From: Vincent Guittot <vincent.guittot@linaro.org>
To: Morten Rasmussen <morten.rasmussen@arm.com>
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 v2 07/13] sched/fair: Let asymmetric cpu configurations balance at wake-up
Date: Thu, 14 Jul 2016 15:45:17 +0200 [thread overview]
Message-ID: <CAKfTPtDE1NsA_q4zGrwBXBBHfdzjFwzPohO+iLqeooZ0S8-jQQ@mail.gmail.com> (raw)
In-Reply-To: <20160713161441.GB21816@e105550-lin.cambridge.arm.com>
On 13 July 2016 at 18:14, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
> On Wed, Jul 13, 2016 at 02:56:41PM +0200, Vincent Guittot wrote:
>> On 22 June 2016 at 19:03, Morten Rasmussen <morten.rasmussen@arm.com> wrote:
>> > Currently, SD_WAKE_AFFINE always takes priority over wakeup balancing if
>> > SD_BALANCE_WAKE is set on the sched_domains. For asymmetric
>> > configurations SD_WAKE_AFFINE is only desirable if the waking task's
>> > compute demand (utilization) is suitable for all the cpu capacities
>> > available within the SD_WAKE_AFFINE sched_domain. If not, let wakeup
>>
>> instead of "suitable for all the cpu capacities available within the
>> SD_WAKE_AFFINE sched_domain", should it be "suitable for local cpu and
>> prev cpu" becasue you only check the capacity of these 2 CPUs.
>
> Good point. I currently make the implicit assumption that capacity of local cpu
> and prev cpu represent the capacity for all cpus their SD_WAKE_AFFINE
> domains. It breaks if you should choose to have SD_WAKE_AFFINE on a
> domain that spans both little and big cpus, as if local/prev cpu happens
> to be big we assume that they are all big and let select_idle_sibling()
> handle the task placement even for big tasks if local/prev cpu are both
> big.
Isn't the sd_llc used in select_idle_sibling and not the
SD_WAKE_AFFINE domian so if CPUs in the sd_llc are homogeneous, we are
safe
>
> I don't see why anybody would want that kind of setup, but I think the
> assumption should still be written down somewhere, either here or in a
> comment in wake_cap() or both.
>
> The next paragraph in the commit message mentions that we actually only
> check waker cpu and prev_cpu capacity. Would it be more clear if we
> extend that to something like:
>
> This patch makes affine wake-ups conditional on whether both the waker
> cpu and prev_cpu has sufficient capacity for the waking task, or
> not, assuming that the cpu capacities within an SD_WAKE_AFFINE
> domain are homogeneous.
>
> Thoughts?
>
>>
>> Other than this comment for the commit message, the patch looks good to me
>> Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
>
> Thanks,
> Morten
>
>
>>
>> > balancing take over (find_idlest_{group, cpu}()).
>> >
>> > This patch makes affine wake-ups conditional on whether both the waker
>> > cpu and prev_cpu has sufficient capacity for the waking task, or not.
>> >
>> > It is assumed that the sched_group(s) containing the waker cpu and
>> > prev_cpu only contain cpu with the same capacity (homogeneous).
>> >
>> > Ideally, we shouldn't set 'want_affine' in the first place, but we don't
>> > know if SD_BALANCE_WAKE is enabled on the sched_domain(s) until we start
>> > traversing them.
>> >
>> > cc: Ingo Molnar <mingo@redhat.com>
>> > cc: Peter Zijlstra <peterz@infradead.org>
>> >
>> > Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
>> > ---
>> > kernel/sched/fair.c | 28 +++++++++++++++++++++++++++-
>> > 1 file changed, 27 insertions(+), 1 deletion(-)
>> >
>> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> > index 216db302e87d..dba02c7b57b3 100644
>> > --- a/kernel/sched/fair.c
>> > +++ b/kernel/sched/fair.c
>> > @@ -114,6 +114,12 @@ unsigned int __read_mostly sysctl_sched_shares_window = 10000000UL;
>> > unsigned int sysctl_sched_cfs_bandwidth_slice = 5000UL;
>> > #endif
>> >
>> > +/*
>> > + * The margin used when comparing utilization with cpu capacity:
>> > + * util * 1024 < capacity * margin
>> > + */
>> > +unsigned int capacity_margin = 1280; /* ~20% */
>> > +
>> > static inline void update_load_add(struct load_weight *lw, unsigned long inc)
>> > {
>> > lw->weight += inc;
>> > @@ -5260,6 +5266,25 @@ static int cpu_util(int cpu)
>> > return (util >= capacity) ? capacity : util;
>> > }
>> >
>> > +static inline int task_util(struct task_struct *p)
>> > +{
>> > + return p->se.avg.util_avg;
>> > +}
>> > +
>> > +static int wake_cap(struct task_struct *p, int cpu, int prev_cpu)
>> > +{
>> > + long min_cap, max_cap;
>> > +
>> > + min_cap = min(capacity_orig_of(prev_cpu), capacity_orig_of(cpu));
>> > + max_cap = cpu_rq(cpu)->rd->max_cpu_capacity;
>> > +
>> > + /* Minimum capacity is close to max, no need to abort wake_affine */
>> > + if (max_cap - min_cap < max_cap >> 3)
>> > + return 0;
>> > +
>> > + return min_cap * 1024 < task_util(p) * capacity_margin;
>> > +}
>> > +
>> > /*
>> > * select_task_rq_fair: Select target runqueue for the waking task in domains
>> > * that have the 'sd_flag' flag set. In practice, this is SD_BALANCE_WAKE,
>> > @@ -5283,7 +5308,8 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
>> >
>> > if (sd_flag & SD_BALANCE_WAKE) {
>> > record_wakee(p);
>> > - want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
>> > + want_affine = !wake_wide(p) && !wake_cap(p, cpu, prev_cpu)
>> > + && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
>> > }
>> >
>> > rcu_read_lock();
>> > --
>> > 1.9.1
>> >
next prev parent reply other threads:[~2016-07-14 13:45 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-22 17:03 [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 01/13] sched: Fix power to capacity renaming in comment Morten Rasmussen
2016-08-10 18:03 ` [tip:sched/core] sched/core: " tip-bot for Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 02/13] sched/fair: Consistent use of prev_cpu in wakeup path Morten Rasmussen
2016-06-22 18:04 ` Rik van Riel
2016-06-23 9:56 ` Morten Rasmussen
2016-06-23 12:24 ` Rik van Riel
2016-08-10 18:03 ` [tip:sched/core] sched/fair: Make the use of prev_cpu consistent in the " tip-bot for Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 03/13] sched/fair: Optimize find_idlest_cpu() when there is no choice Morten Rasmussen
2016-07-13 12:20 ` Vincent Guittot
2016-08-10 18:03 ` [tip:sched/core] " tip-bot for Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 04/13] sched: Introduce SD_ASYM_CPUCAPACITY sched_domain topology flag Morten Rasmussen
2016-07-11 9:55 ` Peter Zijlstra
2016-07-11 10:42 ` Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 05/13] sched: Enable SD_BALANCE_WAKE for asymmetric capacity systems Morten Rasmussen
2016-07-11 10:04 ` Peter Zijlstra
2016-07-11 10:37 ` Morten Rasmussen
2016-07-11 11:04 ` Morten Rasmussen
2016-07-11 11:24 ` Peter Zijlstra
2016-07-12 14:26 ` Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 06/13] sched: Store maximum per-cpu capacity in root domain Morten Rasmussen
2016-07-11 10:18 ` Peter Zijlstra
2016-07-11 16:16 ` Dietmar Eggemann
2016-07-12 11:42 ` Peter Zijlstra
2016-07-13 11:18 ` Dietmar Eggemann
2016-07-13 12:40 ` Vincent Guittot
2016-07-13 13:48 ` Dietmar Eggemann
2016-07-13 16:37 ` Morten Rasmussen
2016-07-14 13:25 ` Vincent Guittot
2016-07-14 15:15 ` Morten Rasmussen
2016-07-15 11:46 ` Morten Rasmussen
2016-07-15 13:39 ` Vincent Guittot
2016-07-15 16:02 ` Morten Rasmussen
2016-07-18 12:48 ` Vincent Guittot
2016-07-18 15:11 ` Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 07/13] sched/fair: Let asymmetric cpu configurations balance at wake-up Morten Rasmussen
2016-07-11 11:13 ` Peter Zijlstra
2016-07-11 12:32 ` Morten Rasmussen
2016-07-13 12:56 ` Vincent Guittot
2016-07-13 16:14 ` Morten Rasmussen
2016-07-14 13:45 ` Vincent Guittot [this message]
2016-07-15 8:37 ` Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 08/13] sched/fair: Compute task/cpu utilization at wake-up more correctly Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 09/13] sched/fair: Consider spare capacity in find_idlest_group() Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 10/13] sched: Add per-cpu max capacity to sched_group_capacity Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 11/13] sched/fair: Avoid pulling tasks from non-overloaded higher capacity groups Morten Rasmussen
2016-06-23 21:20 ` Sai Gurrappadi
2016-06-30 7:49 ` Morten Rasmussen
2016-07-14 16:39 ` Sai Gurrappadi
2016-07-15 8:39 ` Morten Rasmussen
2016-07-12 12:59 ` Peter Zijlstra
2016-07-12 14:34 ` Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 12/13] arm: Set SD_ASYM_CPUCAPACITY for big.LITTLE platforms Morten Rasmussen
2016-06-22 17:03 ` [PATCH v2 13/13] arm: Update arch_scale_cpu_capacity() to reflect change to define Morten Rasmussen
2016-06-28 10:20 ` [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support Koan-Sin Tan
2016-06-30 7:53 ` Morten Rasmussen
2016-07-08 7:35 ` KEITA KOBAYASHI
2016-07-08 8:18 ` Morten Rasmussen
2016-07-11 8:33 ` Morten Rasmussen
2016-07-11 12:44 ` Vincent Guittot
2016-07-12 13:25 ` Peter Zijlstra
2016-07-12 14:39 ` Morten Rasmussen
2016-07-13 12:06 ` Vincent Guittot
2016-07-13 15:54 ` 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=CAKfTPtDE1NsA_q4zGrwBXBBHfdzjFwzPohO+iLqeooZ0S8-jQQ@mail.gmail.com \
--to=vincent.guittot@linaro.org \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgalbraith@suse.de \
--cc=mingo@redhat.com \
--cc=morten.rasmussen@arm.com \
--cc=peterz@infradead.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).