linux-ext4 - Re: xfstests #62 broken on ext4

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20110813201419.GA6442@infradead.org>
Date:	Sat, 13 Aug 2011 16:14:19 -0400
From:	Christoph Hellwig <hch@...radead.org>
To:	Theodore Ts'o <tytso@....edu>
Cc:	linux-ext4@...r.kernel.org, xfs@....sgi.com,
	Stefan Behrens <sbehrens@...ntdisaster.de>
Subject: Re: xfstests #62 broken on ext4

On Sat, Aug 13, 2011 at 10:38:51AM -0400, Theodore Ts'o wrote:
> 
> Xfstests #62 was recently enabled by commit b2b36d0a4.
> 
> However, this test is failing for ext4 because ext4 doesn't support
> extended attributes on anything other than regular files and
> directories.  This is behavior is documented in the attr(5) man page:

As explained to you before that is not the reason why it fails.  The
reason is that Andreas Gruenbacher changed return values for certain
xattr operations in commit 55b23bde19c08f14127a27d461a4e079942c7258.

> It looks like XFS is not conformant to documented behaviour and the file
> system generic code.

Care to explain that conclusion?  A simple test and/or a look at the
code will reveal that this is not true.

Below is the patch from Andreas that makes the test work properly again
on Linux 3.0+ on XFS, although other filesystem still have some issues
with it.  For ext4 that mostly seems the presence of the lost+found
directory, and a different ordering from readdir than expected.

View attachment "2.diff" of type "text/plain" (7387 bytes)