refactor: tidy attention params and launcher interfaces

- rename output pointer field o to o_ptr for consistency with q_ptr/k_ptr/v_ptr - regroup AttentionParams fields by responsibility and fix misleading comments - drop unused max_seq_len/total_q fields and paged decode max_seq_len arg - drop redundant group_size param from decode launchers (computed from p)
2026-08-09 20:52:06 +08:00
parent a5a3cc1fc2
commit cd31f1f62f
14 changed files with 69 additions and 84 deletions
@@ -545,7 +545,6 @@ class CudaBackend(AttentionBackend):
            kv_cache.req_to_token,
            kv_cache.req_pool_indices,
            kv_indptr,
-            kv_cache.max_len,
            is_causal=True,
            o_part_buf=kv_cache.decode_o_part,
            ml_part_buf=kv_cache.decode_ml_part,