On Tue, 22 Feb 2011 16:06:14 +0530, nithin s <nithin19484(a)gmail.com> wrote:
is there any way to interpret a gpuarray's dtype as a
understand that it tied to numpy.dtype. But I also saw code in
pycuda/tools.py that interprets pycuda::complex<float> .
That said, you can of course fake something like it by having an array
with a minor dimension of 4. (float3 is 4-float-aligned IIRC)