Date: Sun, 12 Aug 2012 03:36:29 +0200 From: magnum <john.magnum@...hmail.com> To: john-dev@...ts.openwall.com Subject: Re: pwsafe-gpu On 2012-08-11 22:53, Lukas Odzioba wrote: > 2012/8/11 magnum <john.magnum@...hmail.com>: >> -#define KEYS_PER_CRYPT 1024 >> +#define KEYS_PER_CRYPT 512*112 >> >> magnum@...l:src [1.7.9-jumbo-6-fixes]$ ../run/john -t -fo:pwsafe-opencl >> OpenCL platform 0: NVIDIA CUDA, 1 device(s). >> Using device 0: GeForce GTX 570 >> Benchmarking: Password Safe SHA-256 [OpenCL]... DONE >> Raw: 128862 c/s real, 128862 c/s virtual >> >> I just picked the number used in CUDA - I suppose it can be even better. >> >> magnum > > I'll try to make it faster later, now we have more important formats > that needs tweaking. Sure. > Faster OpenCL code is nothing new for me (cl compiler does better job > here and dummy code is near always faster on OpenCL), after proper > optimizations they should have similar speed. > 40% - you meant memset or w=0 ? The 40% (actually it was >41%) was after applying Solar's patch to pwsafe-cuda, and then making the opencl code (both kernel and fmt) very close to the cuda one (running "meld opencl_pwsafe_fmt.c cuda_pwsafe_fmt.c" and "meld opencl/pwsafe_kernel.cl cuda/pwsafe.cu"). So it was more than just Solar's changes. magnum View attachment "opencl-pwsafe.diff" of type "text/x-patch" (5389 bytes)
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.