[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHC9VhTCafjmo7k=qGnOMDm4s4UHL9g=o5KgRfg9_xwWCzUhcw@mail.gmail.com>
Date: Mon, 3 Apr 2023 15:28:11 -0400
From: Paul Moore <paul@...l-moore.com>
To: Valentin Vidić <vvidic@...entin-vidic.from.hr>
Cc: James Morris <jmorris@...ei.org>,
"Serge E. Hallyn" <serge@...lyn.com>,
Christian Brauner <brauner@...nel.org>,
Dave Chinner <dchinner@...hat.com>,
Casey Schaufler <casey@...aufler-ca.com>,
John Johansen <john.johansen@...onical.com>,
Micah Morton <mortonm@...omium.org>,
Günther Noack <gnoack3000@...il.com>,
Al Viro <viro@...iv.linux.org.uk>,
Mickaël Salaün <mic@...ikod.net>,
linux-kernel@...r.kernel.org,
linux-security-module@...r.kernel.org,
Vivek Goyal <vgoyal@...hat.com>
Subject: Re: [PATCH] security, lsm: security_old_inode_init_security() Handle
multi LSM registration
On Sun, Apr 2, 2023 at 4:03 PM Valentin Vidić
<vvidic@...entin-vidic.from.hr> wrote:
>
> On Sun, Apr 02, 2023 at 11:14:33AM -0400, Paul Moore wrote:
> > If you have the time, could you try a patch we have queued up in the
> > lsm/next branch? We are in the process of removing
> > security_old_inode_init_security() and transitioning all the callers
> > over to security_inode_init_security(), and I believe the ocfs2 patch
> > for this should solve the problem you are seeing, can you test it on
> > your system and let us know?
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/lsm.git/patch/?id=de3004c874e740304cc4f4a83d6200acb511bbda
>
> Great, thanks for the pointer. This patch also works for me as I don't
> see the crash anymore. Can it also be included in the 6.1 LTS kernel
> since this is were I first noticed the problem?
I'm glad that solved your problem, thanks for taking the time to test it out.
I think backporting it to the stable kernels would be okay, but I'd
prefer to let it get some more testing in linux-next first if that's
okay with you. Since we are currently at v6.3-rc5 and this patch is
scheduled to go up to Linus during the next merge window, it might
make the most sense to give this two more weeks in -next, let it land
in Linus' tree, and they ask the stable team for a backport ("Option
2" in the stable kernel docs):
https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
Thoughts?
In addition to the ocfs2 patch mentioned above, there is a similar
reiserfs patch which should probably also be backported:
https://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/lsm.git/patch/?id=52ca4b6435a493e47aaa98e7345e19e1e8710b13
--
paul-moore.com
Powered by blists - more mailing lists