|
Message-ID: <4FA2E7E8.8010109@gmail.com>
Date: Thu, 03 May 2012 17:17:44 -0300
From: Claudio André <claudioandre.br@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: cryptsha512-opencl (plans)
Em 03-05-2012 16:21, magnum escreveu:
> Strange. Do we know if all of your older versions behaved the same?
> Maybe we can dissect. It even works on AMD CPU, right? I think I
> recall it does. magnum
The test i did worked fine on my X6 CPU, on bull's CPU, on GTX 570 and
on Radeon HD 6770. Only failed on 7970.
I can create a version that compiles fine on 7970, but with some kind of
strange behavior.
(i just selected this example because it shows that something is not good).
Now the situation (commited at
https://github.com/claudioandre/magnum-jumbo):
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
=> Using device 0: GeForce GTX 570
Local work size (LWS) 32, Keys per crypt (KPC) 7168
Benchmarking: crypt SHA-512 (rounds=5000) [OpenCL]... DONE
Raw: 3592 c/s real, 3584 c/s virtual
OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
=> Using device 1: AMD FX(tm)-8120 Eight-Core Processor
Local work size (LWS) 1, Keys per crypt (KPC) 1024
Benchmarking: crypt SHA-512 (rounds=5000) [OpenCL]... DONE
Raw: 1600 c/s real, 200 c/s virtual
OpenCL platform 0: AMD Accelerated Parallel Processing, 2 device(s).
=> Using device 1: AMD Phenom(tm) II X6 1075T Processor
Local work size (LWS) 1, Keys per crypt (KPC) 768
Benchmarking: crypt SHA-512 (rounds=5000) [OpenCL]... DONE
Raw: 1725 c/s real, 290 c/s virtual
OpenCL platform 0: AMD Accelerated Parallel Processing, 2 device(s).
=> Using device 0: Juniper
Local work size (LWS) 32, Keys per crypt (KPC) 5120
Benchmarking: crypt SHA-512 (rounds=5000) [OpenCL]... DONE
Raw: 1385 c/s real, 512000 c/s virtual
Claudio.
Content of type "text/html" skipped
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.