[Top] [All Lists]

Re: [PATCH 2/2] mkfs: proto file creation does not set ftype correctly

To: Eric Sandeen <sandeen@xxxxxxxxxxx>
Subject: Re: [PATCH 2/2] mkfs: proto file creation does not set ftype correctly
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 28 Feb 2014 06:40:37 +1100
Cc: xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <530F4907.5010808@xxxxxxxxxxx>
References: <1393494344-30056-1-git-send-email-david@xxxxxxxxxxxxx> <1393494344-30056-3-git-send-email-david@xxxxxxxxxxxxx> <530F4907.5010808@xxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Feb 27, 2014 at 08:17:43AM -0600, Eric Sandeen wrote:
> On 2/27/14, 3:45 AM, Dave Chinner wrote:
> > From: Dave Chinner <dchinner@xxxxxxxxxx>
> > 
> > Hence running xfs_repair on a ftype enable filesystem that has
> > contents created by a proto file will throw warnings on mismatched
> > ftype entries and correct them. xfs/031 fails due to this problem.
> > 
> > Fix it by settin gup the xname structure with the correct type
> "setting up"
> > fields.
> > 
> > Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
> huh, surprised that hung out there for so long ;)

Just shows how long it took my backlog down far enough to deal with
non-critical xfstests failures on CRC enabled filesystems...

> I guess the root inode is handled by core mkfs code, right?
> (around line 544)

Yes - and the root inode is indexed in a directory (it's indexed by
the superblock), so there's not ftype entry to be set up.

> And I notice that "r" filetypes aren't documented in the
> manpage but that's a different issue.

Patch? ;)

> As long as I'm right about the root inode,
> Reviewed-by: Eric Sandeen <sandeen@xxxxxxxxxx>


Dave Chinner

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