From: Lucas De Marchi <lucas.de.marchi@gmail.com>
To: Fabio Checconi <fchecconi@gmail.com>
Cc: Anirban Sinha <ASinha@zeugmasystems.com>,
linux-kernel@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: question on sched-rt group allocation cap: sched_rt_runtime_us
Date: Sun, 6 Sep 2009 00:40:29 +0200 [thread overview]
Message-ID: <193b0f820909051540v1ad63c24xc822dbdef5bd0c1c@mail.gmail.com> (raw)
In-Reply-To: <20090905204037.GE4953@gandalf.sssup.it>
> You say you pin the threads to a single core: how many cores does your
> system have?
>
> If this is the case, this behavior is the expected one, the scheduler
> tries to reduce the number of migrations, concentrating the bandwidth
> of rt tasks on a single core. With your workload it doesn't work well
> because runtime migration has freed the other core(s) from rt bandwidth,
> so these cores are available to SCHED_OTHER ones, but your SCHED_OTHER
> thread is pinned and cannot make use of them.
Indeed. I've tested this same test program in a single core machine and it
produces the expected behavior:
rt_runtime_us / rt_period_us % loops executed in SCHED_OTHER
95% 4.48%
60% 54.84%
50% 86.03%
40% OTHER completed first
Lucas De Marchi
next prev parent reply other threads:[~2009-09-05 22:40 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <Acotv57vW6nkRxQOQLuBf8W5yfJIlwAAtMAw>
2009-09-05 0:55 ` Anirban Sinha
2009-09-05 17:43 ` Lucas De Marchi
2009-09-05 20:40 ` Fabio Checconi
2009-09-05 22:40 ` Lucas De Marchi [this message]
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC481@exchange.ZeugmaSystems.local>
2009-09-06 0:47 ` Anirban Sinha
2009-09-06 13:21 ` Fabio Checconi
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC484@exchange.ZeugmaSystems.local>
2009-09-07 0:26 ` Anirban Sinha
2009-09-08 17:26 ` Anirban Sinha
2009-09-08 21:37 ` Anirban Sinha
2009-09-05 17:13 Anirban Sinha
[not found] <dgRNo-3uc-5@gated-at.bofh.it>
[not found] ` <dhb9j-1hp-5@gated-at.bofh.it>
[not found] ` <dhcf5-263-13@gated-at.bofh.it>
2009-09-06 2:32 ` Ani
2009-09-06 6:32 ` Mike Galbraith
2009-09-06 10:18 ` Mike Galbraith
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC482@exchange.ZeugmaSystems.local>
2009-09-06 15:09 ` Mike Galbraith
2009-09-07 0:41 ` Anirban Sinha
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC483@exchange.ZeugmaSystems.local>
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC485@exchange.ZeugmaSystems.local>
2009-09-07 0:28 ` Anirban Sinha
2009-09-07 6:54 ` Mike Galbraith
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC489@exchange.ZeugmaSystems.local>
2009-09-08 7:10 ` Anirban Sinha
2009-09-08 9:26 ` Mike Galbraith
2009-09-07 7:59 ` Peter Zijlstra
2009-09-07 8:24 ` Mike Galbraith
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC487@exchange.ZeugmaSystems.local>
2009-09-08 7:08 ` Anirban Sinha
2009-09-08 8:42 ` Peter Zijlstra
2009-09-08 14:41 ` Anirban Sinha
[not found] ` <DDFD17CC94A9BD49A82147DDF7D545C54DC48B@exchange.ZeugmaSystems.local>
2009-09-08 17:41 ` Anirban Sinha
2009-09-08 19:06 ` Mike Galbraith
2009-09-08 19:34 ` Anirban Sinha
2009-09-09 4:10 ` Mike Galbraith
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=193b0f820909051540v1ad63c24xc822dbdef5bd0c1c@mail.gmail.com \
--to=lucas.de.marchi@gmail.com \
--cc=ASinha@zeugmasystems.com \
--cc=a.p.zijlstra@chello.nl \
--cc=fchecconi@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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
all inboxes | Powered by JetHome®