From: Morten Rasmussen <morten.rasmussen@arm.com>
To: Lei Wen <adrian.wenl@gmail.com>
Cc: Alex Shi <alex.shi@intel.com>,
"mingo@redhat.com" <mingo@redhat.com>,
"peterz@infradead.org" <peterz@infradead.org>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
"bp@alien8.de" <bp@alien8.de>, "pjt@google.com" <pjt@google.com>,
"namhyung@kernel.org" <namhyung@kernel.org>,
"efault@gmx.de" <efault@gmx.de>,
"vincent.guittot@linaro.org" <vincent.guittot@linaro.org>,
"preeti@linux.vnet.ibm.com" <preeti@linux.vnet.ibm.com>,
"viresh.kumar@linaro.org" <viresh.kumar@linaro.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"mgorman@suse.de" <mgorman@suse.de>,
"riel@redhat.com" <riel@redhat.com>,
"wangyun@linux.vnet.ibm.com" <wangyun@linux.vnet.ibm.com>,
Jason Low <jason.low2@hp.com>,
Changlong Xie <changlongx.xie@intel.com>,
"sgruszka@redhat.com" <sgruszka@redhat.com>,
"fweisbec@gmail.com" <fweisbec@gmail.com>
Subject: Re: [patch v8 3/9] sched: set initial value of runnable avg for new forked task
Date: Thu, 20 Jun 2013 11:23:34 +0100 [thread overview]
Message-ID: <20130620102334.GI5460@e103034-lin> (raw)
In-Reply-To: <CALZhoSTxKUD2oE7YUFvwASKqnpJBgPP6mnJG3KhFUre5OqYBFQ@mail.gmail.com>
On Sat, Jun 15, 2013 at 01:09:12PM +0100, Lei Wen wrote:
> On Fri, Jun 14, 2013 at 9:59 PM, Alex Shi <alex.shi@intel.com> wrote:
> > On 06/14/2013 06:02 PM, Lei Wen wrote:
> >>> > enqueue_entity
> >>> > enqueue_entity_load_avg
> >>> >
> >>> > and make forking balancing imbalance since incorrect load_avg_contrib.
> >>> >
> >>> > Further more, Morten Rasmussen notice some tasks were not launched at
> >>> > once after created. So Paul and Peter suggest giving a start value for
> >>> > new task runnable avg time same as sched_slice().
> >> I am confused at this comment, how set slice to runnable avg would change
> >> the behavior of "some tasks were not launched at once after created"?
> >
> > I also don't know the details on Morten's machine. but just guess, there
> > are much tasks on in the run queue. the minimum load avg make the new
> > task wait its time...
>
> Is there some possibility that since task structure is allocated without being
> set to 0, and it cause the imbalance between runqueues. Then the new forked
> is migrated to other cpus, so that it cause its execution being delayed?
>
> It is better for Morten to give us more details here. :)
>
I think Peter's reply pretty much covers it. The problem is when a task
is not running (other task has lower vruntime or blocked for other
reasons) shortly after the task was created. The runnable_avg_period is
very small, so the load_contrib is very sensitive.
Say if a task runs for 1 ms then is blocked for 1 ms and then runs
again, the load_contrib will go from 100% to 50% instantly and then ramp
back up again. So the task load may be quite different from the true
load of the task depending on when you calculate the load_contrib.
Preloading runnable_avg_period should make the load_contrib a little
less sensitive to this behaviour.
Morten
> Thanks,
> Lei
>
> >>
> >> IMHO, I could only tell that for the new forked task, it could be run if current
> >> task already be set as need_resched, and preempt_schedule or
> >> preempt_schedule_irq
> >> is called.
> >>
> >> Since the set slice to avg behavior would not affect this task's vruntime,
> >> and hence cannot make current running task be need_sched, if
> >> previously it cannot.
> >>
> >> Could you help correct if I am wrong at somewhere? ....
> >>
> >> Thanks,
> >
> >
> > --
> > Thanks
> > Alex
>
next prev parent reply other threads:[~2013-06-20 10:23 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-07 7:20 [patch 0/9] sched: use runnable load avg in balance Alex Shi
2013-06-07 7:20 ` [patch v8 1/9] Revert "sched: Introduce temporary FAIR_GROUP_SCHED dependency for load-tracking" Alex Shi
2013-06-07 7:20 ` [patch v8 2/9] sched: move few runnable tg variables into CONFIG_SMP Alex Shi
2013-06-17 12:26 ` Paul Turner
2013-06-17 15:32 ` Alex Shi
2013-06-07 7:20 ` [patch v8 3/9] sched: set initial value of runnable avg for new forked task Alex Shi
2013-06-10 1:51 ` Gu Zheng
2013-06-14 10:02 ` Lei Wen
2013-06-14 13:59 ` Alex Shi
2013-06-15 12:09 ` Lei Wen
2013-06-17 0:33 ` Alex Shi
2013-06-20 10:23 ` Morten Rasmussen [this message]
2013-06-21 2:57 ` Lei Wen
2013-06-17 9:20 ` Peter Zijlstra
2013-06-17 12:26 ` Lei Wen
2013-06-17 12:33 ` Peter Zijlstra
2013-06-14 11:09 ` Paul Turner
2013-06-14 14:16 ` Alex Shi
2013-06-17 9:21 ` Peter Zijlstra
2013-06-17 9:39 ` Paul Turner
2013-06-17 13:00 ` Peter Zijlstra
2013-06-17 9:57 ` Alex Shi
2013-06-17 13:07 ` Peter Zijlstra
2013-06-17 13:23 ` Alex Shi
2013-06-07 7:20 ` [patch v8 4/9] sched: fix slept time double counting in enqueue entity Alex Shi
2013-06-17 11:51 ` Paul Turner
2013-06-17 15:41 ` Alex Shi
2013-06-20 1:43 ` Lei Wen
2013-06-20 1:46 ` Alex Shi
2013-06-20 2:46 ` Lei Wen
2013-06-20 14:59 ` Alex Shi
2013-06-21 2:30 ` Lei Wen
2013-06-21 2:39 ` Alex Shi
2013-06-21 2:50 ` Lei Wen
2013-06-21 8:56 ` Alex Shi
2013-06-21 9:18 ` Lei Wen
2013-06-21 11:09 ` Alex Shi
2013-06-21 13:26 ` Lei Wen
2013-06-07 7:20 ` [patch v8 5/9] sched: update cpu load after task_tick Alex Shi
2013-06-17 11:54 ` Paul Turner
2013-06-07 7:20 ` [patch v8 6/9] sched: compute runnable load avg in cpu_load and cpu_avg_load_per_task Alex Shi
2013-06-10 1:49 ` Gu Zheng
2013-06-10 2:01 ` Alex Shi
2013-06-10 2:05 ` Gu Zheng
2013-06-17 10:51 ` Paul Turner
2013-06-17 12:17 ` Paul Turner
2013-06-17 13:39 ` Peter Zijlstra
2013-06-17 13:59 ` Alex Shi
2013-06-17 13:57 ` Alex Shi
2013-06-17 23:00 ` Paul Turner
2013-06-18 3:44 ` Alex Shi
2013-06-18 9:44 ` Alex Shi
2013-06-19 8:15 ` Alex Shi
2013-06-20 0:33 ` Alex Shi
2013-06-17 14:57 ` Alex Shi
2013-06-17 15:21 ` Alex Shi
2013-06-07 7:20 ` [patch v8 7/9] math64: add div64_ul macro Alex Shi
2013-06-07 7:20 ` [patch v8 8/9] sched: consider runnable load average in move_tasks Alex Shi
2013-06-17 10:58 ` Paul Turner
2013-06-17 14:01 ` Alex Shi
2013-06-17 14:15 ` Alex Shi
2013-06-17 13:59 ` Peter Zijlstra
2013-06-17 14:29 ` Alex Shi
2013-06-07 7:20 ` [patch v8 9/9] sched/tg: remove blocked_load_avg in balance Alex Shi
2013-06-17 9:38 ` Peter Zijlstra
2013-06-17 12:20 ` Paul Turner
2013-06-17 14:01 ` Peter Zijlstra
2013-06-19 9:49 ` Alex Shi
2013-06-20 1:33 ` Alex Shi
2013-06-08 2:37 ` [patch 0/9] sched: use runnable load avg " Alex Shi
2013-06-10 1:35 ` Alex Shi
2013-06-10 15:01 ` Peter Zijlstra
2013-06-11 3:30 ` Alex Shi
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=20130620102334.GI5460@e103034-lin \
--to=morten.rasmussen@arm.com \
--cc=adrian.wenl@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=alex.shi@intel.com \
--cc=bp@alien8.de \
--cc=changlongx.xie@intel.com \
--cc=efault@gmx.de \
--cc=fweisbec@gmail.com \
--cc=jason.low2@hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=preeti@linux.vnet.ibm.com \
--cc=riel@redhat.com \
--cc=sgruszka@redhat.com \
--cc=tglx@linutronix.de \
--cc=vincent.guittot@linaro.org \
--cc=viresh.kumar@linaro.org \
--cc=wangyun@linux.vnet.ibm.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).