Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Sun, 5 Feb 2012 09:42:29 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: Bit slice implementation of DES based hashes

On Sat, Feb 04, 2012 at 08:56:44PM +0530, Piyush Mittal wrote:
> Moving to 1.7.9 is not a big issue that I can move later easily.

Yes, but like I said postponing this might mean that you're currently
running into problems that you would not even have.

Since you don't yet have any new code that you'd actually use (all you
have are some unsuccessful tests), I think starting from scratch based
on 1.7.9-jumbo-5 is the best option.

> Whole implementation is almost on my hand. According to oracle hashes we
> need to apply DES encryption two times so it will be more appropriate if I
> check result of first encryption if it is coming right or not?

While you're debugging, sure.

> > Are you only testing the "Key values.." printout from DES_bs_crypt_Oracle()?
> 
> Yes that is what I want first because I am not getting correct sequence of
> 768 bits (you also please don't consider other things now). And in order to
> check these round key values I have taken an example from Behrouz A.
> Forouzan. The key that I am using after parity drop according to the given
> example is as follows:
> 
>    key="\xAB\x74\x20\xC2\x66\xF3\x6E";

Like I said, I have no idea if the information given in that book is
correct or not and if you're interpreting it correctly or not.  Instead,
it might be better for you to compare the result of the first DES
encryption as you say.

> You can see this example from the links given below:
> 
> https://picasaweb.google.com/104472102705685602064/February42012#5705299936162326818
> https://picasaweb.google.com/104472102705685602064/February42012#5705300078962977538

Somehow these give me "Sorry, that page was not found."  I guess they
were meant to be pages from the book or something; I'd rather not spend
time verifying and analyzing them anyway.

Alexander

Powered by blists - more mailing lists

Your e-mail address:

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