Open dennisai opened 6 years ago
Hi, I looked through the CuBLAS API and discovered a method called gemmStridedBatched.
https://docs.nvidia.com/cuda/cublas/#cublas-lt-t-gt-gemmstridedbatched https://devblogs.nvidia.com/cublas-strided-batched-matrix-multiply/
There does not appear to be a wrapper of this method in ManagedCuda, is this something that can be supported in the future?
It seems that I missed that one... It's only implemented in the native methods part. I'll add it to CudaBlasHandler.cs. Anything else missing?
Cheers, Michael
I see, nope, not that I'm aware of right now!
Hi, I looked through the CuBLAS API and discovered a method called gemmStridedBatched.
https://docs.nvidia.com/cuda/cublas/#cublas-lt-t-gt-gemmstridedbatched https://devblogs.nvidia.com/cublas-strided-batched-matrix-multiply/
There does not appear to be a wrapper of this method in ManagedCuda, is this something that can be supported in the future?