From: Andrew Morton <akpm@zip.com.au>
To: Marcelo Tosatti <marcelo@conectiva.com.br>
Cc: Andreas Dilger <adilger@turbolabs.com>,
"'lkml'" <linux-kernel@vger.kernel.org>
Subject: Re: Unresponiveness of 2.4.16
Date: Wed, 28 Nov 2001 13:26:55 -0800 [thread overview]
Message-ID: <3C05569F.7A712167@zip.com.au> (raw)
In-Reply-To: <3C05533D.98DCE6D1@zip.com.au> <Pine.LNX.4.21.0111281802280.15737-100000@freak.distro.conectiva>
Marcelo Tosatti wrote:
>
> > ...
> > But so little code is actually using READA at this stage that I didn't
> > bother - I first need to go through those paths and make sure that they
> > are in fact complete, working and useful...
>
> I've done some experiments in the past which have shown that doing this
> will cause us to almost _never_ do readahead on IO intensive workloads,
> which ended up decreasing performance instead increasing it.
Interesting. Thanks.
One _could_ make the first readahead page non-READA, and then
make the rest READA. That way, all block-contiguous requests
will be merged, and any non-contiguous requests will be dropped on
the floor if the request queue is full. Which is probably what
we want to happen anyway.
Of course the alternative is to slot a little bmap() call into
the readhead logic :)
> Please make sure to extensively test the propagation of READA through the
> pagecache when you do so...
Extensivelytest is my middle name.
-
next prev parent reply other threads:[~2001-11-28 21:28 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-11-26 22:02 Unresponiveness of 2.4.16 Nathan G. Grennan
2001-11-26 22:17 ` Alan Cox
2001-11-26 23:34 ` Nicolas Pitre
2001-11-27 0:05 ` Steve Lion
2001-11-27 9:12 ` Ahmed Masud
2001-11-27 17:12 ` Andrew Morton
2001-11-27 20:31 ` Mike Fedyk
2001-11-27 20:57 ` Andrew Morton
2001-11-27 21:19 ` Martin Eriksson
2001-11-27 21:24 ` Mike Fedyk
2001-11-28 18:24 ` Marcelo Tosatti
2001-11-28 18:57 ` Marcelo Tosatti
2001-11-28 20:31 ` Andrew Morton
2001-11-28 20:56 ` Andreas Dilger
2001-11-28 21:12 ` Andrew Morton
2001-11-28 20:04 ` Marcelo Tosatti
2001-11-28 21:26 ` Andrew Morton [this message]
2001-11-26 23:59 ` Rik van Riel
2001-11-27 0:36 ` Andrew Morton
2001-11-27 0:46 ` Rik van Riel
2001-11-27 4:38 ` Mike Fedyk
2001-11-27 4:45 ` Andrew Morton
2001-11-27 1:45 ` Andrea Arcangeli
2001-11-26 22:21 ` Andrew Morton
2001-11-27 7:42 ` Jens Axboe
2001-11-27 7:58 ` Mike Fedyk
2001-11-27 8:01 ` Jens Axboe
2001-11-27 8:31 ` Andrew Morton
2001-11-27 8:38 ` Jens Axboe
2001-11-26 22:44 ` Lincoln Dale
2001-11-27 4:34 ` GOTO Masanori
2001-11-27 0:44 ` Lost Logic
2001-11-27 0:57 ` Lost Logic
2001-11-27 3:49 ` Sean Elble
2001-11-27 3:56 ` Doug Ledford
2001-11-27 4:00 ` Sean Elble
-- strict thread matches above, loose matches on Subject: below --
2001-11-27 9:56 willy tarreau
2001-11-27 10:57 ` Heinz Diehl
2001-11-28 0:33 Torrey Hoffman
2001-11-28 0:48 ` Andrew Morton
2001-11-28 18:09 ` Marcelo Tosatti
2001-11-28 19:38 ` Andrew Morton
2001-11-28 1:31 Dieter Nützel
2001-11-28 2:13 ` Andrew Morton
2001-11-28 2:34 ` Mike Fedyk
2001-11-28 2:48 ` Andrew Morton
2001-11-28 20:21 ` Roger Larsson
2001-11-28 3:53 ` Dieter Nützel
[not found] ` <200111280353.fAS3rEB05638@zero.tech9.net>
2001-11-28 4:14 ` Robert Love
2001-11-28 18:56 Torrey Hoffman
2001-11-28 19:31 ` Andrew Morton
2001-11-28 19:42 Torrey Hoffman
2001-11-28 20:51 ` Dieter Nützel
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=3C05569F.7A712167@zip.com.au \
--to=akpm@zip.com.au \
--cc=adilger@turbolabs.com \
--cc=linux-kernel@vger.kernel.org \
--cc=marcelo@conectiva.com.br \
/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