All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Chinner <dgc@sgi.com>
To: Torsten Kaiser <just.for.lkml@googlemail.com>
Cc: Fengguang Wu <wfg@mail.ustc.edu.cn>,
	Peter Zijlstra <peterz@infradead.org>,
	Maxim Levitsky <maximlevitsky@gmail.com>,
	linux-kernel@vger.kernel.org,
	Andrew Morton <akpm@linux-foundation.org>,
	David Chinner <dgc@sgi.com>,
	linux-fsdevel@vger.kernel.org
Subject: Re: writeout stalls in current -git
Date: Wed, 7 Nov 2007 10:31:14 +1100	[thread overview]
Message-ID: <20071106233114.GB995458@sgi.com> (raw)
In-Reply-To: <64bb37e0711061353g4a8b881cgd78fef3a11378b9c@mail.gmail.com>

On Tue, Nov 06, 2007 at 10:53:25PM +0100, Torsten Kaiser wrote:
> On 11/6/07, David Chinner <dgc@sgi.com> wrote:
> > Rather than vmstat, can you use something like iostat to show how busy your
> > disks are?  i.e. are we seeing RMW cycles in the raid5 or some such issue.
> 
> Both "vmstat 10" and "iostat -x 10" output from this test:
> procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
>  r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
>  2  0      0 3700592      0  85424    0    0    31    83  108  244  2  1 95  1
> -> emerge reads something, don't knwo for sure what...
>  1  0      0 3665352      0  87940    0    0   239     2  343  585  2  1 97  0
....
> 
> The last 20% of the btrace look more or less completely like this, no
> other programs do any IO...
> 
> 253,0    3   104626   526.293450729   974  C  WS 79344288 + 8 [0]
> 253,0    3   104627   526.293455078   974  C  WS 79344296 + 8 [0]
> 253,0    1    36469   444.513863133  1068  Q  WS 154998480 + 8 [xfssyncd]
> 253,0    1    36470   444.513863135  1068  Q  WS 154998488 + 8 [xfssyncd]
                                                ^^
Apparently we are doing synchronous writes. That would explain why
it is slow. We shouldn't be doing synchronous writes here. I'll see if
I can reproduce this.

<goes off and looks>

Yes, I can reproduce the sync writes coming out of xfssyncd. I'll
look into this further and send a patch when I have something concrete.

Cheers,

Dave.
-- 
Dave Chinner
Principal Engineer
SGI Australian Software Group

  reply	other threads:[~2007-11-06 23:31 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-22  6:22 100% iowait on one of cpus in current -git Maxim Levitsky
2007-10-22  9:11 ` Paolo Ornati
2007-10-22  9:43   ` Maxim Levitsky
2007-10-22  9:41 ` Peter Zijlstra
2007-10-22  9:59   ` Maxim Levitsky
2007-10-22 10:22     ` Peter Zijlstra
2007-10-22 10:40       ` Maxim Levitsky
2007-10-22 10:55         ` Fengguang Wu
2007-10-22 10:55           ` Fengguang Wu
2007-10-22 10:58             ` Maxim Levitsky
2007-10-22 11:19               ` Fengguang Wu
2007-10-22 11:19                 ` Fengguang Wu
2007-10-22 12:21                   ` Maxim Levitsky
2007-10-22 12:37                     ` Fengguang Wu
2007-10-22 12:37                       ` Fengguang Wu
2007-10-22 13:05                         ` Maxim Levitsky
2007-10-22 13:10                           ` Fengguang Wu
2007-10-22 13:10                             ` Fengguang Wu
2007-10-22 13:28                               ` Maxim Levitsky
2007-10-22 13:41                             ` Fengguang Wu
2007-10-22 13:41                               ` Fengguang Wu
2007-10-31 15:22                                 ` Torsten Kaiser
2007-11-01  7:57                                   ` Fengguang Wu
2007-11-01  7:57                                     ` Fengguang Wu
2007-11-01 18:20                                       ` Torsten Kaiser
2007-11-01 19:00                                         ` Torsten Kaiser
2007-11-02  2:21                                           ` writeout stalls " Fengguang Wu
2007-11-02  2:21                                             ` Fengguang Wu
2007-11-02  7:50                                               ` Torsten Kaiser
2007-11-02 10:15                                             ` Peter Zijlstra
2007-11-02 10:33                                               ` Fengguang Wu
2007-11-02 10:33                                                 ` Fengguang Wu
2007-11-05 23:57                                                   ` Andrew Morton
2007-11-06 10:20                                                     ` Peter Zijlstra
2007-11-06 16:25                                                     ` Patch tags [was writeout stalls in current -git] Jonathan Corbet
2007-11-06 17:03                                                       ` Balbir Singh
2007-11-06 23:26                                                       ` Adrian Bunk
2007-11-09 16:10                                                         ` Jonathan Corbet
2007-11-09 16:19                                                           ` Adrian Bunk
2007-11-02 19:22                                               ` writeout stalls in current -git Torsten Kaiser
2007-11-02 20:43                                                 ` David Chinner
2007-11-02 21:02                                                   ` Torsten Kaiser
2007-11-04 11:19                                                   ` Torsten Kaiser
2007-11-05  1:45                                                     ` David Chinner
2007-11-05  7:01                                                       ` Torsten Kaiser
2007-11-05 18:27                                                       ` Torsten Kaiser
2007-11-06  4:25                                                         ` David Chinner
2007-11-06  7:10                                                           ` Torsten Kaiser
2007-11-06 19:01                                                           ` Peter Zijlstra
2007-11-06 20:26                                                             ` Torsten Kaiser
2007-11-06  9:17                                                 ` Fengguang Wu
2007-11-06  9:17                                                   ` Fengguang Wu
2007-11-06 21:53                                                     ` Torsten Kaiser
2007-11-06 23:31                                                       ` David Chinner [this message]
2007-11-07  2:13                                                         ` David Chinner
2007-11-07  7:15                                                           ` Torsten Kaiser
2007-11-08  0:38                                                             ` David Chinner
2007-11-20 13:16                                                               ` Damien Wyart
2007-11-20 21:09                                                                 ` David Chinner
2007-11-02  1:54                                         ` Fengguang Wu
2007-11-02  1:54                                           ` Fengguang Wu
2007-11-02  7:42                                             ` Torsten Kaiser
2007-11-02  7:52                                               ` Fengguang Wu
2007-11-02  7:52                                                 ` Fengguang Wu
2007-11-02 17:47                                                   ` Torsten Kaiser
2007-10-23  7:55                     ` [PATCH] reiserfs: don't drop PG_dirty when releasing sub-page-sized dirty file Fengguang Wu
2007-10-23  7:55                       ` Fengguang Wu
2007-10-23 10:07                         ` Peter Zijlstra
2007-10-23 11:56                           ` Fengguang Wu
2007-10-23 11:56                             ` Fengguang Wu
2007-10-23 14:10                               ` Chris Mason
2007-10-23 14:40                                 ` Fengguang Wu
2007-10-23 14:40                                   ` Fengguang Wu
2007-10-23 10:17                         ` Maxim Levitsky
2007-10-23 14:41                           ` Fengguang Wu
2007-10-23 14:41                             ` Fengguang Wu

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=20071106233114.GB995458@sgi.com \
    --to=dgc@sgi.com \
    --cc=akpm@linux-foundation.org \
    --cc=just.for.lkml@googlemail.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maximlevitsky@gmail.com \
    --cc=peterz@infradead.org \
    --cc=wfg@mail.ustc.edu.cn \
    /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.