Merge tag 'drm-fixes-2024-11-09' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
 "Weekly fixes, usual leaders in amdgpu and xe, then a panel quirk, and
  some fixes to imagination and panthor drivers. Seems around the usual
  level for this time and don't know of any big problems.

  amdgpu:
   - Brightness fix
   - DC vbios parsing fix
   - ACPI fix
   - SMU 14.x fix
   - Power workload profile fix
   - GC partitioning fix
   - Debugfs fixes

  imagination:
   - Track PVR context per file
   - Break ref-counting cycle

  panel-orientation-quirks:
   - Fix matching Lenovo Yoga Tab 3 X90F

  panthor:
   - Lock VM array
   - Be strict about I/O mapping flags

  xe:
   - Fix ccs_mode setting for Xe2 and later
   - Synchronize ccs_mode setting with client creation
   - Apply scheduling WA for LNL in additional places as needed
   - Fix leak and lock handling in error paths of xe_exec ioctl
   - Fix GGTT allocation leak leading to eventual crash in SR-IOV
   - Move run_ticks update out of job handling to avoid synchronization
     with reader"

* tag 'drm-fixes-2024-11-09' of https://gitlab.freedesktop.org/drm/kernel: (23 commits)
  drm/panthor: Be stricter about IO mapping flags
  drm/panthor: Lock XArray when getting entries for the VM
  drm: panel-orientation-quirks: Make Lenovo Yoga Tab 3 X90F DMI match less strict
  drm/xe: Stop accumulating LRC timestamp on job_free
  drm/xe/pf: Fix potential GGTT allocation leak
  drm/xe: Drop VM dma-resv lock on xe_sync_in_fence_get failure in exec IOCTL
  drm/xe: Fix possible exec queue leak in exec IOCTL
  drm/amdgpu: add missing size check in amdgpu_debugfs_gprwave_read()
  drm/amdgpu: Adjust debugfs eviction and IB access permissions
  drm/amdgpu: Adjust debugfs register access permissions
  drm/amdgpu: Fix DPX valid mode check on GC 9.4.3
  drm/amd/pm: correct the workload setting
  drm/amd/pm: always pick the pptable from IFWI
  drm/amdgpu: prevent NULL pointer dereference if ATIF is not supported
  drm/amd/display: parse umc_info or vram_info based on ASIC
  drm/amd/display: Fix brightness level not retained over reboot
  drm/xe/guc/tlb: Flush g2h worker in case of tlb timeout
  drm/xe/ufence: Flush xe ordered_wq in case of ufence timeout
  drm/xe: Move LNL scheduling WA to xe_device.h
  drm/xe: Use the filelist from drm for ccs_mode change
  ...

38 files changed, 257 insertions, 155 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 1f5a296f5ed2..7dd55ed57c1d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -172,8 +172,8 @@ static union acpi_object *amdgpu_atif_call(struct amdgpu_atif *atif,
 				      &buffer);
 	obj = (union acpi_object *)buffer.pointer;
 
-	/* Fail if calling the method fails and ATIF is supported */
-	if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
+	/* Fail if calling the method fails */
+	if (ACPI_FAILURE(status)) {
 		DRM_DEBUG_DRIVER("failed to evaluate ATIF got %s\n",
 				 acpi_format_exception(status));
 		kfree(obj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index cbef720de779..9da4414de617 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -402,7 +402,7 @@ static ssize_t amdgpu_debugfs_gprwave_read(struct file *f, char __user *buf, siz
 	int r;
 	uint32_t *data, x;
 
-	if (size & 0x3 || *pos & 0x3)
+	if (size > 4096 || size & 0x3 || *pos & 0x3)
 		return -EINVAL;
 
 	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
@@ -1648,7 +1648,7 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev)
 
 	for (i = 0; i < ARRAY_SIZE(debugfs_regs); i++) {
 		ent = debugfs_create_file(debugfs_regs_names[i],
-					  S_IFREG | 0444, root,
+					  S_IFREG | 0400, root,
 					  adev, debugfs_regs[i]);
 		if (!i && !IS_ERR_OR_NULL(ent))
 			i_size_write(ent->d_inode, adev->rmmio_size);
@@ -2100,11 +2100,11 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 	amdgpu_securedisplay_debugfs_init(adev);
 	amdgpu_fw_attestation_debugfs_init(adev);
 
-	debugfs_create_file("amdgpu_evict_vram", 0444, root, adev,
+	debugfs_create_file("amdgpu_evict_vram", 0400, root, adev,
 			    &amdgpu_evict_vram_fops);
-	debugfs_create_file("amdgpu_evict_gtt", 0444, root, adev,
+	debugfs_create_file("amdgpu_evict_gtt", 0400, root, adev,
 			    &amdgpu_evict_gtt_fops);
-	debugfs_create_file("amdgpu_test_ib", 0444, root, adev,
+	debugfs_create_file("amdgpu_test_ib", 0400, root, adev,
 			    &amdgpu_debugfs_test_ib_fops);
 	debugfs_create_file("amdgpu_vm_info", 0444, root, adev,
 			    &amdgpu_debugfs_vm_info_fops);
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
index 5e8833e4fed2..ccfd2a4b4acc 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
@@ -482,7 +482,7 @@ static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 	case AMDGPU_SPX_PARTITION_MODE:
 		return adev->gmc.num_mem_partitions == 1 && num_xcc > 0;
 	case AMDGPU_DPX_PARTITION_MODE:
-		return adev->gmc.num_mem_partitions != 8 && (num_xcc % 4) == 0;
+		return adev->gmc.num_mem_partitions <= 2 && (num_xcc % 4) == 0;
 	case AMDGPU_TPX_PARTITION_MODE:
 		return (adev->gmc.num_mem_partitions == 1 ||
 			adev->gmc.num_mem_partitions == 3) &&
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 13421a58210d..07e9ce99694f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -9429,6 +9429,7 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state,
 	bool mode_set_reset_required = false;
 	u32 i;
 	struct dc_commit_streams_params params = {dc_state->streams, dc_state->stream_count};
+	bool set_backlight_level = false;
 
 	/* Disable writeback */
 	for_each_old_connector_in_state(state, connector, old_con_state, i) {
@@ -9548,6 +9549,7 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state,
 			acrtc->hw_mode = new_crtc_state->mode;
 			crtc->hwmode = new_crtc_state->mode;
 			mode_set_reset_required = true;
+			set_backlight_level = true;
 		} else if (modereset_required(new_crtc_state)) {
 			drm_dbg_atomic(dev,
 				       "Atomic commit: RESET. crtc id %d:[%p]\n",
@@ -9599,6 +9601,19 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state,
 				acrtc->otg_inst = status->primary_otg_inst;
 		}
 	}
+
+	/* During boot up and resume the DC layer will reset the panel brightness
+	 * to fix a flicker issue.
+	 * It will cause the dm->actual_brightness is not the current panel brightness
+	 * level. (the dm->brightness is the correct panel level)
+	 * So we set the backlight level with dm->brightness value after set mode
+	 */
+	if (set_backlight_level) {
+		for (i = 0; i < dm->num_of_edps; i++) {
+			if (dm->backlight_dev[i])
+				amdgpu_dm_backlight_set_level(dm, i, dm->brightness[i]);
+		}
+	}
 }
 
 static void dm_set_writeback(struct amdgpu_display_manager *dm,
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
index 0d8498ab9b23..be8fbb04ad98 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -3127,7 +3127,9 @@ static enum bp_result bios_parser_get_vram_info(
 	struct atom_data_revision revision;
 
 	// vram info moved to umc_info for DCN4x
-	if (info && DATA_TABLES(umc_info)) {
+	if (dcb->ctx->dce_version >= DCN_VERSION_4_01 &&
+		dcb->ctx->dce_version < DCN_VERSION_MAX &&
+		info && DATA_TABLES(umc_info)) {
 		header = GET_IMAGE(struct atom_common_table_header,
 					DATA_TABLES(umc_info));
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 80e60ea2d11e..ee1bcfaae3e3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -1259,26 +1259,33 @@ static int smu_sw_init(void *handle)
 	smu->watermarks_bitmap = 0;
 	smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
 	smu->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	smu->user_dpm_profile.user_workload_mask = 0;
 
 	atomic_set(&smu->smu_power.power_gate.vcn_gated, 1);
 	atomic_set(&smu->smu_power.power_gate.jpeg_gated, 1);
 	atomic_set(&smu->smu_power.power_gate.vpe_gated, 1);
 	atomic_set(&smu->smu_power.power_gate.umsch_mm_gated, 1);
 
-	smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT] = 0;
-	smu->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D] = 1;
-	smu->workload_prority[PP_SMC_POWER_PROFILE_POWERSAVING] = 2;
-	smu->workload_prority[PP_SMC_POWER_PROFILE_VIDEO] = 3;
-	smu->workload_prority[PP_SMC_POWER_PROFILE_VR] = 4;
-	smu->workload_prority[PP_SMC_POWER_PROFILE_COMPUTE] = 5;
-	smu->workload_prority[PP_SMC_POWER_PROFILE_CUSTOM] = 6;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT] = 0;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_FULLSCREEN3D] = 1;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_POWERSAVING] = 2;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_VIDEO] = 3;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_VR] = 4;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_COMPUTE] = 5;
+	smu->workload_priority[PP_SMC_POWER_PROFILE_CUSTOM] = 6;
 
 	if (smu->is_apu ||
-	    !smu_is_workload_profile_available(smu, PP_SMC_POWER_PROFILE_FULLSCREEN3D))
-		smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT];
-	else
-		smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D];
+	    !smu_is_workload_profile_available(smu, PP_SMC_POWER_PROFILE_FULLSCREEN3D)) {
+		smu->driver_workload_mask =
+			1 << smu->workload_priority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT];
+	} else {
+		smu->driver_workload_mask =
+			1 << smu->workload_priority[PP_SMC_POWER_PROFILE_FULLSCREEN3D];
+		smu->default_power_profile_mode = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
+	}
 
+	smu->workload_mask = smu->driver_workload_mask |
+							smu->user_dpm_profile.user_workload_mask;
 	smu->workload_setting[0] = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
 	smu->workload_setting[1] = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
 	smu->workload_setting[2] = PP_SMC_POWER_PROFILE_POWERSAVING;
@@ -2348,17 +2355,20 @@ static int smu_switch_power_profile(void *handle,
 		return -EINVAL;
 
 	if (!en) {
-		smu->workload_mask &= ~(1 << smu->workload_prority[type]);
+		smu->driver_workload_mask &= ~(1 << smu->workload_priority[type]);
 		index = fls(smu->workload_mask);
 		index = index > 0 && index <= WORKLOAD_POLICY_MAX ? index - 1 : 0;
 		workload[0] = smu->workload_setting[index];
 	} else {
-		smu->workload_mask |= (1 << smu->workload_prority[type]);
+		smu->driver_workload_mask |= (1 << smu->workload_priority[type]);
 		index = fls(smu->workload_mask);
 		index = index <= WORKLOAD_POLICY_MAX ? index - 1 : 0;
 		workload[0] = smu->workload_setting[index];
 	}
 
+	smu->workload_mask = smu->driver_workload_mask |
+						 smu->user_dpm_profile.user_workload_mask;
+
 	if (smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL &&
 		smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_PERF_DETERMINISM)
 		smu_bump_power_profile_mode(smu, workload, 0);
@@ -3049,12 +3059,23 @@ static int smu_set_power_profile_mode(void *handle,
 				      uint32_t param_size)
 {
 	struct smu_context *smu = handle;
+	int ret;
 
 	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled ||
 	    !smu->ppt_funcs->set_power_profile_mode)
 		return -EOPNOTSUPP;
 
-	return smu_bump_power_profile_mode(smu, param, param_size);
+	if (smu->user_dpm_profile.user_workload_mask &
+	   (1 << smu->workload_priority[param[param_size]]))
+	   return 0;
+
+	smu->user_dpm_profile.user_workload_mask =
+		(1 << smu->workload_priority[param[param_size]]);
+	smu->workload_mask = smu->user_dpm_profile.user_workload_mask |
+		smu->driver_workload_mask;
+	ret = smu_bump_power_profile_mode(smu, param, param_size);
+
+	return ret;
 }
 
 static int smu_get_fan_control_mode(void *handle, u32 *fan_mode)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index b44a185d07e8..d60d9a12a47e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -240,6 +240,7 @@ struct smu_user_dpm_profile {
 	/* user clock state information */
 	uint32_t clk_mask[SMU_CLK_COUNT];
 	uint32_t clk_dependency;
+	uint32_t user_workload_mask;
 };
 
 #define SMU_TABLE_INIT(tables, table_id, s, a, d)	\
@@ -557,7 +558,8 @@ struct smu_context {
 	bool disable_uclk_switch;
 
 	uint32_t workload_mask;
-	uint32_t workload_prority[WORKLOAD_POLICY_MAX];
+	uint32_t driver_workload_mask;
+	uint32_t workload_priority[WORKLOAD_POLICY_MAX];
 	uint32_t workload_setting[WORKLOAD_POLICY_MAX];
 	uint32_t power_profile_mode;
 	uint32_t default_power_profile_mode;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
index c0f6b59369b7..31fe512028f4 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
@@ -1455,7 +1455,6 @@ static int arcturus_set_power_profile_mode(struct smu_context *smu,
 		return -EINVAL;
 	}
 
-
 	if ((profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) &&
 	     (smu->smc_fw_version >= 0x360d00)) {
 		if (size != 10)
@@ -1523,14 +1522,14 @@ static int arcturus_set_power_profile_mode(struct smu_context *smu,
 
 	ret = smu_cmn_send_smc_msg_with_param(smu,
 					  SMU_MSG_SetWorkloadMask,
-					  1 << workload_type,
+					  smu->workload_mask,
 					  NULL);
 	if (ret) {
 		dev_err(smu->adev->dev, "Fail to set workload type %d\n", workload_type);
 		return ret;
 	}
 
-	smu->power_profile_mode = profile_mode;
+	smu_cmn_assign_power_profile(smu);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
index 16af1a329621..12223f507977 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
@@ -2081,10 +2081,13 @@ static int navi10_set_power_profile_mode(struct smu_context *smu, long *input, u
 						       smu->power_profile_mode);
 	if (workload_type < 0)
 		return -EINVAL;
+
 	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask,
-				    1 << workload_type, NULL);
+				    smu->workload_mask, NULL);
 	if (ret)
 		dev_err(smu->adev->dev, "[%s] Failed to set work load mask!", __func__);
+	else
+		smu_cmn_assign_power_profile(smu);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index 9c3c48297cba..3b7b2ec8319a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -1786,10 +1786,13 @@ static int sienna_cichlid_set_power_profile_mode(struct smu_context *smu, long *
 						       smu->power_profile_mode);
 	if (workload_type < 0)
 		return -EINVAL;
+
 	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask,
-				    1 << workload_type, NULL);
+				    smu->workload_mask, NULL);
 	if (ret)
 		dev_err(smu->adev->dev, "[%s] Failed to set work load mask!", __func__);
+	else
+		smu_cmn_assign_power_profile(smu);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
index 1fe020f1f4db..952ee22cbc90 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
@@ -1079,7 +1079,7 @@ static int vangogh_set_power_profile_mode(struct smu_context *smu, long *input,
 	}
 
 	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_ActiveProcessNotify,
-				    1 << workload_type,
+				    smu->workload_mask,
 				    NULL);
 	if (ret) {
 		dev_err_once(smu->adev->dev, "Fail to set workload type %d\n",
@@ -1087,7 +1087,7 @@ static int vangogh_set_power_profile_mode(struct smu_context *smu, long *input,
 		return ret;
 	}
 
-	smu->power_profile_mode = profile_mode;
+	smu_cmn_assign_power_profile(smu);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
index cc0504b063fa..62316a6707ef 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
@@ -890,14 +890,14 @@ static int renoir_set_power_profile_mode(struct smu_context *smu, long *input, u
 	}
 
 	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_ActiveProcessNotify,
-				    1 << workload_type,
+				    smu->workload_mask,
 				    NULL);
 	if (ret) {
 		dev_err_once(smu->adev->dev, "Fail to set workload type %d\n", workload_type);
 		return ret;
 	}
 
-	smu->power_profile_mode = profile_mode;
+	smu_cmn_assign_power_profile(smu);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index d53e162dcd8d..5dd7ceca64fe 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -2485,7 +2485,7 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu,
 	DpmActivityMonitorCoeffInt_t *activity_monitor =
 		&(activity_monitor_external.DpmActivityMonitorCoeffInt);
 	int workload_type, ret = 0;
-	u32 workload_mask, selected_workload_mask;
+	u32 workload_mask;
 
 	smu->power_profile_mode = input[size];
 
@@ -2552,7 +2552,7 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu,
 	if (workload_type < 0)
 		return -EINVAL;
 
-	selected_workload_mask = workload_mask = 1 << workload_type;
+	workload_mask = 1 << workload_type;
 
 	/* Add optimizations for SMU13.0.0/10.  Reuse the power saving profile */
 	if ((amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0) &&
@@ -2567,12 +2567,22 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu,
 			workload_mask |= 1 << workload_type;
 	}
 
+	smu->workload_mask |= workload_mask;
 	ret = smu_cmn_send_smc_msg_with_param(smu,
 					       SMU_MSG_SetWorkloadMask,
-					       workload_mask,
+					       smu->workload_mask,
 					       NULL);
-	if (!ret)
-		smu->workload_mask = selected_workload_mask;
+	if (!ret) {
+		smu_cmn_assign_power_profile(smu);
+		if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_POWERSAVING) {
+			workload_type = smu_cmn_to_asic_specific_index(smu,
+							       CMN2ASIC_MAPPING_WORKLOAD,
+							       PP_SMC_POWER_PROFILE_FULLSCREEN3D);
+			smu->power_profile_mode = smu->workload_mask & (1 << workload_type)
+										? PP_SMC_POWER_PROFILE_FULLSCREEN3D
+										: PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+		}
+	}
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index b891a5e0a396..9d0b19419de0 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -2499,13 +2499,14 @@ static int smu_v13_0_7_set_power_profile_mode(struct smu_context *smu, long *inp
 						       smu->power_profile_mode);
 	if (workload_type < 0)
 		return -EINVAL;
+
 	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask,
-				    1 << workload_type, NULL);
+				    smu->workload_mask, NULL);
 
 	if (ret)
 		dev_err(smu->adev->dev, "[%s] Failed to set work load mask!", __func__);
 	else
-		smu->workload_mask = (1 << workload_type);
+		smu_cmn_assign_power_profile(smu);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
index e83ea2bc7f9c..1aa13d32ceb2 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
@@ -367,54 +367,6 @@ static int smu_v14_0_2_store_powerplay_table(struct smu_context *smu)
 	return 0;
 }
 
-#ifndef atom_smc_dpm_info_table_14_0_0
-struct atom_smc_dpm_info_table_14_0_0 {
-	struct atom_common_table_header table_header;
-	BoardTable_t BoardTable;
-};
-#endif
-
-static int smu_v14_0_2_append_powerplay_table(struct smu_context *smu)
-{
-	struct smu_table_context *table_context = &smu->smu_table;
-	PPTable_t *smc_pptable = table_context->driver_pptable;
-	struct atom_smc_dpm_info_table_14_0_0 *smc_dpm_table;
-	BoardTable_t *BoardTable = &smc_pptable->BoardTable;
-	int index, ret;
-
-	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
-					    smc_dpm_info);
-
-	ret = amdgpu_atombios_get_data_table(smu->adev, index, NULL, NULL, NULL,
-					     (uint8_t **)&smc_dpm_table);
-	if (ret)
-		return ret;
-
-	memcpy(BoardTable, &smc_dpm_table->BoardTable, sizeof(BoardTable_t));
-
-	return 0;
-}
-
-#if 0
-static int smu_v14_0_2_get_pptable_from_pmfw(struct smu_context *smu,
-					     void **table,
-					     uint32_t *size)
-{
-	struct smu_table_context *smu_table = &smu->smu_table;
-	void *combo_pptable = smu_table->combo_pptable;
-	int ret = 0;
-
-	ret = smu_cmn_get_combo_pptable(smu);
-	if (ret)
-		return ret;
-
-	*table = combo_pptable;
-	*size = sizeof(struct smu_14_0_powerplay_table);
-
-	return 0;
-}
-#endif
-
 static int smu_v14_0_2_get_pptable_from_pmfw(struct smu_context *smu,
 					     void **table,
 					     uint32_t *size)
@@ -436,16 +388,12 @@ static int smu_v14_0_2_get_pptable_from_pmfw(struct smu_context *smu,
 static int smu_v14_0_2_setup_pptable(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
-	struct amdgpu_device *adev = smu->adev;
 	int ret = 0;
 
 	if (amdgpu_sriov_vf(smu->adev))
 		return 0;
 
-	if (!adev->scpm_enabled)
-		ret = smu_v14_0_setup_pptable(smu);
-	else
-		ret = smu_v14_0_2_get_pptable_from_pmfw(smu,
+	ret = smu_v14_0_2_get_pptable_from_pmfw(smu,
 							&smu_table->power_play_table,
 							&smu_table->power_play_table_size);
 	if (ret)
@@ -455,16 +403,6 @@ static int smu_v14_0_2_setup_pptable(struct smu_context *smu)
 	if (ret)
 		return ret;
 
-	/*
-	 * With SCPM enabled, the operation below will be handled
-	 * by PSP. Driver involvment is unnecessary and useless.
-	 */
-	if (!adev->scpm_enabled) {
-		ret = smu_v14_0_2_append_powerplay_table(smu);
-		if (ret)
-			return ret;
-	}
-
 	ret = smu_v14_0_2_check_powerplay_table(smu);
 	if (ret)
 		return ret;
@@ -1869,12 +1807,11 @@ static int smu_v14_0_2_set_power_profile_mode(struct smu_context *smu,
 	if (workload_type < 0)
 		return -EINVAL;
 
-	ret = smu_cmn_send_smc_msg_with_param(smu,
-					       SMU_MSG_SetWorkloadMask,
-					       1 << workload_type,
-					       NULL);
+	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetWorkloadMask,
+										  smu->workload_mask, NULL);
+
 	if (!ret)
-		smu->workload_mask = 1 << workload_type;
+		smu_cmn_assign_power_profile(smu);
 
 	return ret;
 }
@@ -2799,7 +2736,6 @@ static const struct pptable_funcs smu_v14_0_2_ppt_funcs = {
 	.check_fw_status = smu_v14_0_check_fw_status,
 	.setup_pptable = smu_v14_0_2_setup_pptable,
 	.check_fw_version = smu_v14_0_check_fw_version,
-	.write_pptable = smu_cmn_write_pptable,
 	.set_driver_table_location = smu_v14_0_set_driver_table_location,
 	.system_features_control = smu_v14_0_system_features_control,
 	.set_allowed_mask = smu_v14_0_set_allowed_mask,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
index 91ad434bcdae..bdfc5e617333 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
@@ -1138,6 +1138,14 @@ int smu_cmn_set_mp1_state(struct smu_context *smu,
 	return ret;
 }
 
+void smu_cmn_assign_power_profile(struct smu_context *smu)
+{
+	uint32_t index;
+	index = fls(smu->workload_mask);
+	index = index > 0 && index <= WORKLOAD_POLICY_MAX ? index - 1 : 0;
+	smu->power_profile_mode = smu->workload_setting[index];
+}
+
 bool smu_cmn_is_audio_func_enabled(struct amdgpu_device *adev)
 {
 	struct pci_dev *p = NULL;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
index 1de685defe85..8a801e389659 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
@@ -130,6 +130,8 @@ void smu_cmn_init_soft_gpu_metrics(void *table, uint8_t frev, uint8_t crev);
 int smu_cmn_set_mp1_state(struct smu_context *smu,
 			  enum pp_mp1_state mp1_state);
 
+void smu_cmn_assign_power_profile(struct smu_context *smu);
+
 /*
  * Helper function to make sysfs_emit_at() happy. Align buf to
  * the current page boundary and record the offset.
diff --git a/drivers/gpu/drm/drm_panel_orientation_quirks.c b/drivers/gpu/drm/drm_panel_orientation_quirks.c
index 0830cae9a4d0..2d84d7ea1ab7 100644
--- a/drivers/gpu/drm/drm_panel_orientation_quirks.c
+++ b/drivers/gpu/drm/drm_panel_orientation_quirks.c
@@ -403,7 +403,6 @@ static const struct dmi_system_id orientation_data[] = {
 	}, {	/* Lenovo Yoga Tab 3 X90F */
 		.matches = {
 		 DMI_MATCH(DMI_SYS_VENDOR, "Intel Corporation"),
-		 DMI_MATCH(DMI_PRODUCT_NAME, "CHERRYVIEW D1 PLATFORM"),
 		 DMI_MATCH(DMI_PRODUCT_VERSION, "Blade3-10A-001"),
 		},
 		.driver_data = (void *)&lcd1600x2560_rightside_up,
diff --git a/drivers/gpu/drm/imagination/pvr_context.c b/drivers/gpu/drm/imagination/pvr_context.c
index eded5e955cc0..4cb3494c0bb2 100644
--- a/drivers/gpu/drm/imagination/pvr_context.c
+++ b/drivers/gpu/drm/imagination/pvr_context.c
@@ -17,10 +17,14 @@
 
 #include <drm/drm_auth.h>
 #include <drm/drm_managed.h>
+
+#include <linux/bug.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
+#include <linux/spinlock.h>
 #include <linux/string.h>
 #include <linux/types.h>
 #include <linux/xarray.h>
@@ -354,6 +358,10 @@ int pvr_context_create(struct pvr_file *pvr_file, struct drm_pvr_ioctl_create_co
 		return err;
 	}
 
+	spin_lock(&pvr_dev->ctx_list_lock);
+	list_add_tail(&ctx->file_link, &pvr_file->contexts);
+	spin_unlock(&pvr_dev->ctx_list_lock);
+
 	return 0;
 
 err_destroy_fw_obj:
@@ -380,6 +388,11 @@ pvr_context_release(struct kref *ref_count)
 		container_of(ref_count, struct pvr_context, ref_count);
 	struct pvr_device *pvr_dev = ctx->pvr_dev;
 
+	WARN_ON(in_interrupt());
+	spin_lock(&pvr_dev->ctx_list_lock);
+	list_del(&ctx->file_link);
+	spin_unlock(&pvr_dev->ctx_list_lock);
+
 	xa_erase(&pvr_dev->ctx_ids, ctx->ctx_id);
 	pvr_context_destroy_queues(ctx);
 	pvr_fw_object_destroy(ctx->fw_obj);
@@ -437,11 +450,30 @@ pvr_context_destroy(struct pvr_file *pvr_file, u32 handle)
  */
 void pvr_destroy_contexts_for_file(struct pvr_file *pvr_file)
 {
+	struct pvr_device *pvr_dev = pvr_file->pvr_dev;
 	struct pvr_context *ctx;
 	unsigned long handle;
 
 	xa_for_each(&pvr_file->ctx_handles, handle, ctx)
 		pvr_context_destroy(pvr_file, handle);
+
+	spin_lock(&pvr_dev->ctx_list_lock);
+	ctx = list_first_entry(&pvr_file->contexts, struct pvr_context, file_link);
+
+	while (!list_entry_is_head(ctx, &pvr_file->contexts, file_link)) {
+		list_del_init(&ctx->file_link);
+
+		if (pvr_context_get_if_referenced(ctx)) {
+			spin_unlock(&pvr_dev->ctx_list_lock);
+
+			pvr_vm_unmap_all(ctx->vm_ctx);
+
+			pvr_context_put(ctx);
+			spin_lock(&pvr_dev->ctx_list_lock);
+		}
+		ctx = list_first_entry(&pvr_file->contexts, struct pvr_context, file_link);
+	}
+	spin_unlock(&pvr_dev->ctx_list_lock);
 }
 
 /**
@@ -451,6 +483,7 @@ void pvr_destroy_contexts_for_file(struct pvr_file *pvr_file)
 void pvr_context_device_init(struct pvr_device *pvr_dev)
 {
 	xa_init_flags(&pvr_dev->ctx_ids, XA_FLAGS_ALLOC1);
+	spin_lock_init(&pvr_dev->ctx_list_lock);
 }
 
 /**
diff --git a/drivers/gpu/drm/imagination/pvr_context.h b/drivers/gpu/drm/imagination/pvr_context.h
index 0c7b97dfa6ba..07afa179cdf4 100644
--- a/drivers/gpu/drm/imagination/pvr_context.h
+++ b/drivers/gpu/drm/imagination/pvr_context.h
@@ -85,6 +85,9 @@ struct pvr_context {
 		/** @compute: Transfer queue. */
 		struct pvr_queue *transfer;
 	} queues;
+
+	/** @file_link: pvr_file PVR context list link. */
+	struct list_head file_link;
 };
 
 static __always_inline struct pvr_queue *
@@ -124,6 +127,24 @@ pvr_context_get(struct pvr_context *ctx)
 }
 
 /**
+ * pvr_context_get_if_referenced() - Take an additional reference on a still
+ * referenced context.
+ * @ctx: Context pointer.
+ *
+ * Call pvr_context_put() to release.
+ *
+ * Returns:
+ *  * True on success, or
+ *  * false if no context pointer passed, or the context wasn't still
+ *  * referenced.
+ */
+static __always_inline bool
+pvr_context_get_if_referenced(struct pvr_context *ctx)
+{
+	return ctx != NULL && kref_get_unless_zero(&ctx->ref_count) != 0;
+}
+
+/**
  * pvr_context_lookup() - Lookup context pointer from handle and file.
  * @pvr_file: Pointer to pvr_file structure.
  * @handle: Context handle.
diff --git a/drivers/gpu/drm/imagination/pvr_device.h b/drivers/gpu/drm/imagination/pvr_device.h
index b574e23d484b..6d0dfacb677b 100644
--- a/drivers/gpu/drm/imagination/pvr_device.h
+++ b/drivers/gpu/drm/imagination/pvr_device.h
@@ -23,6 +23,7 @@
 #include <linux/kernel.h>
 #include <linux/math.h>
 #include <linux/mutex.h>
+#include <linux/spinlock_types.h>
 #include <linux/timer.h>
 #include <linux/types.h>
 #include <linux/wait.h>
@@ -293,6 +294,12 @@ struct pvr_device {
 
 	/** @sched_wq: Workqueue for schedulers. */
 	struct workqueue_struct *sched_wq;
+
+	/**
+	 * @ctx_list_lock: Lock to be held when accessing the context list in
+	 *  struct pvr_file.
+	 */
+	spinlock_t ctx_list_lock;
 };
 
 /**
@@ -344,6 +351,9 @@ struct pvr_file {
 	 * This array is used to allocate handles returned to userspace.
 	 */
 	struct xarray vm_ctx_handles;
+
+	/** @contexts: PVR context list. */
+	struct list_head contexts;
 };
 
 /**
diff --git a/drivers/gpu/drm/imagination/pvr_drv.c b/drivers/gpu/drm/imagination/pvr_drv.c
index 1a0cb7aa9cea..fb17196e05f4 100644
--- a/drivers/gpu/drm/imagination/pvr_drv.c
+++ b/drivers/gpu/drm/imagination/pvr_drv.c
@@ -28,6 +28,7 @@
 #include <linux/export.h>
 #include <linux/fs.h>
 #include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -1326,6 +1327,8 @@ pvr_drm_driver_open(struct drm_device *drm_dev, struct drm_file *file)
 	 */
 	pvr_file->pvr_dev = pvr_dev;
 
+	INIT_LIST_HEAD(&pvr_file->contexts);
+
 	xa_init_flags(&pvr_file->ctx_handles, XA_FLAGS_ALLOC1);
 	xa_init_flags(&pvr_file->free_list_handles, XA_FLAGS_ALLOC1);
 	xa_init_flags(&pvr_file->hwrt_handles, XA_FLAGS_ALLOC1);
diff --git a/drivers/gpu/drm/imagination/pvr_vm.c b/drivers/gpu/drm/imagination/pvr_vm.c
index 97c0f772ed65..7bd6ba4c6e8a 100644
--- a/drivers/gpu/drm/imagination/pvr_vm.c
+++ b/drivers/gpu/drm/imagination/pvr_vm.c
@@ -14,6 +14,7 @@
 #include <drm/drm_gem.h>
 #include <drm/drm_gpuvm.h>
 
+#include <linux/bug.h>
 #include <linux/container_of.h>
 #include <linux/err.h>
 #include <linux/errno.h>
@@ -597,12 +598,26 @@ err_free:
 }
 
 /**
- * pvr_vm_context_release() - Teardown a VM context.
- * @ref_count: Pointer to reference counter of the VM context.
+ * pvr_vm_unmap_all() - Unmap all mappings associated with a VM context.
+ * @vm_ctx: Target VM context.
  *
  * This function ensures that no mappings are left dangling by unmapping them
  * all in order of ascending device-virtual address.
  */
+void
+pvr_vm_unmap_all(struct pvr_vm_context *vm_ctx)
+{
+	WARN_ON(pvr_vm_unmap(vm_ctx, vm_ctx->gpuvm_mgr.mm_start,
+			     vm_ctx->gpuvm_mgr.mm_range));
+}
+
+/**
+ * pvr_vm_context_release() - Teardown a VM context.
+ * @ref_count: Pointer to reference counter of the VM context.
+ *
+ * This function also ensures that no mappings are left dangling by calling
+ * pvr_vm_unmap_all.
+ */
 static void
 pvr_vm_context_release(struct kref *ref_count)
 {
@@ -612,8 +627,7 @@ pvr_vm_context_release(struct kref *ref_count)
 	if (vm_ctx->fw_mem_ctx_obj)
 		pvr_fw_object_destroy(vm_ctx->fw_mem_ctx_obj);
 
-	WARN_ON(pvr_vm_unmap(vm_ctx, vm_ctx->gpuvm_mgr.mm_start,
-			     vm_ctx->gpuvm_mgr.mm_range));
+	pvr_vm_unmap_all(vm_ctx);
 
 	pvr_mmu_context_destroy(vm_ctx->mmu_ctx);
 	drm_gem_private_object_fini(&vm_ctx->dummy_gem);
diff --git a/drivers/gpu/drm/imagination/pvr_vm.h b/drivers/gpu/drm/imagination/pvr_vm.h
index f2a6463f2b05..79406243617c 100644
--- a/drivers/gpu/drm/imagination/pvr_vm.h
+++ b/drivers/gpu/drm/imagination/pvr_vm.h
@@ -39,6 +39,7 @@ int pvr_vm_map(struct pvr_vm_context *vm_ctx,
 	       struct pvr_gem_object *pvr_obj, u64 pvr_obj_offset,
 	       u64 device_addr, u64 size);
 int pvr_vm_unmap(struct pvr_vm_context *vm_ctx, u64 device_addr, u64 size);
+void pvr_vm_unmap_all(struct pvr_vm_context *vm_ctx);
 
 dma_addr_t pvr_vm_get_page_table_root_addr(struct pvr_vm_context *vm_ctx);
 struct dma_resv *pvr_vm_get_dma_resv(struct pvr_vm_context *vm_ctx);
diff --git a/drivers/gpu/drm/panthor/panthor_device.c b/drivers/gpu/drm/panthor/panthor_device.c
index 4082c8f2951d..6fbff516c1c1 100644
--- a/drivers/gpu/drm/panthor/panthor_device.c
+++ b/drivers/gpu/drm/panthor/panthor_device.c
@@ -390,11 +390,15 @@ int panthor_device_mmap_io(struct panthor_device *ptdev, struct vm_area_struct *
 {
 	u64 offset = (u64)vma->vm_pgoff << PAGE_SHIFT;
 
+	if ((vma->vm_flags & VM_SHARED) == 0)
+		return -EINVAL;
+
 	switch (offset) {
 	case DRM_PANTHOR_USER_FLUSH_ID_MMIO_OFFSET:
 		if (vma->vm_end - vma->vm_start != PAGE_SIZE ||
 		    (vma->vm_flags & (VM_WRITE | VM_EXEC)))
 			return -EINVAL;
+		vm_flags_clear(vma, VM_MAYWRITE);
 
 		break;
 
diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index 5d5e25b1be95..7db2edb3374c 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -1580,7 +1580,9 @@ panthor_vm_pool_get_vm(struct panthor_vm_pool *pool, u32 handle)
 {
 	struct panthor_vm *vm;
 
+	xa_lock(&pool->xa);
 	vm = panthor_vm_get(xa_load(&pool->xa, handle));
+	xa_unlock(&pool->xa);
 
 	return vm;
 }
diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
index 00ad34ed73a5..bd604b9f08e4 100644
--- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
@@ -517,7 +517,7 @@
  *   [4-6]     RSVD
  *   [7]       Disabled
  */
-#define CCS_MODE				XE_REG(0x14804)
+#define CCS_MODE				XE_REG(0x14804, XE_REG_OPTION_MASKED)
 #define   CCS_MODE_CSLICE_0_3_MASK		REG_GENMASK(11, 0) /* 3 bits per cslice */
 #define   CCS_MODE_CSLICE_MASK			0x7 /* CCS0-3 + rsvd */
 #define   CCS_MODE_CSLICE_WIDTH			ilog2(CCS_MODE_CSLICE_MASK + 1)
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 10fd4601b9f2..a1987b554a8d 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -87,10 +87,6 @@ static int xe_file_open(struct drm_device *dev, struct drm_file *file)
 	mutex_init(&xef->exec_queue.lock);
 	xa_init_flags(&xef->exec_queue.xa, XA_FLAGS_ALLOC1);
 
-	spin_lock(&xe->clients.lock);
-	xe->clients.count++;
-	spin_unlock(&xe->clients.lock);
-
 	file->driver_priv = xef;
 	kref_init(&xef->refcount);
 
@@ -107,17 +103,12 @@ static int xe_file_open(struct drm_device *dev, struct drm_file *file)
 static void xe_file_destroy(struct kref *ref)
 {
 	struct xe_file *xef = container_of(ref, struct xe_file, refcount);
-	struct xe_device *xe = xef->xe;
 
 	xa_destroy(&xef->exec_queue.xa);
 	mutex_destroy(&xef->exec_queue.lock);
 	xa_destroy(&xef->vm.xa);
 	mutex_destroy(&xef->vm.lock);
 
-	spin_lock(&xe->clients.lock);
-	xe->clients.count--;
-	spin_unlock(&xe->clients.lock);
-
 	xe_drm_client_put(xef->client);
 	kfree(xef->process_name);
 	kfree(xef);
@@ -333,7 +324,6 @@ struct xe_device *xe_device_create(struct pci_dev *pdev,
 	xe->info.force_execlist = xe_modparam.force_execlist;
 
 	spin_lock_init(&xe->irq.lock);
-	spin_lock_init(&xe->clients.lock);
 
 	init_waitqueue_head(&xe->ufence_wq);
 
diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
index 894f04770454..34620ef855c0 100644
--- a/drivers/gpu/drm/xe/xe_device.h
+++ b/drivers/gpu/drm/xe/xe_device.h
@@ -178,4 +178,18 @@ void xe_device_declare_wedged(struct xe_device *xe);
 struct xe_file *xe_file_get(struct xe_file *xef);
 void xe_file_put(struct xe_file *xef);
 
+/*
+ * Occasionally it is seen that the G2H worker starts running after a delay of more than
+ * a second even after being queued and activated by the Linux workqueue subsystem. This
+ * leads to G2H timeout error. The root cause of issue lies with scheduling latency of
+ * Lunarlake Hybrid CPU. Issue disappears if we disable Lunarlake atom cores from BIOS
+ * and this is beyond xe kmd.
+ *
+ * TODO: Drop this change once workqueue scheduling delay issue is fixed on LNL Hybrid CPU.
+ */
+#define LNL_FLUSH_WORKQUEUE(wq__) \
+	flush_workqueue(wq__)
+#define LNL_FLUSH_WORK(wrk__) \
+	flush_work(wrk__)
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 09d731a9125c..687f3a9039bb 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -353,15 +353,6 @@ struct xe_device {
 		struct workqueue_struct *wq;
 	} sriov;
 
-	/** @clients: drm clients info */
-	struct {
-		/** @clients.lock: Protects drm clients info */
-		spinlock_t lock;
-
-		/** @clients.count: number of drm clients */
-		u64 count;
-	} clients;
-
 	/** @usm: unified memory state */
 	struct {
 		/** @usm.asid: convert a ASID to VM */
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index f23ac1e2ed88..756b492f13b0 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -132,12 +132,16 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	if (XE_IOCTL_DBG(xe, !q))
 		return -ENOENT;
 
-	if (XE_IOCTL_DBG(xe, q->flags & EXEC_QUEUE_FLAG_VM))
-		return -EINVAL;
+	if (XE_IOCTL_DBG(xe, q->flags & EXEC_QUEUE_FLAG_VM)) {
+		err = -EINVAL;
+		goto err_exec_queue;
+	}
 
 	if (XE_IOCTL_DBG(xe, args->num_batch_buffer &&
-			 q->width != args->num_batch_buffer))
-		return -EINVAL;
+			 q->width != args->num_batch_buffer)) {
+		err = -EINVAL;
+		goto err_exec_queue;
+	}
 
 	if (XE_IOCTL_DBG(xe, q->ops->reset_status(q))) {
 		err = -ECANCELED;
@@ -220,6 +224,7 @@ retry:
 			fence = xe_sync_in_fence_get(syncs, num_syncs, q, vm);
 			if (IS_ERR(fence)) {
 				err = PTR_ERR(fence);
+				xe_vm_unlock(vm);
 				goto err_unlock_list;
 			}
 			for (i = 0; i < num_syncs; i++)
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index d098d2dd1b2d..fd0f3b3c9101 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -260,8 +260,14 @@ void xe_exec_queue_fini(struct xe_exec_queue *q)
 {
 	int i;
 
+	/*
+	 * Before releasing our ref to lrc and xef, accumulate our run ticks
+	 */
+	xe_exec_queue_update_run_ticks(q);
+
 	for (i = 0; i < q->width; ++i)
 		xe_lrc_put(q->lrc[i]);
+
 	__xe_exec_queue_free(q);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c
index d2e4dc3aaf61..ffcbd05671fc 100644
--- a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c
+++ b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c
@@ -68,6 +68,12 @@ static void __xe_gt_apply_ccs_mode(struct xe_gt *gt, u32 num_engines)
 		}
 	}
 
+	/*
+	 * Mask bits need to be set for the register. Though only Xe2+
+	 * platforms require setting of mask bits, it won't harm for older
+	 * platforms as these bits are unused there.
+	 */
+	mode |= CCS_MODE_CSLICE_0_3_MASK << 16;
 	xe_mmio_write32(gt, CCS_MODE, mode);
 
 	xe_gt_dbg(gt, "CCS_MODE=%x config:%08x, num_engines:%d, num_slices:%d\n",
@@ -133,9 +139,10 @@ ccs_mode_store(struct device *kdev, struct device_attribute *attr,
 	}
 
 	/* CCS mode can only be updated when there are no drm clients */
-	spin_lock(&xe->clients.lock);
-	if (xe->clients.count) {
-		spin_unlock(&xe->clients.lock);
+	mutex_lock(&xe->drm.filelist_mutex);
+	if (!list_empty(&xe->drm.filelist)) {
+		mutex_unlock(&xe->drm.filelist_mutex);
+		xe_gt_dbg(gt, "Rejecting compute mode change as there are active drm clients\n");
 		return -EBUSY;
 	}
 
@@ -146,7 +153,7 @@ ccs_mode_store(struct device *kdev, struct device_attribute *attr,
 		xe_gt_reset_async(gt);
 	}
 
-	spin_unlock(&xe->clients.lock);
+	mutex_unlock(&xe->drm.filelist_mutex);
 
 	return count;
 }
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index 8250ef71e685..afdb477ecf83 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -387,6 +387,8 @@ static void pf_release_ggtt(struct xe_tile *tile, struct xe_ggtt_node *node)
 		 * the xe_ggtt_clear() called by below xe_ggtt_remove_node().
 		 */
 		xe_ggtt_node_remove(node, false);
+	} else {
+		xe_ggtt_node_fini(node);
 	}
 }
 
@@ -442,7 +444,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size)
 	config->ggtt_region = node;
 	return 0;
 err:
-	xe_ggtt_node_fini(node);
+	pf_release_ggtt(tile, node);
 	return err;
 }
 
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
index bbb9e411d21f..9d82ea30f4df 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
@@ -72,6 +72,8 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_gt_tlb_invalidation_fence *fence, *next;
 
+	LNL_FLUSH_WORK(&gt->uc.guc.ct.g2h_worker);
+
 	spin_lock_irq(&gt->tlb_invalidation.pending_lock);
 	list_for_each_entry_safe(fence, next,
 				 &gt->tlb_invalidation.pending_fences, link) {
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 17986bfd8818..9c505d3517cd 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -897,17 +897,8 @@ retry_same_fence:
 
 	ret = wait_event_timeout(ct->g2h_fence_wq, g2h_fence.done, HZ);
 
-	/*
-	 * Occasionally it is seen that the G2H worker starts running after a delay of more than
-	 * a second even after being queued and activated by the Linux workqueue subsystem. This
-	 * leads to G2H timeout error. The root cause of issue lies with scheduling latency of
-	 * Lunarlake Hybrid CPU. Issue dissappears if we disable Lunarlake atom cores from BIOS
-	 * and this is beyond xe kmd.
-	 *
-	 * TODO: Drop this change once workqueue scheduling delay issue is fixed on LNL Hybrid CPU.
-	 */
 	if (!ret) {
-		flush_work(&ct->g2h_worker);
+		LNL_FLUSH_WORK(&ct->g2h_worker);
 		if (g2h_fence.done) {
 			xe_gt_warn(gt, "G2H fence %u, action %04x, done\n",
 				   g2h_fence.seqno, action[0]);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index f903b0772722..4f5d00aea716 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -745,8 +745,6 @@ static void guc_exec_queue_free_job(struct drm_sched_job *drm_job)
 {
 	struct xe_sched_job *job = to_xe_sched_job(drm_job);
 
-	xe_exec_queue_update_run_ticks(job->q);
-
 	trace_xe_sched_job_free(job);
 	xe_sched_job_put(job);
 }
diff --git a/drivers/gpu/drm/xe/xe_wait_user_fence.c b/drivers/gpu/drm/xe/xe_wait_user_fence.c
index f5deb81eba01..5b4264ea38bd 100644
--- a/drivers/gpu/drm/xe/xe_wait_user_fence.c
+++ b/drivers/gpu/drm/xe/xe_wait_user_fence.c
@@ -155,6 +155,13 @@ int xe_wait_user_fence_ioctl(struct drm_device *dev, void *data,
 		}
 
 		if (!timeout) {
+			LNL_FLUSH_WORKQUEUE(xe->ordered_wq);
+			err = do_compare(addr, args->value, args->mask,
+					 args->op);
+			if (err <= 0) {
+				drm_dbg(&xe->drm, "LNL_FLUSH_WORKQUEUE resolved ufence timeout\n");
+				break;
+			}
 			err = -ETIME;
 			break;
 		}