From: Ric Wheeler <rwheeler@redhat.com>
To: Dave Chinner <david@fromorbit.com>
Cc: Lawrence Greenfield <leg@google.com>, "Ted Ts'o" <tytso@mit.edu>,
Josef Bacik <josef@redhat.com>,
linux-kernel@vger.kernel.org, linux-btrfs@vger.kernel.org,
linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org,
xfs@oss.sgi.com, joel.becker@oracle.com, cmm@us.ibm.com,
cluster-devel@redhat.com
Subject: Re: [PATCH 1/6] fs: add hole punching to fallocate
Date: Fri, 28 Jan 2011 13:13:16 -0500 [thread overview]
Message-ID: <4D43073C.1040100@redhat.com> (raw)
In-Reply-To: <20110112124431.GP28803@dastard>
On 01/12/2011 07:44 AM, Dave Chinner wrote:
> On Tue, Jan 11, 2011 at 04:13:42PM -0500, Lawrence Greenfield wrote:
>> On Tue, Nov 9, 2010 at 6:40 PM, Dave Chinner<david@fromorbit.com> wrote:
>>> The historical reason for such behaviour existing in XFS was that in
>>> 1997 the CPU and IO latency cost of unwritten extent conversion was
>>> significant,
> .....
>
>>>> (Take for example a trusted cluster filesystem backend that checks the
>>>> object checksum before returning any data to the user; and if the
>>>> check fails the cluster file system will try to use some other replica
>>>> stored on some other server.)
>>> IOWs, all they want to do is avoid the unwritten extent conversion
>>> overhead. Time has shown that a bad security/performance tradeoff
>>> decision was made 13 years ago in XFS, so I see little reason to
>>> repeat it for ext4 today....
>> I'd make use of FALLOC_FL_EXPOSE_OLD_DATA. It's not the CPU overhead
>> of extent conversion. It's that extent conversion causes more metadata
>> operations than what you'd have otherwise,
> Yes, that's the "IO latency" part of the cost I mentioned above.
>
>> which means systems that
>> want to use O_DIRECT and make sure the data doesn't go away either
>> have to write O_DIRECT|O_DSYNC or need to call fdatasync().
> Seriously, we tell application writers _all the time_ that they
> *must* use fsync/fdatasync to guarantee their data is on stable
> storage and that they cannot rely on side-effects of filesystem or
> storage specific behaviours (like ext3 ordered mode) to do that job
> for them.
>
> You're suggesting that by introducing FALLOC_FL_EXPOSE_OLD_DATA,
> applications can rely on filesystem/storage specific behaviour to
> guarantee data is on stable storage without the use of
> fdatasync/fsync. Wht you describe is definitely storage specific,
> because volatile write caches still needs the fdatasync to issue a
> cache flush.
>
> Do you see the same conflict here that I do?
>
The very concept seems quite "non-enterprise". I also agree that the cost of
maintaining extra mount options (and code) for something that no sane end user
would ever do seems to be a loss.
Why wouldn't you want to convert the punched hole to an unwritten extent?
Thanks!
Ric
next prev parent reply other threads:[~2011-01-28 18:13 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-08 20:32 [PATCH 1/6] fs: add hole punching to fallocate Josef Bacik
2010-11-08 20:32 ` [PATCH 2/6] XFS: handle hole punching via fallocate properly Josef Bacik
2010-11-09 1:22 ` Dave Chinner
2010-11-09 2:05 ` Josef Bacik
2010-11-09 4:21 ` Dave Chinner
2010-11-08 20:32 ` [PATCH 3/6] Ocfs2: " Josef Bacik
2010-11-08 20:32 ` [PATCH 4/6] Ext4: fail if we try to use hole punch Josef Bacik
2010-11-08 20:32 ` [PATCH 5/6] Btrfs: " Josef Bacik
2010-11-09 10:05 ` Will Newton
2010-11-09 12:53 ` Josef Bacik
2010-11-08 20:32 ` [PATCH 6/6] Gfs2: " Josef Bacik
2010-11-09 1:12 ` [PATCH 1/6] fs: add hole punching to fallocate Dave Chinner
2010-11-09 2:10 ` Josef Bacik
2010-11-09 3:30 ` Ted Ts'o
2010-11-09 4:42 ` Dave Chinner
2010-11-09 21:41 ` Ted Ts'o
2010-11-09 21:53 ` Jan Kara
2010-11-09 23:40 ` Dave Chinner
2011-01-11 21:13 ` Lawrence Greenfield
2011-01-11 21:30 ` Ted Ts'o
2011-01-12 11:48 ` Dave Chinner
2011-01-12 12:44 ` Dave Chinner
2011-01-28 18:13 ` Ric Wheeler [this message]
2010-11-09 20:51 ` Josef Bacik
2010-11-15 17:05 Hole Punching V2 Josef Bacik
2010-11-15 17:05 ` [PATCH 1/6] fs: add hole punching to fallocate Josef Bacik
2010-11-16 11:16 ` Jan Kara
2010-11-16 11:43 ` Jan Kara
2010-11-16 12:52 ` Josef Bacik
2010-11-16 13:14 ` Jan Kara
2010-11-17 0:22 ` Andreas Dilger
2010-11-17 2:11 ` Dave Chinner
2010-11-17 2:28 ` Josef Bacik
2010-11-17 2:34 ` Josef Bacik
2010-11-17 9:30 ` Andreas Dilger
2010-11-17 9:19 ` Andreas Dilger
2010-11-16 12:53 ` Josef Bacik
2010-11-18 1:46 Hole Punching V3 Josef Bacik
2010-11-18 1:46 ` [PATCH 1/6] fs: add hole punching to fallocate Josef Bacik
2010-11-18 23:43 ` Jan Kara
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=4D43073C.1040100@redhat.com \
--to=rwheeler@redhat.com \
--cc=cluster-devel@redhat.com \
--cc=cmm@us.ibm.com \
--cc=david@fromorbit.com \
--cc=joel.becker@oracle.com \
--cc=josef@redhat.com \
--cc=leg@google.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tytso@mit.edu \
--cc=xfs@oss.sgi.com \
/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).