Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1525444
| From | luca abeni <luca.abeni@unitn.it> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [RFC v3 1/6] Track the active utilisation |
| Date | 2016-11-18 16:10 +0100 |
| Message-ID | <sESXD-l4-13@gated-at.bofh.it> (permalink) |
| References | <svO6S-ak-7@gated-at.bofh.it> <svO6S-ak-23@gated-at.bofh.it> <sERRU-7R7-9@gated-at.bofh.it> |
| Organization | university of trento |
Hi Peter,
On Fri, 18 Nov 2016 14:55:54 +0100
Peter Zijlstra <peterz@infradead.org> wrote:
> On Mon, Oct 24, 2016 at 04:06:33PM +0200, Luca Abeni wrote:
>
> > @@ -498,6 +514,8 @@ static void update_dl_entity(struct
> > sched_dl_entity *dl_se, struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
> > struct rq *rq = rq_of_dl_rq(dl_rq);
> >
> > + add_running_bw(dl_se, dl_rq);
> > +
> > if (dl_time_before(dl_se->deadline, rq_clock(rq)) ||
> > dl_entity_overflow(dl_se, pi_se, rq_clock(rq))) {
> > dl_se->deadline = rq_clock(rq) +
> > pi_se->dl_deadline; @@ -947,14 +965,19 @@ static void
> > enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
> > return; }
> >
> > + if (p->on_rq == TASK_ON_RQ_MIGRATING)
> > + add_running_bw(&p->dl, &rq->dl);
> > +
> > /*
> > * If p is throttled, we do nothing. In fact, if it
> > exhausted
> > * its budget it needs a replenishment and, since it now
> > is on
> > * its rq, the bandwidth timer callback (which clearly has
> > not
> > * run yet) will take care of this.
> > */
> > - if (p->dl.dl_throttled && !(flags & ENQUEUE_REPLENISH))
> > + if (p->dl.dl_throttled && !(flags & ENQUEUE_REPLENISH)) {
> > + add_running_bw(&p->dl, &rq->dl);
> > return;
> > + }
> >
> > enqueue_dl_entity(&p->dl, pi_se, flags);
> >
>
> I realize the enqueue path is a bit of a maze, but this hurts my head.
>
> Isn't there anything we can do to streamline this a bit?
>
> Maybe move the add_running_bw() from update_dl_entity() to the
> ENQUEUE_WAKEUP branch in enqueue_dl_entity()? Because that's what you
> really want, isn't it? Its not actually related to recomputing the
> absolute deadline.
Right... I'll change the code in this way. I am currently modifying
this code (because after Juri's review I realized that there is an
issue - see last email exchange with Juri); I'll make this change as
soon as the code stabilizes.
Thanks,
Luca
Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread
Re: [RFC v3 1/6] Track the active utilisation Peter Zijlstra <peterz@infradead.org> - 2016-11-18 15:00 +0100 Re: [RFC v3 1/6] Track the active utilisation luca abeni <luca.abeni@unitn.it> - 2016-11-18 16:10 +0100
csiph-web