On 29/09/2022 19:01, Christian König wrote:
> Otherwise we would crash if the job is not resubmitted.
>
> Signed-off-by: Christian König <christian.koenig@xxxxxxx>
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> index 4f2395d1a791..23e5e8275dc7 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -829,7 +829,8 @@ drm_sched_get_cleanup_job(struct drm_gpu_scheduler *sched)
> job = list_first_entry_or_null(&sched->pending_list,
> struct drm_sched_job, list);
>
> - if (job && dma_fence_is_signaled(job->s_fence->parent)) {
> + if (job && (!job->s_fence->parent ||
> + dma_fence_is_signaled(job->s_fence->parent))) {
> /* remove job from pending_list */
> list_del_init(&job->list);
>
Sadly there's another use further down, so this isn't a complete fix:
> /* cancel this job's TO timer */
> cancel_delayed_work(&sched->work_tdr);
> /* make the scheduled timestamp more accurate */
> next = list_first_entry_or_null(&sched->pending_list,
> typeof(*next), list);
>
> if (next) {
> next->s_fence->scheduled.timestamp =
> job->s_fence->parent->timestamp;
>From the comment I presume that just skipping this section (when parent
is NULL) would be fine.
Steve
