CUTLASS 2.1 (#83)

CUTLASS 2.1 contributes:
- BLAS-style host-side API added to CUTLASS Library
- Planar Complex GEMM kernels targeting Volta and Turing Tensor Cores
- Minor enhancements and bug fixes

include/cutlass/layout/tensor.h

@@ -33,7 +33,11 @@
     defined in cutlass/tensor_ref.h.
 */
 #pragma once
+#if defined(__CUDACC_RTC__)
+#include <cuda/std/cassert>
+#else
 #include "assert.h"
+#endif
 #include "cutlass/cutlass.h"
 #include "cutlass/fast_math.h"
 #include "cutlass/layout/matrix.h"