On Wed, Feb 06, 2013, Brad House wrote:

> On 02/06/2013 12:30 PM, Brad House wrote:
> ...<snip>...
> > I should also note that currently I am using
> >OpenSSL 1.0.1d-fips 5 Feb 2013
> >On an Intel(R) Core(TM) i7-3770S CPU @ 3.10GHz running Ubuntu 12.04 64bit
> >(so presumably I'm using AES-NI ... noticed changes to that in the
> >changelog).
> >
> >I have not yet tried to compile it in non-FIPS mode to rule
> >that out, but I am not running it in an active FIPS mode.  I
> >have reproduced this issue on Linux x64 and Windows x86 thus far,
> >I haven't tested it on any other system.
> ...<snip>...
> 
> I have reproduced the issue with a stock build as well following
> the procedure below.  The stock build does seem to have a couple
> of behavioral differences.  First, I can reproduce the issue
> now on the server side without specifying the cipher as long as
> the client side does specify the cipher suite.  Next, I can only
> get the actual error message to be reported when running under
> valgrind... however, the corruption is ALWAYS there when not
> running under valgrind.
> 
> I can see it exhibited by a random character before the word
> "DONE" on the server side after my "Hello World" is printed out
> that was sent from the client side.  Using older versions did not
> exhibit this behavior.
> 
> Perhaps the issue is in the  ECDHE-RSA-AES256-SHA cipher suite
> which is being chosen... when it uses ECDHE-RSA-AES256-GCM-SHA384
> when no cipher suite is specified, everything is OK (e.g. no
> valgrind errors and no random character).
> 

A possibility is the AESNI+SHA1 stitched code which is handled as a special
case. You'd only see that with AES+SHA1 ciphersuites on AES-NI supporting
processors.

Try disabling AES-NI with OPENSSL_ia32cap=~0x200000200000000 also try entering
FIPS mode for a FIPS build with OPENSSL_FIPS=1.

Finally you could try reverting the last changes to e_aes_cbc_hmac_sha1.c for
test purposes: note this will also make you vulnerable to CVE-2013-0169

Steve.
--
Dr Stephen N. Henson. OpenSSL project core developer.
Commercial tech support now available see: http://www.openssl.org
______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       openssl-dev@openssl.org
Automated List Manager                           majord...@openssl.org

Reply via email to