public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Paul Jackson <pj@sgi.com>
To: David Rientjes <rientjes@google.com>
Cc: clameter@sgi.com, akpm@linux-foundation.org, ak@suse.de,
	clameter@cthulhu.engr.sgi.com, linux-kernel@vger.kernel.org
Subject: Re: [patch] cpusets: do not allow TIF_MEMDIE tasks to allocate globally
Date: Tue, 5 Jun 2007 16:55:38 -0700	[thread overview]
Message-ID: <20070605165538.0a3ce6fe.pj@sgi.com> (raw)
In-Reply-To: <alpine.DEB.0.99.0706051633590.17626@chino.kir.corp.google.com>

> If that fails, we can't allocate elsewhere because then we have taken 
> exclusive memory from other applications and is contrary to the definition 
> of mem_exclusive. 

Well, I can't speak to the 'real' meaning of TIF_MEMDIE with authority,
but I can speak to the meaning of cpuset flags.

The mem_exclusive flag doesn't mean this.

It means that you cannot overlap the memory of a sibling cpuset.

You will, necessarily, still overlap the memory of your ancestor cpusets.

Whether or not you make any use of the mem_exclusive flag, you still
get the same (limited) guarantees of memory usage -- namely that your
memory won't be used by tasks in non-overlapping cpusets, with some
exceptions, such as:
 1) memory handed out to interrupt code,
 2) memory handed out for GFP_ATOMIC requests, and
 3) tasks marked PF_EXITING -- will soon free up memory

Tasks in cpusets ancestor to your tasks cpuset can always, easily,
use memory on the same nodes your task is on.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@sgi.com> 1.925.600.0401

  reply	other threads:[~2007-06-05 23:55 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-06-05 22:39 [patch] cpusets: do not allow TIF_MEMDIE tasks to allocate globally David Rientjes
2007-06-05 22:40 ` Christoph Lameter
2007-06-05 22:42   ` David Rientjes
2007-06-05 23:01 ` Paul Jackson
2007-06-05 23:16   ` David Rientjes
2007-06-05 23:19     ` Paul Jackson
2007-06-05 23:20     ` Christoph Lameter
2007-06-05 23:25       ` David Rientjes
2007-06-05 23:32         ` Christoph Lameter
2007-06-05 23:44           ` David Rientjes
2007-06-05 23:55             ` Paul Jackson [this message]
2007-06-06  1:17               ` David Rientjes
2007-06-06  1:20                 ` Paul Jackson
2007-06-05 23:57             ` Christoph Lameter
2007-06-06  1:23               ` David Rientjes
2007-06-06  1:32                 ` Christoph Lameter
2007-06-06  1:40                   ` David Rientjes
2007-06-06  1:54                     ` Christoph Lameter
2007-06-06  3:29                       ` David Rientjes
2007-06-06  6:20 ` Peter Zijlstra
2007-06-06  6:42   ` David Rientjes
2007-06-06  7:09     ` Peter Zijlstra
2007-06-06  7:18       ` David Rientjes
2007-06-06  7:34         ` Paul Jackson
2007-06-06  7:39           ` Andrew Morton
2007-06-06  7:48           ` David Rientjes
2007-06-06  7:56             ` Paul Jackson
2007-06-06  8:00               ` Andrew Morton
2007-06-06  8:03                 ` Peter Zijlstra
2007-06-06  7:56             ` Peter Zijlstra

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=20070605165538.0a3ce6fe.pj@sgi.com \
    --to=pj@sgi.com \
    --cc=ak@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=clameter@cthulhu.engr.sgi.com \
    --cc=clameter@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rientjes@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