1 files changed, 47 insertions, 36 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index bdfb80377e6a..7a899fb4de29 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5072,6 +5072,10 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev)
 	if (!adev->in_s4 && (adev->flags & AMD_IS_APU))
 		return 0;
 
+	/* No need to evict when going to S5 through S4 callbacks */
+	if (system_state == SYSTEM_POWER_OFF)
+		return 0;
+
 	ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM);
 	if (ret) {
 		dev_warn(adev->dev, "evicting device resources failed\n");
@@ -5196,7 +5200,7 @@ 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)
+		if (!adev->in_runpm)
 			amdgpu_amdkfd_suspend_process(adev);
 		amdgpu_virt_fini_data_exchange(adev);
 		r = amdgpu_virt_request_full_gpu(adev, false);
@@ -5216,10 +5220,8 @@ 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, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
-		amdgpu_userq_suspend(adev);
-	}
+	amdgpu_amdkfd_suspend(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
+	amdgpu_userq_suspend(adev);
 
 	r = amdgpu_device_evict_resources(adev);
 	if (r)
@@ -5314,15 +5316,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
 		goto exit;
 	}
 
-	if (!adev->in_s0ix) {
-		r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
-		if (r)
-			goto exit;
+	r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
+	if (r)
+		goto exit;
 
-		r = amdgpu_userq_resume(adev);
-		if (r)
-			goto exit;
-	}
+	r = amdgpu_userq_resume(adev);
+	if (r)
+		goto exit;
 
 	r = amdgpu_device_ip_late_init(adev);
 	if (r)
@@ -5335,7 +5335,7 @@ exit:
 		amdgpu_virt_init_data_exchange(adev);
 		amdgpu_virt_release_full_gpu(adev, true);
 
-		if (!adev->in_s0ix && !r && !adev->in_runpm)
+		if (!r && !adev->in_runpm)
 			r = amdgpu_amdkfd_resume_process(adev);
 	}
 
@@ -6389,23 +6389,28 @@ static int amdgpu_device_sched_resume(struct list_head *device_list,
 		if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled)
 			drm_helper_resume_force_mode(adev_to_drm(tmp_adev));
 
-		if (tmp_adev->asic_reset_res)
-			r = tmp_adev->asic_reset_res;
-
-		tmp_adev->asic_reset_res = 0;
-
-		if (r) {
+		if (tmp_adev->asic_reset_res) {
 			/* bad news, how to tell it to userspace ?
 			 * for ras error, we should report GPU bad status instead of
 			 * reset failure
 			 */
 			if (reset_context->src != AMDGPU_RESET_SRC_RAS ||
 			    !amdgpu_ras_eeprom_check_err_threshold(tmp_adev))
-				dev_info(tmp_adev->dev, "GPU reset(%d) failed\n",
-					atomic_read(&tmp_adev->gpu_reset_counter));
-			amdgpu_vf_error_put(tmp_adev, AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0, r);
+				dev_info(
+					tmp_adev->dev,
+					"GPU reset(%d) failed with error %d \n",
+					atomic_read(
+						&tmp_adev->gpu_reset_counter),
+					tmp_adev->asic_reset_res);
+			amdgpu_vf_error_put(tmp_adev,
+					    AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0,
+					    tmp_adev->asic_reset_res);
+			if (!r)
+				r = tmp_adev->asic_reset_res;
+			tmp_adev->asic_reset_res = 0;
 		} else {
-			dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&tmp_adev->gpu_reset_counter));
+			dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n",
+				 atomic_read(&tmp_adev->gpu_reset_counter));
 			if (amdgpu_acpi_smart_shift_update(tmp_adev,
 							   AMDGPU_SS_DEV_D0))
 				dev_warn(tmp_adev->dev,
@@ -6937,7 +6942,8 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
+	struct amdgpu_hive_info *hive __free(xgmi_put_hive) =
+		amdgpu_get_xgmi_hive(adev);
 	struct amdgpu_reset_context reset_context;
 	struct list_head device_list;
 
@@ -6976,10 +6982,8 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 		amdgpu_device_recovery_get_reset_lock(adev, &device_list);
 		amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list,
 					      hive, false);
-		if (hive) {
+		if (hive)
 			mutex_unlock(&hive->hive_lock);
-			amdgpu_put_xgmi_hive(hive);
-		}
 		return PCI_ERS_RESULT_NEED_RESET;
 	case pci_channel_io_perm_failure:
 		/* Permanent error, prepare for device removal */
@@ -7158,28 +7162,35 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 
 static void amdgpu_device_cache_switch_state(struct amdgpu_device *adev)
 {
-	struct pci_dev *parent = pci_upstream_bridge(adev->pdev);
+	struct pci_dev *swus, *swds;
 	int r;
 
-	if (parent->vendor != PCI_VENDOR_ID_ATI)
+	swds = pci_upstream_bridge(adev->pdev);
+	if (!swds || swds->vendor != PCI_VENDOR_ID_ATI ||
+	    pci_pcie_type(swds) != PCI_EXP_TYPE_DOWNSTREAM)
+		return;
+	swus = pci_upstream_bridge(swds);
+	if (!swus ||
+	    (swus->vendor != PCI_VENDOR_ID_ATI &&
+	     swus->vendor != PCI_VENDOR_ID_AMD) ||
+	    pci_pcie_type(swus) != PCI_EXP_TYPE_UPSTREAM)
 		return;
 
 	/* If already saved, return */
 	if (adev->pcie_reset_ctx.swus)
 		return;
 	/* Upstream bridge is ATI, assume it's SWUS/DS architecture */
-	r = pci_save_state(parent);
+	r = pci_save_state(swds);
 	if (r)
 		return;
-	adev->pcie_reset_ctx.swds_pcistate = pci_store_saved_state(parent);
+	adev->pcie_reset_ctx.swds_pcistate = pci_store_saved_state(swds);
 
-	parent = pci_upstream_bridge(parent);
-	r = pci_save_state(parent);
+	r = pci_save_state(swus);
 	if (r)
 		return;
-	adev->pcie_reset_ctx.swus_pcistate = pci_store_saved_state(parent);
+	adev->pcie_reset_ctx.swus_pcistate = pci_store_saved_state(swus);
 
-	adev->pcie_reset_ctx.swus = parent;
+	adev->pcie_reset_ctx.swus = swus;
 }
 
 static void amdgpu_device_load_switch_state(struct amdgpu_device *adev)