drm/amdgpu: save error count in RAS poison handler

Otherwise the RAS error count couldn't be queried from sysfs.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Tao Zhou
2021-12-20 16:36:54 +08:00
committed by Alex Deucher
parent 45e3d1db7d
commit fec8c5244f
3 changed files with 97 additions and 76 deletions

View File

@@ -727,7 +727,7 @@ void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bo
/* CPU MCA will handle page retirement if connected_to_cpu is 1 */
if (!adev->gmc.xgmi.connected_to_cpu)
amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset);
amdgpu_umc_poison_handler(adev, &err_data, reset);
else if (reset)
amdgpu_amdkfd_gpu_reset(adev);
}