Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit a053fb7e authored by Grazvydas Ignotas's avatar Grazvydas Ignotas Committed by Alex Deucher
Browse files

drm/amdgpu: fix sched fence slab teardown 



To free fences, call_rcu() is used, which calls amd_sched_fence_free()
after a grace period. During teardown, there is no guarantee all
callbacks have finished, so sched_fence_slab may be destroyed before
all fences have been freed. If we are lucky, this results in some slab
warnings, if not, we get a crash in one of rcu threads because callback
is called after amdgpu has already been unloaded.

Fix it with a rcu_barrier().

Fixes: 189e0fb7 ("drm/amdgpu: RCU protected amd_sched_fence_release")
Acked-by: default avatarChunming Zhou <david1.zhou@amd.com>
Reviewed-by: default avatarChristian König <christian.koenig@amd.com>
Signed-off-by: default avatarGrazvydas Ignotas <notasas@gmail.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent 9dc79965

drivers/gpu/drm/amd/scheduler/gpu_scheduler.c

+1 −0
Original line number Diff line number Diff line
@@ -645,6 +645,7 @@ void amd_sched_fini(struct amd_gpu_scheduler *sched) 
{

	if (sched->thread)

		kthread_stop(sched->thread);

	rcu_barrier();

	if (atomic_dec_and_test(&sched_fence_slab_ref))

		kmem_cache_destroy(sched_fence_slab);

}