On Tue, 07 Jun 2011 09:22:25 +0200, Gerald Schweighofer <gerald.schweigho...@alumni.tugraz.at> wrote: > It is now ( since cuda 4.0 ) possible to share global memory across > host threads. > > Does someone know if it is possible to share global memory also across > host processes? > This would be useful in using multiprocessing in python! > Is this somehow supported by pycuda?
Agreed--this would be useful. I don't think there's a possibility for that in CUDA though. If there were, we'd support it. Andreas
pgprMehFgLPbX.pgp
Description: PGP signature
_______________________________________________ PyCUDA mailing list PyCUDA@tiker.net http://lists.tiker.net/listinfo/pycuda