1 files changed, 77 insertions, 10 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index 6fa9fa11c8f3..5bf9be073cdd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -47,7 +47,7 @@ static int amdgpu_mes_doorbell_init(struct amdgpu_device *adev)
 	/* Bitmap for dynamic allocation of kernel doorbells */
 	mes->doorbell_bitmap = bitmap_zalloc(PAGE_SIZE / sizeof(u32), GFP_KERNEL);
 	if (!mes->doorbell_bitmap) {
-		DRM_ERROR("Failed to allocate MES doorbell bitmap\n");
+		dev_err(adev->dev, "Failed to allocate MES doorbell bitmap\n");
 		return -ENOMEM;
 	}
 
@@ -191,6 +191,20 @@ int amdgpu_mes_init(struct amdgpu_device *adev)
 	if (r)
 		goto error_doorbell;
 
+	if (adev->mes.hung_queue_db_array_size) {
+		r = amdgpu_bo_create_kernel(adev,
+					    adev->mes.hung_queue_db_array_size * sizeof(u32),
+					    PAGE_SIZE,
+					    AMDGPU_GEM_DOMAIN_GTT,
+					    &adev->mes.hung_queue_db_array_gpu_obj,
+					    &adev->mes.hung_queue_db_array_gpu_addr,
+					    &adev->mes.hung_queue_db_array_cpu_addr);
+		if (r) {
+			dev_warn(adev->dev, "failed to create MES hung db array buffer (%d)", r);
+			goto error_doorbell;
+		}
+	}
+
 	return 0;
 
 error_doorbell:
@@ -216,6 +230,10 @@ void amdgpu_mes_fini(struct amdgpu_device *adev)
 {
 	int i;
 
+	amdgpu_bo_free_kernel(&adev->mes.hung_queue_db_array_gpu_obj,
+			      &adev->mes.hung_queue_db_array_gpu_addr,
+			      &adev->mes.hung_queue_db_array_cpu_addr);
+
 	amdgpu_bo_free_kernel(&adev->mes.event_log_gpu_obj,
 			      &adev->mes.event_log_gpu_addr,
 			      &adev->mes.event_log_cpu_addr);
@@ -256,7 +274,7 @@ int amdgpu_mes_suspend(struct amdgpu_device *adev)
 	r = adev->mes.funcs->suspend_gang(&adev->mes, &input);
 	amdgpu_mes_unlock(&adev->mes);
 	if (r)
-		DRM_ERROR("failed to suspend all gangs");
+		dev_err(adev->dev, "failed to suspend all gangs");
 
 	return r;
 }
@@ -280,7 +298,7 @@ int amdgpu_mes_resume(struct amdgpu_device *adev)
 	r = adev->mes.funcs->resume_gang(&adev->mes, &input);
 	amdgpu_mes_unlock(&adev->mes);
 	if (r)
-		DRM_ERROR("failed to resume all gangs");
+		dev_err(adev->dev, "failed to resume all gangs");
 
 	return r;
 }
@@ -304,7 +322,7 @@ int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev,
 	r = adev->mes.funcs->map_legacy_queue(&adev->mes, &queue_input);
 	amdgpu_mes_unlock(&adev->mes);
 	if (r)
-		DRM_ERROR("failed to map legacy queue\n");
+		dev_err(adev->dev, "failed to map legacy queue\n");
 
 	return r;
 }
@@ -329,7 +347,7 @@ int amdgpu_mes_unmap_legacy_queue(struct amdgpu_device *adev,
 	r = adev->mes.funcs->unmap_legacy_queue(&adev->mes, &queue_input);
 	amdgpu_mes_unlock(&adev->mes);
 	if (r)
-		DRM_ERROR("failed to unmap legacy queue\n");
+		dev_err(adev->dev, "failed to unmap legacy queue\n");
 
 	return r;
 }
@@ -361,7 +379,54 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
 	r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input);
 	amdgpu_mes_unlock(&adev->mes);
 	if (r)
-		DRM_ERROR("failed to reset legacy queue\n");
+		dev_err(adev->dev, "failed to reset legacy queue\n");
+
+	return r;
+}
+
+int amdgpu_mes_get_hung_queue_db_array_size(struct amdgpu_device *adev)
+{
+	return adev->mes.hung_queue_db_array_size;
+}
+
+int amdgpu_mes_detect_and_reset_hung_queues(struct amdgpu_device *adev,
+					    int queue_type,
+					    bool detect_only,
+					    unsigned int *hung_db_num,
+					    u32 *hung_db_array)
+
+{
+	struct mes_detect_and_reset_queue_input input;
+	u32 *db_array = adev->mes.hung_queue_db_array_cpu_addr;
+	int r, i;
+
+	if (!hung_db_num || !hung_db_array)
+		return -EINVAL;
+
+	if ((queue_type != AMDGPU_RING_TYPE_GFX) &&
+	    (queue_type != AMDGPU_RING_TYPE_COMPUTE) &&
+	    (queue_type != AMDGPU_RING_TYPE_SDMA))
+		return -EINVAL;
+
+	/* Clear the doorbell array before detection */
+	memset(adev->mes.hung_queue_db_array_cpu_addr, 0,
+		adev->mes.hung_queue_db_array_size * sizeof(u32));
+	input.queue_type = queue_type;
+	input.detect_only = detect_only;
+
+	r = adev->mes.funcs->detect_and_reset_hung_queues(&adev->mes,
+							  &input);
+	if (r) {
+		dev_err(adev->dev, "failed to detect and reset\n");
+	} else {
+		*hung_db_num = 0;
+		for (i = 0; i < adev->mes.hung_queue_db_array_size; i++) {
+			if (db_array[i] != AMDGPU_MES_INVALID_DB_OFFSET) {
+				hung_db_array[i] = db_array[i];
+				*hung_db_num += 1;
+			}
+		}
+	}
 
 	return r;
 }
@@ -469,7 +534,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 	int r;
 
 	if (!adev->mes.funcs->misc_op) {
-		DRM_ERROR("mes set shader debugger is not supported!\n");
+		dev_err(adev->dev,
+			"mes set shader debugger is not supported!\n");
 		return -EINVAL;
 	}
 
@@ -493,7 +559,7 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
 	if (r)
-		DRM_ERROR("failed to set_shader_debugger\n");
+		dev_err(adev->dev, "failed to set_shader_debugger\n");
 
 	amdgpu_mes_unlock(&adev->mes);
 
@@ -507,7 +573,8 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev,
 	int r;
 
 	if (!adev->mes.funcs->misc_op) {
-		DRM_ERROR("mes flush shader debugger is not supported!\n");
+		dev_err(adev->dev,
+			"mes flush shader debugger is not supported!\n");
 		return -EINVAL;
 	}
 
@@ -519,7 +586,7 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev,
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
 	if (r)
-		DRM_ERROR("failed to set_shader_debugger\n");
+		dev_err(adev->dev, "failed to set_shader_debugger\n");
 
 	amdgpu_mes_unlock(&adev->mes);