linux-kernel - Re: [PATCH] selinux: Mark array 'labeling_behaviors' as __maybe

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] [day] [month] [year] [list]

Message-ID: <20170519165259.GJ141096@google.com>
Date:   Fri, 19 May 2017 09:52:59 -0700
From:   Matthias Kaehlcke <mka@...omium.org>
To:     Stephen Smalley <sds@...ho.nsa.gov>
Cc:     Paul Moore <paul@...l-moore.com>,
        James Morris <james.l.morris@...cle.com>,
        linux-kernel@...r.kernel.org,
        linux-security-module@...r.kernel.org, selinux@...ho.nsa.gov
Subject: Re: [PATCH] selinux: Mark array 'labeling_behaviors' as
 __maybe_unused

El Fri, May 19, 2017 at 11:45:19AM -0400 Stephen Smalley ha dit:

> On Fri, 2017-05-19 at 11:09 -0400, Paul Moore wrote:
> > On Thu, May 18, 2017 at 3:07 PM, Matthias Kaehlcke <mka@...omium.org>
> > wrote:
> > > The array is only referenced in an ARRAY_SIZE() statement. Adding
> > > the
> > > attribute fixes the following warning when building with clang:
> > > 
> > > security/selinux/hooks.c:338:20: error: variable
> > > 'labeling_behaviors'
> > >     is not needed and will not be emitted
> > >     [-Werror,-Wunneeded-internal-declaration]
> > > 
> > > Signed-off-by: Matthias Kaehlcke <mka@...omium.org>
> > > ---
> > >  security/selinux/hooks.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > The fact that we only reference labeling_behaviors in one spot, and
> > even then we only use it as a parameter to ARRAY_SIZE(), makes me
> > believe we may be able to get rid of labeling_behaviors and use
> > SECURITY_FS_USE_MAX in its place.
> > 
> > Anyone working on any patches which make use of labeling_behaviors?
> 
> I think you could just remove both the array and the code that
> referenced it; it only made sense before commit
> 2088d60e3b2f53d0c9590a0202eeff85b288b1eb.  We already check that the
> policy doesn't contain any behavior > SECURITY_FS_USE_MAX during policy
> load, so this cannot occur (modulo memory corruption), and it was only
> there to make sure we didn't try to dereference off the end of the
> array prior to the aforementioned commit.

Thanks for your comments.

I will send out a patch that removes the array shortly.

Matthias