Hi Michael,

Michael McNeil Forbes <[email protected]> writes:
> On Jul 18, 2013, at 10:46 PM, Michael McNeil Forbes 
> <[email protected]> wrote:
>
>> What is the recommended way of preparing ElementwiseKernel instances for 
>> repeated calling on the same GPU arrays for performance?
>
> Here is one attempt… is this headed in the right direction or is there a 
> better/safer way?

I'm not in principle opposed to including such a thing. But I do have
one question: Have you measured that this is really a
performance-limiting issue for you?

Andreas

Attachment: pgp0HUO8T6Peq.pgp
Description: PGP signature

_______________________________________________
PyCUDA mailing list
[email protected]
http://lists.tiker.net/listinfo/pycuda

Reply via email to