drm/amdgpu: Retain job->vm in amdgpu_job_prepare_job
authorYuanShang <YuanShang.Mao@amd.com>
Wed, 23 Jul 2025 08:44:49 +0000 (16:44 +0800)
committerAlex Deucher <alexander.deucher@amd.com>
Mon, 4 Aug 2025 19:41:09 +0000 (15:41 -0400)
commitc00d8b79fd2167c6ac65e096619535acdf8678d5
tree18c9e3fdce375db7f1c63da743ce5d5ddc8f42d4
parent1c8dc3e088e09531bcdfc9fe348204abc3decb6c
drm/amdgpu: Retain job->vm in amdgpu_job_prepare_job

The field job->vm is used in function amdgpu_job_run to get the page
table re-generation counter and decide whether the job should be skipped.

Specifically, function amdgpu_vm_generation checks if the VM is valid for this job to use.
For instance, if a gfx job depends on a cancelled sdma job from entity vm->delayed,
then the gfx job should be skipped.

Fixes: 26c95e838e63 ("drm/amdgpu: set the VM pointer to NULL in amdgpu_job_prepare")
Signed-off-by: YuanShang <YuanShang.Mao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ed76936c6b10b547c6df4ca75412331e9ef6d339)
Cc: stable@vger.kernel.org
drivers/gpu/drm/amd/amdgpu/amdgpu_job.c