Home
last modified time | relevance | path

Searched defs:gmem_thr_copy_QKV (Results 1 – 2 of 2) sorted by relevance

/external/pytorch/aten/src/ATen/native/transformers/cuda/flash_attn/
Dflash_fwd_kernel.h152 auto gmem_thr_copy_QKV = gmem_tiled_copy_QKV.get_thread_slice(tidx); in compute_attn_1rowblock() local
598 auto gmem_thr_copy_QKV = gmem_tiled_copy_QKV.get_thread_slice(tidx); in compute_attn_1rowblock_splitkv() local
Dflash_bwd_kernel.h178 auto gmem_thr_copy_QKV = gmem_tiled_copy_QKV.get_thread_slice(tidx); in compute_dq_dk_dv_1colblock() local