drm/amdgpu: set RAS fed status for more cases

[ Upstream commit 09a3d82 ]

Indicate fatal error for each RAS block and NBIO.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Author: Tao Zhou

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

@@ -2112,6 +2112,7 @@ static void amdgpu_ras_interrupt_umc_handler(struct ras_manager *obj,
 	/* Let IP handle its data, maybe we need get the output
 	 * from the callback to update the error type/count, etc
 	 */
+	amdgpu_ras_set_fed(obj->adev, true);
 	ret = data->cb(obj->adev, &err_data, entry);
 	/* ue will trigger an interrupt, and in that case
 	 * we need do a reset to recovery the whole system.

drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c

@@ -418,6 +418,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device
 		/* ras_controller_int is dedicated for nbif ras error,
 		 * not the global interrupt for sync flood
 		 */
+		amdgpu_ras_set_fed(adev, true);
 		amdgpu_ras_reset_gpu(adev);
 	}