Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 26 May 2015 00:43:09 +0300
From: Solar Designer <>
Subject: Re: Agnieszka's weekly report #4

On Mon, May 25, 2015 at 10:57:49PM +0200, Agnieszka Bielec wrote:
> speed for parallel-opencl:
> [ run]$ ./john --test --format=parallel-opencl  --dev=1 --cost=0:0
> Device 1: Tahiti [AMD Radeon HD 7900 Series]
> Building the kernel, this could take a while
> Build log: LOOP UNROLL: pragma unroll (line 106)
>     Unrolled as requested!
> LOOP UNROLL: pragma unroll (line 87)
>     Unrolled as requested!
> Benchmarking: parallel-opencl, parallel SHA-512 [ ]... DONE
> Speed for cost 1 (N) of 0
> Many salts:     45936 c/s real, 4915K c/s virtual
> Only one salt:  46369 c/s real, 4915K c/s virtual

How do these speeds compare to CPU?  Your current CPU implementation is
still SIMD-less, correct?

"parallel SHA-512 [ ]" and "Speed for cost 1 (N)" look weird.

magnum, what's jumbo's current convention on format names (not labels)
and on algorithm names?  What should be reported in this case?

Agnieszka, I guess calling Parallel's cost parameter "N" got inherited
from scrypt or something.  Would you rename it?  In fact, would you
perhaps have JtR report Parallel's parallel and sequential (aka upgrade)
costs as two separate cost parameters?


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.