D | TensorDeviceGpu.h | 359 #define LAUNCH_GPU_KERNEL(kernel, gridsize, blocksize, sharedmem, device, ...) \ argument 360 …hipLaunchKernelGGL(kernel, dim3(gridsize), dim3(blocksize), (sharedmem), (device).stream(), __VA_A… 365 #define LAUNCH_GPU_KERNEL(kernel, gridsize, blocksize, sharedmem, device, ...) \ argument 366 (kernel) <<< (gridsize), (blocksize), (sharedmem), (device).stream() >>> (__VA_ARGS__); \
|