Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 14 Jul 2012 11:22:30 -0500
From: "jfoug" <jfoug@....net>
To: <john-dev@...ts.openwall.com>
Subject: RE: prepare() and source() are in CVS

>From: Solar Designer [mailto:solar@...nwall.com]
>
>Jim, magnum, all -
>
>I've committed my changes introducing prepare() and source() format
>methods to CVS.  This is John version 1.7.9.5, which only exists as part
>of Owl-current right now.

Would have been nice to wait for a little peer review on this, but that's
your call.

>Please merge these into bleeding.
>
>I intend to proceed experimenting with further formats interface
>changes, but I wanted to eliminate these differences between core and
>jumbo first.
>
>Some of my further changes may be relevant to myrice's work on
>offloading hash comparisons and candidate password generation to GPU.
>Talking to myrice yesterday, I suggested basing such experimental trees
>on bleeding.  Since my further changes will initially be for core rather
>than bleeding, it might help myrice if bleeding's formats interface is
>more similar to core's.

I have followed along on some of these interface changes, and many seem very
good. Others, I simply do not know the 'why' or behind the scene need, so I
will simply watch them being used.

>BTW, bleeding currently lacks source() for LM hashes (I guess since this
>was a bit non-trivial to implement, requiring DES final permutation), 

That was exactly why.  LM would have been good, since there are some larger
lists, but I was uncomfortable with the complexity needed.

>so
>merging the changes from core (which include this) will also be an
>improvement for bleeding.

Agreed.

Jim.

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.