diff --git a/examples/cute/tutorial/sgemm_sm80.cu b/examples/cute/tutorial/sgemm_sm80.cu index 29fb51e5..bd7c6fa4 100644 --- a/examples/cute/tutorial/sgemm_sm80.cu +++ b/examples/cute/tutorial/sgemm_sm80.cu @@ -496,8 +496,8 @@ gemm_nt(int m, int n, int k, size(ceil_div(N, bN))); gemm_device<<>> (prob_shape, cta_tiler, - A, dA, sA, copyA, AutoVectorizingCopy{}, - B, dB, sB, copyB, AutoVectorizingCopy{}, + A, dA, sA, copyA, Copy_Atom{}, + B, dB, sB, copyB, Copy_Atom{}, C, dC, sC, mmaC, alpha, beta); } @@ -573,8 +573,8 @@ gemm_tn(int m, int n, int k, size(ceil_div(N, bN))); gemm_device<<>> (prob_shape, cta_tiler, - A, dA, sA, copyA, AutoVectorizingCopy{}, - B, dB, sB, copyB, AutoVectorizingCopy{}, + A, dA, sA, copyA, Copy_Atom{}, + B, dB, sB, copyB, Copy_Atom{}, C, dC, sC, mmaC, alpha, beta); }