Date: Mon, 25 Jun 2012 21:15:34 +0200 From: magnum <john.magnum@...hmail.com> To: john-dev@...ts.openwall.com Subject: Re: Jumbo candidate vs Test Suite On 2012-06-25 09:58, Lukas Odzioba wrote: > 2012/6/25 magnum<john.magnum@...hmail.com>: >> I'm pretty sure we have had vectorized phpass pass TS before. This is not >> only slower, it also does not work: >> >> $ ../run/john -t -fo:phpass-opencl >> OpenCL platform 0: NVIDIA CUDA, 1 device(s). >> Using device 0: GeForce GTX 570 >> Compilation log: >> ptxas info : Compiling entry function 'phpass' for 'sm_20' >> ptxas info : Function properties for phpass >> 32 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads >> ptxas info : Used 35 registers, 44 bytes cmem >> OpenCL error (CL_OUT_OF_RESOURCES) in file (opencl_phpass_fmt.c) at line >> (165) - (Run kernel) > > This should be easy to fix. I'll try to find first vectorized version on github. Lukas, I played around with this but could not find any vectorized that worked. We now have a working version in the tree that is your older non-vectorized format, hand-edited for new names/labels/etc and shared opencl_find_best_workgroup(). It works fine with GTX570. Please try to get the cuda mscash formats working. From what is committed now, you need to #include "unicode.h" for a starter. I'll have a look and see if I can give you some hints (in case it's just Unicode problems). magnum
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.