From: Morten Rasmussen <morten.rasmussen@arm.com>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
"mingo@redhat.com" <mingo@redhat.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Yuyang Du <yuyang.du@intel.com>,
mgalbraith@suse.de, linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 06/16] sched: Disable WAKE_AFFINE for asymmetric configurations
Date: Wed, 25 May 2016 10:12:01 +0100 [thread overview]
Message-ID: <20160525091200.GJ27946@e105550-lin.cambridge.arm.com> (raw)
In-Reply-To: <CAKfTPtB7bjy7skRQ1JyWdCZ1ZAM1pgsF7f1c_GLOffte=2p4zQ@mail.gmail.com>
On Tue, May 24, 2016 at 05:53:27PM +0200, Vincent Guittot wrote:
> On 24 May 2016 at 17:02, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
> > On Tue, May 24, 2016 at 03:52:00PM +0200, Vincent Guittot wrote:
> >> On 24 May 2016 at 15:36, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
> >> > On Tue, May 24, 2016 at 03:27:05PM +0200, Vincent Guittot wrote:
> >> >> On 24 May 2016 at 15:16, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
> >> >> > On Tue, May 24, 2016 at 02:12:38PM +0200, Vincent Guittot wrote:
> >> >> >> On 24 May 2016 at 12:29, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
> >> >> >> > On Tue, May 24, 2016 at 11:10:28AM +0200, Vincent Guittot wrote:
> >> >> >> >> On 23 May 2016 at 12:58, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
> >> >> >> >> > If the system has cpu of different compute capacities (e.g. big.LITTLE)
> >> >> >> >> > let affine wakeups be constrained to cpus of the same type.
> >> >> >> >>
> >> >> >> >> Can you explain why you don't want wake affine with cpus with
> >> >> >> >> different compute capacity ?
> >> >> >> >
> >> >> >> > I should have made the overall idea a bit more clear. The idea is to
> >> >> >> > deal with cross-capacity migrations in the find_idlest_{group, cpu}{}
> >> >> >> > path so we don't have to touch select_idle_sibling().
> >> >> >> > select_idle_sibling() is critical for wake-up latency, and I'm assumed
> >> >> >> > that people wouldn't like adding extra overhead in there to deal with
> >> >> >> > capacity and utilization.
> >> >> >>
> >> >> >> So this means that we will never use the quick path of
> >> >> >> select_idle_sibling for cross capacity migration but always the one
> >> >> >> with extra overhead?
> >> >> >
> >> >> > Yes. select_idle_sibling() is only used to choose among equal capacity
> >> >> > cpus (capacity_orig).
> >> >> >
> >> >> >> Patch 9 adds more tests for enabling wake_affine path. Can't it also
> >> >> >> be used for cross capacity migration ? so we can use wake_affine if
> >> >> >> the task or the cpus (even with different capacity) doesn't need this
> >> >> >> extra overhead
> >> >> >
> >> >> > The test in patch 9 is to determine whether we are happy with the
> >> >> > capacity of the previous cpu, or we should go look for one with more
> >> >> > capacity. I don't see how we can use select_idle_sibling() unmodified
> >> >> > for sched domains containing cpus of different capacity to select an
> >> >> > appropriate cpu. It is just picking an idle cpu, it might have high
> >> >> > capacity or low, it wouldn't care.
> >> >> >
> >> >> > How would you avoid the overhead of checking capacity and utilization of
> >> >> > the cpus and still pick an appropriate cpu?
> >> >>
> >> >> My point is that there is some wake up case where we don't care about
> >> >> the capacity and utilization of cpus even for cross capacity migration
> >> >> and we will never take benefit of this fast path.
> >> >> You have added an extra check for setting want_affine in patch 9 which
> >> >> uses capacity and utilization of cpu to disable this fast path when a
> >> >> task needs more capacity than available. Can't you use this function
> >> >> to disable the want_affine for cross-capacity migration situation that
> >> >> cares of the capacity and need the full scan of sched_domain but keep
> >> >> it enable for other cases ?
> >> >
> >> > It is not clear to me what the other cases are. What kind of cases do
> >> > you have in mind?
> >>
> >> As an example, you have a task A that have to be on a big CPU because
> >> of the requirement of compute capacity, that wakes up a task B that
> >> can run on any cpu according to its utilization. The fast wake up path
> >> is fine for task B whatever prev cpu is.
> >
> > In that case, we will take always take fast path (select_idle_sibling())
> > for task B if wake_wide() allows it, which should be fine.
>
> Even if want_affine is set, the wake up of task B will not use the fast path.
> The affine_sd will not be set because the sched_domain, which have
> both cpus, will not have the SD_WAKE_AFFINE flag according to this
> patch, isn't it ?
> So task B can't use the fast path whereas nothing prevent him to take
> benefit of it
>
> Am I missing something ?
No, I think you are right. Very good point. The cpumask test with
sched_domain_span() will of cause return false. So yes, in this case the
slow path is taken. It isn't wrong as such, just slower for asymmetric
capacity systems :-)
It is clearly not as optimized for asymmetric capacity systems as it
could be, but my focus was to not ruin existing behaviour and minimize
overhead for others. There are a lot of different routes through those
conditions in the first half of select_task_rq_fair() that aren't
obvious. I worry that some users depend on them and that I don't
see/understand all of them.
If people agree on changing things, it is fine with me. I just tried to
avoid getting the patches shot down on that account ;-)
next prev parent reply other threads:[~2016-05-25 9:11 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-23 10:58 [PATCH 00/16] sched: Clean-ups and asymmetric cpu capacity support Morten Rasmussen
2016-05-23 10:58 ` [PATCH 01/16] sched: Fix power to capacity renaming in comment Morten Rasmussen
2016-05-23 10:58 ` [PATCH 02/16] sched/fair: Consistent use of prev_cpu in wakeup path Morten Rasmussen
2016-06-01 19:49 ` Peter Zijlstra
2016-05-23 10:58 ` [PATCH 03/16] sched/fair: Disregard idle task wakee_flips in wake_wide Morten Rasmussen
2016-05-23 11:12 ` Mike Galbraith
2016-05-23 12:00 ` Morten Rasmussen
2016-05-23 13:00 ` Mike Galbraith
2016-05-23 14:10 ` Morten Rasmussen
2016-05-23 15:42 ` Mike Galbraith
2016-05-23 23:17 ` Yuyang Du
2016-05-23 23:04 ` Yuyang Du
2016-06-01 19:57 ` Peter Zijlstra
2016-06-02 8:05 ` Peter Zijlstra
2016-06-07 12:08 ` Morten Rasmussen
2016-05-23 10:58 ` [PATCH 04/16] sched/fair: Optimize find_idlest_cpu() when there is no choice Morten Rasmussen
2016-05-24 6:29 ` Mike Galbraith
2016-05-24 8:05 ` Morten Rasmussen
2016-05-24 8:12 ` Mike Galbraith
2016-06-01 19:59 ` Peter Zijlstra
2016-06-07 14:25 ` Morten Rasmussen
2016-05-23 10:58 ` [PATCH 05/16] sched: Introduce SD_ASYM_CPUCAPACITY sched_domain topology flag Morten Rasmussen
2016-05-23 10:58 ` [PATCH 06/16] sched: Disable WAKE_AFFINE for asymmetric configurations Morten Rasmussen
2016-05-24 9:10 ` Vincent Guittot
2016-05-24 10:29 ` Morten Rasmussen
2016-05-24 12:12 ` Vincent Guittot
2016-05-24 13:16 ` Morten Rasmussen
2016-05-24 13:27 ` Vincent Guittot
2016-05-24 13:36 ` Morten Rasmussen
2016-05-24 13:52 ` Vincent Guittot
2016-05-24 15:02 ` Morten Rasmussen
2016-05-24 15:53 ` Vincent Guittot
2016-05-25 9:12 ` Morten Rasmussen [this message]
2016-05-26 6:45 ` Vincent Guittot
2016-06-07 16:50 ` Morten Rasmussen
2016-05-23 10:58 ` [PATCH 07/16] sched: Make SD_BALANCE_WAKE a topology flag Morten Rasmussen
2016-05-24 23:52 ` Yuyang Du
2016-05-25 9:27 ` Morten Rasmussen
2016-06-01 20:18 ` Peter Zijlstra
2016-06-08 8:45 ` Morten Rasmussen
2016-05-23 10:58 ` [PATCH 08/16] sched: Store maximum per-cpu capacity in root domain Morten Rasmussen
2016-05-23 10:58 ` [PATCH 09/16] sched/fair: Let asymmetric cpu configurations balance at wake-up Morten Rasmussen
2016-05-24 0:04 ` Yuyang Du
2016-05-24 8:10 ` Morten Rasmussen
2016-05-24 7:03 ` Mike Galbraith
2016-05-24 7:15 ` Mike Galbraith
2016-05-25 6:57 ` Wanpeng Li
2016-05-25 9:49 ` Morten Rasmussen
2016-05-25 10:29 ` Wanpeng Li
2016-05-25 10:54 ` Morten Rasmussen
2016-05-25 11:18 ` Wanpeng Li
2016-06-02 14:21 ` Peter Zijlstra
2016-06-08 11:29 ` Morten Rasmussen
2016-06-08 14:36 ` Peter Zijlstra
2016-05-23 10:58 ` [PATCH 10/16] sched/fair: Compute task/cpu utilization at wake-up more correctly Morten Rasmussen
2016-05-23 10:58 ` [PATCH 11/16] sched/fair: Consider spare capacity in find_idlest_group() Morten Rasmussen
2016-05-23 10:58 ` [PATCH 12/16] sched: Add per-cpu max capacity to sched_group_capacity Morten Rasmussen
2016-05-23 10:58 ` [PATCH 13/16] sched/fair: Avoid pulling tasks from non-overloaded higher capacity groups Morten Rasmussen
2016-05-23 10:58 ` [PATCH 14/16] arm: Set SD_ASYM_CPUCAPACITY for big.LITTLE platforms Morten Rasmussen
2016-05-23 10:58 ` [PATCH 15/16] arm: Set SD_BALANCE_WAKE flag for asymmetric capacity systems Morten Rasmussen
2016-05-23 10:58 ` [PATCH 16/16] arm: Update arch_scale_cpu_capacity() to reflect change to define Morten Rasmussen
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=20160525091200.GJ27946@e105550-lin.cambridge.arm.com \
--to=morten.rasmussen@arm.com \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgalbraith@suse.de \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=vincent.guittot@linaro.org \
--cc=yuyang.du@intel.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).