Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130716144819.GA9132@openwall.com>
Date: Tue, 16 Jul 2013 18:48:19 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: Parallella: bcrypt

Katja,

Here are some other observations for your current code (as "stolen" from
~kmalvoni/integration/JohnTheRipper/):

Somehow printing of the status line on keypress does not work when
running the bcrypt-parallella format.  Why is that?  Can you fix it?

You're actively polling the Epiphany chip for results, thereby wasting
one ARM core.  Are there good alternatives to this polling approach?

If polling is to stay, we could consider inter-mixing it with
computation on the ARM core... although it's tricky to do well,
especially given that one crypt_all() call should compute a whole number
of bcrypt's (we could have to increase max_keys_per_crypt substantially,
but this has undesirable side-effects on usability).

Alexander

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.