On Wed, 26 Nov 2025 14:57:32 -0800 Chia-I Wu <[email protected]> wrote:
> On Thu, Nov 13, 2025 at 7:32 AM Boris Brezillon > <[email protected]> wrote: > > > > If the group is already assigned a slot but was idle before this job > > submission, we need to make sure the priority rotation happens in the > > future. Extract the existing logic living in group_schedule_locked() > > and call this new sched_resume_tick() helper from the "group is > > assigned a slot" path. > > > > v2: > > - Add R-b > > > > Fixes: de8548813824 ("drm/panthor: Add the scheduler logical block") > > Signed-off-by: Boris Brezillon <[email protected]> > > Reviewed-by: Steven Price <[email protected]> > > --- > > drivers/gpu/drm/panthor/panthor_sched.c | 43 +++++++++++++++++++------ > > 1 file changed, 34 insertions(+), 9 deletions(-) > > > > diff --git a/drivers/gpu/drm/panthor/panthor_sched.c > > b/drivers/gpu/drm/panthor/panthor_sched.c > > index 0394f377c284..d1484f4a1c5b 100644 > > --- a/drivers/gpu/drm/panthor/panthor_sched.c > > +++ b/drivers/gpu/drm/panthor/panthor_sched.c > > @@ -2543,14 +2543,33 @@ static void sync_upd_work(struct work_struct *work) > > sched_queue_delayed_work(sched, tick, 0); > > } > > > > +static void sched_resume_tick(struct panthor_device *ptdev) > > +{ > > + struct panthor_scheduler *sched = ptdev->scheduler; > > + u64 delay_jiffies, now; > > + > > + drm_WARN_ON(&ptdev->base, sched->resched_target != U64_MAX); > > + > > + /* Scheduler tick was off, recalculate the resched_target based on > > the > > + * last tick event, and queue the scheduler work. > > + */ > > + now = get_jiffies_64(); > > + sched->resched_target = sched->last_tick + sched->tick_period; > > + if (sched->used_csg_slot_count == sched->csg_slot_count && > > + time_before64(now, sched->resched_target)) > > + delay_jiffies = min_t(unsigned long, sched->resched_target > > - now, ULONG_MAX); > > + else > > + delay_jiffies = 0; > > + > > + sched_queue_delayed_work(sched, tick, delay_jiffies); > > +} > > + > > static void group_schedule_locked(struct panthor_group *group, u32 > > queue_mask) > > { > > struct panthor_device *ptdev = group->ptdev; > > struct panthor_scheduler *sched = ptdev->scheduler; > > struct list_head *queue = &sched->groups.runnable[group->priority]; > > - u64 delay_jiffies = 0; > > bool was_idle; > > - u64 now; > > > > if (!group_can_run(group)) > > return; > > @@ -2595,13 +2614,7 @@ static void group_schedule_locked(struct > > panthor_group *group, u32 queue_mask) > > /* Scheduler tick was off, recalculate the resched_target based on > > the > > * last tick event, and queue the scheduler work. > > */ > > - now = get_jiffies_64(); > > - sched->resched_target = sched->last_tick + sched->tick_period; > > - if (sched->used_csg_slot_count == sched->csg_slot_count && > > - time_before64(now, sched->resched_target)) > > - delay_jiffies = min_t(unsigned long, sched->resched_target > > - now, ULONG_MAX); > > - > > - sched_queue_delayed_work(sched, tick, delay_jiffies); > > + sched_resume_tick(ptdev); > > } > > > > static void queue_stop(struct panthor_queue *queue, > > @@ -3202,6 +3215,18 @@ queue_run_job(struct drm_sched_job *sched_job) > > > > group_schedule_locked(group, BIT(job->queue_idx)); > > } else { > > + u32 queue_mask = BIT(job->queue_idx); > > + bool resume_tick = group_is_idle(group) && > > + (group->idle_queues & queue_mask) && > > + !(group->blocked_queues & queue_mask) && > > + sched->resched_target == U64_MAX; > The logic here should be the same as the first part of > group_schedule_locked. I wonder if we can refactor that as well. Could be, yes. I'll give it a try. > > > + > > + /* We just added something to the queue, so it's no longer > > idle. */ > > + group->idle_queues &= ~BIT(job->queue_idx); > group->idle_queues &= queue_mask; You mean group->idle_queues &= ~queue_mask; I guess. > > > + > > + if (resume_tick) > > + sched_resume_tick(ptdev); > > + > > gpu_write(ptdev, CSF_DOORBELL(queue->doorbell_id), 1); > > if (!sched->pm.has_ref && > > !(group->blocked_queues & BIT(job->queue_idx))) { > > -- > > 2.51.1 > >
