From: "Martin K. Petersen" <martin.petersen@oracle.com>
To: James Bottomley <James.Bottomley@suse.de>
Cc: Chris Mason <chris.mason@oracle.com>,
Christof Schmitt <christof.schmitt@de.ibm.com>,
Boaz Harrosh <bharrosh@panasas.com>,
"Martin K. Petersen" <martin.petersen@oracle.com>,
linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-fsdevel@vger.kernel.org
Subject: Re: Wrong DIF guard tag on ext2 write
Date: Tue, 01 Jun 2010 09:50:29 -0400 [thread overview]
Message-ID: <yq1hblmkgka.fsf@sermon.lab.mkp.net> (raw)
In-Reply-To: <1275399637.21962.11.camel@mulgrave.site> (James Bottomley's message of "Tue, 01 Jun 2010 08:40:37 -0500")
>>>>> "James" == James Bottomley <James.Bottomley@suse.de> writes:
James> Would it be too much work in the fs to mark the page dirty before
James> you begin altering it (and again after you finish, just in case
James> some cleaner noticed and initiated a write)? Or some other flag
James> that indicates page under modification? All the process
James> controlling the writeout (which is pretty high up in the stack)
James> needs to know is if we triggered the check error by altering the
James> page while it was in flight.
James> I agree that a block based retry would close all the holes ... it
James> just doesn't look elegant to me that the fs will already be
James> repeating the I/O if it changed the page and so will block.
I experimented with this approach a while back. However, I quickly got
into a situation where frequently updated blocks never made it to disk
because the page was constantly being updated. And all writes failed
with a guard tag error.
--
Martin K. Petersen Oracle Linux Engineering
next prev parent reply other threads:[~2010-06-01 13:50 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-05-31 11:28 Wrong DIF guard tag on ext2 write Christof Schmitt
2010-05-31 11:34 ` Christof Schmitt
2010-05-31 14:20 ` Martin K. Petersen
2010-05-31 14:46 ` Christof Schmitt
2010-06-01 13:16 ` Martin K. Petersen
2010-06-02 13:37 ` Christof Schmitt
2010-06-02 23:20 ` Dave Chinner
2010-06-04 1:34 ` Martin K. Petersen
2010-06-04 2:32 ` Dave Chinner
2010-06-07 16:20 ` Martin K. Petersen
2010-06-07 17:22 ` Boaz Harrosh
2010-06-07 17:40 ` Martin K. Petersen
2010-06-08 7:15 ` Christof Schmitt
2010-06-08 8:47 ` Dave Chinner
2010-06-08 8:52 ` Nick Piggin
2010-05-31 14:49 ` Nick Piggin
2010-06-01 13:17 ` Martin K. Petersen
2010-05-31 15:01 ` James Bottomley
2010-05-31 15:30 ` Boaz Harrosh
2010-05-31 15:49 ` Nick Piggin
2010-05-31 16:25 ` Boaz Harrosh
2010-06-01 13:22 ` Martin K. Petersen
2010-06-01 10:30 ` Christof Schmitt
2010-06-01 10:49 ` Boaz Harrosh
2010-06-01 13:03 ` Chris Mason
2010-06-01 13:50 ` Christof Schmitt
2010-06-01 13:50 ` Christof Schmitt
2010-06-01 13:50 ` Christof Schmitt
2010-06-01 13:58 ` Chris Mason
2010-06-08 7:18 ` Christof Schmitt
2010-06-08 7:18 ` Christof Schmitt
2010-06-08 7:18 ` Christof Schmitt
2010-06-01 14:26 ` Nick Piggin
2010-06-01 13:27 ` James Bottomley
2010-06-01 13:33 ` Chris Mason
2010-06-01 13:40 ` James Bottomley
2010-06-01 13:49 ` Chris Mason
2010-06-01 16:29 ` Matthew Wilcox
2010-06-01 16:29 ` Matthew Wilcox
2010-06-01 16:47 ` Chris Mason
2010-06-01 16:54 ` James Bottomley
2010-06-01 18:09 ` Chris Mason
2010-06-01 18:46 ` Nick Piggin
2010-06-01 19:35 ` Chris Mason
2010-06-02 3:20 ` Nick Piggin
2010-06-02 3:20 ` Nick Piggin
2010-06-02 13:17 ` Martin K. Petersen
2010-06-02 13:41 ` Nick Piggin
2010-06-03 15:46 ` Chris Mason
2010-06-03 16:27 ` Nick Piggin
2010-06-03 16:27 ` Nick Piggin
2010-06-04 1:46 ` Martin K. Petersen
2010-06-04 3:09 ` Nick Piggin
2010-06-03 16:27 ` Nick Piggin
2010-06-04 2:02 ` Dave Chinner
2010-06-04 2:02 ` Dave Chinner
2010-06-04 2:02 ` Dave Chinner
2010-06-04 15:32 ` Jan Kara
2010-06-04 1:30 ` Martin K. Petersen
2010-06-02 3:20 ` Nick Piggin
2010-06-01 18:46 ` Nick Piggin
2010-06-01 18:46 ` Nick Piggin
2010-06-01 21:07 ` James Bottomley
2010-06-01 22:49 ` Chris Mason
2010-06-01 16:29 ` Matthew Wilcox
2010-06-01 13:50 ` Martin K. Petersen [this message]
2010-06-01 14:28 ` Nick Piggin
2010-06-01 14:32 ` James Bottomley
2010-06-01 14:54 ` Martin K. Petersen
2010-06-03 11:20 ` Vladislav Bolkhovitin
2010-06-03 12:07 ` Boaz Harrosh
2010-06-03 12:41 ` Vladislav Bolkhovitin
2010-06-03 12:46 ` Vladislav Bolkhovitin
2010-06-09 15:58 ` Vladislav Bolkhovitin
2010-06-03 13:06 ` Boaz Harrosh
2010-06-03 13:23 ` Vladislav Bolkhovitin
2010-07-23 17:59 ` Gennadiy Nerubayev
2010-07-23 17:59 ` Gennadiy Nerubayev
2010-07-23 19:16 ` Vladislav Bolkhovitin
2010-07-23 20:51 ` Gennadiy Nerubayev
2010-07-26 12:22 ` Vladislav Bolkhovitin
2010-07-26 17:00 ` Gennadiy Nerubayev
2010-07-26 19:26 ` Vladislav Bolkhovitin
2010-07-24 1:03 ` Dave Chinner
2010-06-01 2:40 ` FUJITA Tomonori
2010-06-03 16:09 ` [LFS/VM TOPIC] Stable pages while IO (was Wrong DIF guard tag on ext2 write) Boaz Harrosh
2010-06-03 16:09 ` Boaz Harrosh
2010-06-03 16:09 ` Boaz Harrosh
2010-06-03 16:30 ` [Lsf10-pc] " J. Bruce Fields
2010-06-03 17:41 ` Vladislav Bolkhovitin
2010-06-04 16:23 ` Jan Kara
2010-06-04 16:30 ` [Lsf10-pc] " J. Bruce Fields
2010-06-04 17:11 ` Jan Kara
2010-06-06 9:35 ` Boaz Harrosh
2010-06-06 23:37 ` Jan Kara
2010-06-07 8:30 ` Boaz Harrosh
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=yq1hblmkgka.fsf@sermon.lab.mkp.net \
--to=martin.petersen@oracle.com \
--cc=James.Bottomley@suse.de \
--cc=bharrosh@panasas.com \
--cc=chris.mason@oracle.com \
--cc=christof.schmitt@de.ibm.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.