From: Juri Lelli <juri.lelli@redhat.com>
To: luca abeni <luca.abeni@santannapisa.it>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>,
Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Vincent Guittot <vincent.guittot@linaro.org>,
Steven Rostedt <rostedt@goodmis.org>,
Daniel Bristot de Oliveira <bristot@redhat.com>,
Wei Wang <wvw@google.com>, Quentin Perret <qperret@google.com>,
Alessio Balsini <balsini@google.com>,
Pavan Kondeti <pkondeti@codeaurora.org>,
Patrick Bellasi <patrick.bellasi@matbug.net>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Valentin Schneider <valentin.schneider@arm.com>,
Qais Yousef <qais.yousef@arm.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/4] sched/deadline: Make DL capacity-aware
Date: Thu, 16 Apr 2020 15:19:38 +0200 [thread overview]
Message-ID: <20200416131938.GI9767@localhost.localdomain> (raw)
In-Reply-To: <20200415184203.50862783@sweethome>
On 15/04/20 18:42, luca abeni wrote:
> Hi Juri,
>
> On Wed, 15 Apr 2020 15:20:04 +0200
> Juri Lelli <juri.lelli@redhat.com> wrote:
> [...]
> > > > I'm thinking that, while dl_task_fits_capacity() works well when
> > > > selecting idle cpus, in this case we should consider the fact
> > > > that curr might be deadline as well and already consuming some of
> > > > the rq capacity.
> > > >
> > > > Do you think we should try to take that into account, maybe using
> > > > dl_rq->this_bw ?
> > >
> > > So you're saying that cpudl_find(..., later_mask) could return 1 (w/
> > > best_cpu (cp->elements[0].cpu) in later_mask).
> > >
> > > And that this best_cpu could be a non-fitting CPU for p.
> > >
> > > This could happen if cp->free_cpus is empty (no idle CPUs) so we
> > > take cpudl_find()'s else path and in case p's deadline <
> > > cp->elements[0] deadline.
> > >
> > > We could condition the 'return 1' on best_cpu fitting p.
> > >
> > > But should we do this for cpudl_find(..., NULL) calls from
> > > check_preempt_equal_dl() as well or will this break GEDF?
> >
> > So, even by not returning best_cpu, as above, if it doesn't fit p's bw
> > requirement, I think we would be breaking GEDF, which however doesn't
> > take asym capacities into account.
>
> Well, gEDF could take asymmetric capacities into account by scheduling
> the earliest deadline task on the fastest CPU (and the task with the
> second earliest deadline on the second fastest CPU, and so on...)
>
> But this could cause a lot of unneeded migrations (I tried to discuss
> this issue in a previous OSPM presentation). My original approach to
> work around this issue was to schedule a task on the slowest core on
> which the task can fit (some experiments revealed that this heuristic
> can approximate the gEDF behaviour without causing too many
> migrations)... But this patch is not included on the current patchset,
> and will be proposed later, after the most important patches have been
> merged.
OK, makes sense to me. And I'm ok also with a 2 steps approach. Asym
idle now and asym busy with a later series.
Best,
Juri
next prev parent reply other threads:[~2020-04-16 13:19 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-08 9:50 [PATCH 0/4] Capacity awareness for SCHED_DEADLINE Dietmar Eggemann
2020-04-08 9:50 ` [PATCH 1/4] sched/topology: Store root domain CPU capacity sum Dietmar Eggemann
2020-04-08 12:29 ` Vincent Guittot
2020-04-08 16:30 ` Dietmar Eggemann
2020-04-08 17:03 ` Vincent Guittot
2020-04-09 13:50 ` Dietmar Eggemann
2020-04-09 14:13 ` Vincent Guittot
2020-04-14 9:20 ` Dietmar Eggemann
2020-04-14 12:45 ` Quentin Perret
2020-04-14 15:27 ` Dietmar Eggemann
2020-04-14 15:43 ` Vincent Guittot
2020-04-08 9:50 ` [PATCH 2/4] sched/deadline: Improve admission control for asymmetric CPU capacities Dietmar Eggemann
2020-04-08 10:42 ` Valentin Schneider
2020-04-08 12:26 ` Dietmar Eggemann
2020-04-08 13:30 ` luca abeni
2020-04-08 14:23 ` Qais Yousef
2020-04-08 15:01 ` Valentin Schneider
2020-04-09 17:29 ` Dietmar Eggemann
2020-04-14 11:40 ` Qais Yousef
2020-04-14 14:29 ` Valentin Schneider
2020-04-14 15:41 ` Qais Yousef
2020-04-14 14:28 ` Valentin Schneider
2020-04-17 12:19 ` Juri Lelli
2020-04-17 14:55 ` Dietmar Eggemann
2020-04-17 15:08 ` Juri Lelli
2020-04-17 15:47 ` Juri Lelli
2020-04-08 9:50 ` [PATCH 3/4] sched/deadline: Make DL capacity-aware Dietmar Eggemann
2020-04-10 12:52 ` Juri Lelli
2020-04-15 9:39 ` Dietmar Eggemann
2020-04-15 13:20 ` Juri Lelli
2020-04-15 16:42 ` luca abeni
2020-04-16 13:19 ` Juri Lelli [this message]
2020-04-08 9:50 ` [PATCH 4/4] sched/deadline: Implement fallback mechanism for !fit case Dietmar Eggemann
2020-04-09 10:25 ` Qais Yousef
2020-04-09 13:00 ` luca abeni
2020-04-09 14:55 ` Qais Yousef
2020-04-09 18:43 ` Dietmar Eggemann
2020-04-14 11:29 ` Qais Yousef
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200416131938.GI9767@localhost.localdomain \
--to=juri.lelli@redhat.com \
--cc=balsini@google.com \
--cc=bristot@redhat.com \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luca.abeni@santannapisa.it \
--cc=mingo@redhat.com \
--cc=morten.rasmussen@arm.com \
--cc=patrick.bellasi@matbug.net \
--cc=peterz@infradead.org \
--cc=pkondeti@codeaurora.org \
--cc=qais.yousef@arm.com \
--cc=qperret@google.com \
--cc=rostedt@goodmis.org \
--cc=valentin.schneider@arm.com \
--cc=vincent.guittot@linaro.org \
--cc=wvw@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).