john-users - Re: OpenCL performance numbers on Radeon HD 5870

Openwall GNU/*/Linux 3.0 - a small security-enhanced Linux distro for servers

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Thu, 22 Sep 2011 19:42:19 -0400
From: Erik Winkler <ewinkler@...ls.com>
To: john-users@...ts.openwall.com
Subject: Re: OpenCL performance numbers on Radeon HD 5870

Lukas,

To compile your code on MacOS 10.7, I needed to make the following changes:

1) Add the following code anywhere #include <CL/cl.h> exists:

#ifdef __APPLE__
#include <OpenCL/opencl.h>
#else
#include <CL/cl.h>

2) Add the following new build target to the Makefile:

macosx-x86-64-opencl:
	$(LN) x86-64.h arch.h
	$(MAKE) $(PROJ) \
		JOHN_OBJS="$(JOHN_OBJS) x86-64.o" \
		ASFLAGS="$(ASFLAGS) -m64 -DUNDERSCORES -DBSD -DALIGN_LOG" \
		CFLAGS="$(CFLAGS) -m64" \
		LDFLAGS="$(LDFLAGS) -m64 -framework OpenCL"

Feel free to add these changes to your patch.

Erik

On Sep 22, 2011, at 12:02 PM, Lukas Odzioba wrote:

> I uploaded newest version fixing some bugs and faster phpass.
> On phpass@...0 I've got ~300k c/s what is still slow comparing to
> gtx460 with ~245k c/s.
> 
> I had a problem with this revision. Sha256crypt halts my computer
> during .cl compilation. I would apreciate if someone could confirm it.
> 
> It happens only on AMD devices, not on nVidia's.
> Direct link to patch:
> http://tnij.org/jtr-opencl02patch
> 
> Lukas