Closed cdeterman closed 6 years ago
Here is an example of how to use custom CUDA kernels: https://github.com/viennacl/viennacl-dev/blob/master/examples/tutorial/custom-cuda.cu
Thanks, as I am thinking about these kernels am I correct that the padding in OpenCL is also applicable to the CUDA memory backend?
yes, padding is needed for the CUDA backend as well.
I am familiar with the means to implement custom OpenCL kernels from the documentation. What about custom CUDA kernels? I know it's not quite as simple but perhaps there is either plans or thoughts on how to accomplish this?