From: Matthew Wilcox <matthew@wil.cx>
To: Ric Wheeler <rwheeler@redhat.com>
Cc: Konrad Rzeszutek <konrad@virtualiron.com>,
Jens Axboe <jens.axboe@oracle.com>,
"Martin K. Petersen" <martin.petersen@oracle.com>,
snitzer@redhat.com, jeff@garzik.org, neilb@suse.de,
James.Bottomley@hansenpartnership.com, dgilbert@interlog.com,
linux-ide@vger.kernel.org, linux-scsi@vger.kernel.org
Subject: Re: [PATCH 2 of 8] block: Export I/O topology for block devices and partitions
Date: Thu, 23 Apr 2009 12:44:19 -0600 [thread overview]
Message-ID: <20090423184418.GT1926@parisc-linux.org> (raw)
In-Reply-To: <49F0B2CD.2020706@redhat.com>
On Thu, Apr 23, 2009 at 02:26:21PM -0400, Ric Wheeler wrote:
> >Should there be a comment/Documentation blurb on how to remediate a
> >misaligment? Such as how to modify the partition tables and what to look
> >for in the BIOS (if anything) to take advantage of 4K sectors?
> >
> >Is there a corresponding patch for fdisk that understands 4K sectors? Or
> >is that already covered by it doing ioctl calls on the block device?
>
> We need to update the various tools to deal with this data and prevent
> misalignment, some of which is going on already.
>
> I am afraid that if you are misaligned, the only remedy would be to tar up
> your data, recreate an aligned device and then restore :-)
We could write a dd_move command (analagous to memmove) that reads a chunk
of your block device into memory, then writes it back shifted slightly.
Of course, if you lose power midway through, you're completely hosed.
A better solution might be for the filesystem to perform subsequent
writes to the device that are aligned. I'm thinking about the case
where you've been using a filesystem on one device, you dd it to another
device (accidentally misaligning it) and then start modifying it on the
new device. If all writes on the new drive go to an aligned address,
the performance implications shouldn't be too bad.
--
Matthew Wilcox Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours. We can't possibly take such
a retrograde step."
next prev parent reply other threads:[~2009-04-23 18:44 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-04-23 5:29 [PATCH 0 of 8] I/O topology patch kit Martin K. Petersen
2009-04-23 5:29 ` [PATCH 1 of 8] block: Expose stacked device queues in sysfs Martin K. Petersen
2009-04-23 5:29 ` [PATCH 2 of 8] block: Export I/O topology for block devices and partitions Martin K. Petersen
2009-04-23 10:51 ` Jens Axboe
2009-04-23 11:49 ` Matthew Wilcox
2009-04-23 11:55 ` Jens Axboe
2009-04-23 13:22 ` Matthew Wilcox
2009-04-23 13:30 ` Matthew Wilcox
2009-04-23 13:17 ` Martin K. Petersen
2009-04-23 18:13 ` Konrad Rzeszutek
2009-04-23 18:26 ` Ric Wheeler
2009-04-23 18:44 ` Matthew Wilcox [this message]
2009-04-23 18:34 ` Martin K. Petersen
2009-04-23 5:29 ` [PATCH 3 of 8] MD: Use new topology calls to indicate alignment and I/O sizes Martin K. Petersen
2009-04-23 5:29 ` [PATCH 4 of 8] sd: Physical block size and alignment support Martin K. Petersen
2009-04-23 16:37 ` Konrad Rzeszutek
2009-04-23 18:25 ` Martin K. Petersen
2009-04-23 18:44 ` Konrad Rzeszutek
2009-04-23 19:02 ` Martin K. Petersen
2009-04-23 5:29 ` [PATCH 5 of 8] sd: Detect non-rotational devices Martin K. Petersen
2009-04-23 10:52 ` Jens Axboe
2009-04-23 11:09 ` Jeff Garzik
2009-04-23 11:13 ` Jens Axboe
2009-04-23 11:22 ` Jeff Garzik
2009-04-23 11:38 ` Matthew Wilcox
2009-04-23 11:58 ` Jeff Garzik
2009-04-23 12:03 ` Jens Axboe
2009-04-23 13:16 ` Martin K. Petersen
2009-04-23 13:33 ` Jeff Garzik
2009-04-23 14:10 ` James Bottomley
2009-04-23 14:16 ` Matthew Wilcox
2009-04-23 14:39 ` Jeff Garzik
2009-04-23 17:25 ` Matthew Wilcox
2009-04-23 17:37 ` James Bottomley
2009-04-23 5:29 ` [PATCH 6 of 8] sd: Block limits VPD support Martin K. Petersen
2009-04-23 5:29 ` [PATCH 7 of 8] scsi_debug: Add support for physical block exponent and alignment Martin K. Petersen
2009-04-23 5:29 ` [PATCH 8 of 8] libata: Report disk alignment and physical block size Martin K. Petersen
2009-04-23 13:46 ` Matthew Wilcox
2009-04-23 14:05 ` Martin K. Petersen
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=20090423184418.GT1926@parisc-linux.org \
--to=matthew@wil.cx \
--cc=James.Bottomley@hansenpartnership.com \
--cc=dgilbert@interlog.com \
--cc=jeff@garzik.org \
--cc=jens.axboe@oracle.com \
--cc=konrad@virtualiron.com \
--cc=linux-ide@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=neilb@suse.de \
--cc=rwheeler@redhat.com \
--cc=snitzer@redhat.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).