Re: [RFC PATCH v15 6/7] sched: Fix proxy/current (push,pull)ability

[K Prateek Nayak]

Hello John, Peter,

On 3/17/2025 7:37 PM, Peter Zijlstra wrote:
> On Wed, Mar 12, 2025 at 03:11:36PM -0700, John Stultz wrote:
> > diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> > index b4f7b14f62a24..3596244f613f8 100644
> > --- a/kernel/sched/core.c
> > +++ b/kernel/sched/core.c
> > @@ -6722,6 +6722,23 @@ find_proxy_task(struct rq *rq, struct task_struct *donor, struct rq_flags *rf)
> >   }
> >   #endif /* SCHED_PROXY_EXEC */
> >   
> > +static inline void proxy_tag_curr(struct rq *rq, struct task_struct *owner)
> > +{
> > +	if (!sched_proxy_exec())
> > +		return;
> > +	/*
> > +	 * pick_next_task() calls set_next_task() on the chosen task
> > +	 * at some point, which ensures it is not push/pullable.
> > +	 * However, the chosen/donor task *and* the mutex owner form an
> > +	 * atomic pair wrt push/pull.
> > +	 *
> > +	 * Make sure owner we run is not pushable. Unfortunately we can
> > +	 * only deal with that by means of a dequeue/enqueue cycle. :-/
> > +	 */
> > +	dequeue_task(rq, owner, DEQUEUE_NOCLOCK | DEQUEUE_SAVE);
> > +	enqueue_task(rq, owner, ENQUEUE_NOCLOCK | ENQUEUE_RESTORE);
> > +}
>
> So this is probably fine at this point; but we should eventually look at
> fixing this.
>
> We can probably look at (ab)using sched_class::set_cpus_allowed() for
> this.

Isn't the net result in this case just "dequeue_pushable_task()"?

Can it be defined as a class callback to avoid a full dequeue + requeue
overhead for the fair class during proxy?

--
Thanks and Regards,
Prateek