On Fri, May 27, 2011 at 4:00 PM, Anthony LaTorre <tlatorre9(a)gmail.com> wrote:
Is it possible to allocate, copy, and bind memory to a layered
pycuda? If so, is there any example code?
PyCUDA mailing list
I'm new here but I'll venture an answer. I believe there is a way to
do it using a CUDA array. CUDA arrays are discussed here:
However, I don't know of any code examples. I'm not very well versed
in the docs, so others please correct me if I'm wrong. The nVidia docs
cover the CUDA Driver interface and would probably give you a good
idea of what the Python bindings do since they are a pretty thin
wrapper. See the CUDA C Programming Guide Version 4.0, page 69.
Sent via my carrier pigeon.