Re: [PATCH] sched: let __sched_period() use rq cfs's nr_running

[Byungchul Park]

On Tue, Jul 07, 2015 at 01:28:18PM -0700, Cong Wang wrote:
> On Mon, Jul 6, 2015 at 9:40 PM,  <byungchul.park@xxxxxxx> wrote:
> > From: Byungchul Park <byungchul.park@xxxxxxx>
> >
> > __sched_period() returns a period which a rq can have. the period has to be
> > stretched by the number of task *the rq has*, when nr_running > nr_latency.
> > otherwise, task slice can be very smaller than sysctl_sched_min_granularity
> > depending on the position of tg hierarchy when CONFIG_FAIR_GROUP_SCHED.
> 
> 
> Hmm, rq_of(cfs_rq)->cfs should point to the root cfs_rq whose
> ->nr_running is number of running entities at root level? This
> doesn't look right.

hello, cong wang,

__sched_period() returns a root level period now. stretching the period with
a local task group's nr_running is meaningless for getting this period.
is there something i missed? :)

thank you,
byungchul

> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/