Topic: CUDA
Why not going parallel
cuFFT
cuBLAS
cuSPARSE..?
I wonder if its possible via ISO_C_BINDING
mechanism.
--
Carlos
For discussions of all Approximatrix products and related software
You are not logged in. Please login or register.
Approximatrix Forums → Package Requests → CUDA
Why not going parallel
cuFFT
cuBLAS
cuSPARSE..?
I wonder if its possible via ISO_C_BINDING
mechanism.
--
Carlos
Carlos,
It should be possible to link to the packages you've mentioned, but it would be a bit of a "hack." I say this because Nvidia does not support the GNU compilers on Windows for compiling CUDA applications. The pre-compiled packages you've listed might be usable with MinGW, but it wouldn't be particularly simple.
Using OpenCL rather than CUDA might be a more promising possibility. I'll look into it and post what I find.
Approximatrix Forums → Package Requests → CUDA
Powered by PunBB, supported by Informer Technologies, Inc.