How to configure cache and shared memory on Fermi

Hello,

I’ve been searching about Fermi Architecture and one of the great things about it is that you can configure how much shared memory and cache you want, up to 64kb. Using CUDA, it’s possible to configure it by calling the function cudaFuncSetCacheConfig(), but how do I do that using OpenCL?

Thanks
Samuel

Using CUDA, it’s possible to configure it by calling the function cudaFuncSetCacheConfig(), but how do I do that using OpenCL?

That’s really a question for NVidia. There’s no way to do that with standard OpenCL.

Ok David, thank you.

Sincerely,
Samuel