drm/amdgpu: add an INFO query for monitoring VRAM CPU page faults

v2: bump the DRM version

Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -1529,6 +1529,7 @@ struct amdgpu_device {
 	atomic64_t			gtt_usage;
 	atomic64_t			num_bytes_moved;
 	atomic64_t			num_evictions;
+	atomic64_t			num_vram_cpu_page_faults;
 	atomic_t			gpu_reset_counter;
 	atomic_t			vram_lost_counter;
 

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

@@ -66,9 +66,10 @@
  * - 3.14.0 - Fix race in amdgpu_ctx_get_fence() and note new functionality
  * - 3.15.0 - Export more gpu info for gfx9
  * - 3.16.0 - Add reserved vmid support
+ * - 3.17.0 - Add AMDGPU_NUM_VRAM_CPU_PAGE_FAULTS.
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	16
+#define KMS_DRIVER_MINOR	17
 #define KMS_DRIVER_PATCHLEVEL	0
 
 int amdgpu_vram_limit = 0;

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

@@ -417,6 +417,9 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 	case AMDGPU_INFO_NUM_EVICTIONS:
 		ui64 = atomic64_read(&adev->num_evictions);
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
+	case AMDGPU_INFO_NUM_VRAM_CPU_PAGE_FAULTS:
+		ui64 = atomic64_read(&adev->num_vram_cpu_page_faults);
+		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
 	case AMDGPU_INFO_VRAM_USAGE:
 		ui64 = atomic64_read(&adev->vram_usage);
 		return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;

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

@@ -960,6 +960,7 @@ int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 		return -EINVAL;
 
 	/* hurrah the memory is not visible ! */
+	atomic64_inc(&adev->num_vram_cpu_page_faults);
 	amdgpu_ttm_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_VRAM);
 	lpfn =	adev->mc.visible_vram_size >> PAGE_SHIFT;
 	for (i = 0; i < abo->placement.num_placement; i++) {

include/uapi/drm/amdgpu_drm.h

@@ -603,6 +603,8 @@ struct drm_amdgpu_cs_chunk_data {
 	#define AMDGPU_INFO_SENSOR_VDDNB		0x6
 	/* Subquery id: Query graphics voltage */
 	#define AMDGPU_INFO_SENSOR_VDDGFX		0x7
+/* Number of VRAM page faults on CPU access. */
+#define AMDGPU_INFO_NUM_VRAM_CPU_PAGE_FAULTS	0x1E
 
 #define AMDGPU_INFO_MMR_SE_INDEX_SHIFT	0
 #define AMDGPU_INFO_MMR_SE_INDEX_MASK	0xff