drm/amdkfd: reset queue which consumes RAS poison (v2)
authorTao Zhou <tao.zhou1@amd.com>
Mon, 6 Dec 2021 07:54:54 +0000 (15:54 +0800)
committerAlex Deucher <alexander.deucher@amd.com>
Tue, 28 Dec 2021 21:02:59 +0000 (16:02 -0500)
commitb6485bed40d7859735bdbfedbd55dcc8366a88a7
tree367edc60dcecfb6eaca11767011032e81bccae5e
parentdec63443380cf2ce6bc41258159cda4165fa4279
drm/amdkfd: reset queue which consumes RAS poison (v2)

CP supports unmap queue with reset mode which only destroys specific queue without affecting others.
Replacing whole gpu reset with reset queue mode for RAS poison consumption
saves much time, and we can also fallback to gpu reset solution if reset
queue fails.

v2: Return directly if process is NULL;
    Reset queue solution is not applicable to SDMA, fallback to legacy
way;
    Call kfd_unref_process after lookup process.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
drivers/gpu/drm/amd/amdkfd/kfd_priv.h