[Top] [All Lists]

Re: write 'O_DIRECT' file w/odd amount of data: desirable result?

To: "Linda A. Walsh" <lkml@xxxxxxxxx>
Subject: Re: write 'O_DIRECT' file w/odd amount of data: desirable result?
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Thu, 24 Feb 2011 20:26:25 +1100
Cc: PXXdraig Brady <P@xxxxxxxxxxxxxx>, LKML <linux-kernel@xxxxxxxxxxxxxxx>, xfs-oss <xfs@xxxxxxxxxxx>
In-reply-to: <4D654C2E.2000703@xxxxxxxxx>
References: <4D648D7D.7040500@xxxxxxxxx> <4D64E2BB.7010000@xxxxxxxxxxxxxx> <4D654C2E.2000703@xxxxxxxxx>
User-agent: Mutt/1.5.20 (2009-06-14)
On Wed, Feb 23, 2011 at 10:04:30AM -0800, Linda A. Walsh wrote:
> FWIW -- xfs-oss, included as 'last line' was of minor interest; known bug on
> this kernel?:
> Linux Ishtar #2 SMP PREEMPT Mon Oct 11
> 17:19:41 PDT 2010 x86_64 x86_64 x86_64 GNU/Linux
> Of *SIGNIFICANT* note.  In trying to create an empty file of the size
> used, from scratch, using 'xfs_mkfile', I got an error:
> > xfs_mkfile 5776419696 testfile
> pwrite64: Invalid argument

xfs_mkfile does not create an "empty" file. It creates a file that
is full of zeros.

iAnd you're getting that erro because:

5776419696 / 512 = 11,282,069.7188

the last write is not a multiple of the sector size and xfs_mkfile
uses direct IO. It has always failed when you try to do this. If you
want to create allocated, zeroed files of abitrary size, then use:

xfs_io -f -c "truncate $size" -c "resvsp 0 $size" $filename

to preallocate it. it'll be much, much faster than xfs_mkfile.


Dave Chinner

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