From: Russell King <rmk@arm.linux.org.uk>
To: Robert Love <rml@tech9.net>
Cc: Andrew Morton <akpm@zip.com.au>,
Linus Torvalds <torvalds@transmeta.com>,
viro@math.psu.edu, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] 2.5: push BKL out of llseek
Date: Wed, 30 Jan 2002 10:36:40 +0000 [thread overview]
Message-ID: <20020130103640.D16937@flint.arm.linux.org.uk> (raw)
In-Reply-To: <Pine.LNX.4.33.0201291602510.1747-100000@penguin.transmeta.com>, <Pine.LNX.4.33.0201291602510.1747-100000@penguin.transmeta.com> <1012351309.813.56.camel@phantasy> <3C574BD1.E5343312@zip.com.au> <1012357211.817.67.camel@phantasy>
In-Reply-To: <1012357211.817.67.camel@phantasy>; from rml@tech9.net on Tue, Jan 29, 2002 at 09:20:10PM -0500
On Tue, Jan 29, 2002 at 09:20:10PM -0500, Robert Love wrote:
> On Tue, 2002-01-29 at 20:26, Andrew Morton wrote:
> > Just a little word of caution here. Remember the
> > apache-flock-synchronisation fiasco, where removal
> > of the BKL halved Apache throughput on 8-way x86.
> >
> > This was because the BKL removal turned serialisation
> > on a quick codepath from a spinlock into a schedule().
>
> I feared this too, but eventually I decided it was worth it and
> benchmarks backed that up. If nothing else this is yet-another-excuse
> for locks that can spin-then-sleep.
>
> I posted dbench results, which show a positive gain even on 2-way for
> multiple client loads.
Luckily, apache on its own doesn't seem to use lseek() when sending the
file - it seems to be an open, mmap, write. (apache 1.3.22)
However, php with apache does do an lseek on the target script. Now
remember the /. effect... You'll be accessing the same file from
several apache or php processes.
--
Russell King (rmk@arm.linux.org.uk) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
next prev parent reply other threads:[~2002-01-30 10:37 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-01-30 0:00 [PATCH] 2.5: push BKL out of llseek Robert Love
2002-01-30 0:09 ` Linus Torvalds
2002-01-30 0:41 ` Robert Love
2002-01-30 0:52 ` Linus Torvalds
2002-01-30 2:24 ` Robert Love
2002-01-30 1:26 ` Andrew Morton
2002-01-30 2:16 ` Linus Torvalds
2002-01-30 2:20 ` Robert Love
2002-01-30 2:20 ` Andrew Morton
2002-01-30 2:21 ` Dave Jones
2002-01-30 2:37 ` Robert Love
2002-01-30 2:50 ` Nigel Gamble
2002-01-30 3:19 ` Andrew Morton
2002-01-30 9:34 ` Nigel Gamble
2002-01-30 10:36 ` Russell King [this message]
2002-01-30 4:54 ` Alexander Viro
2002-01-30 8:00 ` Trond Myklebust
2002-01-30 13:39 ` Robert Love
2002-01-30 4:50 ` Anton Blanchard
2002-01-30 5:03 ` Robert Love
-- strict thread matches above, loose matches on Subject: below --
2002-01-30 21:14 Martin Wirth
2002-01-31 15:39 Martin Wirth
2002-01-31 21:06 ` Nigel Gamble
2002-02-01 19:29 John Hawkes
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=20020130103640.D16937@flint.arm.linux.org.uk \
--to=rmk@arm.linux.org.uk \
--cc=akpm@zip.com.au \
--cc=linux-kernel@vger.kernel.org \
--cc=rml@tech9.net \
--cc=torvalds@transmeta.com \
--cc=viro@math.psu.edu \
/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