev *rtwdev, IEEE80211_HE_PHY_CAP9_RX_1024_QAM_LESS_THAN_242_TONE_RU | IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_COMP_SIGB | IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_NON_COMP_SIGB | - IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_16US; + u8_encode_bits(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US, + IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK); if (i == NL80211_IFTYPE_STATION) phy_cap_info[9] |= IEEE80211_HE_PHY_CAP9_TX_1024_QAM_LESS_THAN_242_TONE_RU; he_cap->he_mcs_nss_supp.rx_mcs_80 = cpu_to_le16(mcs_map); diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c index 9756d75ef24e..22bd1d03e722 100644 --- a/drivers/net/wireless/realtek/rtw89/debug.c +++ b/drivers/net/wireless/realtek/rtw89/debug.c @@ -2,6 +2,8 @@ /* Copyright(c) 2019-2020 Realtek Corporation */ +#include <linux/vmalloc.h> + #include "coex.h" #include "debug.h" #include "fw.h" diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index d4b59fbe7365..8a57b75b07c0 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -781,7 +781,7 @@ static void __get_sta_he_pkt_padding(struct rtw89_dev *rtwdev, if (!ppe_th) { u8 pad; - pad = FIELD_GET(IEEE80211_HE_PHY_CAP9_NOMIMAL_PKT_PADDING_MASK, + pad = FIELD_GET(IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK, sta->he_cap.he_cap_elem.phy_cap_info[9]); for (i = 0; i < RTW89_PPE_BW_NUM; i++) |