[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 09 Dec 2014 11:18:53 -0500
From: Steve Grubb <sgrubb@...hat.com>
To: oss-security@...ts.openwall.com
Cc: Hanno Böck <hanno@...eck.de>
Subject: Re: Offset2lib: bypassing full ASLR on 64bit Linux
On Friday, December 05, 2014 01:54:35 PM Hanno Böck wrote:
> On Thu, 04 Dec 2014 21:19:04 +0100
>
> Hector Marco <hecmargi@....es> wrote:
> > This is a disclosure of a weakness of the ASLR Linux implementation.
> > The problem appears when the executable is PIE compiled and it has an
> > address leak belonging to the executable. We named this weakness:
> > offset2lib.
>
> Thanks for that.
>
> Two things on that:
>
> Cynics might say "most linux distros aren't vulnerable to ASLR bypass
> because they don't use ASLR at all".
>
> Can we please take this as an opportunity to discuss the state of ASLR
> on Linux in general?
I studied this area 2 years ago for a gray hat talk and in preparation to help
set the policy going forward for Fedora and RHEL. The general reason I've
heard mentioned about why its not used as fully as possible is that it adds
memory pages that can't be coalesced or consolidated because they are not the
same.
For Fedora and RHEL 7, the intended policy is PIE for all daemons, privileged
apps, network facing, and parsers of untrusted media. Enforcement is not so
easy. How do you identify in an automated way parsers of untrusted media? I
have a script that can grade an installed system that uses rpms:
http://people.redhat.com/sgrubb/files/rpm-chksec
It has options to grade the system or an individual rpm for compliance with
the intended policy.
During my research, I found a couple interesting things. These ASLR related
tidbits below are pulled from the speech I gave about when open source
security mechanisms don't work as intended (all measured on a 64 bit system):
1) On non-PIE applications, the heap doesn't get much randomization. Just 14
bits.
2) Also non-PIE applications seems to be some bias in the numbers chosen.
This could be an effect of 14 bits of randomization. It did follow a bell curve
such that guessing some addresses was much luckier than others. (I did not get
a sample size large enough on PIE apps to see if the same bias could be
measured.)
3) When using PIE, you pretty much got 29 bits of randomness everywhere. That
lead to the question of why the heap on non-PIE is so limited in address
scope. As I remember, there was some coupling with sbrk() that caused
this....which might need revisiting.
4) Then I started wondering about the heap when you use other memory manager
libraries such as jemalloc. This turned out to be interesting. You get about
19 bits of randomness using it. Its not as bad as non-PIE glibc but not as
good as PIE glibc. You also got the same amount of randomness whether the app
was PIE or not. This is an area ripe for more experimenting, exploiting, and
patching. Supposedly some of these heap managers use mmap as the underlying
allocator. So, why aren't they getting 29 bits, too? :-)
Here's the current numbers from 3.17.4 kernel + glibc-2.18:
$ ./all-bits
heap 14 bits
exec No randomization
mmap 29 bits
so No randomization
stack 28 bits
pie-exec 29 bits
pie-heap 29 bits
pie-so 29 bits
$ ./all-mask
heap 0x0000000003FFF000
exec 0x0000000000000000
mmap 0x000001FFFFFFF000
so 0x0000000000000000
stack 0x00000000FFFFFFF0
pie-exec 0x000001FFFFFFF000
pie-heap 0x000001FFFFFFF000
pie-so 0x000001FFFFFFF000
-Steve
> It's pretty sad, afaik Linux was one of the first
> to have ASLR (in the form of pax) back in 2001. Today everyone uses
> ASLR by default except Linux.
>
> Most distros don't ship pic/pie executables by default. Why? I haven't
> done benchmarks, the saying is that this has a notable performance hit
> on 32 bit but almost none on 64 bit. If this is true then could we at
> least have all major distros enable it on 64 bit?
>
>
> Second:
> I wrote a small test .c to print out offset diffs. As expected
> printf-main offset is static on normal Linux with pic/pie and random
> on a pax-enabled system.
>
> What i found notable: diff-ing two function offsets from different
> libraries (I use printf-sin) is alway static, even on Pax. Is this by
> design? Can't different libraries be loaded at different offsets in ram?
Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)
Powered by blists - more mailing lists
Please check out the
Open Source Software Security Wiki, which is counterpart to this
mailing list.
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
