On 11/7/25 10:04, Pierre-Eric Pelloux-Prayer wrote: > Currently, the scheduler score is incremented when a job is pushed to an > entity and when an entity is attached to the scheduler. > > This leads to some bad scheduling decision where the score value is > largely made of idle entities. > > For instance, a scenario with 2 schedulers and where 10 entities submit > a single job, then do nothing, each scheduler will probably end up with > a score of 5. > Now, let's imagine 5 userspace apps exit, so their entities will be dropped. > In > the worst case, these apps' entities were all attached to the same > scheduler and we end up with score=5 (the 5 remaining entities) and > score=0, despite the 2 schedulers being idle. > When new entities show up, they will all select the second scheduler > based on its low score value, instead of alternating between the 2. > > Some amdgpu rings depended on this feature, but the previous commit > implemented the same thing in amdgpu directly so it can be safely > removed from drm/sched. > > Signed-off-by: Pierre-Eric Pelloux-Prayer <[email protected]> > Reviewed-by: Tvrtko Ursulin <[email protected]> > Acked-by: Tomeu Vizoso <[email protected]> > Acked-by: Philipp Stanner <[email protected]>
Reviewed-by: Christian König <[email protected]> > --- > drivers/gpu/drm/scheduler/sched_main.c | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/drivers/gpu/drm/scheduler/sched_main.c > b/drivers/gpu/drm/scheduler/sched_main.c > index c39f0245e3a9..8a3d99a86090 100644 > --- a/drivers/gpu/drm/scheduler/sched_main.c > +++ b/drivers/gpu/drm/scheduler/sched_main.c > @@ -206,7 +206,6 @@ void drm_sched_rq_add_entity(struct drm_sched_rq *rq, > if (!list_empty(&entity->list)) > return; > > - atomic_inc(rq->sched->score); > list_add_tail(&entity->list, &rq->entities); > } > > @@ -228,7 +227,6 @@ void drm_sched_rq_remove_entity(struct drm_sched_rq *rq, > > spin_lock(&rq->lock); > > - atomic_dec(rq->sched->score); > list_del_init(&entity->list); > > if (rq->current_entity == entity)
