Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 31 Aug 2012 12:28:05 +0200
From: Camille Mougey <commial@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Benchmark format error

2012/8/30 magnum <john.magnum@...hmail.com>

> On 30 Aug, 2012, at 15:06 , Camille Mougey <commial@...il.com> wrote:
>
> > Indeed, it doesn't still work. Surprisingly, the hash is cracked when it
> is with nine others, and still non guessed (but detected) with hundred or
> thousand others.
> > However, the only function able to compute the hash part is crypt_all.
> What's happened?
>
> Just spray debug output over crypt_all() and the get_hash*() functions and
> you will probably spot the problem in no time.
>
> In Jumbo, we have a couple prefab debug functions in memory.c:
>
> #include memory.h
> ...
> dump_stuff_msg("crypt_all result", binary, 16);
>
> This will dump 16 bytes of hex output from binary. Do this right before
> crypt_all() returns, and a similar print of the exact value used by
> get_hash_2(), for a starter.
>
> magnum
>

Thanks to your help, I've spot the bug. I had misunderstood the relations
between get_hash* and cmp_*. Actually, the default format function fix the
problem :).

I hope I'll be able to put online the patch in further days.

Have a nice day,
Camille

Content of type "text/html" skipped

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.