ViperEkura 60d7ee614a fix: improve attention kernel numerical stability and test precision checks ...

- use fmaf() for V-accumulation in scalar decode paths to reduce rounding
- delay scale multiplication to after dot-product in scalar prefill
- unify __expf/expf across MMA and scalar paths for consistent numerics
- harmonize divide-by-zero guards to 1e-20f
- add both absolute and relative error checks in standalone tests (atol=0.01, rtol=0.01)

2026-07-21 23:05:17 +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_split_kv_mma.cuh

refactor: adopt FA2-style KernelTraits + compile-time causal/mask dispatch

2026-07-21 21:52:46 +08:00

attn_decode_split_kv.cuh

fix: improve attention kernel numerical stability and test precision checks

2026-07-21 23:05:17 +08:00

attn_decode.cu

refactor: extract shared dispatcher header, unify MMA/scalar dispatch format

2026-07-21 22:21:39 +08:00

attn_dispatchers.cuh

refactor: extract shared dispatcher header, unify MMA/scalar dispatch format

2026-07-21 22:21:39 +08:00

attn_entry_utils.cuh

refactor: extract shared dispatcher header, unify MMA/scalar dispatch format

2026-07-21 22:21:39 +08:00

attn_mma_utils.cuh

refactor: adopt FA2-style KernelTraits + compile-time causal/mask dispatch

2026-07-21 21:52:46 +08:00

attn_paged_decode_split_kv_mma.cuh

refactor: adopt FA2-style KernelTraits + compile-time causal/mask dispatch

2026-07-21 21:52:46 +08:00

attn_paged_decode_split_kv.cuh

fix: improve attention kernel numerical stability and test precision checks

2026-07-21 23:05:17 +08:00

attn_paged_decode.cu

refactor: extract shared dispatcher header, unify MMA/scalar dispatch format

2026-07-21 22:21:39 +08:00

attn_prefill_split_q_mma.cuh

refactor: adopt FA2-style KernelTraits + compile-time causal/mask dispatch

2026-07-21 21:52:46 +08:00

attn_prefill_split_q.cuh

fix: improve attention kernel numerical stability and test precision checks

2026-07-21 23:05:17 +08:00

attn_prefill.cu

refactor: extract shared dispatcher header, unify MMA/scalar dispatch format

2026-07-21 22:21:39 +08:00