[AMD Official Use Only - AMD Internal Distribution Only] Reviewed-by: Jesse Zhang <[email protected]>
> -----Original Message----- > From: amd-gfx <[email protected]> On Behalf Of Alex > Deucher > Sent: Wednesday, January 21, 2026 11:01 AM > To: [email protected] > Cc: Deucher, Alexander <[email protected]> > Subject: [PATCH 09/10] drm/amdgpu: Call drm_sched_increase_karma() for ring > resets > > Mark the entity as guilty. Align with adapter resets. > This only makes sense now that the CS IOCTL returns -ETIME rather than - > ECANCELED. Without that patch, this would break userspace. > > Signed-off-by: Alex Deucher <[email protected]> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c > b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c > index a323071762822..3c4b148d4f6ca 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c > @@ -137,6 +137,8 @@ static enum drm_gpu_sched_stat > amdgpu_job_timedout(struct drm_sched_job *s_job) > s_job->sched->name); > /* Stop the scheduler to prevent anybody else from touching the > ring > buffer. */ > drm_sched_wqueue_stop(&ring->sched); > + if (job->vm) > + drm_sched_increase_karma(&job->base); > r = amdgpu_ring_reset(ring, job->vmid, job->hw_fence); > if (!r) { > /* add the job back to the pending list */ > -- > 2.52.0
