Re: [PATCH 2/2 v4] sched: Rewrite per entity runnable load average tracking

[Peter Zijlstra]

On Tue, Jul 29, 2014 at 11:12:37AM +0200, Vincent Guittot wrote:
> On 27 July 2014 19:36, Yuyang Du <yuyang.du@xxxxxxxxx> wrote:
> > Hi Vincent,
> >
> > On Fri, Jul 18, 2014 at 11:43:00AM +0200, Vincent Guittot wrote:
> >> > @@ -2291,23 +2299,24 @@ static __always_inline int __update_entity_runnable_avg(u64 now,
> >> >         delta >>= 10;
> >> >         if (!delta)
> >> >                 return 0;
> >> > -       sa->last_runnable_update = now;
> >> > +       sa->last_update_time = now;
> >> >
> >> >         /* delta_w is the amount already accumulated against our next period */
> >> > -       delta_w = sa->runnable_avg_period % 1024;
> >> > +       delta_w = sa->period_contrib;
> >> >         if (delta + delta_w >= 1024) {
> >> > -               /* period roll-over */
> >> >                 decayed = 1;
> >> >
> >> > +               /* how much left for next period will start over, we don't know yet */
> >> > +               sa->period_contrib = 0;
> >> > +
> >> >                 /*
> >> >                  * Now that we know we're crossing a period boundary, figure
> >> >                  * out how much from delta we need to complete the current
> >> >                  * period and accrue it.
> >> >                  */
> >> >                 delta_w = 1024 - delta_w;
> >> > -               if (runnable)
> >> > -                       sa->runnable_avg_sum += delta_w;
> >> > -               sa->runnable_avg_period += delta_w;
> >> > +               if (w)
> >> > +                       sa->load_sum += w * delta_w;
> >>
> >> Do you really need to have *w for computing the load_sum ? can't you
> >> only use it when computing the load_avg ?
> >>
> >> sa->load_avg = div_u64(sa->load_sum * w , LOAD_AVG_MAX)
> >>
> >
> > For task, assuming its load.weight does not change much, yes, we can. But in theory, task's
> 
> I would even say that the load_avg of a task should not be impacted by
> an old priority value. Once, the priority of a task is changed, we
> should only take into account this new priority to weight the load_avg
> of the task

So for tasks I would immediately agree, and I think for groups too,
seeing how the group weight is based off of this avg, if you then
include the old weight we'll get a feedback loop. This might not be
desired as it would counteract the SMP movement of tasks.
--
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/