Re: CVE request: kernel: buffer overflow and DoS issues in agp

[akuster <akuster () mvista com>]

I am a bit confused.

https://bugzilla.redhat.com/show_bug.cgi?id=698999 references
https://lkml.org/lkml/2011/4/14/294

 which is assigned to CVE-2011-1746 not CVE-2011-1747.

is there a patch for CVE-2011-1747?

- Armin

On 04/22/2011 05:32 AM, Vasiliy Kulikov wrote:
> On Fri, Apr 22, 2011 at 11:11 -0400, Petr Matousek wrote:
> > > Another problem in agp code is not addressed in the patch - kernel
> > > memory
> > > exhaustion (AGPIOC_RESERVE and AGPIOC_ALLOCATE ioctls). It is not
> > > checked
> > > whether requested pid is a pid of the caller (no check in
> > > agpioc_reserve_wrap()).
> > > Each allocation is limited to 16KB, though, there is no per-process
> > > limit.
> > > This might lead to OOM situation, which is not even solved in case of
> > > the
> > > caller death by OOM killer - the memory is allocated for another
> > > (faked)
> > > process."
> >
> > Please use CVE-2011-1747.
>
> In https://bugzilla.redhat.com/show_bug.cgi?id=698999 it is said
> "Reference and patch:", but there is no patch for the issue (as I said
> in the patch description).  I have no agp hardware and I cannot test
> whether forcing the requested pid to the current pid is a good idea (it
> might not).
>
> Thanks,