From: Glauber Costa <glommer@parallels.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Balbir Singh <bsingharora@gmail.com>,
Paul Turner <pjt@google.com>, <linux-kernel@vger.kernel.org>,
<paul@paulmenage.org>, <lizf@cn.fujitsu.com>,
<daniel.lezcano@free.fr>, <jbottomley@parallels.com>,
<fweisbec@gmail.com>
Subject: Re: [PATCH v2 14/14] Change CPUACCT to default n
Date: Thu, 24 Nov 2011 14:07:32 -0200 [thread overview]
Message-ID: <4ECE6BC4.7090600@parallels.com> (raw)
In-Reply-To: <1322141073.2921.51.camel@twins>
On 11/24/2011 11:24 AM, Peter Zijlstra wrote:
> On Mon, 2011-11-21 at 10:59 +0900, KAMEZAWA Hiroyuki wrote:
>> Could you share your analysis why cpuacct cgroup is slow ?
>
> It adds accounting to all scheduler hot paths, accounting that is mostly
> duplicate of accounting already done. It also does another cgroup
> hierarchy walk, separate from existing controllers, hitting all cold
> cachelines again.
>
> IOW, it sucks chunks and I should never have allowed it to be merged.
OTOH, if the use case for it includes separating processes for the cpu
and cpuacct cgroups in an independent manner - which apparently it does,
I've just learned, there isn't much we can do except try to make it cheaper.
The direction I am going right now wrt this is to clean this up and
reduce the impact of it when the root cgroup is the only one active. All
other users are using the functionality, so let them pay...
next prev parent reply other threads:[~2011-11-24 16:08 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-01 21:19 [PATCH v2 00/14] per-cgroup /proc/stat Glauber Costa
2011-11-01 21:19 ` [PATCH v2 01/14] trivial: initialize root cgroup's sibling list Glauber Costa
2011-11-11 21:34 ` Paul Turner
2011-11-14 19:44 ` Glauber Costa
2011-11-14 21:44 ` Peter Zijlstra
2011-11-18 23:42 ` [tip:sched/core] sched, trivial: Initialize " tip-bot for Glauber Costa
2011-11-01 21:19 ` [PATCH v2 02/14] Change cpustat fields to an array Glauber Costa
2011-11-01 21:19 ` [PATCH v2 03/14] Move /proc/stat logic inside sched.c Glauber Costa
2011-11-12 1:35 ` Paul Turner
2011-11-12 10:27 ` Glauber Costa
2011-11-01 21:19 ` [PATCH v2 04/14] split kernel stat in two Glauber Costa
2011-11-01 21:19 ` [PATCH v2 05/14] Display /proc/stat information per cgroup Glauber Costa
2011-11-01 21:19 ` [PATCH v2 06/14] Make total_forks per-cgroup Glauber Costa
2011-11-01 21:19 ` [PATCH v2 07/14] per-cgroup boot time Glauber Costa
2011-11-01 21:19 ` [PATCH v2 08/14] Report steal time for cgroup Glauber Costa
2011-11-01 21:19 ` [PATCH v2 09/14] Keep nr_iowait per cgroup Glauber Costa
2011-11-10 10:27 ` Andrew Wagin
2011-11-01 21:19 ` [PATCH v2 10/14] Keep number of context switches per-cgroup Glauber Costa
2011-11-01 21:19 ` [PATCH v2 11/14] provide a version of cpuacct statistics inside cpu cgroup Glauber Costa
2011-11-01 21:19 ` [PATCH v2 12/14] Keep number of running processes per-cgroup Glauber Costa
2011-11-14 14:42 ` Andrew Wagin
2011-11-01 21:19 ` [PATCH v2 13/14] provide a version of cpuusage statistics inside cpu cgroup Glauber Costa
2011-11-09 11:51 ` Andrew Wagin
2011-11-09 11:58 ` Glauber Costa
2011-11-09 14:18 ` Andrew Wagin
2011-11-09 15:30 ` Peter Zijlstra
2011-11-09 16:51 ` Glauber Costa
2011-11-10 8:59 ` Andrew Wagin
2011-11-01 21:19 ` [PATCH v2 14/14] Change CPUACCT to default n Glauber Costa
2011-11-11 21:33 ` Paul Turner
2011-11-12 10:29 ` Glauber Costa
2011-11-15 11:02 ` Paul Turner
2011-11-16 10:21 ` Balbir Singh
2011-11-16 23:52 ` KAMEZAWA Hiroyuki
2011-11-17 2:49 ` Glauber Costa
2011-11-17 2:58 ` Balbir Singh
2011-11-17 15:58 ` Glauber Costa
2011-11-21 1:59 ` KAMEZAWA Hiroyuki
2011-11-24 13:24 ` Peter Zijlstra
2011-11-24 16:07 ` Glauber Costa [this message]
2011-11-24 16:29 ` Peter Zijlstra
2011-11-24 16:38 ` Glauber Costa
2011-11-24 16:58 ` Peter Zijlstra
2011-11-25 5:38 ` Balbir Singh
2011-11-25 10:19 ` Peter Zijlstra
2011-11-26 13:18 ` Paul Turner
2011-11-28 8:29 ` Balbir Singh
2011-11-25 2:05 ` Li Zefan
2011-11-25 10:09 ` Peter Zijlstra
2011-11-26 13:07 ` Paul Turner
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=4ECE6BC4.7090600@parallels.com \
--to=glommer@parallels.com \
--cc=a.p.zijlstra@chello.nl \
--cc=bsingharora@gmail.com \
--cc=daniel.lezcano@free.fr \
--cc=fweisbec@gmail.com \
--cc=jbottomley@parallels.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizf@cn.fujitsu.com \
--cc=paul@paulmenage.org \
--cc=pjt@google.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