Merge tag 'amd-drm-next-6.17-2025-07-01' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.17-2025-07-01:

amdgpu:
- FAMS2 fixes
- OLED fixes
- Misc cleanups
- AUX fixes
- DMCUB updates
- SR-IOV hibernation support
- RAS updates
- DP tunneling fixes
- DML2 fixes
- Backlight improvements
- Suspend improvements
- Use scaling for non-native modes on eDP
- SDMA 4.4.x fixes
- PCIe DPM fixes
- SDMA 5.x fixes
- Cleaner shader updates for GC 9.x
- Remove fence slab
- ISP genpd support
- Parition handling rework
- SDMA FW checks for userq support
- Add missing firmware declaration
- Fix leak in amdgpu_ctx_mgr_entity_fini()
- Freesync fix
- Ring reset refactoring
- Legacy dpm verbosity changes

amdkfd:
- GWS fix
- mtype fix for ext coherent system memory
- MMU notifier fix
- gfx7/8 fix

radeon:
- CS validation support for additional GL extensions
- Bump driver version for new CS validation checks

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250701194707.32905-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>

1 files changed, 302 insertions, 153 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index a59f194e3360..d282c0753b14 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -232,7 +232,7 @@ static int amdgpu_device_attr_sysfs_init(struct amdgpu_device *adev)
 {
 	int ret = 0;
 
-	if (!amdgpu_sriov_vf(adev))
+	if (amdgpu_nbio_is_replay_cnt_supported(adev))
 		ret = sysfs_create_file(&adev->dev->kobj,
 					&dev_attr_pcie_replay_count.attr);
 
@@ -241,7 +241,7 @@ static int amdgpu_device_attr_sysfs_init(struct amdgpu_device *adev)
 
 static void amdgpu_device_attr_sysfs_fini(struct amdgpu_device *adev)
 {
-	if (!amdgpu_sriov_vf(adev))
+	if (amdgpu_nbio_is_replay_cnt_supported(adev))
 		sysfs_remove_file(&adev->dev->kobj,
 				  &dev_attr_pcie_replay_count.attr);
 }
@@ -1288,14 +1288,14 @@ u32 amdgpu_device_get_rev_id(struct amdgpu_device *adev)
  */
 static uint32_t amdgpu_invalid_rreg(struct amdgpu_device *adev, uint32_t reg)
 {
-	DRM_ERROR("Invalid callback to read register 0x%04X\n", reg);
+	dev_err(adev->dev, "Invalid callback to read register 0x%04X\n", reg);
 	BUG();
 	return 0;
 }
 
 static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg)
 {
-	DRM_ERROR("Invalid callback to read register 0x%llX\n", reg);
+	dev_err(adev->dev, "Invalid callback to read register 0x%llX\n", reg);
 	BUG();
 	return 0;
 }
@@ -1312,15 +1312,17 @@ static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg
  */
 static void amdgpu_invalid_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
 {
-	DRM_ERROR("Invalid callback to write register 0x%04X with 0x%08X\n",
-		  reg, v);
+	dev_err(adev->dev,
+		"Invalid callback to write register 0x%04X with 0x%08X\n", reg,
+		v);
 	BUG();
 }
 
 static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, uint32_t v)
 {
-	DRM_ERROR("Invalid callback to write register 0x%llX with 0x%08X\n",
-		  reg, v);
+	dev_err(adev->dev,
+		"Invalid callback to write register 0x%llX with 0x%08X\n", reg,
+		v);
 	BUG();
 }
 
@@ -1336,14 +1338,15 @@ static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, ui
  */
 static uint64_t amdgpu_invalid_rreg64(struct amdgpu_device *adev, uint32_t reg)
 {
-	DRM_ERROR("Invalid callback to read 64 bit register 0x%04X\n", reg);
+	dev_err(adev->dev, "Invalid callback to read 64 bit register 0x%04X\n",
+		reg);
 	BUG();
 	return 0;
 }
 
 static uint64_t amdgpu_invalid_rreg64_ext(struct amdgpu_device *adev, uint64_t reg)
 {
-	DRM_ERROR("Invalid callback to read register 0x%llX\n", reg);
+	dev_err(adev->dev, "Invalid callback to read register 0x%llX\n", reg);
 	BUG();
 	return 0;
 }
@@ -1360,15 +1363,17 @@ static uint64_t amdgpu_invalid_rreg64_ext(struct amdgpu_device *adev, uint64_t r
  */
 static void amdgpu_invalid_wreg64(struct amdgpu_device *adev, uint32_t reg, uint64_t v)
 {
-	DRM_ERROR("Invalid callback to write 64 bit register 0x%04X with 0x%08llX\n",
-		  reg, v);
+	dev_err(adev->dev,
+		"Invalid callback to write 64 bit register 0x%04X with 0x%08llX\n",
+		reg, v);
 	BUG();
 }
 
 static void amdgpu_invalid_wreg64_ext(struct amdgpu_device *adev, uint64_t reg, uint64_t v)
 {
-	DRM_ERROR("Invalid callback to write 64 bit register 0x%llX with 0x%08llX\n",
-		  reg, v);
+	dev_err(adev->dev,
+		"Invalid callback to write 64 bit register 0x%llX with 0x%08llX\n",
+		reg, v);
 	BUG();
 }
 
@@ -1386,8 +1391,9 @@ static void amdgpu_invalid_wreg64_ext(struct amdgpu_device *adev, uint64_t reg,
 static uint32_t amdgpu_block_invalid_rreg(struct amdgpu_device *adev,
 					  uint32_t block, uint32_t reg)
 {
-	DRM_ERROR("Invalid callback to read register 0x%04X in block 0x%04X\n",
-		  reg, block);
+	dev_err(adev->dev,
+		"Invalid callback to read register 0x%04X in block 0x%04X\n",
+		reg, block);
 	BUG();
 	return 0;
 }
@@ -1407,8 +1413,9 @@ static void amdgpu_block_invalid_wreg(struct amdgpu_device *adev,
 				      uint32_t block,
 				      uint32_t reg, uint32_t v)
 {
-	DRM_ERROR("Invalid block callback to write register 0x%04X in block 0x%04X with 0x%08X\n",
-		  reg, block, v);
+	dev_err(adev->dev,
+		"Invalid block callback to write register 0x%04X in block 0x%04X with 0x%08X\n",
+		reg, block, v);
 	BUG();
 }
 
@@ -1694,7 +1701,9 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev)
 
 	/* PCI_EXT_CAP_ID_VNDR extended capability is located at 0x100 */
 	if (!pci_find_ext_capability(adev->pdev, PCI_EXT_CAP_ID_VNDR))
-		DRM_WARN("System can't access extended configuration space, please check!!\n");
+		dev_warn(
+			adev->dev,
+			"System can't access extended configuration space, please check!!\n");
 
 	/* skip if the bios has already enabled large BAR */
 	if (adev->gmc.real_vram_size &&
@@ -1734,9 +1743,10 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev)
 
 	r = pci_resize_resource(adev->pdev, 0, rbar_size);
 	if (r == -ENOSPC)
-		DRM_INFO("Not enough PCI address space for a large BAR.");
+		dev_info(adev->dev,
+			 "Not enough PCI address space for a large BAR.");
 	else if (r && r != -ENOTSUPP)
-		DRM_ERROR("Problem resizing BAR0 (%d).", r);
+		dev_err(adev->dev, "Problem resizing BAR0 (%d).", r);
 
 	pci_assign_unassigned_bus_resources(adev->pdev->bus);
 
@@ -1838,8 +1848,8 @@ bool amdgpu_device_seamless_boot_supported(struct amdgpu_device *adev)
 	case 0:
 		return false;
 	default:
-		DRM_ERROR("Invalid value for amdgpu.seamless: %d\n",
-			  amdgpu_seamless);
+		dev_err(adev->dev, "Invalid value for amdgpu.seamless: %d\n",
+			amdgpu_seamless);
 		return false;
 	}
 
@@ -2015,7 +2025,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev)
 		return;
 
 	if (!is_os_64) {
-		DRM_WARN("Not 64-bit OS, feature not supported\n");
+		dev_warn(adev->dev, "Not 64-bit OS, feature not supported\n");
 		goto def_value;
 	}
 	si_meminfo(&si);
@@ -2030,7 +2040,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev)
 		if (total_memory < dram_size_seven_GB)
 			goto def_value1;
 	} else {
-		DRM_WARN("Smu memory pool size not supported\n");
+		dev_warn(adev->dev, "Smu memory pool size not supported\n");
 		goto def_value;
 	}
 	adev->pm.smu_prv_buffer_size = amdgpu_smu_memory_pool_size << 28;
@@ -2038,7 +2048,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev)
 	return;
 
 def_value1:
-	DRM_WARN("No enough system memory\n");
+	dev_warn(adev->dev, "No enough system memory\n");
 def_value:
 	adev->pm.smu_prv_buffer_size = 0;
 }
@@ -2202,12 +2212,13 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev,
 		amdgpu_device_load_pci_state(pdev);
 		r = pci_enable_device(pdev);
 		if (r)
-			DRM_WARN("pci_enable_device failed (%d)\n", r);
+			dev_warn(&pdev->dev, "pci_enable_device failed (%d)\n",
+				 r);
 		amdgpu_device_resume(dev, true);
 
 		dev->switch_power_state = DRM_SWITCH_POWER_ON;
 	} else {
-		pr_info("switched off\n");
+		dev_info(&pdev->dev, "switched off\n");
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
 		amdgpu_device_prepare(dev);
 		amdgpu_device_suspend(dev, true);
@@ -2274,8 +2285,9 @@ int amdgpu_device_ip_set_clockgating_state(void *dev,
 		r = adev->ip_blocks[i].version->funcs->set_clockgating_state(
 			&adev->ip_blocks[i], state);
 		if (r)
-			DRM_ERROR("set_clockgating_state of IP block <%s> failed %d\n",
-				  adev->ip_blocks[i].version->funcs->name, r);
+			dev_err(adev->dev,
+				"set_clockgating_state of IP block <%s> failed %d\n",
+				adev->ip_blocks[i].version->funcs->name, r);
 	}
 	return r;
 }
@@ -2308,8 +2320,9 @@ int amdgpu_device_ip_set_powergating_state(void *dev,
 		r = adev->ip_blocks[i].version->funcs->set_powergating_state(
 			&adev->ip_blocks[i], state);
 		if (r)
-			DRM_ERROR("set_powergating_state of IP block <%s> failed %d\n",
-				  adev->ip_blocks[i].version->funcs->name, r);
+			dev_err(adev->dev,
+				"set_powergating_state of IP block <%s> failed %d\n",
+				adev->ip_blocks[i].version->funcs->name, r);
 	}
 	return r;
 }
@@ -2525,9 +2538,11 @@ static void amdgpu_device_enable_virtual_display(struct amdgpu_device *adev)
 			}
 		}
 
-		DRM_INFO("virtual display string:%s, %s:virtual_display:%d, num_crtc:%d\n",
-			 amdgpu_virtual_display, pci_address_name,
-			 adev->enable_virtual_display, adev->mode_info.num_crtc);
+		dev_info(
+			adev->dev,
+			"virtual display string:%s, %s:virtual_display:%d, num_crtc:%d\n",
+			amdgpu_virtual_display, pci_address_name,
+			adev->enable_virtual_display, adev->mode_info.num_crtc);
 
 		kfree(pciaddstr);
 	}
@@ -2538,8 +2553,9 @@ void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev)
 	if (amdgpu_sriov_vf(adev) && !adev->enable_virtual_display) {
 		adev->mode_info.num_crtc = 1;
 		adev->enable_virtual_display = true;
-		DRM_INFO("virtual_display:%d, num_crtc:%d\n",
-			 adev->enable_virtual_display, adev->mode_info.num_crtc);
+		dev_info(adev->dev, "virtual_display:%d, num_crtc:%d\n",
+			 adev->enable_virtual_display,
+			 adev->mode_info.num_crtc);
 	}
 }
 
@@ -2773,21 +2789,29 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 	if (!amdgpu_device_pcie_dynamic_switching_supported(adev))
 		adev->pm.pp_feature &= ~PP_PCIE_DPM_MASK;
 
+	adev->virt.is_xgmi_node_migrate_enabled = false;
+	if (amdgpu_sriov_vf(adev)) {
+		adev->virt.is_xgmi_node_migrate_enabled =
+			amdgpu_ip_version((adev), GC_HWIP, 0) == IP_VERSION(9, 4, 4);
+	}
+
 	total = true;
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		ip_block = &adev->ip_blocks[i];
 
 		if ((amdgpu_ip_block_mask & (1 << i)) == 0) {
-			DRM_WARN("disabled ip block: %d <%s>\n",
-				  i, adev->ip_blocks[i].version->funcs->name);
+			dev_warn(adev->dev, "disabled ip block: %d <%s>\n", i,
+				 adev->ip_blocks[i].version->funcs->name);
 			adev->ip_blocks[i].status.valid = false;
 		} else if (ip_block->version->funcs->early_init) {
 			r = ip_block->version->funcs->early_init(ip_block);
 			if (r == -ENOENT) {
 				adev->ip_blocks[i].status.valid = false;
 			} else if (r) {
-				DRM_ERROR("early_init of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_err(adev->dev,
+					"early_init of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 				total = false;
 			} else {
 				adev->ip_blocks[i].status.valid = true;
@@ -2868,8 +2892,10 @@ static int amdgpu_device_ip_hw_init_phase1(struct amdgpu_device *adev)
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) {
 			r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]);
 			if (r) {
-				DRM_ERROR("hw_init of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_err(adev->dev,
+					"hw_init of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 				return r;
 			}
 			adev->ip_blocks[i].status.hw = true;
@@ -2893,8 +2919,9 @@ static int amdgpu_device_ip_hw_init_phase2(struct amdgpu_device *adev)
 			continue;
 		r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]);
 		if (r) {
-			DRM_ERROR("hw_init of IP block <%s> failed %d\n",
-				  adev->ip_blocks[i].version->funcs->name, r);
+			dev_err(adev->dev,
+				"hw_init of IP block <%s> failed %d\n",
+				adev->ip_blocks[i].version->funcs->name, r);
 			return r;
 		}
 		adev->ip_blocks[i].status.hw = true;
@@ -2932,8 +2959,11 @@ static int amdgpu_device_fw_loading(struct amdgpu_device *adev)
 			} else {
 				r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]);
 				if (r) {
-					DRM_ERROR("hw_init of IP block <%s> failed %d\n",
-							  adev->ip_blocks[i].version->funcs->name, r);
+					dev_err(adev->dev,
+						"hw_init of IP block <%s> failed %d\n",
+						adev->ip_blocks[i]
+							.version->funcs->name,
+						r);
 					return r;
 				}
 				adev->ip_blocks[i].status.hw = true;
@@ -2988,25 +3018,29 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev)
 
 		r = drm_sched_init(&ring->sched, &args);
 		if (r) {
-			DRM_ERROR("Failed to create scheduler on ring %s.\n",
-				  ring->name);
+			dev_err(adev->dev,
+				"Failed to create scheduler on ring %s.\n",
+				ring->name);
 			return r;
 		}
 		r = amdgpu_uvd_entity_init(adev, ring);
 		if (r) {
-			DRM_ERROR("Failed to create UVD scheduling entity on ring %s.\n",
-				  ring->name);
+			dev_err(adev->dev,
+				"Failed to create UVD scheduling entity on ring %s.\n",
+				ring->name);
 			return r;
 		}
 		r = amdgpu_vce_entity_init(adev, ring);
 		if (r) {
-			DRM_ERROR("Failed to create VCE scheduling entity on ring %s.\n",
-				  ring->name);
+			dev_err(adev->dev,
+				"Failed to create VCE scheduling entity on ring %s.\n",
+				ring->name);
 			return r;
 		}
 	}
 
-	amdgpu_xcp_update_partition_sched_list(adev);
+	if (adev->xcp_mgr)
+		amdgpu_xcp_update_partition_sched_list(adev);
 
 	return 0;
 }
@@ -3038,8 +3072,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 		if (adev->ip_blocks[i].version->funcs->sw_init) {
 			r = adev->ip_blocks[i].version->funcs->sw_init(&adev->ip_blocks[i]);
 			if (r) {
-				DRM_ERROR("sw_init of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_err(adev->dev,
+					"sw_init of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 				goto init_failed;
 			}
 		}
@@ -3053,7 +3089,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			/* need to do common hw init early so everything is set up for gmc */
 			r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]);
 			if (r) {
-				DRM_ERROR("hw_init %d failed %d\n", i, r);
+				dev_err(adev->dev, "hw_init %d failed %d\n", i,
+					r);
 				goto init_failed;
 			}
 			adev->ip_blocks[i].status.hw = true;
@@ -3065,17 +3102,21 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 
 			r = amdgpu_device_mem_scratch_init(adev);
 			if (r) {
-				DRM_ERROR("amdgpu_mem_scratch_init failed %d\n", r);
+				dev_err(adev->dev,
+					"amdgpu_mem_scratch_init failed %d\n",
+					r);
 				goto init_failed;
 			}
 			r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]);
 			if (r) {
-				DRM_ERROR("hw_init %d failed %d\n", i, r);
+				dev_err(adev->dev, "hw_init %d failed %d\n", i,
+					r);
 				goto init_failed;
 			}
 			r = amdgpu_device_wb_init(adev);
 			if (r) {
-				DRM_ERROR("amdgpu_device_wb_init failed %d\n", r);
+				dev_err(adev->dev,
+					"amdgpu_device_wb_init failed %d\n", r);
 				goto init_failed;
 			}
 			adev->ip_blocks[i].status.hw = true;
@@ -3087,14 +3128,16 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 							       AMDGPU_GEM_DOMAIN_GTT,
 							       AMDGPU_CSA_SIZE);
 				if (r) {
-					DRM_ERROR("allocate CSA failed %d\n", r);
+					dev_err(adev->dev,
+						"allocate CSA failed %d\n", r);
 					goto init_failed;
 				}
 			}
 
 			r = amdgpu_seq64_init(adev);
 			if (r) {
-				DRM_ERROR("allocate seq64 failed %d\n", r);
+				dev_err(adev->dev, "allocate seq64 failed %d\n",
+					r);
 				goto init_failed;
 			}
 		}
@@ -3284,8 +3327,10 @@ int amdgpu_device_set_cg_state(struct amdgpu_device *adev,
 			r = adev->ip_blocks[i].version->funcs->set_clockgating_state(&adev->ip_blocks[i],
 										     state);
 			if (r) {
-				DRM_ERROR("set_clockgating_state(gate) of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_err(adev->dev,
+					"set_clockgating_state(gate) of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 				return r;
 			}
 		}
@@ -3321,8 +3366,10 @@ int amdgpu_device_set_pg_state(struct amdgpu_device *adev,
 			r = adev->ip_blocks[i].version->funcs->set_powergating_state(&adev->ip_blocks[i],
 											state);
 			if (r) {
-				DRM_ERROR("set_powergating_state(gate) of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_err(adev->dev,
+					"set_powergating_state(gate) of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 				return r;
 			}
 		}
@@ -3388,8 +3435,10 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 		if (adev->ip_blocks[i].version->funcs->late_init) {
 			r = adev->ip_blocks[i].version->funcs->late_init(&adev->ip_blocks[i]);
 			if (r) {
-				DRM_ERROR("late_init of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_err(adev->dev,
+					"late_init of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 				return r;
 			}
 		}
@@ -3398,7 +3447,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 
 	r = amdgpu_ras_late_init(adev);
 	if (r) {
-		DRM_ERROR("amdgpu_ras_late_init failed %d", r);
+		dev_err(adev->dev, "amdgpu_ras_late_init failed %d", r);
 		return r;
 	}
 
@@ -3412,7 +3461,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 
 	r = amdgpu_device_enable_mgpu_fan_boost();
 	if (r)
-		DRM_ERROR("enable mgpu fan boost failed (%d).\n", r);
+		dev_err(adev->dev, "enable mgpu fan boost failed (%d).\n", r);
 
 	/* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */
 	if (amdgpu_passthrough(adev) &&
@@ -3445,7 +3494,9 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 				r = amdgpu_xgmi_set_pstate(gpu_instance->adev,
 						AMDGPU_XGMI_PSTATE_MIN);
 				if (r) {
-					DRM_ERROR("pstate setting failed (%d).\n", r);
+					dev_err(adev->dev,
+						"pstate setting failed (%d).\n",
+						r);
 					break;
 				}
 			}
@@ -3459,17 +3510,19 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 
 static void amdgpu_ip_block_hw_fini(struct amdgpu_ip_block *ip_block)
 {
+	struct amdgpu_device *adev = ip_block->adev;
 	int r;
 
 	if (!ip_block->version->funcs->hw_fini) {
-		DRM_ERROR("hw_fini of IP block <%s> not defined\n",
-			  ip_block->version->funcs->name);
+		dev_err(adev->dev, "hw_fini of IP block <%s> not defined\n",
+			ip_block->version->funcs->name);
 	} else {
 		r = ip_block->version->funcs->hw_fini(ip_block);
 		/* XXX handle errors */
 		if (r) {
-			DRM_DEBUG("hw_fini of IP block <%s> failed %d\n",
-				  ip_block->version->funcs->name, r);
+			dev_dbg(adev->dev,
+				"hw_fini of IP block <%s> failed %d\n",
+				ip_block->version->funcs->name, r);
 		}
 	}
 
@@ -3510,15 +3563,16 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev)
 
 		r = adev->ip_blocks[i].version->funcs->early_fini(&adev->ip_blocks[i]);
 		if (r) {
-			DRM_DEBUG("early_fini of IP block <%s> failed %d\n",
-				  adev->ip_blocks[i].version->funcs->name, r);
+			dev_dbg(adev->dev,
+				"early_fini of IP block <%s> failed %d\n",
+				adev->ip_blocks[i].version->funcs->name, r);
 		}
 	}
 
 	amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE);
 	amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE);
 
-	amdgpu_amdkfd_suspend(adev, false);
+	amdgpu_amdkfd_suspend(adev, true);
 	amdgpu_userq_suspend(adev);
 
 	/* Workaround for ASICs need to disable SMC first */
@@ -3533,7 +3587,8 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev)
 
 	if (amdgpu_sriov_vf(adev)) {
 		if (amdgpu_virt_release_full_gpu(adev, false))
-			DRM_ERROR("failed to release exclusive mode on fini\n");
+			dev_err(adev->dev,
+				"failed to release exclusive mode on fini\n");
 	}
 
 	return 0;
@@ -3581,8 +3636,10 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 			r = adev->ip_blocks[i].version->funcs->sw_fini(&adev->ip_blocks[i]);
 			/* XXX handle errors */
 			if (r) {
-				DRM_DEBUG("sw_fini of IP block <%s> failed %d\n",
-					  adev->ip_blocks[i].version->funcs->name, r);
+				dev_dbg(adev->dev,
+					"sw_fini of IP block <%s> failed %d\n",
+					adev->ip_blocks[i].version->funcs->name,
+					r);
 			}
 		}
 		adev->ip_blocks[i].status.sw = false;
@@ -3615,7 +3672,7 @@ static void amdgpu_device_delayed_init_work_handler(struct work_struct *work)
 
 	r = amdgpu_ib_ring_tests(adev);
 	if (r)
-		DRM_ERROR("ib ring test failed (%d).\n", r);
+		dev_err(adev->dev, "ib ring test failed (%d).\n", r);
 }
 
 static void amdgpu_device_delay_enable_gfx_off(struct work_struct *work)
@@ -3756,8 +3813,9 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev)
 			if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) {
 				r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state);
 				if (r) {
-					DRM_ERROR("SMC failed to set mp1 state %d, %d\n",
-							adev->mp1_state, r);
+					dev_err(adev->dev,
+						"SMC failed to set mp1 state %d, %d\n",
+						adev->mp1_state, r);
 					return r;
 				}
 			}
@@ -4041,12 +4099,14 @@ static void amdgpu_device_detect_sriov_bios(struct amdgpu_device *adev)
 /**
  * amdgpu_device_asic_has_dc_support - determine if DC supports the asic
  *
+ * @pdev : pci device context
  * @asic_type: AMD asic type
  *
  * Check if there is DC (new modesetting infrastructre) support for an asic.
  * returns true if DC has support, false if not.
  */
-bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type)
+bool amdgpu_device_asic_has_dc_support(struct pci_dev *pdev,
+				       enum amd_asic_type asic_type)
 {
 	switch (asic_type) {
 #ifdef CONFIG_DRM_AMDGPU_SI
@@ -4089,7 +4149,9 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type)
 #else
 	default:
 		if (amdgpu_dc > 0)
-			DRM_INFO_ONCE("Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n");
+			dev_info_once(
+				&pdev->dev,
+				"Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n");
 		return false;
 #endif
 	}
@@ -4108,7 +4170,7 @@ bool amdgpu_device_has_dc_support(struct amdgpu_device *adev)
 	    (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK))
 		return false;
 
-	return amdgpu_device_asic_has_dc_support(adev->asic_type);
+	return amdgpu_device_asic_has_dc_support(adev->pdev, adev->asic_type);
 }
 
 static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
@@ -4150,7 +4212,8 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
 
 fail:
 	if (adev->asic_reset_res)
-		DRM_WARN("ASIC reset failed with error, %d for drm dev, %s",
+		dev_warn(adev->dev,
+			 "ASIC reset failed with error, %d for drm dev, %s",
 			 adev->asic_reset_res, adev_to_drm(adev)->unique);
 	amdgpu_put_xgmi_hive(hive);
 }
@@ -4274,7 +4337,7 @@ static void amdgpu_device_set_mcbp(struct amdgpu_device *adev)
 		adev->gfx.mcbp = true;
 
 	if (adev->gfx.mcbp)
-		DRM_INFO("MCBP is enabled\n");
+		dev_info(adev->dev, "MCBP is enabled\n");
 }
 
 /**
@@ -4342,9 +4405,11 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	adev->audio_endpt_rreg = &amdgpu_block_invalid_rreg;
 	adev->audio_endpt_wreg = &amdgpu_block_invalid_wreg;
 
-	DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X 0x%02X).\n",
-		 amdgpu_asic_name[adev->asic_type], pdev->vendor, pdev->device,
-		 pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision);
+	dev_info(
+		adev->dev,
+		"initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X 0x%02X).\n",
+		amdgpu_asic_name[adev->asic_type], pdev->vendor, pdev->device,
+		pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision);
 
 	/* mutex initialization are all done here so we
 	 * can recall function without having locking issues
@@ -4461,8 +4526,10 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	if (!adev->rmmio)
 		return -ENOMEM;
 
-	DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base);
-	DRM_INFO("register mmio size: %u\n", (unsigned int)adev->rmmio_size);
+	dev_info(adev->dev, "register mmio base: 0x%08X\n",
+		 (uint32_t)adev->rmmio_base);
+	dev_info(adev->dev, "register mmio size: %u\n",
+		 (unsigned int)adev->rmmio_size);
 
 	/*
 	 * Reset domain needs to be present early, before XGMI hive discovered
@@ -4599,7 +4666,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 			r = -EINVAL;
 			goto failed;
 		}
-		DRM_INFO("GPU posting now...\n");
+		dev_info(adev->dev, "GPU posting now...\n");
 		r = amdgpu_device_asic_init(adev);
 		if (r) {
 			dev_err(adev->dev, "gpu post error!\n");
@@ -4709,12 +4776,12 @@ fence_driver_init:
 
 	r = amdgpu_pm_sysfs_init(adev);
 	if (r)
-		DRM_ERROR("registering pm sysfs failed (%d).\n", r);
+		dev_err(adev->dev, "registering pm sysfs failed (%d).\n", r);
 
 	r = amdgpu_ucode_sysfs_init(adev);
 	if (r) {
 		adev->ucode_sysfs_en = false;
-		DRM_ERROR("Creating firmware sysfs failed (%d).\n", r);
+		dev_err(adev->dev, "Creating firmware sysfs failed (%d).\n", r);
 	} else
 		adev->ucode_sysfs_en = true;
 
@@ -4963,7 +5030,7 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev)
 
 	ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM);
 	if (ret)
-		DRM_WARN("evicting device resources failed\n");
+		dev_warn(adev->dev, "evicting device resources failed\n");
 	return ret;
 }
 
@@ -5035,6 +5102,28 @@ int amdgpu_device_prepare(struct drm_device *dev)
 }
 
 /**
+ * amdgpu_device_complete - complete power state transition
+ *
+ * @dev: drm dev pointer
+ *
+ * Undo the changes from amdgpu_device_prepare. This will be
+ * called on all resume transitions, including those that failed.
+ */
+void amdgpu_device_complete(struct drm_device *dev)
+{
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int i;
+
+	for (i = 0; i < adev->num_ip_blocks; i++) {
+		if (!adev->ip_blocks[i].status.valid)
+			continue;
+		if (!adev->ip_blocks[i].version->funcs->complete)
+			continue;
+		adev->ip_blocks[i].version->funcs->complete(&adev->ip_blocks[i]);
+	}
+}
+
+/**
  * amdgpu_device_suspend - initiate device suspend
  *
  * @dev: drm dev pointer
@@ -5055,6 +5144,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 	adev->in_suspend = true;
 
 	if (amdgpu_sriov_vf(adev)) {
+		if (!adev->in_s0ix && !adev->in_runpm)
+			amdgpu_amdkfd_suspend_process(adev);
 		amdgpu_virt_fini_data_exchange(adev);
 		r = amdgpu_virt_request_full_gpu(adev, false);
 		if (r)
@@ -5062,7 +5153,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 	}
 
 	if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3))
-		DRM_WARN("smart shift update failed\n");
+		dev_warn(adev->dev, "smart shift update failed\n");
 
 	if (notify_clients)
 		drm_client_dev_suspend(adev_to_drm(adev), false);
@@ -5074,7 +5165,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 	amdgpu_device_ip_suspend_phase1(adev);
 
 	if (!adev->in_s0ix) {
-		amdgpu_amdkfd_suspend(adev, adev->in_runpm);
+		amdgpu_amdkfd_suspend(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
 		amdgpu_userq_suspend(adev);
 	}
 
@@ -5098,6 +5189,32 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 	return 0;
 }
 
+static inline int amdgpu_virt_resume(struct amdgpu_device *adev)
+{
+	int r;
+	unsigned int prev_physical_node_id = adev->gmc.xgmi.physical_node_id;
+
+	/* During VM resume, QEMU programming of VF MSIX table (register GFXMSIX_VECT0_ADDR_LO)
+	 * may not work. The access could be blocked by nBIF protection as VF isn't in
+	 * exclusive access mode. Exclusive access is enabled now, disable/enable MSIX
+	 * so that QEMU reprograms MSIX table.
+	 */
+	amdgpu_restore_msix(adev);
+
+	r = adev->gfxhub.funcs->get_xgmi_info(adev);
+	if (r)
+		return r;
+
+	dev_info(adev->dev, "xgmi node, old id %d, new id %d\n",
+		prev_physical_node_id, adev->gmc.xgmi.physical_node_id);
+
+	adev->vm_manager.vram_base_offset = adev->gfxhub.funcs->get_mc_fb_offset(adev);
+	adev->vm_manager.vram_base_offset +=
+		adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
+
+	return 0;
+}
+
 /**
  * amdgpu_device_resume - initiate device resume
  *
@@ -5119,6 +5236,12 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
 			return r;
 	}
 
+	if (amdgpu_virt_xgmi_migrate_enabled(adev)) {
+		r = amdgpu_virt_resume(adev);
+		if (r)
+			goto exit;
+	}
+
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
@@ -5140,7 +5263,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
 	}
 
 	if (!adev->in_s0ix) {
-		r = amdgpu_amdkfd_resume(adev, adev->in_runpm);
+		r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
 		if (r)
 			goto exit;
 
@@ -5159,6 +5282,9 @@ exit:
 	if (amdgpu_sriov_vf(adev)) {
 		amdgpu_virt_init_data_exchange(adev);
 		amdgpu_virt_release_full_gpu(adev, true);
+
+		if (!adev->in_s0ix && !r && !adev->in_runpm)
+			r = amdgpu_amdkfd_resume_process(adev);
 	}
 
 	if (r)
@@ -5196,7 +5322,7 @@ exit:
 	adev->in_suspend = false;
 
 	if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D0))
-		DRM_WARN("smart shift update failed\n");
+		dev_warn(adev->dev, "smart shift update failed\n");
 
 	return 0;
 }
@@ -5727,7 +5853,9 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context)
 					amdgpu_coredump(tmp_adev, false, vram_lost, reset_context->job);
 
 				if (vram_lost) {
-					DRM_INFO("VRAM is lost due to GPU reset!\n");
+					dev_info(
+						tmp_adev->dev,
+						"VRAM is lost due to GPU reset!\n");
 					amdgpu_inc_vram_lost(tmp_adev);
 				}
 
@@ -6006,29 +6134,20 @@ static int amdgpu_device_health_check(struct list_head *device_list_handle)
 {
 	struct amdgpu_device *tmp_adev;
 	int ret = 0;
-	u32 status;
 
 	list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
-		pci_read_config_dword(tmp_adev->pdev, PCI_COMMAND, &status);
-		if (PCI_POSSIBLE_ERROR(status)) {
-			dev_err(tmp_adev->dev, "device lost from bus!");
-			ret = -ENODEV;
-		}
+		ret |= amdgpu_device_bus_status_check(tmp_adev);
 	}
 
 	return ret;
 }
 
-static int amdgpu_device_halt_activities(struct amdgpu_device *adev,
-			      struct amdgpu_job *job,
-			      struct amdgpu_reset_context *reset_context,
-			      struct list_head *device_list,
-			      struct amdgpu_hive_info *hive,
-			      bool need_emergency_restart)
+static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev,
+					  struct list_head *device_list,
+					  struct amdgpu_hive_info *hive)
 {
-	struct list_head *device_list_handle =  NULL;
 	struct amdgpu_device *tmp_adev = NULL;
-	int i, r = 0;
+	int r;
 
 	/*
 	 * Build list of devices to reset.
@@ -6045,26 +6164,55 @@ static int amdgpu_device_halt_activities(struct amdgpu_device *adev,
 		}
 		if (!list_is_first(&adev->reset_list, device_list))
 			list_rotate_to_front(&adev->reset_list, device_list);
-		device_list_handle = device_list;
 	} else {
 		list_add_tail(&adev->reset_list, device_list);
-		device_list_handle = device_list;
 	}
 
 	if (!amdgpu_sriov_vf(adev) && (!adev->pcie_reset_ctx.occurs_dpc)) {
-		r = amdgpu_device_health_check(device_list_handle);
+		r = amdgpu_device_health_check(device_list);
 		if (r)
 			return r;
 	}
 
-	/* We need to lock reset domain only once both for XGMI and single device */
-	tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device,
-				    reset_list);
+	return 0;
+}
+
+static void amdgpu_device_recovery_get_reset_lock(struct amdgpu_device *adev,
+						  struct list_head *device_list)
+{
+	struct amdgpu_device *tmp_adev = NULL;
+
+	if (list_empty(device_list))
+		return;
+	tmp_adev =
+		list_first_entry(device_list, struct amdgpu_device, reset_list);
 	amdgpu_device_lock_reset_domain(tmp_adev->reset_domain);
+}
 
-	/* block all schedulers and reset given job's ring */
-	list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
+static void amdgpu_device_recovery_put_reset_lock(struct amdgpu_device *adev,
+						  struct list_head *device_list)
+{
+	struct amdgpu_device *tmp_adev = NULL;
+
+	if (list_empty(device_list))
+		return;
+	tmp_adev =
+		list_first_entry(device_list, struct amdgpu_device, reset_list);
+	amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain);
+}
 
+static void amdgpu_device_halt_activities(struct amdgpu_device *adev,
+					  struct amdgpu_job *job,
+					  struct amdgpu_reset_context *reset_context,
+					  struct list_head *device_list,
+					  struct amdgpu_hive_info *hive,
+					  bool need_emergency_restart)
+{
+	struct amdgpu_device *tmp_adev = NULL;
+	int i;
+
+	/* block all schedulers and reset given job's ring */
+	list_for_each_entry(tmp_adev, device_list, reset_list) {
 		amdgpu_device_set_mp1_state(tmp_adev);
 
 		/*
@@ -6113,8 +6261,6 @@ static int amdgpu_device_halt_activities(struct amdgpu_device *adev,
 		}
 		atomic_inc(&tmp_adev->gpu_reset_counter);
 	}
-
-	return r;
 }
 
 static int amdgpu_device_asic_reset(struct amdgpu_device *adev,
@@ -6220,7 +6366,8 @@ static int amdgpu_device_sched_resume(struct list_head *device_list,
 		} else {
 			dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&tmp_adev->gpu_reset_counter));
 			if (amdgpu_acpi_smart_shift_update(adev_to_drm(tmp_adev), AMDGPU_SS_DEV_D0))
-				DRM_WARN("smart shift update failed\n");
+				dev_warn(tmp_adev->dev,
+					 "smart shift update failed\n");
 		}
 	}
 
@@ -6252,11 +6399,6 @@ static void amdgpu_device_gpu_resume(struct amdgpu_device *adev,
 		amdgpu_ras_set_error_query_ready(tmp_adev, true);
 
 	}
-
-	tmp_adev = list_first_entry(device_list, struct amdgpu_device,
-					    reset_list);
-	amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain);
-
 }
 
 
@@ -6306,7 +6448,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 	 */
 	if (need_emergency_restart && amdgpu_ras_get_context(adev) &&
 		amdgpu_ras_get_context(adev)->reboot) {
-		DRM_WARN("Emergency reboot.");
+		dev_warn(adev->dev, "Emergency reboot.");
 
 		ksys_sync_helper();
 		emergency_restart();
@@ -6324,11 +6466,14 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 	reset_context->hive = hive;
 	INIT_LIST_HEAD(&device_list);
 
-	r = amdgpu_device_halt_activities(adev, job, reset_context, &device_list,
-					 hive, need_emergency_restart);
-	if (r)
+	if (amdgpu_device_recovery_prepare(adev, &device_list, hive))
 		goto end_reset;
 
+	/* We need to lock reset domain only once both for XGMI and single device */
+	amdgpu_device_recovery_get_reset_lock(adev, &device_list);
+
+	amdgpu_device_halt_activities(adev, job, reset_context, &device_list,
+				      hive, need_emergency_restart);
 	if (need_emergency_restart)
 		goto skip_sched_resume;
 	/*
@@ -6337,7 +6482,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 	 *
 	 * job->base holds a reference to parent fence
 	 */
-	if (job && dma_fence_is_signaled(&job->hw_fence)) {
+	if (job && dma_fence_is_signaled(&job->hw_fence.base)) {
 		job_signaled = true;
 		dev_info(adev->dev, "Guilty job already signaled, skipping HW reset");
 		goto skip_hw_reset;
@@ -6345,13 +6490,15 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 
 	r = amdgpu_device_asic_reset(adev, &device_list, reset_context);
 	if (r)
-		goto end_reset;
+		goto reset_unlock;
 skip_hw_reset:
 	r = amdgpu_device_sched_resume(&device_list, reset_context, job_signaled);
 	if (r)
-		goto end_reset;
+		goto reset_unlock;
 skip_sched_resume:
 	amdgpu_device_gpu_resume(adev, &device_list, need_emergency_restart);
+reset_unlock:
+	amdgpu_device_recovery_put_reset_lock(adev, &device_list);
 end_reset:
 	if (hive) {
 		mutex_unlock(&hive->hive_lock);
@@ -6747,7 +6894,6 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
 	struct amdgpu_reset_context reset_context;
 	struct list_head device_list;
-	int r = 0;
 
 	dev_info(adev->dev, "PCI error: detected callback!!\n");
 
@@ -6772,14 +6918,14 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 		memset(&reset_context, 0, sizeof(reset_context));
 		INIT_LIST_HEAD(&device_list);
 
-		r = amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list,
-					 hive, false);
+		amdgpu_device_recovery_prepare(adev, &device_list, hive);
+		amdgpu_device_recovery_get_reset_lock(adev, &device_list);
+		amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list,
+					      hive, false);
 		if (hive) {
 			mutex_unlock(&hive->hive_lock);
 			amdgpu_put_xgmi_hive(hive);
 		}
-		if (r)
-			return PCI_ERS_RESULT_DISCONNECT;
 		return PCI_ERS_RESULT_NEED_RESET;
 	case pci_channel_io_perm_failure:
 		/* Permanent error, prepare for device removal */
@@ -6889,8 +7035,8 @@ out:
 		if (hive) {
 			list_for_each_entry(tmp_adev, &device_list, reset_list)
 				amdgpu_device_unset_mp1_state(tmp_adev);
-			amdgpu_device_unlock_reset_domain(adev->reset_domain);
 		}
+		amdgpu_device_recovery_put_reset_lock(adev, &device_list);
 	}
 
 	if (hive) {
@@ -6936,6 +7082,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 
 	amdgpu_device_sched_resume(&device_list, NULL, NULL);
 	amdgpu_device_gpu_resume(adev, &device_list, false);
+	amdgpu_device_recovery_put_reset_lock(adev, &device_list);
 	adev->pcie_reset_ctx.occurs_dpc = false;
 
 	if (hive) {
@@ -6960,11 +7107,11 @@ bool amdgpu_device_cache_pci_state(struct pci_dev *pdev)
 		adev->pci_state = pci_store_saved_state(pdev);
 
 		if (!adev->pci_state) {
-			DRM_ERROR("Failed to store PCI saved state");
+			dev_err(adev->dev, "Failed to store PCI saved state");
 			return false;
 		}
 	} else {
-		DRM_WARN("Failed to save PCI state, err:%d\n", r);
+		dev_warn(adev->dev, "Failed to save PCI state, err:%d\n", r);
 		return false;
 	}
 
@@ -6985,7 +7132,7 @@ bool amdgpu_device_load_pci_state(struct pci_dev *pdev)
 	if (!r) {
 		pci_restore_state(pdev);
 	} else {
-		DRM_WARN("Failed to load PCI state, err:%d\n", r);
+		dev_warn(adev->dev, "Failed to load PCI state, err:%d\n", r);
 		return false;
 	}
 
@@ -7231,7 +7378,7 @@ struct dma_fence *amdgpu_device_enforce_isolation(struct amdgpu_device *adev,
 	dep = amdgpu_sync_peek_fence(&isolation->prev, ring);
 	r = amdgpu_sync_fence(&isolation->active, &f->finished, GFP_NOWAIT);
 	if (r)
-		DRM_WARN("OOM tracking isolation\n");
+		dev_warn(adev->dev, "OOM tracking isolation\n");
 
 out_grab_ref:
 	dma_fence_get(dep);
@@ -7299,9 +7446,11 @@ uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev,
 		tmp_ = RREG32(reg_addr);
 		loop--;
 		if (!loop) {
-			DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn",
-				  inst, reg_name, (uint32_t)expected_value,
-				  (uint32_t)(tmp_ & (mask)));
+			dev_warn(
+				adev->dev,
+				"Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn",
+				inst, reg_name, (uint32_t)expected_value,
+				(uint32_t)(tmp_ & (mask)));
 			ret = -ETIMEDOUT;
 			break;
 		}