ViperEkura
/
AstrAI
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
AstrAI/csrc/kernels
History
ViperEkura 0654b4b916 refactor: template combine kernel, fix mask bug, unify dispatch 
- Template combine kernel, share macros, extract entry_utils helpers
- Fix mask indexing (pass stride not pre-multiplied base)
- Remove !p.use_mask — MMA handles mask
 		2026-07-15 21:44:17 +08:00
..
attn_common.h refactor: stride-based attn interface with layout and causal mask 2026-07-14 21:34:42 +08:00
attn_decode.cu refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00
attn_decode_split_kv.cuh refactor: stride-based attn interface with layout and causal mask 2026-07-14 21:34:42 +08:00
attn_decode_split_kv_mma.cuh refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00
attn_entry_utils.cuh refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00
attn_mma_utils.cuh refactor: extract load_q_mma_frags template, unify comment style 2026-07-15 19:07:18 +08:00
attn_paged_decode.cu refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00
attn_paged_decode_split_kv.cuh refactor: stride-based attn interface with layout and causal mask 2026-07-14 21:34:42 +08:00
attn_paged_decode_split_kv_mma.cuh refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00
attn_prefill.cu refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00
attn_prefill_split_q.cuh refactor: stride-based attn interface with layout and causal mask 2026-07-14 21:34:42 +08:00
attn_prefill_split_q_mma.cuh refactor: template combine kernel, fix mask bug, unify dispatch 2026-07-15 21:44:17 +08:00