From: Rik van Riel <riel@redhat.com>
To: "Pierre-Loup A. Griffais" <pgriffais@valvesoftware.com>
Cc: hannes@cmpxchg.org, linux-kernel@vger.kernel.org,
torvalds@linux-foundation.org, sonnyrao@chromium.org,
kamezawa.hiroyu@jp.fujitsu.com, akpm@linux-foundation.org
Subject: Re: IO regression after ab8fabd46f on x86 kernels with high memory
Date: Fri, 26 Apr 2013 21:53:56 -0400 [thread overview]
Message-ID: <517B2FB4.30605@redhat.com> (raw)
In-Reply-To: <517B1153.8000401@valvesoftware.com>
On 04/26/2013 07:44 PM, Pierre-Loup A. Griffais wrote:
> I initially observed this between kernels 3.2 and 3.5: on 3.2, copying a
> 180M shared object on the same ext4 filesystem takes 0.6s. On 3.5, it
> takes between two and three minutes. It looks like a similar throughput
> regression happens on any machine running an i386 PAE kernel with high
> amounts of memory; the threshold seems to be 16G; passing mem=15G to the
> kernel commandline fixes it.
If you have that much memory in the system, you will
want to run a 64 bit kernel to avoid all kinds of
memory management corner cases.
> I bisected it to the following change:
>
> commit ab8fabd46f811d5153d8a0cd2fac9a0d41fb593d
> Author: Johannes Weiner <jweiner@redhat.com>
> Date: Tue Jan 10 15:07:42 2012 -0800
>
> mm: exclude reserved pages from dirtyable memory
>
> I realize running x86 kernels against high amounts of memory is not
> advised for various reasons, but I would assume that such a big
> regression in basic functionality to not be part of them. Is that
> accurate, or are these configurations expected to become unusable from
> 3.3 onwards?
Reverting that patch would probably break i686 PAE systems with
lots of memory at a different threshold.
With more than 8-12GB of memory, an i686 kernel is between a
rock and a hard place. Whether you move it closer to the rock,
or closer to the hard place, all you do is change the way in
which it breaks.
> Also CCing Sonny since it looks like he tried to fix an overflow issue
> related to the same change with commit c8b74c2f66049, but I'm still
> experiencing the problem with a kernel built from master.
>
> Thanks,
> - Pierre-Loup
--
All rights reversed
next prev parent reply other threads:[~2013-04-27 1:54 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-26 23:44 IO regression after ab8fabd46f on x86 kernels with high memory Pierre-Loup A. Griffais
2013-04-27 1:53 ` Rik van Riel [this message]
2013-04-27 2:42 ` Johannes Weiner
2013-04-29 21:53 ` Pierre-Loup A. Griffais
2013-04-29 22:03 ` Linus Torvalds
2013-04-29 22:08 ` Pierre-Loup A. Griffais
2013-05-02 4:37 ` Sonny Rao
2013-04-30 0:48 ` Rik van Riel
2013-04-30 1:06 ` Pierre-Loup A. Griffais
2013-05-02 1:34 ` Steven Rostedt
2013-05-02 2:46 ` [PATCH] mm,x86: limit 32 bit kernel to 12GB memory Rik van Riel
2013-05-02 7:37 ` Pierre-Loup A. Griffais
2013-05-02 20:03 ` Linus Torvalds
2013-05-11 9:16 ` Yuhong Bao
2013-05-08 19:10 ` IO regression after ab8fabd46f on x86 kernels with high memory H. Peter Anvin
2013-06-03 1:17 ` Yuhong Bao
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=517B2FB4.30605@redhat.com \
--to=riel@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=pgriffais@valvesoftware.com \
--cc=sonnyrao@chromium.org \
--cc=torvalds@linux-foundation.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 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).