Merge tag 'amd-drm-fixes-5.12-2021-03-31' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-5.12-2021-03-31:

amdgpu:
- Polaris idle power fix
- VM fix
- Vangogh S3 fix
- Fixes for non-4K page sizes

amdkfd:
- dqm fence memory corruption fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210401020057.17831-1-alexander.deucher@amd.com

drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c

@@ -778,9 +778,9 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 			dev_info->high_va_offset = AMDGPU_GMC_HOLE_END;
 			dev_info->high_va_max = AMDGPU_GMC_HOLE_END | vm_size;
 		}
-		dev_info->virtual_address_alignment = max((int)PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE);
+		dev_info->virtual_address_alignment = max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE);
 		dev_info->pte_fragment_size = (1 << adev->vm_manager.fragment_size) * AMDGPU_GPU_PAGE_SIZE;
-		dev_info->gart_page_size = AMDGPU_GPU_PAGE_SIZE;
+		dev_info->gart_page_size = max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE);
 		dev_info->cu_active_number = adev->gfx.cu_info.number;
 		dev_info->cu_ao_mask = adev->gfx.cu_info.ao_cu_mask;
 		dev_info->ce_ram_size = adev->gfx.ce_ram_size;

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

@@ -2197,8 +2197,8 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	uint64_t eaddr;
 
 	/* validate the parameters */
-	if (saddr & AMDGPU_GPU_PAGE_MASK || offset & AMDGPU_GPU_PAGE_MASK ||
-	    size == 0 || size & AMDGPU_GPU_PAGE_MASK)
+	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK ||
+	    size == 0 || size & ~PAGE_MASK)
 		return -EINVAL;
 
 	/* make sure object fit at this offset */
@@ -2263,8 +2263,8 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 	int r;
 
 	/* validate the parameters */
-	if (saddr & AMDGPU_GPU_PAGE_MASK || offset & AMDGPU_GPU_PAGE_MASK ||
-	    size == 0 || size & AMDGPU_GPU_PAGE_MASK)
+	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK ||
+	    size == 0 || size & ~PAGE_MASK)
 		return -EINVAL;
 
 	/* make sure object fit at this offset */
@@ -2409,7 +2409,7 @@ int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev,
 			after->start = eaddr + 1;
 			after->last = tmp->last;
 			after->offset = tmp->offset;
-			after->offset += after->start - tmp->start;
+			after->offset += (after->start - tmp->start) << PAGE_SHIFT;
 			after->flags = tmp->flags;
 			after->bo_va = tmp->bo_va;
 			list_add(&after->list, &tmp->bo_va->invalids);

drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c

@@ -155,7 +155,7 @@ static int dbgdev_diq_submit_ib(struct kfd_dbgdev *dbgdev,
 
 	/* Wait till CP writes sync code: */
 	status = amdkfd_fence_wait_timeout(
-			(unsigned int *) rm_state,
+			rm_state,
 			QUEUESTATE__ACTIVE, 1500);
 
 	kfd_gtt_sa_free(dbgdev->dev, mem_obj);

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c

@@ -1167,7 +1167,7 @@ static int start_cpsch(struct device_queue_manager *dqm)
 	if (retval)
 		goto fail_allocate_vidmem;
 
-	dqm->fence_addr = dqm->fence_mem->cpu_ptr;
+	dqm->fence_addr = (uint64_t *)dqm->fence_mem->cpu_ptr;
 	dqm->fence_gpu_addr = dqm->fence_mem->gpu_addr;
 
 	init_interrupts(dqm);
@@ -1340,8 +1340,8 @@ out:
 	return retval;
 }
 
-int amdkfd_fence_wait_timeout(unsigned int *fence_addr,
-				unsigned int fence_value,
+int amdkfd_fence_wait_timeout(uint64_t *fence_addr,
+				uint64_t fence_value,
 				unsigned int timeout_ms)
 {
 	unsigned long end_jiffies = msecs_to_jiffies(timeout_ms) + jiffies;

drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h

@@ -192,7 +192,7 @@ struct device_queue_manager {
 	uint16_t		vmid_pasid[VMID_NUM];
 	uint64_t		pipelines_addr;
 	uint64_t		fence_gpu_addr;
-	unsigned int		*fence_addr;
+	uint64_t		*fence_addr;
 	struct kfd_mem_obj	*fence_mem;
 	bool			active_runlist;
 	int			sched_policy;

drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c

@@ -347,7 +347,7 @@ fail_create_runlist_ib:
 }
 
 int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address,
-			uint32_t fence_value)
+			uint64_t fence_value)
 {
 	uint32_t *buffer, size;
 	int retval = 0;

drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c

@@ -283,7 +283,7 @@ static int pm_unmap_queues_v9(struct packet_manager *pm, uint32_t *buffer,
 }
 
 static int pm_query_status_v9(struct packet_manager *pm, uint32_t *buffer,
-			uint64_t fence_address,	uint32_t fence_value)
+			uint64_t fence_address,	uint64_t fence_value)
 {
 	struct pm4_mes_query_status *packet;
 

drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c

@@ -263,7 +263,7 @@ static int pm_unmap_queues_vi(struct packet_manager *pm, uint32_t *buffer,
 }
 
 static int pm_query_status_vi(struct packet_manager *pm, uint32_t *buffer,
-			uint64_t fence_address,	uint32_t fence_value)
+			uint64_t fence_address,	uint64_t fence_value)
 {
 	struct pm4_mes_query_status *packet;
 

drivers/gpu/drm/amd/amdkfd/kfd_priv.h

@@ -1003,8 +1003,8 @@ int pqm_get_wave_state(struct process_queue_manager *pqm,
 		       u32 *ctl_stack_used_size,
 		       u32 *save_area_used_size);
 
-int amdkfd_fence_wait_timeout(unsigned int *fence_addr,
-			      unsigned int fence_value,
+int amdkfd_fence_wait_timeout(uint64_t *fence_addr,
+			      uint64_t fence_value,
 			      unsigned int timeout_ms);
 
 /* Packet Manager */
@@ -1040,7 +1040,7 @@ struct packet_manager_funcs {
 			uint32_t filter_param, bool reset,
 			unsigned int sdma_engine);
 	int (*query_status)(struct packet_manager *pm, uint32_t *buffer,
-			uint64_t fence_address,	uint32_t fence_value);
+			uint64_t fence_address,	uint64_t fence_value);
 	int (*release_mem)(uint64_t gpu_addr, uint32_t *buffer);
 
 	/* Packet sizes */
@@ -1062,7 +1062,7 @@ int pm_send_set_resources(struct packet_manager *pm,
 				struct scheduling_resources *res);
 int pm_send_runlist(struct packet_manager *pm, struct list_head *dqm_queues);
 int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address,
-				uint32_t fence_value);
+				uint64_t fence_value);
 
 int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_queue_type type,
 			enum kfd_unmap_queues_filter mode,

drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c

@@ -3330,7 +3330,8 @@ static int smu7_apply_state_adjust_rules(struct pp_hwmgr *hwmgr,
 
 	disable_mclk_switching_for_display = ((1 < hwmgr->display_config->num_display) &&
 						!hwmgr->display_config->multi_monitor_in_sync) ||
-						smu7_vblank_too_short(hwmgr, hwmgr->display_config->min_vblank_time);
+						(hwmgr->display_config->num_display &&
+						smu7_vblank_too_short(hwmgr, hwmgr->display_config->min_vblank_time));
 
 	disable_mclk_switching = disable_mclk_switching_for_frame_lock ||
 					 disable_mclk_switching_for_display;

drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c

@@ -384,10 +384,15 @@ static int vangogh_dpm_set_jpeg_enable(struct smu_context *smu, bool enable)
 
 static bool vangogh_is_dpm_running(struct smu_context *smu)
 {
+	struct amdgpu_device *adev = smu->adev;
 	int ret = 0;
 	uint32_t feature_mask[2];
 	uint64_t feature_enabled;
 
+	/* we need to re-init after suspend so return false */
+	if (adev->in_suspend)
+		return false;
+
 	ret = smu_cmn_get_enabled_32_bits_mask(smu, feature_mask, 2);
 
 	if (ret)