All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mike Galbraith <umgwanakikbuti-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Cc: Peter Zijlstra <peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
	Rik van Riel <riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Zefan Li <lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Clark Williams <williams-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Ingo Molnar <mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Luiz Capitulino
	<lcapitulino-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	David Rientjes <rientjes-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH v4 2/2] cpusets,isolcpus: add file to show isolated cpus in cpuset
Date: Mon, 02 Mar 2015 18:35:38 +0100	[thread overview]
Message-ID: <1425317738.9880.17.camel@gmail.com> (raw)
In-Reply-To: <20150302160903.GE17694-piEFEHQLUPpN0TnZuCh8vA@public.gmane.org>

On Mon, 2015-03-02 at 11:09 -0500, Tejun Heo wrote:
> On Mon, Mar 02, 2015 at 05:02:57PM +0100, Mike Galbraith wrote:
> > Well, those are the only ones we can do anything about.  Dirt simple
> > diddling of the workqueue default mask as sched domains are
> > added/removed should do it I think.  Automatically moving any existing
> > unbound worker away from isolated cores at the same time would be a
> > bonus, most important is that no new threads sneak in.
> 
> Worker pools are immutable once created and configuraiton changes are
> achieved by creating new pools and draining old ones but at any rate
> making it follow config changes is almost trivial.  Figuring out
> configuration policy might take a bit of effort tho.  Can you point me
> to what specific configuration it should be following?

For cpusets, an exclusive set should become taboo to unbound workers
when load balancing is turned off.  The user making sched domains go
away is a not so subtle hint that he wants no random interference, as he
is trying to assume full responsibility for task placement therein.

In my trees, I let the user turn rt cpupri/push/pull off as well, as
that further reduces jitter.

	-Mike

WARNING: multiple messages have this Message-ID (diff)
From: Mike Galbraith <umgwanakikbuti@gmail.com>
To: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Rik van Riel <riel@redhat.com>, Zefan Li <lizefan@huawei.com>,
	linux-kernel@vger.kernel.org,
	Clark Williams <williams@redhat.com>,
	Ingo Molnar <mingo@redhat.com>,
	Luiz Capitulino <lcapitulino@redhat.com>,
	David Rientjes <rientjes@google.com>,
	cgroups@vger.kernel.org
Subject: Re: [PATCH v4 2/2] cpusets,isolcpus: add file to show isolated cpus in cpuset
Date: Mon, 02 Mar 2015 18:35:38 +0100	[thread overview]
Message-ID: <1425317738.9880.17.camel@gmail.com> (raw)
In-Reply-To: <20150302160903.GE17694@htj.duckdns.org>

On Mon, 2015-03-02 at 11:09 -0500, Tejun Heo wrote:
> On Mon, Mar 02, 2015 at 05:02:57PM +0100, Mike Galbraith wrote:
> > Well, those are the only ones we can do anything about.  Dirt simple
> > diddling of the workqueue default mask as sched domains are
> > added/removed should do it I think.  Automatically moving any existing
> > unbound worker away from isolated cores at the same time would be a
> > bonus, most important is that no new threads sneak in.
> 
> Worker pools are immutable once created and configuraiton changes are
> achieved by creating new pools and draining old ones but at any rate
> making it follow config changes is almost trivial.  Figuring out
> configuration policy might take a bit of effort tho.  Can you point me
> to what specific configuration it should be following?

For cpusets, an exclusive set should become taboo to unbound workers
when load balancing is turned off.  The user making sched domains go
away is a not so subtle hint that he wants no random interference, as he
is trying to assume full responsibility for task placement therein.

In my trees, I let the user turn rt cpupri/push/pull off as well, as
that further reduces jitter.

	-Mike


  parent reply	other threads:[~2015-03-02 17:35 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-25 16:38 [PATCH -v2 0/2] cpusets,isolcpus: resolve conflict between cpusets and isolcpus riel
2015-02-25 16:38 ` [PATCH 1/2] cpusets,isolcpus: exclude isolcpus from load balancing in cpusets riel
     [not found]   ` <1424882288-2910-2-git-send-email-riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2015-02-27  9:32     ` Peter Zijlstra
2015-02-27  9:32       ` Peter Zijlstra
     [not found]       ` <20150227093231.GG21418-ndre7Fmf5hadTX5a5knrm8zTDFooKrT+cvkQGrU6aU0@public.gmane.org>
2015-02-27 17:08         ` [PATCH 3/2] cpusets,isolcpus: document relationship between cpusets & isolcpus Rik van Riel
2015-02-27 17:08           ` Rik van Riel
2015-02-27 21:15           ` David Rientjes
     [not found]           ` <20150227120803.07db15fe-XDFsmL65zczMw9PXcw3sFBL4W9x8LtSr@public.gmane.org>
2015-02-28  3:23             ` Zefan Li
2015-02-28  3:23               ` Zefan Li
2015-02-28  3:21   ` [PATCH 1/2] cpusets,isolcpus: exclude isolcpus from load balancing in cpusets Zefan Li
2015-02-28  3:21     ` Zefan Li
2015-02-25 16:38 ` [PATCH 2/2] cpusets,isolcpus: add file to show isolated cpus in cpuset riel
     [not found]   ` <1424882288-2910-3-git-send-email-riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2015-02-25 21:09     ` David Rientjes
2015-02-25 21:09       ` David Rientjes
2015-02-25 21:21       ` Rik van Riel
2015-02-25 21:32     ` [PATCH v3 " Rik van Riel
2015-02-25 21:32       ` Rik van Riel
2015-02-26 11:05     ` [PATCH " Zefan Li
2015-02-26 11:05       ` Zefan Li
2015-02-26 15:24       ` Rik van Riel
2015-02-26 17:12       ` [PATCH v4 " Rik van Riel
2015-02-26 17:12         ` Rik van Riel
     [not found]         ` <20150226121231.6fcba7e8-XDFsmL65zczMw9PXcw3sFBL4W9x8LtSr@public.gmane.org>
2015-02-28  3:22           ` Zefan Li
2015-02-28  3:22             ` Zefan Li
2015-03-02  6:15           ` Zefan Li
2015-03-02  6:15             ` Zefan Li
2015-03-02  9:12             ` Peter Zijlstra
2015-03-03  9:51               ` Zefan Li
2015-03-03  9:51                 ` Zefan Li
2015-03-02  9:09           ` Peter Zijlstra
2015-03-02  9:09             ` Peter Zijlstra
2015-03-02 12:44             ` Mike Galbraith
     [not found]               ` <1425300290.5863.17.camel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2015-03-02 14:35                 ` Rik van Riel
2015-03-02 14:35                   ` Rik van Riel
     [not found]                   ` <54F47534.7070204-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2015-03-02 14:54                     ` Mike Galbraith
2015-03-02 14:54                       ` Mike Galbraith
2015-03-02 15:29                 ` Tejun Heo
2015-03-02 15:29                   ` Tejun Heo
2015-03-02 16:02                   ` Mike Galbraith
     [not found]                     ` <1425312177.12094.15.camel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2015-03-02 16:09                       ` Tejun Heo
2015-03-02 16:09                         ` Tejun Heo
     [not found]                         ` <20150302160903.GE17694-piEFEHQLUPpN0TnZuCh8vA@public.gmane.org>
2015-03-02 17:35                           ` Mike Galbraith [this message]
2015-03-02 17:35                             ` Mike Galbraith
2015-03-02 17:01         ` Tejun Heo
     [not found]           ` <20150302170116.GJ17694-piEFEHQLUPpN0TnZuCh8vA@public.gmane.org>
2015-03-02 17:31             ` Tejun Heo
2015-03-02 17:31               ` Tejun Heo

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=1425317738.9880.17.camel@gmail.com \
    --to=umgwanakikbuti-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
    --cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=lcapitulino-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=lizefan-hv44wF8Li93QT0dZR+AlfA@public.gmane.org \
    --cc=mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=peterz-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
    --cc=riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=rientjes-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
    --cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
    --cc=williams-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.