Re: [REGRESSION] 2.6.19/2.6.20-rc3 buffered write slowdown

All of lore.kernel.org
 help / color / mirror / Atom feed

From: David Chinner <dgc@sgi.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: David Chinner <dgc@sgi.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [REGRESSION] 2.6.19/2.6.20-rc3 buffered write slowdown
Date: Thu, 11 Jan 2007 10:08:55 +1100	[thread overview]
Message-ID: <20070110230855.GF44411608@melbourne.sgi.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0701101503310.22578@schroedinger.engr.sgi.com>

On Wed, Jan 10, 2007 at 03:04:15PM -0800, Christoph Lameter wrote:
> On Thu, 11 Jan 2007, David Chinner wrote:
> 
> > The performance and smoothness is fully restored on 2.6.20-rc3
> > by setting dirty_ratio down to 10 (from the default 40), so
> > something in the VM is not working as well as it used to....
> 
> dirty_background_ratio is left as is at 10?

Yes.

> So you gain performance by switching off background writes via pdflush?

Well, pdflush appears to be doing very little on both 2.6.18 and
2.6.20-rc3. In both cases kswapd is consuming 10-20% of a CPU and
all of the pdflush threads combined (I've seen up to 7 active at
once) use maybe 1-2% of cpu time. This occurs regardless of the
dirty_ratio setting.

Cheers,

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

WARNING: multiple messages have this Message-ID (diff)

From: David Chinner <dgc@sgi.com>
To: Christoph Lameter <clameter@sgi.com>
Cc: David Chinner <dgc@sgi.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [REGRESSION] 2.6.19/2.6.20-rc3 buffered write slowdown
Date: Thu, 11 Jan 2007 10:08:55 +1100	[thread overview]
Message-ID: <20070110230855.GF44411608@melbourne.sgi.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0701101503310.22578@schroedinger.engr.sgi.com>

On Wed, Jan 10, 2007 at 03:04:15PM -0800, Christoph Lameter wrote:
> On Thu, 11 Jan 2007, David Chinner wrote:
> 
> > The performance and smoothness is fully restored on 2.6.20-rc3
> > by setting dirty_ratio down to 10 (from the default 40), so
> > something in the VM is not working as well as it used to....
> 
> dirty_background_ratio is left as is at 10?

Yes.

> So you gain performance by switching off background writes via pdflush?

Well, pdflush appears to be doing very little on both 2.6.18 and
2.6.20-rc3. In both cases kswapd is consuming 10-20% of a CPU and
all of the pdflush threads combined (I've seen up to 7 active at
once) use maybe 1-2% of cpu time. This occurs regardless of the
dirty_ratio setting.

Cheers,

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

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

next prev parent reply	other threads:[~2007-01-10 23:09 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-01-10 22:37 [REGRESSION] 2.6.19/2.6.20-rc3 buffered write slowdown David Chinner
2007-01-10 22:37 ` David Chinner
2007-01-10 23:04 ` Christoph Lameter
2007-01-10 23:04   ` Christoph Lameter
2007-01-10 23:08   ` David Chinner [this message]
2007-01-10 23:08     ` David Chinner
2007-01-10 23:12     ` Christoph Lameter
2007-01-10 23:12       ` Christoph Lameter
2007-01-10 23:18       ` David Chinner
2007-01-10 23:18         ` David Chinner
2007-01-10 23:13     ` Nick Piggin
2007-01-10 23:13       ` Nick Piggin
2007-01-11  0:31       ` David Chinner
2007-01-11  0:43         ` Christoph Lameter
2007-01-11  0:43           ` Christoph Lameter
2007-01-11  1:06           ` David Chinner
2007-01-11  1:06             ` David Chinner
2007-01-11  1:40             ` Christoph Lameter
2007-01-11  1:40               ` Christoph Lameter
2007-01-11  2:57               ` David Chinner
2007-01-11  2:57                 ` David Chinner
2007-01-11  1:08         ` Nick Piggin
2007-01-11  1:08           ` Nick Piggin
2007-01-11  1:24           ` David Chinner
2007-01-11  1:24             ` David Chinner
2007-01-11  9:27             ` Nick Piggin
2007-01-11  9:27               ` Nick Piggin
2007-01-11 17:51               ` Christoph Lameter
2007-01-11 17:51                 ` Christoph Lameter
2007-01-12  0:06                 ` Nick Piggin
2007-01-12  0:06                   ` Nick Piggin
2007-01-12  3:04                   ` Christoph Lameter
2007-01-12  3:04                     ` Christoph Lameter
     [not found]           ` <20070111063555.GB33919298@melbourne.sgi.com>
2007-01-11  9:23             ` Nick Piggin
2007-01-11  9:23               ` Nick Piggin
2007-01-11  1:11     ` David Chinner
2007-01-11  1:11       ` David Chinner

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=20070110230855.GF44411608@melbourne.sgi.com \
    --to=dgc@sgi.com \
    --cc=clameter@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.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 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.