From: Mike Galbraith <efault@gmx.de>
To: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
Ingo Molnar <mingo@kernel.org>,
linux-man <linux-man@vger.kernel.org>,
lkml <linux-kernel@vger.kernel.org>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: RFC: documentation of the autogroup feature [v2]
Date: Fri, 25 Nov 2016 14:02:53 +0100 [thread overview]
Message-ID: <1480078973.4075.58.camel@gmx.de> (raw)
In-Reply-To: <c5a9acd3-6b60-192c-312e-2777f2d537c2@gmail.com>
On Thu, 2016-11-24 at 22:41 +0100, Michael Kerrisk (man-pages) wrote:
> Suppose that there are two autogroups competing for the same
> CPU. The first group contains ten CPU-bound processes from a
> kernel build started with make -j10. The other contains a sin‐
> gle CPU-bound process: a video player. The effect of auto‐
> grouping is that the two groups will each receive half of the
> CPU cycles. That is, the video player will receive 50% of the
> CPU cycles, rather just 9% of the cycles, which would likely
> lead to degraded video playback. Or to put things another way:
> an autogroup that contains a large number of CPU-bound pro‐
> cesses does not end up overwhelming the CPU at the expense of
> the other jobs on the system.
I'd say something more wishy-washy here, like cycles are distributed
fairly across groups and leave it at that, as your detailed example is
incorrect due to SMP fairness (which I don't like much because [very
unlikely] worst case scenario renders a box sized group incapable of
utilizing more that a single CPU total). For example, if a group of
NR_CPUS size competes with a singleton, load balancing will try to give
the singleton a full CPU of its very own. If groups intersect for
whatever reason on say my quad lappy, distribution is 80/20 in favor of
the singleton.
> ┌─────────────────────────────────────────────────────┐
> │FIXME │
> ├─────────────────────────────────────────────────────┤
> │How do the nice value of a process and the nice │
> │value of an autogroup interact? Which has priority? │
> │ │
> │It *appears* that the autogroup nice value is used │
> │for CPU distribution between task groups, and that │
> │the process nice value has no effect there. (I.e., │
> │suppose two autogroups each contain a CPU-bound │
> │process, with one process having nice==0 and the │
> │other having nice==19. It appears that they each │
> │get 50% of the CPU.) It appears that the process │
> │nice value has effect only with respect to schedul‐ │
> │ing relative to other processes in the *same* auto‐ │
> │group. Is this correct? │
> └─────────────────────────────────────────────────────┘
Yup, entity nice level affects distribution among peer entities.
-Mike
next prev parent reply other threads:[~2016-11-25 13:21 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-22 15:59 RFC: documentation of the autogroup feature Michael Kerrisk (man-pages)
2016-11-23 10:33 ` [patch] sched/autogroup: Fix 64bit kernel nice adjustment Mike Galbraith
2016-11-23 13:47 ` Michael Kerrisk (man-pages)
2016-11-23 14:12 ` Mike Galbraith
2016-11-23 14:20 ` Michael Kerrisk (man-pages)
2016-11-23 15:55 ` Mike Galbraith
2016-11-24 6:24 ` [tip:sched/urgent] sched/autogroup: Fix 64-bit kernel nice level adjustment tip-bot for Mike Galbraith
2016-11-23 11:39 ` RFC: documentation of the autogroup feature Mike Galbraith
2016-11-23 13:54 ` Michael Kerrisk (man-pages)
2016-11-23 15:33 ` Mike Galbraith
2016-11-23 16:04 ` Michael Kerrisk (man-pages)
2016-11-23 17:11 ` Mike Galbraith
2016-11-24 21:41 ` RFC: documentation of the autogroup feature [v2] Michael Kerrisk (man-pages)
2016-11-25 12:52 ` Afzal Mohammed
2016-11-25 13:04 ` Michael Kerrisk (man-pages)
2016-11-25 13:02 ` Mike Galbraith [this message]
2016-11-25 15:04 ` Michael Kerrisk (man-pages)
2016-11-25 15:48 ` Michael Kerrisk (man-pages)
2016-11-25 15:51 ` Mike Galbraith
2016-11-25 16:08 ` Michael Kerrisk (man-pages)
2016-11-25 16:18 ` Peter Zijlstra
2016-11-25 16:34 ` Michael Kerrisk (man-pages)
2016-11-25 20:54 ` Michael Kerrisk (man-pages)
2016-11-25 21:49 ` Peter Zijlstra
2016-11-29 7:43 ` Michael Kerrisk (man-pages)
2016-11-29 11:46 ` Peter Zijlstra
2016-11-29 13:44 ` Michael Kerrisk (man-pages)
2016-11-25 16:04 ` Peter Zijlstra
2016-11-25 16:13 ` Peter Zijlstra
2016-11-25 16:33 ` Michael Kerrisk (man-pages)
2016-11-25 22:48 ` Peter Zijlstra
2016-11-23 16:05 ` RFC: documentation of the autogroup feature Michael Kerrisk (man-pages)
2016-11-23 17:19 ` Mike Galbraith
2016-11-23 22:12 ` Michael Kerrisk (man-pages)
2016-11-27 21:13 ` Michael Kerrisk (man-pages)
2016-11-28 1:46 ` Mike Galbraith
[not found] ` <1127218a-dd9b-71a8-845d-3a83969632fc@gmail.com>
2016-11-29 9:10 ` Michael Kerrisk (man-pages)
2016-11-29 13:46 ` 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=1480078973.4075.58.camel@gmx.de \
--to=efault@gmx.de \
--cc=a.p.zijlstra@chello.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-man@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=mtk.manpages@gmail.com \
--cc=tglx@linutronix.de \
/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).