linux-ext4 - Re: remove _supported

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: <20241210130033.GA1839653@mit.edu>
Date: Tue, 10 Dec 2024 08:00:33 -0500
From: "Theodore Ts'o" <tytso@....edu>
To: Christoph Hellwig <hch@....de>
Cc: Zorro Lang <zlang@...nel.org>, Brian Foster <bfoster@...hat.com>,
        fstests@...r.kernel.org, linux-ext4@...r.kernel.org
Subject: Re: remove _supported_fs

On Tue, Dec 10, 2024 at 07:58:24AM +0100, Christoph Hellwig wrote:
> this series removes the remaining _supported_fs calls and replaces them
> with a new _exclude_fs call.
> 
> The first patch removes a _supported_fs for a relatively new test from
> Brian that fails on other file systems.  We should still run it so that
> people have a chance to fix the corruption, so I think this make sense.
> 
> Then the ext4 directory is split so that the shared extN tests have their
> own directory, and then it finally does the switch over now that now many
> _supported_fs calls are left.

Hmm, instead of doing this (would require hard-coding support for ext2
and ext3 file systems needing to use ext-common), why not just have
special-case code which causes ext2 and ext3 file systems to include
the ext4 group, and then we'll have _exclude_fs declaractions as
needed for ext2 and ext3?

After all, ext3 has been removed except for the very oldest LTS
kernels (and I dount anyone is actually testing ext3 using xfstests
these days), and ext2 is not used by most distributions (they use
CONFIG_EXT4_USE_EXT2) and the reason why we've kept it around is that
it's a realtively simple file system that still uses the more modern,
non-legacy vfs/mm interfaces.

So it might not be worth it to move a bunch of tests and creating a
new (somewhat ugly) group, ext4-common, IMO.

Cheers,

						- Ted