Fwd: 3.15: kernel BUG at kernel/auditsc.c:1525!

[Andy Lutomirski]

Steve Grubb pointed out that I forgot to cc: linux-audit.

--Andy


---------- Forwarded message ----------
From: Andy Lutomirski <luto at amacapital.net>
Date: Mon, Jun 16, 2014 at 2:35 PM
Subject: Re: 3.15: kernel BUG at kernel/auditsc.c:1525!
To: Richard Weinberger <richard at nod.at>, "H. Peter Anvin"
<hpa at zytor.com>, X86 ML <x86 at kernel.org>
Cc: Toralf Förster <toralf.foerster at gmx.de>, Eric Paris
<eparis at redhat.com>, Linux Kernel <linux-kernel at vger.kernel.org>


[cc: hpa, x86 list]

On Mon, Jun 16, 2014 at 1:43 PM, Richard Weinberger <richard at nod.at> wrote:
> Am 16.06.2014 22:41, schrieb Toralf Förster:
>> Well, might be the mail:subject should be adapted, b/c the issue can be triggered in a 3.13.11 kernel too.
>> Unfortunately it does not appear within an UML guest, therefore an automated bisecting isn't possible I fear.
>
> You could try KVM. :)

Before you do that, just to clarify:

What bitness is your kernel?  That is, are you on 32-bit or 64-bit kernel?

What bitness is your test case?  'file a.out' will say.

What does /proc/cpuinfo say in flags?

Can you try the attached patch?  It's only compile-tested.

To hpa, etc:  It appears that entry_32.S is missing any call to the
audit exit hook on the badsys path.  If I'm diagnosing this bug report
correctly, this causes OOPSes.

The the world at large: it's increasingly apparent that no one (except
maybe the blackhats) has ever scrutinized the syscall auditing code.
This is two old severe bugs in the code that have probably been there
for a long time.

--Andy

--
Andy Lutomirski
AMA Capital Management, LLC
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-x86_32-entry-Fix-badsys-paths.patch
Type: text/x-patch
Size: 1582 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/linux-audit/attachments/20140619/410fc4ce/attachment.bin>