linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Martin Bligh <mbligh@mbligh.org>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Arjan van de Ven <arjan@infradead.org>,
	Erik Mouw <erik@harddisk-recovery.com>,
	Andrew Morton <akpm@osdl.org>,
	Jason Schoonover <jasons@pioneer-pra.com>,
	linux-kernel@vger.kernel.org
Subject: Re: High load average on disk I/O on 2.6.17-rc3
Date: Mon, 08 May 2006 19:02:18 -0700	[thread overview]
Message-ID: <445FF82A.2080106@mbligh.org> (raw)
In-Reply-To: <445FF714.4050803@yahoo.com.au>

Nick Piggin wrote:
> Arjan van de Ven wrote:
> 
>>> ... except that any kernel < 2.6 didn't account tasks waiting for disk
>>> IO.
>>>
>>
>> they did. It was "D" state, which counted into load average.
>>
> 
> Perhaps kernel threads in D state should not contribute toward load avg.
> 
> Userspace does not care whether there are 2 or 20 pdflush threads waiting
> for IO. However, when the network/disks can no longer keep up, userspace
> processes will end up going to sleep in writeback or reclaim -- *that* is
> when we start feeling the load.

Personally I'd be far happier having separated counters for both. Then
we can see what the real bottleneck is. Whilst we're at it, on a per-cpu
and per-elevator-queue basis ;-)

M.

  reply	other threads:[~2006-05-09  2:02 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-05-05 17:10 High load average on disk I/O on 2.6.17-rc3 Jason Schoonover
2006-05-06 23:03 ` bert hubert
2006-05-07  1:02   ` Jason Schoonover
2006-05-07 10:54     ` bert hubert
2006-05-07 16:50 ` Andrew Morton
2006-05-07 17:24   ` Jason Schoonover
2006-05-08 11:13   ` Erik Mouw
2006-05-08 11:22     ` Arjan van de Ven
2006-05-08 11:28       ` Russell King
2006-05-08 11:38         ` Avi Kivity
2006-05-08 12:37         ` Arjan van de Ven
2006-05-09 14:37         ` Bill Davidsen
2006-05-08 14:24   ` Martin J. Bligh
2006-05-08 14:55     ` Arjan van de Ven
2006-05-08 15:22       ` Erik Mouw
2006-05-08 15:25         ` Martin J. Bligh
2006-05-08 15:31         ` Arjan van de Ven
2006-05-08 15:42           ` Erik Mouw
2006-05-08 16:02             ` Martin J. Bligh
2006-05-08 16:02             ` Miquel van Smoorenburg
2006-05-08 16:47             ` Russell King
2006-05-08 17:04               ` Gabor Gombas
2006-05-08 17:18             ` Mike Galbraith
2006-05-09  1:57           ` Nick Piggin
2006-05-09  2:02             ` Martin Bligh [this message]
2006-05-09  2:16               ` Nick Piggin
2006-05-09  4:36             ` Arjan van de Ven
2006-05-09  4:46               ` Nick Piggin
2006-05-09  5:27                 ` Hua Zhong
2006-05-09  5:03               ` David Lang
2006-05-15  7:46                 ` Sander
2006-05-08 22:24         ` Bernd Eckenfels
2006-05-08 22:39           ` Lee Revell
2006-05-09  0:08           ` Peter Williams
2006-05-09 18:33           ` Bill Davidsen
     [not found] <69c8K-3Bu-57@gated-at.bofh.it>
2006-05-05 23:12 ` Robert Hancock
2006-05-06  4:39   ` Jason Schoonover
2006-05-06 17:20     ` Robert Hancock
2006-05-06 18:23       ` Jason Schoonover
2006-05-06 20:01         ` Robert Hancock

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=445FF82A.2080106@mbligh.org \
    --to=mbligh@mbligh.org \
    --cc=akpm@osdl.org \
    --cc=arjan@infradead.org \
    --cc=erik@harddisk-recovery.com \
    --cc=jasons@pioneer-pra.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nickpiggin@yahoo.com.au \
    /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).