Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Sun, 3 Apr 2016 16:44:57 +0200
From: Matus UHLAR - fantomas <uhlar@...tomas.sk>
To: john-users@...ts.openwall.com
Subject: sha512crypt-opencl gives CL_INVALID_COMMAND_QUEUE

Hello,

I have tried to benchmark opencl performance of sha512crypt.

The cuda was faster than my CPU at least in 1.8.0-jumbo1, but recommends
opencl that should be even faster (for md5crypt, it is).

However the sha512crypt-opencl procudes errors, I was unable to help with
LWS and GWS settings.
Any idea what can be wrong? the standard output of git clone below:

% ./john -test -format=sha512crypt-opencl
Device 0: GeForce GT 610
Benchmarking: sha512crypt-opencl, crypt(3) $6$ (rounds=5000) [SHA512 OpenCL]... Build log: 
ptxas info    : 0 bytes gmem, 920 bytes cmem[2]
ptxas info    : Compiling entry function 'kernel_crypt_fast' for 'sm_21'
ptxas info    : Function properties for kernel_crypt_fast
ptxas         .     272 bytes stack frame, 32 bytes spill stores, 20 bytes spill loads
ptxas info    : Used 63 registers, 48 bytes cmem[0], 52 bytes cmem[16]
ptxas info    : Compiling entry function 'kernel_preprocess' for 'sm_21'
ptxas info    : Function properties for kernel_preprocess
ptxas         .     0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 48 bytes cmem[0], 8 bytes cmem[16]
ptxas info    : Compiling entry function 'kernel_crypt_full' for 'sm_21'
ptxas info    : Function properties for kernel_crypt_full
ptxas         .     328 bytes stack frame, 160 bytes spill stores, 84 bytes spill loads
ptxas info    : Used 63 registers, 48 bytes cmem[0], 52 bytes cmem[16]
ptxas info    : Compiling entry function 'kernel_prepare' for 'sm_21'
ptxas info    : Function properties for kernel_prepare
ptxas         .     1856 bytes stack frame, 16 bytes spill stores, 24 bytes spill loads
ptxas info    : Used 63 registers, 44 bytes cmem[0], 24 bytes cmem[16]
ptxas info    : Compiling entry function 'kernel_final' for 'sm_21'
ptxas info    : Function properties for kernel_final
ptxas         .     328 bytes stack frame, 140 bytes spill stores, 72 bytes spill loads
ptxas info    : Used 63 registers, 48 bytes cmem[0], 52 bytes cmem[16]
OpenCL CL_INVALID_COMMAND_QUEUE error in opencl_cryptsha512_fmt_plug.c:632 - Error running loop kernel


-- 
Matus UHLAR - fantomas, uhlar@...tomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Christian Science Programming: "Let God Debug It!".

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.