Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 9 Sep 2005 19:16:07 +0400
From: Solar Designer <solar@...nwall.com>
To: john-users@...ts.openwall.com
Subject: Re: 2 Qs on john1.6 (stable)

On Fri, Sep 09, 2005 at 09:27:02AM +0200, Christian Karpp wrote:
> Q1: Why do some bechmarks on 'john -test' simply fail?
> I've compiled john three times on identical ppc machines on AIX 4.3.3, 
> 5.1 and 5.3. The one running on AIX 5.3 fails on BSDI DES (x725), 
> Kerberos AFS DES and NT LM DES. On AIX 4.3.3 and 5.1 all benchmarks run 
> fine. The compiler runs went thru fine on all machines.

This is typically caused by compiler bugs and/or "unsafe" optimizations
(there may be compiler switches to enable/disable those).  In this case,
DES_std.c is the file which must have gotten miscompiled.

You might get around this by reducing the compiler's optimization level
(e.g., replace the -O2 with -O1) for all or for some source files (you
can compile DES_std.c into DES_std.o invoking cc manually).
Alternatively, you may use a John binary off one of the AIX machines
where it gets compiled correctly.

You might also have better luck with current development versions of
John.  (The failure with "NT LM DES" indicates that you're likely using
John 1.6, which didn't have these hashes bitsliced yet.)  Some code in
DES_std.c has been rearranged to avoid certain known miscompiles.

> Q2: Are you aware of any makefile stanzas for SLES9 on OpenPower?

The current development versions of John, which is what you should be
using, offer these make targets:

linux-ppc32              Linux, PowerPC 32-bit
linux-ppc64              Linux, PowerPC 64-bit
linux-ppc32-altivec      Linux, PowerPC w/AltiVec

I would appreciate it if you give these targets a try and let us all
know which have worked in your environment.  Obviously, the AltiVec one
would be the fastest for DES-based hashes.

Thanks,

-- 
Alexander Peslyak <solar at openwall.com>
GPG key ID: B35D3598  fp: 6429 0D7E F130 C13E C929  6447 73C3 A290 B35D 3598
http://www.openwall.com - bringing security into open computing environments

Was I helpful?  Please give your feedback here: http://rate.affero.net/solar

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.