Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 20 Sep 2012 08:58:54 +0530
From: Sayantan Datta <std2048@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: 1.7.9-jumbo-7

On Thu, Sep 20, 2012 at 8:31 AM, Solar Designer <solar@...nwall.com> wrote:

> On Thu, Sep 20, 2012 at 06:59:19AM +0400, Solar Designer wrote:
> > However, we got two other show-stoppers: mscash2-opencl and
> > sha512crypt-opencl currently fail for me (somewhere between kernel build
> > and start) on NVIDIA.  They work fine on AMD.
> >
> > This problem is not seen in magnum-jumbo.  It is in -fixes only.
>
> $ ../run/john -te -fo=sha512crypt-opencl
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 570
> Building the kernel, this could take a while
> OpenCL error (CL_INVALID_KERNEL_NAME) in file (opencl_cryptsha512_fmt.c)
> at line (480) - (Error creating kernel. Double-check kernel name?)
>
> $ ../run/john -te -fo=mscash2-opencl
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 570
> Create Kernel PBKDF2 FAILED
> Benchmarking: M$ Cache Hash 2 (DCC2) PBKDF2-HMAC-SHA-1 [OpenCL]... OpenCL
> error (CL_INVALID_COMMAND_QUEUE) in file (common_opencl_pbkdf2.c) at line
> (285) - (Copy data to gpu)
>
> Help wanted.  I'll switch to other tasks now.
>
> Alexander
>

Here's the problem. Someone might have added an "inline" specifier to the
kernel PBKDF2 in pbkdf2_kernel.cl. Please remove the "inline" and
everything will be fine.

std2048@...l:~/magnum-jumbo-fixes/run$ ./john -te -fo=mscash2-opencl
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Create Kernel PBKDF2 FAILED
Benchmarking: M$ Cache Hash 2 (DCC2) PBKDF2-HMAC-SHA-1 [OpenCL]... OpenCL
error (CL_INVALID_COMMAND_QUEUE) in file (common_opencl_pbkdf2.c) at line
(285) - (Copy data to gpu)
std2048@...l:~/magnum-jumbo-fixes/run$ ./john -te -fo=mscash2-opencl
OpenCL platform 0: NVIDIA CUDA, 1 device(s).
Using device 0: GeForce GTX 570
Optimal Work Group Size:128
Kernel Execution Speed (Higher is better):0.490985
Benchmarking: M$ Cache Hash 2 (DCC2) PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
Raw:    31794 c/s real, 31755 c/s virtual

Regards,
Sayantan

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.