cuda-api-wrappers
cuda-api-wrappers copied to clipboard
Support asynchronous memory allocation
CUDA 11.2 added asynchronous memory allocation and de-allocation. Let's support that.
API description: here.
cudaError_t cudaFreeAsync ( void* devPtr, cudaStream_t hStream );
cudaError_t cudaMallocAsync ( void** devPtr, size_t size, cudaStream_t hStream );
Note: Need some test/example program coverage...
Not actually fixed... we had only added allocation, not deallocation :-(