[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 26 Jun 2012 03:32:27 +0200
From: Lukas Odzioba <lukas.odzioba@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: phpass OpenCL and CUDA
2012/6/26 Solar Designer <solar@...nwall.com>:
> Lukas -
>
> The reverted opencl/phpass_kernel.cl currently in magnum-jumbo looks a
> bit dirtier than the newer version did. Perhaps you can re-apply some
> minor changes to it, without going all the way for a vectorized
> implementation yet (so that you don't reintroduce whatever problem that
> had)? Specifically:
>
> 1. There's a function called "cuda_md5" - rename it.
>
> 2. Make use of rotate() and bitselect().
I'll send fix for that in a moment. My target is to make vectorized
code work properly so even if code is dirty it should not stay there
for long.
> Are we releasing with these lower speeds?
Yes, there were no any major changes. Lower speeds are because test
vector change.
Previously we're testing just one password, now 15 passwords with
increasing length.
Lukas
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.
