On 09/07/2016 12:53 PM, Kirill Tkhai wrote: > sched_cfs_active_timer() is used to postpone decrementing of > tg->nr_cpus_active and marking cfs_rq as inactive. It's need > to give a possibility for a new tg task to stand on a hot cpu, > where another tg's task gone to sleep a moment ago. > > See for dec_nr_active_cfs_rqs(), which postpone actions it does. > > Signed-off-by: Kirill Tkhai <ktk...@virtuozzo.com>
Reviewed-by: Andrey Ryabinin <aryabi...@virtuozzo.com> > --- > kernel/sched/fair.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > index 515685f..34eb09d 100644 > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -510,7 +510,7 @@ static enum hrtimer_restart sched_cfs_active_timer(struct > hrtimer *timer) > unsigned long flags; > > raw_spin_lock_irqsave(&rq->lock, flags); > - cfs_rq->active = !!list_empty(&cfs_rq->tasks); > + cfs_rq->active = !list_empty(&cfs_rq->tasks); > raw_spin_unlock_irqrestore(&rq->lock, flags); > > atomic_dec(&cfs_rq->tg->nr_cpus_active); > > _______________________________________________ > Devel mailing list > Devel@openvz.org > https://lists.openvz.org/mailman/listinfo/devel > _______________________________________________ Devel mailing list Devel@openvz.org https://lists.openvz.org/mailman/listinfo/devel