From: Chris Mason <chris.mason@oracle.com>
To: Ric Wheeler <ricwheeler@gmail.com>
Cc: Theodore Tso <tytso@mit.edu>, Dave Chinner <david@fromorbit.com>,
lsf <lsf@lists.linux-foundation.org>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
James Bottomley <james.bottomley@hansenpartnership.com>,
device-mapper development <dm-devel@redhat.com>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Ric Wheeler <rwheeler@redhat.com>
Subject: Re: [Lsf] Preliminary Agenda and Activities for LSF
Date: Wed, 30 Mar 2011 10:07:43 -0400 [thread overview]
Message-ID: <1301493824-sup-9308@think> (raw)
In-Reply-To: <4D9313E2.6080006@gmail.com>
Excerpts from Ric Wheeler's message of 2011-03-30 07:28:34 -0400:
> On 03/30/2011 07:13 AM, Theodore Tso wrote:
> > On Mar 29, 2011, at 10:17 PM, Dave Chinner wrote:
> >
> >> Direct IO semantics have always been that the application is allowed
> >> to overlap IO to the same range if it wants to. The result is
> >> undefined (just like issuing overlapping reads and writes to a disk
> >> at the same time) so it's the application's responsibility to avoid
> >> overlapping IO if it is a problem.
> > Even if the overlapping read/writes are taking place in different processes?
> >
> > DIO has never been standardized, and was originally implemented as gentleman's agreements between various database manufacturers and proprietary unix vendors. The lack of formal specifications of what applications are guaranteed to receive is unfortunate....
> >
> > -- Ted
>
> What possible semantics could you have?
>
> If you ever write concurrently from multiple processes without locking, you
> clearly are at the mercy of the scheduler and the underlying storage which could
> fragment a single write into multiple IO's sent to the backend device.
>
> I would agree with Dave, let's not make it overly complicated or try to give
> people "atomic" unbounded size writes just because they set the O_DIRECT flag :)
We've talked about this with the oracle database people at least, any
concurrent O_DIRECT ios to the same area would be considered a db bug.
As long as it doesn't make the kernel crash or hang, we can return
one of these: http://www.youtube.com/watch?v=rX7wtNOkuHo
IBM might have a different answer, but I don't see how you can have good
results from mixing concurrent IOs.
-chris
next prev parent reply other threads:[~2011-03-30 14:07 UTC|newest]
Thread overview: 138+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1301373398.2590.20.camel@mulgrave.site>
2011-03-29 5:14 ` [Lsf] Preliminary Agenda and Activities for LSF Amir Goldstein
2011-03-29 11:16 ` Ric Wheeler
2011-03-29 11:22 ` Matthew Wilcox
2011-03-29 12:17 ` Jens Axboe
2011-03-29 13:09 ` Martin K. Petersen
2011-03-29 13:12 ` Ric Wheeler
2011-03-29 13:38 ` James Bottomley
2011-03-29 17:20 ` Shyam_Iyer
2011-03-29 17:33 ` Vivek Goyal
2011-03-29 18:10 ` Shyam_Iyer
2011-03-29 18:45 ` Vivek Goyal
2011-03-29 19:13 ` Shyam_Iyer
2011-03-29 19:57 ` Vivek Goyal
2011-03-29 19:59 ` Mike Snitzer
2011-03-29 20:12 ` Shyam_Iyer
2011-03-29 20:23 ` Mike Snitzer
2011-03-29 23:09 ` Shyam_Iyer
2011-03-30 5:58 ` [Lsf] " Hannes Reinecke
2011-03-30 14:02 ` James Bottomley
2011-03-30 14:10 ` Hannes Reinecke
2011-03-30 14:26 ` James Bottomley
2011-03-30 14:55 ` Hannes Reinecke
2011-03-30 15:33 ` James Bottomley
2011-03-30 15:46 ` Shyam_Iyer
2011-03-30 20:32 ` Giridhar Malavali
2011-03-30 20:45 ` James Bottomley
2011-03-29 19:47 ` Nicholas A. Bellinger
2011-03-29 20:29 ` Jan Kara
2011-03-29 20:31 ` Ric Wheeler
2011-03-30 0:33 ` Mingming Cao
2011-03-30 2:17 ` Dave Chinner
2011-03-30 11:13 ` Theodore Tso
2011-03-30 11:28 ` Ric Wheeler
2011-03-30 14:07 ` Chris Mason [this message]
2011-04-01 15:19 ` Ted Ts'o
2011-04-01 16:30 ` Amir Goldstein
2011-04-01 21:46 ` Joel Becker
2011-04-02 3:26 ` Amir Goldstein
2011-04-01 21:43 ` Joel Becker
2011-03-30 21:49 ` Mingming Cao
2011-03-31 0:05 ` Matthew Wilcox
2011-03-31 1:00 ` Joel Becker
2011-04-01 21:34 ` Mingming Cao
2011-04-01 21:49 ` Joel Becker
2011-03-29 17:35 ` Chad Talbott
2011-03-29 19:09 ` Vivek Goyal
2011-03-29 20:14 ` Chad Talbott
2011-03-29 20:35 ` Jan Kara
2011-03-29 21:08 ` Greg Thelen
2011-03-30 4:18 ` Dave Chinner
2011-03-30 15:37 ` IO less throttling and cgroup aware writeback (Was: Re: [Lsf] Preliminary Agenda and Activities for LSF) Vivek Goyal
2011-03-30 22:20 ` Dave Chinner
2011-03-30 22:49 ` Chad Talbott
2011-03-31 3:00 ` Dave Chinner
2011-03-31 14:16 ` Vivek Goyal
2011-03-31 14:34 ` Chris Mason
2011-03-31 22:14 ` Dave Chinner
2011-03-31 23:43 ` Chris Mason
2011-04-01 0:55 ` Dave Chinner
2011-04-01 1:34 ` Vivek Goyal
2011-04-01 4:36 ` Dave Chinner
2011-04-01 6:32 ` [Lsf] IO less throttling and cgroup aware writeback (Was: " Christoph Hellwig
2011-04-01 7:23 ` Dave Chinner
2011-04-01 12:56 ` Christoph Hellwig
2011-04-21 15:07 ` Vivek Goyal
2011-04-01 14:49 ` IO less throttling and cgroup aware writeback (Was: Re: [Lsf] " Vivek Goyal
2011-03-31 22:25 ` Vivek Goyal
2011-03-31 14:50 ` [Lsf] IO less throttling and cgroup aware writeback (Was: " Greg Thelen
2011-03-31 22:27 ` Dave Chinner
2011-04-01 17:18 ` Vivek Goyal
2011-04-01 21:49 ` Dave Chinner
2011-04-02 7:33 ` Greg Thelen
2011-04-02 7:34 ` Greg Thelen
2011-04-05 13:13 ` Vivek Goyal
2011-04-05 22:56 ` Dave Chinner
2011-04-06 14:49 ` Curt Wohlgemuth
2011-04-06 15:39 ` Vivek Goyal
2011-04-06 19:49 ` Greg Thelen
2011-04-06 23:07 ` [Lsf] IO less throttling and cgroup aware writeback Greg Thelen
2011-04-06 23:36 ` Dave Chinner
2011-04-07 19:24 ` Vivek Goyal
2011-04-07 20:33 ` Christoph Hellwig
2011-04-07 21:34 ` Vivek Goyal
2011-04-07 23:42 ` Dave Chinner
2011-04-08 0:59 ` Greg Thelen
2011-04-08 1:25 ` Dave Chinner
2011-04-12 3:17 ` KAMEZAWA Hiroyuki
2011-04-08 13:43 ` Vivek Goyal
2011-04-06 23:08 ` [Lsf] IO less throttling and cgroup aware writeback (Was: Re: Preliminary Agenda and Activities for LSF) Dave Chinner
2011-04-07 20:04 ` Vivek Goyal
2011-04-07 23:47 ` Dave Chinner
2011-04-08 13:50 ` Vivek Goyal
2011-04-11 1:05 ` Dave Chinner
2011-04-06 15:37 ` Vivek Goyal
2011-04-06 16:08 ` Vivek Goyal
2011-04-06 17:10 ` Jan Kara
2011-04-06 17:14 ` Curt Wohlgemuth
2011-04-08 1:58 ` Dave Chinner
2011-04-19 14:26 ` Wu Fengguang
2011-04-06 23:50 ` Dave Chinner
2011-04-07 17:55 ` Vivek Goyal
2011-04-11 1:36 ` Dave Chinner
2011-04-15 21:07 ` Vivek Goyal
2011-04-16 3:06 ` Vivek Goyal
2011-04-18 21:58 ` Jan Kara
2011-04-18 22:51 ` cgroup IO throttling and filesystem ordered mode (Was: Re: [Lsf] IO less throttling and cgroup aware writeback (Was: Re: Preliminary Agenda and Activities for LSF)) Vivek Goyal
2011-04-19 0:33 ` Dave Chinner
2011-04-19 14:30 ` Vivek Goyal
2011-04-19 14:45 ` Jan Kara
2011-04-19 17:17 ` Vivek Goyal
2011-04-19 18:30 ` Vivek Goyal
2011-04-21 0:32 ` Dave Chinner
2011-04-21 0:29 ` Dave Chinner
2011-04-19 14:17 ` [Lsf] IO less throttling and cgroup aware writeback (Was: Re: Preliminary Agenda and Activities for LSF) Wu Fengguang
2011-04-19 14:34 ` Vivek Goyal
2011-04-19 14:48 ` Jan Kara
2011-04-19 15:11 ` Vivek Goyal
2011-04-19 15:22 ` Wu Fengguang
2011-04-19 15:31 ` Vivek Goyal
2011-04-19 16:58 ` Wu Fengguang
2011-04-19 17:05 ` Vivek Goyal
2011-04-19 20:58 ` Jan Kara
2011-04-20 1:21 ` Wu Fengguang
2011-04-20 10:56 ` Jan Kara
2011-04-20 11:19 ` Wu Fengguang
2011-04-20 14:42 ` Jan Kara
2011-04-20 1:16 ` Wu Fengguang
2011-04-20 18:44 ` Vivek Goyal
2011-04-20 19:16 ` Jan Kara
2011-04-21 0:17 ` Dave Chinner
2011-04-21 15:06 ` Wu Fengguang
2011-04-21 15:10 ` Wu Fengguang
2011-04-21 17:20 ` Vivek Goyal
2011-04-22 4:21 ` Wu Fengguang
2011-04-22 15:25 ` Vivek Goyal
2011-04-22 16:28 ` Andrea Arcangeli
2011-04-25 18:19 ` Vivek Goyal
2011-04-26 14:37 ` Vivek Goyal
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=1301493824-sup-9308@think \
--to=chris.mason@oracle.com \
--cc=david@fromorbit.com \
--cc=dm-devel@redhat.com \
--cc=james.bottomley@hansenpartnership.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=lsf@lists.linux-foundation.org \
--cc=ricwheeler@gmail.com \
--cc=rwheeler@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).