LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: Ingo Molnar <mingo@elte.hu>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Christoph Lameter <clameter@sgi.com>,
	Bart Van Assche <bart.vanassche@gmail.com>
Subject: Re: quicklists confuse meminfo
Date: Sun, 9 Mar 2008 12:56:11 +0100 (CET)	[thread overview]
Message-ID: <alpine.LFD.1.00.0803091244420.3099@apollo.tec.linutronix.de> (raw)
In-Reply-To: <20080309111456.GA21690@elte.hu>

On Sun, 9 Mar 2008, Ingo Molnar wrote:

> 
> * Thomas Gleixner <tglx@linutronix.de> wrote:
> 
> > Bart reported http://bugzilla.kernel.org/show_bug.cgi?id=9991. He 
> > assumed a memory leak in 32bit kernels when he analyzed the output of 
> > /proc/meminfo.
> > 
> > The leak is not a leak, it's an accounting bug. quicklists keep a 
> > large amount of pages which are accounted as used memory.
> [...]
> > Another strange observation about quicklists is the imbalance of the 
> > quicklists across CPUs. Running the above loop on a 2way machine I can 
> > observe that the quicklist pages are acuumulating on one CPU. Stopping 
> > and restarting the loop a couple of times can shift the accumulation 
> > from one to the other CPU.
> 
> hm. I think we should not let this much RAM hang around in a 
> special-purpose allocator like quicklists. Shouldnt the quicklists be 
> temporary in nature, and be trimmed much more agressively?
> 
> in fact, we have a check_pgt_cache() call in cpu_idle(), which does:
> 
>         quicklist_trim(0, pgd_dtor, 25, 16);
> 
> but it appears we dont do quicklist trimming anywhere else! So if a 
> system has no idle time, the quicklist can grow unbounded, and that's a 
> real memory leak IMO.

Right, also the quicklist_trim() in idle() is freeing at max 16 pages
in one go. According to the quicklist_trim() code we keep up to
(node_free_pages / 16) in the quicklist unconditionally, which seems
rather odd as well.

Thanks,
	tglx

  reply	other threads:[~2008-03-09 11:56 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-03-09 10:19 Thomas Gleixner
2008-03-09 10:26 ` Bart Van Assche
2008-03-09 10:29 ` Andi Kleen
2008-03-09 10:42 ` KOSAKI Motohiro
2008-03-09 12:00   ` Thomas Gleixner
2008-03-09 11:14 ` Ingo Molnar
2008-03-09 11:56   ` Thomas Gleixner [this message]
2008-03-09 12:01     ` Ingo Molnar
2008-03-09 12:49       ` Andi Kleen
2008-03-10 15:51         ` Christoph Lameter
2008-03-09 12:03   ` Johannes Weiner
2008-03-09 12:03   ` KOSAKI Motohiro
2008-03-09 12:09     ` Ingo Molnar
2008-03-09 12:34       ` Ingo Molnar
2008-03-09 12:51         ` KOSAKI Motohiro
2008-03-09 13:20         ` Thomas Gleixner
2008-03-09 18:46         ` Andrew Morton
2008-03-09 20:21           ` Andi Kleen
2008-03-10 15:54           ` Christoph Lameter
2008-03-10 16:43             ` Andi Kleen
2008-03-10 17:19               ` Hugh Dickins
2008-03-10 17:25                 ` Andi Kleen
2008-03-10 17:31                 ` Jeremy Fitzhardinge
2008-03-10 17:53                   ` Andi Kleen
2008-03-10 18:35                     ` Jeremy Fitzhardinge
2008-03-10 19:06                       ` Andi Kleen
2008-03-10 20:54                       ` H. Peter Anvin
2008-03-10 21:26                         ` Jeremy Fitzhardinge
2008-03-11  4:07             ` Nick Piggin
2008-03-21 12:52               ` Bart Van Assche
2008-03-21 14:45                 ` Ingo Molnar
2008-03-26  7:45                   ` Bart Van Assche
2008-03-26  7:53                     ` Andrew Morton
2008-03-26  8:13                     ` Ingo Molnar
2008-03-26 10:37                       ` Bart Van Assche
2008-03-26 16:34                         ` Christoph Lameter
2008-03-27  9:48                           ` Bart Van Assche
2008-03-09 19:11         ` Arjan van de Ven
2008-03-09 19:25           ` Ingo Molnar
2008-03-09 19:27             ` Ingo Molnar
2008-03-09 19:31               ` Ingo Molnar
2008-03-10 15:57               ` Christoph Lameter
2008-03-10 15:55             ` Christoph Lameter
2008-03-09 12:47       ` KOSAKI Motohiro

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=alpine.LFD.1.00.0803091244420.3099@apollo.tec.linutronix.de \
    --to=tglx@linutronix.de \
    --cc=akpm@linux-foundation.org \
    --cc=bart.vanassche@gmail.com \
    --cc=clameter@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=torvalds@linux-foundation.org \
    --subject='Re: quicklists confuse meminfo' \
    /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

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).