[Top] [All Lists]

Bug#661580: mkfs.xfs fails to detect correct sector size

To: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Subject: Bug#661580: mkfs.xfs fails to detect correct sector size
From: Goswin von Brederlow <goswin-v-b@xxxxxx>
Date: Tue, 28 Feb 2012 20:19:54 +0100
Cc: Goswin von Brederlow <goswin-v-b@xxxxxx>, 661580@xxxxxxxxxxxxxxx, Debian Bug Tracking System <submit@xxxxxxxxxxxxxxx>, Carlos Maiolino <cmaiolino@xxxxxxxxxx>
Delivered-to: submit@xxxxxxxxxxxxxxx
In-reply-to: <20120228091151.GA24717@xxxxxxxxxxxxx> (Christoph Hellwig's message of "Tue, 28 Feb 2012 04:11:51 -0500")
References: <20120228074914.31280.78231.reportbug@xxxxxxxxxxxxxxxxx> <20120228091151.GA24717@xxxxxxxxxxxxx>
Reply-to: Goswin von Brederlow <goswin-v-b@xxxxxx>, 661580@xxxxxxxxxxxxxxx
Resent-cc: XFS Development Team <xfs@xxxxxxxxxxx>
Resent-date: Tue, 28 Feb 2012 19:54:02 +0000
Resent-from: Goswin von Brederlow <goswin-v-b@xxxxxx>
Resent-message-id: <handler.661580.B.133045862012203@xxxxxxxxxxxxxxx>
Resent-sender: Debian BTS <debbugs@xxxxxxxxxxxxxxxxx>
Resent-to: debian-bugs-dist@xxxxxxxxxxxxxxxx
User-agent: Gnus/5.110009 (No Gnus v0.9) XEmacs/21.4.22 (linux, no MULE)
Christoph Hellwig <hch@xxxxxxxxxxxxx> writes:

> Carlos, didn't you plan to look into this issue?
> Goswin, how do you determin that mkfs is still doing unaligned I/O
> when forcing the large sevtor size?  Once we set the sector size XFS
> can't do I/O smaller than it.

I'm writing a nbd-server that uses async IO via libaio (the linux kernel
AIO interface). This has some alignment requirements so I set nbd to use
a 4k blocksize and added asserts that all requests are page aligned
(both offset and size of the request). That worked fine for dd and
mkfs.ext4 but fails on mount in general or mkfs.xfs.

So next I added a printf("WARNING: unaligned read/write") and handled
unaligned access in the server to see how extensive the problem is. With
ext4 I get ONE single read request of 512 Byte on mount and everything
else properly 4k aligned. With xfs I get a ton of them.

If you want I can add a log of the read/write requests a mkfs.xfs and
mount generates.


<Prev in Thread] Current Thread [Next in Thread>