123456789 |
- // Copyright (c) 2024, Jay Shah, Ganesh Bikshandi, Ying Zhang, Vijay Thakkar, Pradeep Ramani, Tri Dao.
- // Splitting the different head dimensions to different files to speed up compilation.
- #include "flash_fwd_launch_template.h"
- template<>
- void run_mha_fwd_gqa_<cutlass::half_t, 64, 4>(Flash_fwd_params ¶ms, cudaStream_t stream) {
- run_mha_fwd_hdim64_gqa<cutlass::half_t, 4>(params, stream);
- }
|