flash-attention/hopper/flash_fwd_hdim256_fp16_gqa2_sm90.cu

10 lines
405 B
Plaintext

// 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, 256, 2>(Flash_fwd_params &params, cudaStream_t stream) {
run_mha_fwd_hdim256_gqa<cutlass::half_t, 2>(params, stream);
}