Re: [PATCH 4/7] sched/fair: Take care of group/affinity/sched_class change for throttled task

[Aaron Lu <ziqianlu@bytedance.com>]

On Thu, May 22, 2025 at 02:03:36PM +0200, Peter Zijlstra wrote:
> On Tue, May 20, 2025 at 06:41:07PM +0800, Aaron Lu wrote:
> > On task group change, for tasks whose on_rq equals to TASK_ON_RQ_QUEUED,
> > core will dequeue it and then requeued it.
> > 
> > The throttled task is still considered as queued by core because p->on_rq
> > is still set so core will dequeue it, but since the task is already
> > dequeued on throttle in fair, handle this case properly.
> > 
> > Affinity and sched class change is similar.
> > 
> > Signed-off-by: Aaron Lu <ziqianlu@bytedance.com>
> > ---
> >  kernel/sched/fair.c | 24 ++++++++++++++++++++++++
> >  1 file changed, 24 insertions(+)
> > 
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 74bc320cbc238..4c66fd8d24389 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -5866,6 +5866,10 @@ static void throttle_cfs_rq_work(struct callback_head *work)
> >  		update_rq_clock(rq);
> >  		WARN_ON_ONCE(!list_empty(&p->throttle_node));
> >  		dequeue_task_fair(rq, p, DEQUEUE_SLEEP | DEQUEUE_SPECIAL);
> > +		/*
> > +		 * Must not add it to limbo list before dequeue or dequeue will
> > +		 * mistakenly regard this task as an already throttled one.
> > +		 */
> >  		list_add(&p->throttle_node, &cfs_rq->throttled_limbo_list);
> >  		resched_curr(rq);
> >  	}
> > @@ -5881,6 +5885,20 @@ void init_cfs_throttle_work(struct task_struct *p)
> >  	INIT_LIST_HEAD(&p->throttle_node);
> >  }
> >  
> > +static void dequeue_throttled_task(struct task_struct *p, int flags)
> > +{
> > +	/*
> > +	 * Task is throttled and someone wants to dequeue it again:
> > +	 * it must be sched/core when core needs to do things like
> > +	 * task affinity change, task group change, task sched class
> > +	 * change etc.
> > +	 */
> > +	WARN_ON_ONCE(p->se.on_rq);
> > +	WARN_ON_ONCE(flags & DEQUEUE_SLEEP);
> > +
> > +	list_del_init(&p->throttle_node);
> > +}
> > +
> >  static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags);
> >  static int tg_unthrottle_up(struct task_group *tg, void *data)
> >  {
> > @@ -6834,6 +6852,7 @@ static inline void sync_throttle(struct task_group *tg, int cpu) {}
> >  static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) {}
> >  static void task_throttle_setup_work(struct task_struct *p) {}
> >  static bool task_is_throttled(struct task_struct *p) { return false; }
> > +static void dequeue_throttled_task(struct task_struct *p, int flags) {}
> >  
> >  static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq)
> >  {
> > @@ -7281,6 +7300,11 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags)
> >   */
> >  static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
> >  {
> > +	if (unlikely(task_is_throttled(p))) {
> > +		dequeue_throttled_task(p, flags);
> > +		return true;
> > +	}
> > +
> >  	if (!(p->se.sched_delayed && (task_on_rq_migrating(p) || (flags & DEQUEUE_SAVE))))
> >  		util_est_dequeue(&rq->cfs, p);
> 
> This is asymmetric -- dequeue removes it from that throttle list, but
> the corresponding enqueue will not add it back, what gives?
> 
> Because now we have:
> 
>  p->on_rq=1
>  p->throttle_node on list
> 
> move_queued_task()
>   deactivate_task()
>     dequeue_task_fair()
>       list_del_init(throttle_node)
>     p->on_rq = 2
> 
>   activate_task()
>     enqueue_task_fair()
>       // nothing special, makes the thing runnable
>     p->on_rq = 1;
> 
> and we exit with a task that is on-rq and not throttled ?!?
>
> Why is this? Are we relying on pick_task_fair() to dequeue it again and
> fix up our inconsistencies? If so, that had better have a comment on.

Correct.

Does the following comment look OK?

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 89afa472299b7..4f4d64cf31fb1 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -7147,6 +7147,10 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags)
 static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 {
 	if (unlikely(task_is_throttled(p))) {
+		/*
+		 * Task migrated to new rq will have its throttle work
+		 * added if necessary in pick time.
+		 */
 		dequeue_throttled_task(p, flags);
 		return true;
 	}