From: Yuyang Du <yuyang.du@intel.com>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@kernel.org>,
linux-kernel <linux-kernel@vger.kernel.org>,
Benjamin Segall <bsegall@google.com>,
Paul Turner <pjt@google.com>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Juri Lelli <juri.lelli@arm.com>
Subject: Re: [PATCH 2/4] sched/fair: Drop out incomplete current period when sched averages accrue
Date: Thu, 14 Apr 2016 02:44:12 +0800 [thread overview]
Message-ID: <20160413184412.GO8697@intel.com> (raw)
In-Reply-To: <CAKfTPtCc3t+g7hESEJ53EnwzvOGr+oE_NF=eOvq0YVo-YXN=HQ@mail.gmail.com>
On Wed, Apr 13, 2016 at 05:28:18PM +0200, Vincent Guittot wrote:
> > For a periodic task, the signals really get much more unstable. Even for
> > a steady state (load/util related) periodic task there is a meander
> > pattern which depends on if we for instance hit a dequeue (decay +
> > accrue) or an enqueue (decay only) after the 1ms has elapsed.
> >
> > IMHO, 1ms is too big to create signals describing task and cpu load/util
> > signals given the current scheduler dynamics. We simply see too many
> > signal driving points (e.g. enqueue/dequeue) bailing out of
> > __update_load_avg().
By "bailing out", you mean return without update because the delta is less
than 1ms?
> > Examples of 1 periodic task pinned to a cpu on an ARM64 system, HZ=250
> > in steady state:
> >
> > (1) task runtime = 100us period = 200us
> >
> > pelt load/util signal
> >
> > 1us: 488-491
> >
> > 1ms: 483-534
100us/200us = 50%, so the util should center around 512, it seems in this
regard, it is better, but the variance is undesirable.
> > We get ~2 dequeues (load/util example: 493->504) and ~2 enqueues
> > (load/util example: 496->483) in the meander pattern in the 1ms case.
> >
> > (2) task runtime = 100us period = 1000us
> >
> > pelt load/util signal
> >
> > 1us: 103-105
> >
> > 1ms: 84-145
> >
> > We get ~3-4 dequeues (load/util example: 104->124->134->140) and ~16-20
> > enqueues (load/util example: 137->134->...->99->97) in the meander
> > pattern in the 1ms case.
The same as above.
>
> yes, similarly i have some use cases with 2ms running task in a period
> of 5.12ms. it will be seen either as a 1ms running task or a 2ms
> running tasks depending on how the running is synced with the 1ms
> boundary
>
> so the load will vary between 197-215 up to 396-423 depending of when
> the 1ms boundary occurs in the 2ms running
>
Same as above, and this time, the util is "expected" to be 2/5.242*1024=391
of all the samples. We solve the problem of overly-decay, but the precision
loss is a new problem.
Let me see if we can get to a 2-level period scheme, :)
next prev parent reply other threads:[~2016-04-14 2:26 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-10 22:36 [PATCH 0/4] Optimize sched avg computation and implement flat util hierarchy Yuyang Du
2016-04-10 22:36 ` [PATCH 1/4] sched/fair: Optimize sum computation with a lookup table Yuyang Du
2016-04-11 9:08 ` Vincent Guittot
2016-04-11 10:41 ` Juri Lelli
2016-04-11 19:12 ` Yuyang Du
2016-04-12 10:14 ` Juri Lelli
2016-04-12 18:07 ` Yuyang Du
2016-04-13 9:11 ` Juri Lelli
2016-04-11 16:59 ` Dietmar Eggemann
2016-04-11 19:17 ` Yuyang Du
2016-04-12 14:19 ` Peter Zijlstra
2016-04-12 18:12 ` Yuyang Du
2016-04-11 23:21 ` Joe Perches
2016-04-12 12:02 ` Juri Lelli
2016-04-11 23:07 ` Joe Perches
2016-04-10 22:36 ` [PATCH 2/4] sched/fair: Drop out incomplete current period when sched averages accrue Yuyang Du
2016-04-11 9:08 ` Vincent Guittot
2016-04-11 19:41 ` Yuyang Du
2016-04-12 11:56 ` Vincent Guittot
2016-04-12 21:09 ` Yuyang Du
2016-04-13 11:11 ` Vincent Guittot
2016-04-12 12:02 ` Dietmar Eggemann
2016-04-12 20:14 ` Yuyang Du
2016-04-13 4:04 ` Joe Perches
2016-04-13 8:40 ` Morten Rasmussen
2016-04-13 15:13 ` Dietmar Eggemann
2016-04-13 15:28 ` Vincent Guittot
2016-04-13 16:20 ` Vincent Guittot
2016-04-13 18:44 ` Yuyang Du [this message]
2016-04-14 12:52 ` Dietmar Eggemann
2016-04-14 20:05 ` Yuyang Du
2016-04-18 17:59 ` Yuyang Du
2016-04-10 22:36 ` [PATCH 3/4] sched/fair: Modify accumulated sums for load/util averages Yuyang Du
2016-04-11 17:14 ` Dietmar Eggemann
2016-04-10 22:36 ` [PATCH 4/4] sched/fair: Implement flat hierarchical structure for util_avg Yuyang Du
2016-04-11 12:29 ` Vincent Guittot
2016-04-11 20:37 ` Yuyang Du
2016-04-13 11:27 ` Vincent Guittot
2016-04-13 18:20 ` Yuyang Du
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=20160413184412.GO8697@intel.com \
--to=yuyang.du@intel.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=juri.lelli@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=vincent.guittot@linaro.org \
/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).