What is your question?
I would like to ask if there is support for uint8_t or int8_t matrix calculations in local_join?
I get an error when using the following int8_t. And equipment ：
NVIDIA GeForce RTX 4090 Driver Version: 560.28.03 CUDA Version: 12.6

// wmma::fragment<wmma::matrix_a, WMMA_M, WMMA_N, WMMA_K, half, wmma::row_major> a_frag;
// wmma::fragment<wmma::matrix_b, WMMA_M, WMMA_N, WMMA_K, half, wmma::col_major> b_frag;

wmma::fragment<wmma::matrix_a, WMMA_M, WMMA_N, WMMA_K, int8_t, wmma::row_major> a_frag;
wmma::fragment<wmma::matrix_b, WMMA_M, WMMA_N, WMMA_K, int8_t, wmma::col_major> b_frag;