[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 14 Apr 2015 02:17:57 +0200
From: Agnieszka Bielec <bielecagnieszka8@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: [GSoC] John the Ripper support for PHC finalists
I tried this method
http://stackoverflow.com/questions/6563261/how-to-use-coalesced-memory-access
[a@...er run]$ ./john --test --format=pomelo-opencl --dev=5 --t_cost=2
--m_cost=2
Benchmarking: pomelo-opencl, POMELO [POMELO OpenCL (inefficient,
development use only)]... Device 5: GeForce GTX TITAN
get_max_mem_alloc_size=7149472
Local worksize (LWS) 64, global worksize (GWS) 4096
DONE
Raw: 67603 c/s real, 67603 c/s virtual
[a@...er run]$ ./john --test --format=pomelo-opencl --dev=1 --t_cost=2
--m_cost=2
Benchmarking: pomelo-opencl, POMELO [POMELO OpenCL (inefficient,
development use only)]... Device 1: Tahiti [AMD Radeon HD 7900 Series]
get_max_mem_alloc_size=7149472
Local worksize (LWS) 64, global worksize (GWS) 4096
DONE
Raw: 62415 c/s real, 6553K c/s virtual
in POMELO function H is resistant to coalescing
and it doesn't crash for bigger values
but for CPU the speed is equal to ~ 10k c/s real
I uploaded my code to the branch "coalescing"
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.
