Hi Andreas,
The default tests pass on a system with the following configuration:
OS: Fedora 11 x64
GFX: NVidia GeForce GTX 260
Compute Capability: 1.3
Total Memory: 916800 KB
CUDA: Cuda compilation tools, release 3.0, V0.2.1221
Boost: 1.42
Python: 2.6 (default RedHat)
PyCUDA: 0.94rc
Other dependencies to be installed:
1) python package jinja2
2) python-imaging-tk
3) PIL
Aside from out of device memory and lack of proper input file, most of the
tests from the wiki pass. The following tests did not pass and I've
included the output.
Convolution needs to import compiler and use that instead of the driver to
compile the actual kernel code.
What's a typical input for the SparseSolve example?
===Error messages below===
==Convolution==
Convolution.py
Traceback (most recent call last):
File "Convolution.py", line 243, in <module>
module = cuda.SourceModule(code)
AttributeError: 'module' object has no attribute 'SourceModule'
==GL Interop==
GlInterop.py
Traceback (most recent call last):
File "GlInterop.py", line 6, in <module>
from OpenGL.GL import *
ImportError: No module named OpenGL.GL
==Matrix Transform==
MatrixTranspose.py
Traceback (most recent call last):
File "MatrixTranspose.py", line 219, in <module>
run_benchmark()
File "MatrixTranspose.py", line 177, in run_benchmark
target = gpuarray.empty((size, size), dtype=source.dtype)
File
"/usr/lib64/python2.6/site-packages/pycuda-0.94rc-py2.6-linux-x86_64.egg/pycuda/gpuarray.py",
line 81, in __init__
self.gpudata = self.allocator(self.size * self.dtype.itemsize)
pycuda._driver.MemoryError: cuMemAlloc failed: out of memory
==Sparse Solve==
SparseSolve.py
starting...
Traceback (most recent call last):
File "SparseSolve.py", line 86, in <module>
main_cg()
File "SparseSolve.py", line 26, in main_cg
csr_mat = mmread(args[0]).tocsr().astype(numpy.float32)
IndexError: list index out of range
On Mon, Apr 19, 2010 at 5:47 AM, Ian Ozsvald <i...@ianozsvald.com> wrote:
> Hi Andreas. Sorry for the gap, I'm back with my CUDA client now. I'm
> reporting two successes with today's 0.94rc:
>
> OS: Windows XP 32 bit
> GFX: NVIDIA 9800GT
> CUDA: 2.3
> Boost: 1.41 built from source with MS VC 2008 (Visual Studio 9)
> Python: 2.6.4
> Numpy 1.4.0
> Visual Studio 2008 (AKA Visual Studio 9) Team
> pyCUDA: 0.94rc (today's)
>
> OS: Mac OS X Leopard 10.5.8 (32bit, not Snow Leopard)
> GFX: MacBook's NVIDIA 9400M
> CUDA: 2.3
> Boost: 1.41 built from source
> Python: 2.5.1 (Mac default)
> numpy: 1.4
> pyCUDA: 0.94rc (today's)
>
> For testing I've run hello_gpu.py, dump_properties.py and my
> Mandelbrot example (wiki). Building and installing worked first time
> on both.
>
> Now I'm off to do some testing with the complex number support.
>
> Cheers,
> Ian.
>
>
> On 28 March 2010 18:26, Andreas Klöckner <li...@informa.tiker.net> wrote:
> > Hi all,
> >
> > PyCUDA's present release version (0.93) is starting to show its age, and
> > so I've just rolled a release candidate for 0.94, after tying up a few
> > loose ends--such as complete CUDA 3.0 support.
> >
> > Please help make sure 0.94 is solid. Go to
> > http://pypi.python.org/pypi/pycuda/0.94rc
> > to download the package, see if it works for you, and report back.
> >
> > The change log for 0.94 is here:
> > http://documen.tician.de/pycuda/misc.html#version-0-94
> > but the big-ticket things in this release are:
> > - Support for CUDA 3.0
> > - Sparse matrices
> > - Complex numbers
> >
> > Let's make this another another rockin' release!
> >
> > Thanks very much for your help,
> > Andreas
> >
> > _______________________________________________
> > PyCUDA mailing list
> > pyc...@host304.hostmonster.com
> > http://host304.hostmonster.com/mailman/listinfo/pycuda_tiker.net
> >
> >
>
>
>
> --
> Ian Ozsvald (A.I. researcher, screencaster)
> i...@ianozsvald.com
>
> http://IanOzsvald.com
> http://morconsulting.com/
> http://TheScreencastingHandbook.com
> http://ProCasts.co.uk/examples.html
> http://twitter.com/ianozsvald
>
> _______________________________________________
> PyCUDA mailing list
> pyc...@host304.hostmonster.com
> http://host304.hostmonster.com/mailman/listinfo/pycuda_tiker.net
>
--
Paul Rigor
Pre-doctoral BIT Fellow and Graduate Student
Institute for Genomics and Bioinformatics
Donald Bren School of Information and Computer Sciences
University of California, Irvine
http://www.ics.uci.edu/~prigor
_______________________________________________
PyCUDA mailing list
pyc...@host304.hostmonster.com
http://host304.hostmonster.com/mailman/listinfo/pycuda_tiker.net
