linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Eric Sandeen <sandeen@redhat.com>
Cc: Theodore Tso <tytso@mit.edu>, Andi Kleen <andi@firstfloor.org>,
	linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH 0/4] (RESEND) ext3[34] barrier changes
Date: Sun, 18 May 2008 21:11:40 -0700	[thread overview]
Message-ID: <20080518211140.b29bee30.akpm@linux-foundation.org> (raw)
In-Reply-To: <4830E60A.2010809@redhat.com>

On Sun, 18 May 2008 21:29:30 -0500 Eric Sandeen <sandeen@redhat.com> wrote:

> Theodore Tso wrote:
> ...
> 
> > Given how rarely people have reported problems, I think it's a really
> > good idea to understand what exactly our exposure is for
> > $COMMON_HARDWARE. 
> 
> I'll propose that very close to 0% of users will ever report "having
> barriers off seems to have corrupted my disk on power loss!" even if
> that's exactly what happened.  And it'd be very tricky to identify in a
> post-mortem.  Instead we'd probably see other weird things caught down
> the road during some later fsck or during filesystem use, and then
> suggest that they go check their cables, run memtest86 or something...
> 
> Perhaps it's not the intent of this reply, Ted, but various other bits
> of this thread have struck me as trying to rationalize away the problem.

Not really.  It's a matter of understanding how big the problem is.  We
know what the cost of the solution is, and it's really large.

It's a tradeoff, and it is unobvious where the ideal answer lies,
especially when not all the information is available.

>  If the discussion were about proper locking to avoid corruption, would
> we really be  saying well, gosh, it's a *really* small window, and
> *most* people won't hit it very often, and proper locking would slow
> things down....

If it slowed really really important workloads by 30% then we'd be
running around with our hair on fire fixing that up.

But fixing this one is nowhere near as easy as fixing some locking
thing.

> So I think that as you suggest, looking for ways to make barriers less
> painful is the far better route, rather than sacrificing correctness for
> speed by turning them off by default when we know there is a chance for
> problems.  People running journaling filesystems most likely expect to
> be safe from this sort of thing, not most of the time, but all of the time.

Well.  Reducing the cost would of course make the decision easy.

  reply	other threads:[~2008-05-19  4:11 UTC|newest]

Thread overview: 80+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-05-16 19:02 [PATCH 0/4] (RESEND) ext3[34] barrier changes Eric Sandeen
2008-05-16 19:05 ` [PATCH 1/4] ext3: enable barriers by default Eric Sandeen
2008-05-19  8:58   ` Pavel Machek
2008-05-16 19:07 ` [PATCH 2/4] ext3: call blkdev_issue_flush on fsync Eric Sandeen
2008-05-16 22:15   ` Jamie Lokier
2008-05-16 19:08 ` [PATCH 3/4] ext4: enable barriers by default Eric Sandeen
2008-05-16 19:09 ` [PATCH 4/4] ext4: call blkdev_issue_flush on fsync Eric Sandeen
2008-05-20  2:34   ` Theodore Tso
2008-05-20 15:43     ` Jamie Lokier
2008-05-20 15:52       ` Eric Sandeen
2008-05-20 20:14         ` Jens Axboe
2008-05-20 19:54       ` Jens Axboe
2008-05-20 22:02         ` Jamie Lokier
2008-05-21  7:30           ` Jens Axboe
2008-05-16 20:05 ` [PATCH 0/4] (RESEND) ext3[34] barrier changes Andrew Morton
2008-05-16 20:53   ` Eric Sandeen
2008-05-16 20:58     ` Andrew Morton
2008-05-16 21:45       ` Jamie Lokier
2008-05-16 22:03         ` Eric Sandeen
2008-05-16 22:09           ` Jamie Lokier
2008-05-16 22:03     ` Jamie Lokier
2008-05-16 22:21       ` Eric Sandeen
2008-05-16 22:53         ` Jamie Lokier
2008-05-17  0:20           ` Theodore Tso
2008-05-17  0:35             ` Andrew Morton
2008-05-17 13:43               ` Theodore Tso
2008-05-17 17:59                 ` Andreas Dilger
2008-05-17 20:44                 ` Theodore Tso
2008-05-20 14:45                   ` Jamie Lokier
2008-05-18  0:48               ` Chris Mason
2008-05-18  1:36                 ` Theodore Tso
2008-05-18 14:49                   ` Ric Wheeler
2008-05-20 14:42                     ` Jamie Lokier
2008-05-20 23:48                     ` Jamie Lokier
     [not found]                   ` <4830420D.4080608__28835.4277647615$1211137279$gmane$org@gmail.com>
2008-05-18 19:59                     ` Andi Kleen
2008-05-18 16:07                       ` Ric Wheeler
2008-05-20 23:44                 ` Jamie Lokier
2008-05-18 20:03         ` Andi Kleen
2008-05-19  0:43           ` Theodore Tso
2008-05-19  2:29             ` Eric Sandeen
2008-05-19  4:11               ` Andrew Morton [this message]
2008-05-19 17:16                 ` Chris Mason
2008-05-19 18:39                   ` Chris Mason
2008-05-19 22:39                     ` Jan Kara
2008-05-20  0:29                       ` Chris Mason
2008-05-20  3:29                         ` Timothy Shimmin
2008-05-20 12:04                           ` Chris Mason
2008-05-20  8:25                     ` Jens Axboe
2008-05-20 12:17                       ` Chris Mason
2008-05-21 11:22                     ` Pavel Machek
2008-05-21 12:32                       ` Theodore Tso
2008-05-21 18:03                       ` Andrew Morton
2008-05-21 18:15                         ` Eric Sandeen
2008-05-21 19:43                           ` Jamie Lokier
2008-05-21 18:29                         ` Theodore Tso
2008-05-21 18:49                           ` Andrew Morton
2008-05-21 19:42                             ` Jamie Lokier
2008-05-21 19:36                           ` Jamie Lokier
2008-05-21 19:40                             ` Chris Mason
2008-05-21 19:54                         ` Jamie Lokier
2008-05-20 14:58                   ` Jamie Lokier
2008-05-21 22:30                   ` Daniel Phillips
2008-05-20 23:35               ` Jamie Lokier
2008-05-19  0:28       ` Theodore Tso
2008-05-20 15:13         ` Jamie Lokier
2008-05-21 20:25           ` Greg Smith
2008-05-16 22:30   ` Jamie Lokier
2008-05-18 19:54   ` Andi Kleen
2008-05-19 13:26     ` Chris Mason
2008-05-19 14:46       ` Theodore Tso
2008-05-20  2:51         ` [PATCH, RFC] ext4: Fix use of write barrier in commit logic Theodore Tso
2008-05-20 15:23           ` Jamie Lokier
2008-05-23 18:33         ` [PATCH 0/4] (RESEND) ext3[34] barrier changes Ric Wheeler
2008-05-20 15:36       ` Jamie Lokier
2008-05-20 16:02         ` Chris Mason
2008-05-20 16:27           ` Jamie Lokier
2008-05-20 17:08             ` Chris Mason
2008-05-20 22:26               ` Jamie Lokier
2008-05-19  9:04   ` Pavel Machek
2008-05-29 13:36   ` Eric Sandeen

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=20080518211140.b29bee30.akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sandeen@redhat.com \
    --cc=tytso@mit.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;
as well as URLs for NNTP newsgroup(s).