Bug#546919: ITP: python-pycuda -- module to access Nvidia‘s CUDA parallel computation API

[Yaroslav Halchenko <debian@onerussian.com>]

Package: wnpp
Severity: wishlist
Owner: Yaroslav Halchenko <debian@onerussian.com>


* Package name    : python-pycuda
  Version         : 0.93
  Upstream Author : Andreas Kloeckner <inform@tiker.net>
* URL             : http://mathema.tician.de/software/pycuda
* License         : MIT
  Programming Lang: C++, Python
  Description     : module to access Nvidia‘s CUDA parallel computation API

PyCUDA lets you access Nvidia‘s CUDA parallel computation API from
Python. Several wrappers of the CUDA API already exist–so what’s so special
about PyCUDA?

 * Object cleanup tied to lifetime of objects. This idiom, often called
   RAII in C++, makes it much easier to write correct, leak- and crash-free code.
   PyCUDA knows about dependencies, too, so (for example) it won’t detach from a
   context before all memory allocated in it is also freed.

 * Convenience. Abstractions like pycuda.driver.SourceModule and
   pycuda.gpuarray.GPUArray make CUDA programming even more convenient than with
   Nvidia’s C-based runtime.

 * Completeness. PyCUDA puts the full power of CUDA’s driver API at your
   disposal, if you wish.

 * Automatic Error Checking. All CUDA errors are automatically translated
   into Python exceptions.

 * Speed. PyCUDA’s base layer is written in C++, so all the niceties
   above are virtually free.

 * Helpful Documentation.

To be submitted into contrib section due to dependency on nvcc (non-free, not
yet in Debian either).

Additional things needed to be package prior:
 http://pypi.python.org/pypi/pytools  -- MIT license