Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Wed, 8 May 2013 13:18:49 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Cisco - Password type 4 - SHA256

On 18 Mar, 2013, at 19:18 , magnum <john.magnum@...hmail.com> wrote:
> On 18 Mar, 2013, at 16:34 , Vlatko Kosturjak <kost@...ux.hr> wrote:
>> On Mon, Mar 18, 2013 at 10:08:17AM -0300, Claudio André wrote:
>>> Hi, if you are having problems looking at previous patches (above), i  
>>> attach a new one (easier to deal but might harder to follow/understand).
>>> 
>>> Notice: it contains part of commit (7ac5b370) and i'm not sure if it is  
>>> the right thin
>> 
>> I thought we gave up of this solution of having separate format? 
> 
> I think we only gave up the idea to add a completely separate new format. The version Claudio posted now adds Cisco support to existing raw-sha256-opencl source file, so only a few functions differs and there are two format structs. While it is technically a separate format, it's not a new source file (and I presume it uses the same OpenCL kernel too).
> 
> Just thinking out loud, I wonder how hard it would be to write a single format (as in using *one* format struct) that reads hex or base64 hashes, and writes pot file entries in the format they came - while still recognizing cracked hashes from the "other" encoding. Maybe it would just get messy.


That was easy, using prepare().

I just added (in bleeding) support for native Cisco type 4 hashes (tagged or not) to the existing raw-sha256 format. The .pot entries are compatible with normal, hex-encoded, raw-sha256 hashes.

We should do the same to OpenCL and ditch the (semi-)separate format.

magnum

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ