if app close CTX right after IB submit, gpu recover will failed to find out the entity/ctx behind the guilty job thus lead to bad job skipping in scheduler failed
to fix this corner case just move the job->karma increasing out of the condition that the backing entity was found that way the job itself will be "guilty" anyway Change-Id: Ia30f02df9297a343d6d8dace496e237827dd1548 Signed-off-by: Monk Liu <[email protected]> --- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 7aa6455..720fd1b 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -464,6 +464,7 @@ void amd_sched_hw_job_reset(struct amd_gpu_scheduler *sched, struct amd_sched_jo spin_unlock(&sched->job_list_lock); if (bad) { + atomic_inc(&bad->karma); /* don't increase @bad's karma if it's from KERNEL RQ, * becuase sometimes GPU hang would cause kernel jobs (like VM updating jobs) * corrupt but keep in mind that kernel jobs always considered good. @@ -474,7 +475,7 @@ void amd_sched_hw_job_reset(struct amd_gpu_scheduler *sched, struct amd_sched_jo spin_lock(&rq->lock); list_for_each_entry_safe(entity, tmp, &rq->entities, list) { if (bad->s_fence->scheduled.context == entity->fence_context) { - if (atomic_inc_return(&bad->karma) > bad->sched->hang_limit) + if (atomic_read(&bad->karma) > bad->sched->hang_limit) if (entity->guilty) atomic_set(entity->guilty, 1); break; -- 2.7.4 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
