From: Jens Axboe <axboe@suse.de>
To: Daniel Phillips <phillips@arcor.de>
Cc: Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] ide write barrier support
Date: Tue, 21 Oct 2003 07:40:44 +0200 [thread overview]
Message-ID: <20031021054044.GC1128@suse.de> (raw)
In-Reply-To: <200310210146.35881.phillips@arcor.de>
On Tue, Oct 21 2003, Daniel Phillips wrote:
> On Monday 20 October 2003 21:56, Jens Axboe wrote:
> > There IS a correctness issue, in
> > that drives ship with write back caching enabled. The fs assumes that
> > once wait_on_buffer() returns, the data is on disk. Which is false, can
> > remain false for quite a long number of seconds.
>
> Maybe what you're saying is, there are only two ways to deal with IDE drives
> with non-disablable writeback cache:
>
> 1) flush the cache on every write
> 2) Implement write barriers, add them to all the journalling filesystems,
> and flush only on the write barrier
>
> and (1) is just too slow. Correct?
Yes, that is exactly what I'm saying. It's not just that, though.
Completely disabling write back caching on IDE drives totally kills
performance typically, they are just not meant to be driven this way.
--
Jens Axboe
next prev parent reply other threads:[~2003-10-21 5:40 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-10-13 14:08 [PATCH] ide write barrier support Jens Axboe
2003-10-13 15:23 ` Jeff Garzik
2003-10-13 15:35 ` Jens Axboe
2003-10-13 15:37 ` Jens Axboe
2003-10-13 22:39 ` Matthias Andree
2003-10-14 0:16 ` Jeff Garzik
2003-10-16 10:36 ` Jens Axboe
2003-10-16 10:46 ` Jeff Garzik
2003-10-16 10:48 ` Jens Axboe
2003-10-13 23:07 ` Andrew Morton
2003-10-14 6:48 ` Jens Axboe
2003-10-15 3:40 ` Greg Stark
2003-10-16 7:10 ` Jens Axboe
2003-10-20 17:10 ` Daniel Phillips
2003-10-20 19:56 ` Jens Axboe
2003-10-20 23:46 ` Daniel Phillips
2003-10-21 5:40 ` Jens Axboe [this message]
2003-10-23 16:22 ` Daniel Phillips
2003-10-23 16:23 ` Jens Axboe
2003-10-23 17:20 ` Daniel Phillips
2003-10-23 23:21 ` Nick Piggin
2003-10-26 21:06 ` Daniel Phillips
2003-10-27 10:29 ` Lars Marowsky-Bree
2003-10-27 21:35 ` Daniel Phillips
2003-10-24 9:36 ` Helge Hafting
2003-10-26 15:38 ` Daniel Phillips
-- strict thread matches above, loose matches on Subject: below --
2003-10-16 16:51 Mudama, Eric
2003-10-16 20:43 ` Greg Stark
2003-10-17 6:44 ` Jens Axboe
2003-10-17 6:46 ` Jens Axboe
2003-10-16 20:51 Mudama, Eric
2003-10-17 6:48 ` Jens Axboe
2003-10-17 16:07 Mudama, Eric
2003-10-17 18:08 ` Jens Axboe
2003-10-17 17:59 Manfred Spraul
2003-10-17 18:06 ` Jens Axboe
2003-10-21 0:47 ` Matthias Andree
2003-10-17 18:42 Mudama, Eric
[not found] <IXzh.61g.5@gated-at.bofh.it>
2003-10-21 19:24 ` Anton Ertl
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=20031021054044.GC1128@suse.de \
--to=axboe@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=phillips@arcor.de \
/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).