Re: CVE-2023-52656: io_uring: drop any code related to SCM_RIGHTS

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Sat, May 25, 2024 at 05:09:45PM +0200, Eduardo' Vela" <Nava> wrote:
> On Sat, 25 May 2024, 09:15 Greg Kroah-Hartman, <gregkh@linuxfoundation.org>
> wrote:
> 
> > On Fri, May 24, 2024 at 10:57:07AM -0600, Jens Axboe wrote:
> > > On 5/24/24 10:45 AM, Gabriel Krisman Bertazi wrote:
> > > > Greg Kroah-Hartman <gregkh@linuxfoundation.org> writes:
> > > >
> > > >> Description
> > > >> ===========
> > > >>
> > > >> In the Linux kernel, the following vulnerability has been resolved:
> > > >>
> > > >> io_uring: drop any code related to SCM_RIGHTS
> > > >>
> > > >> This is dead code after we dropped support for passing io_uring fds
> > > >> over SCM_RIGHTS, get rid of it.
> > > >>
> > > >> The Linux kernel CVE team has assigned CVE-2023-52656 to this issue.
> > > >
> > > > Hello Greg,
> > > >
> > > > [+Jens in Cc]
> > > >
> > > > This is stable material, but doesn't deserve CVE status.  There is
> > > > nothing exploitable that is fixed here. Instead, this commit is
> > dropping
> > > > unreachable code after the removal of a feature, following another CVE
> > > > report.  Doing the clean up in the original patch would have made the
> > > > real security fix harder to review.
> > > >
> > > > The real issue was reported as CVE-2023-52654 and handled by a
> > different
> > > > commit.
> > >
> > > FWIW, the same is true for a number of other commits recently. They are
> > > nowhere near CVE material, it's just generic bug fixes.
> >
> > Ok, glad to revoke them if you do not think they are user triggerable
> > issues.  I'll go reject this one right now, thanks.
> >
> 
> Good day!
> 
> So, either I'm completely lost or CVE-2023-52656 shouldn't have been
> rejected. Forgive me for mudding the problem even more.
> 
> I think we need to unreject this CVE (CVE-2023-52656) or CVE-2023-52654
> should be amended to include the dead code removal commit.. that said,
> that'll be weirder than just unrejecting this commit.
> 
> The reason is that the commit "io_uring/af_unix: disable sending io_uring
> over sockets" is not enough to fix the vulnerability in stable branches,
> because e.g. bcedd497b3b4a0be56f3adf7c7542720eced0792 on 5.15 only fixes
> one path (io_sqe_file_register) to reach unix_inflight(), but it is still
> reachable via another path (io_sqe_fileS_register) which is only removed by
> d909d381c3152393421403be4b6435f17a2378b4 ("io_uring: drop any code related
> to SCM_RIGHTS").
> 
> Although that patch claims "it is dead code", this claim was only true on
> upstream, but not on stable branches (or at least on 5.15 where the
> vulnerability was proven to be reachable).
> 
> What a mess! 😄
> 
> My colleague poprdi@google.com sent this analysis to the CNA list, so maybe
> we can continue the discussion there as he also provided some additional
> details there.

Oh yeah, that's right, that's why we issued that!

Jens, any objection for me restoring this CVE?

thanks,

greg k-h