Thanks Andreas. Yes, I'd been updating from git occasionally, seeing some changes.
It has definitely improved. I updated my web page of some testing
) and hope to find some time to do more debugging
of the failure.
On Oct 28, 2011, at 2:08 PM, Andreas Kloeckner wrote:
On Tue, 30 Aug 2011 10:02:01 -0400, Randy Heiland <heiland(a)indiana.edu> wrote:
> Fwiw, I posted the outputs from the 'test_array.py' script at
> I realize the failures (and fixes) that occur here may be due to Andreas's lack
of having an OSX platform to test on, so I don't intend for this post to be a
nuisance. I'm simply curious what people are getting as results (from tests &
examples) on various platforms and thought I'd offer up some from OSX.
> Btw, if there is some guidance on debugging pyopencl, I'd certainly
> welcome it and try to contribute.
Sorry for the long delay on this. PyOpenCL's current git version should
pass all tests on OS X Lion except one, and that one I find truly
mystifying. It's the test for the random number generator (RANLUXCL)
(test_array.py, test_random) running on the CPU. In particular in the
initialization part, the state array gets filled almost exclusively with
zeros--which shouldn't be happening.
Any help debugging this would be much appreciated!