Commit 792be2e2 authored by Tao Zhou's avatar Tao Zhou Committed by Alex Deucher
Browse files

drm/amdgpu: create function to check RAS RMA status 



In the convenience of calling it globally.

Signed-off-by: default avatarTao Zhou <tao.zhou1@amd.com>
Reviewed-by: default avatarHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent 64a90520

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

+16 −6
Original line number Diff line number Diff line
@@ -2153,7 +2153,7 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager * 
	/* gpu reset is fallback for failed and default cases.

	 * For RMA case, amdgpu_umc_poison_handler will handle gpu reset.

	 */

	if (poison_stat && !con->is_rma) {

	if (poison_stat && !amdgpu_ras_is_rma(adev)) {

		event_id = amdgpu_ras_acquire_event_id(adev, type);

		RAS_EVENT_LOG(adev, event_id,

			      "GPU reset for %s RAS poison consumption is issued!\n",
@@ -2945,7 +2945,7 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work) 


	amdgpu_ras_error_data_fini(&err_data);



	if (err_cnt && con->is_rma)

	if (err_cnt && amdgpu_ras_is_rma(adev))

		amdgpu_ras_reset_gpu(adev);



	amdgpu_ras_schedule_retirement_dwork(con,
@@ -3046,7 +3046,7 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev, 
	}



	/* for RMA, amdgpu_ras_poison_creation_handler will trigger gpu reset */

	if (reset_flags && !con->is_rma) {

	if (reset_flags && !amdgpu_ras_is_rma(adev)) {

		if (reset_flags & AMDGPU_RAS_GPU_RESET_MODE1_RESET)

			reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;

		else if (reset_flags & AMDGPU_RAS_GPU_RESET_MODE2_RESET)
@@ -3192,7 +3192,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) 
	 * This calling fails when is_rma is true or

	 * ret != 0.

	 */

	if (con->is_rma || ret)

	if (amdgpu_ras_is_rma(adev) || ret)

		goto free;



	if (con->eeprom_control.ras_num_recs) {
@@ -3241,7 +3241,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) 
	 * Except error threshold exceeding case, other failure cases in this

	 * function would not fail amdgpu driver init.

	 */

	if (!con->is_rma)

	if (!amdgpu_ras_is_rma(adev))

		ret = 0;

	else

		ret = -EINVAL;
@@ -4284,7 +4284,7 @@ int amdgpu_ras_reset_gpu(struct amdgpu_device *adev) 
	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);



	/* mode1 is the only selection for RMA status */

	if (ras->is_rma) {

	if (amdgpu_ras_is_rma(adev)) {

		ras->gpu_reset_flags = 0;

		ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE1_RESET;

	}
@@ -4824,3 +4824,13 @@ void amdgpu_ras_event_log_print(struct amdgpu_device *adev, u64 event_id, 


	va_end(args);

}



bool amdgpu_ras_is_rma(struct amdgpu_device *adev)

{

	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);



	if (!con)

		return false;



	return con->is_rma;

}

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

+1 −0
Original line number Diff line number Diff line
@@ -974,4 +974,5 @@ __printf(3, 4) 
void amdgpu_ras_event_log_print(struct amdgpu_device *adev, u64 event_id,

				const char *fmt, ...);



bool amdgpu_ras_is_rma(struct amdgpu_device *adev);

#endif

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

+1 −1
Original line number Diff line number Diff line
@@ -196,7 +196,7 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, 
	amdgpu_umc_handle_bad_pages(adev, ras_error_status);



	if ((err_data->ue_count || err_data->de_count) &&

	    (reset || (con && con->is_rma))) {

	    (reset || amdgpu_ras_is_rma(adev))) {

		con->gpu_reset_flags |= reset;

		amdgpu_ras_reset_gpu(adev);

	}

drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c

+1 −1
Original line number Diff line number Diff line
@@ -97,7 +97,7 @@ static int gfx_v11_0_3_poison_consumption_handler(struct amdgpu_device *adev, 
			ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET;

		}



		if (con && !con->is_rma)

		if (con && !amdgpu_ras_is_rma(adev))

			amdgpu_ras_reset_gpu(adev);

	}