Dnia 2013-04-23, wto o godzinie 14:41 -0400, Andreas Kloeckner pisze:
Tomasz Rybak <tomasz.rybak(a)post.pl> writes:
I've pulled latest git version and built PyCUDA on Debian
unstable. I've tested on two machines - one with Fermi (GTX 460)
and one with ION (9400M). In both cases all tests pass for Python 2.7.3
and only 2 tests fail for Python 3.2.
Failing tests test transfer of data from numpy to PyCUDA GPUArray;
there is some problem with C++ code not recognizing numpy object type
(bytearray vs. bytes).
a) try to build PyCUDA on Python3 and confirm
b) look at those failing tests?
But the good news is that we have (almost) working PyCUDA on Python3!
Now completely fixed in git.
I've tested on Fermi (GTX 460) and Tesla (ION 9400M) on both Python 2
and Python 3 and all tests pass!
Thanks very much for your help (and
pushing!) towards getting PyCUDA to work on Py3!
No problem - I just wanted to have Python 2 and Python 3 packages
for both PyOpenCL and PyCUDA - just for the symmetry ;-)
On the more serious note I intend to upload new packages into
Debian as soon as possible. Unless something very terrible happens
Debian Wheezy will be released in the first weekend of May (2013-05-05).
I assume that after that there will be some transitions (new glibc,
Python 3.3 will become main Python 3 version, and so on). Those
transitions will cause some unrest, but I hope to most of those
to be over in month or two.
Thus I assume that I will upload new versions of packages in the second
half of June, or beginning of July. Will you release 2013.1 by then?
If not this is not a problem - I can upload 2012.1+git* revision,
but it would be nice to have "round" release.
Tomasz Rybak GPG/PGP key ID: 2AD5 9860
Fingerprint A481 824E 7DD3 9C0E C40A 488E C654 FB33 2AD5 9860