cuda_mm_cm() [2/4]

void syten::CudaDenseTensorImpl::cuda_mm_cm	(	CudaPtr< const float >	at,
		CudaPtr< const float >	bt,
		CudaPtr< float >	rt,
		Size const	dx,
		Size const	dy,
		Size const	dz
	)

CUDA real float matrix-matrix multiplication.