From: Ted Ts'o <tytso@mit.edu>
To: Eric Sandeen <sandeen@redhat.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>,
Mike Snitzer <snitzer@redhat.com>,
Jens Axboe <jaxboe@fusionio.com>,
"James.Bottomley@hansenpartnership.com"
<James.Bottomley@hansenpartnership.com>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
"linux-ext4@vger.kernel.org" <linux-ext4@vger.kernel.org>
Subject: Re: I/O topology fixes for big physical block size
Date: Thu, 30 Sep 2010 12:30:47 -0400 [thread overview]
Message-ID: <20100930163047.GA4098@thunk.org> (raw)
In-Reply-To: <4CA25FEA.6040505@redhat.com>
On Tue, Sep 28, 2010 at 04:36:42PM -0500, Eric Sandeen wrote:
> Ok, then it sounds like mkfs.ext4's refusal to make fs blocksize less
> than device physical sectorsize without -F is broken, and that should
> be removed. I'd say issue a warning in the case but if there's a 16k
> physical device maybe there's no point in warning either?
If the device physical sectorsize is that big, should we perhaps use
that as a hint to align writes to that blocks aligned with that
physical sectorsize? Right now we use the optimal I/O size, but if
the optimal I/O size is not specified and the physical sectorsize is,
say, 16k or 32k, maybe we should use to calculate for
s_raid_stripe_width?
- Ted
next prev parent reply other threads:[~2010-09-30 16:30 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-27 16:41 I/O topology fixes for big physical block size Martin K. Petersen
2010-09-27 16:41 ` [PATCH 1/2] block: Ensure physical block size is unsigned int Martin K. Petersen
2010-09-27 17:40 ` Mike Snitzer
2010-10-08 5:15 ` Martin K. Petersen
2010-10-13 19:12 ` Mike Snitzer
2010-10-13 19:15 ` Jens Axboe
2010-09-27 16:41 ` [PATCH 2/2] sd: Fix overflow with big physical blocks Martin K. Petersen
2010-09-27 17:42 ` Mike Snitzer
2010-09-27 18:13 ` [PATCH] block: eliminate potential for infinite loop in blkdev_issue_discard Mike Snitzer
2010-10-14 21:37 ` Mike Snitzer
2010-10-15 11:05 ` Jens Axboe
2010-09-27 16:54 ` I/O topology fixes for big physical block size Jens Axboe
2010-09-27 17:20 ` Martin K. Petersen
2010-09-27 22:21 ` Jens Axboe
2010-09-27 22:36 ` Martin K. Petersen
2010-09-27 23:15 ` Mike Snitzer
2010-09-28 4:30 ` Jens Axboe
2010-09-28 5:20 ` Eric Sandeen
2010-09-28 14:15 ` Mike Snitzer
2010-09-28 20:57 ` Ted Ts'o
2010-09-28 21:24 ` Martin K. Petersen
2010-09-28 21:36 ` Eric Sandeen
2010-09-30 16:30 ` Ted Ts'o [this message]
2010-09-30 17:07 ` Eric Sandeen
[not found] ` <4CA4C3B6.9000104@redhat.com>
2010-09-30 17:33 ` Mike Snitzer
2010-10-01 14:24 ` Ted Ts'o
2010-10-01 22:19 ` Martin K. Petersen
2010-10-02 2:31 ` Ted Ts'o
2010-10-04 19:49 ` Martin K. Petersen
2010-09-27 17:23 ` Mike Snitzer
2010-09-27 21:58 ` James Bottomley
2010-09-27 22:03 ` Jens Axboe
2010-09-27 22:14 ` Martin K. Petersen
2010-09-27 22:24 ` Jens Axboe
2010-09-28 18:48 ` Martin K. Petersen
2010-09-28 18:54 ` Mike Snitzer
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=20100930163047.GA4098@thunk.org \
--to=tytso@mit.edu \
--cc=James.Bottomley@hansenpartnership.com \
--cc=jaxboe@fusionio.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=sandeen@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).