Revert "[BugFix][SpecDecode] Fix low MTP acceptance rate for SFA+DSA_CP (MTP>1) (vllm-project#10825)" (vllm-project#10868)

### What this PR does / why we need it?
This reverts commit 21e8dcf, which
breaks
tests/e2e/pull_request/two_card/spec_decode/test_spec_decode.py::test_qwen3_eagle3_pcp2_tp1

- vLLM version: v0.22.1
- vLLM main:
vllm-project/vllm@967c5c3

Author: MengqingCao

vllm_ascend/attention/sfa_v1.py

@@ -200,11 +200,6 @@ def __init__(
 
         self.speculative_config = vllm_config.speculative_config
         self.decode_threshold = 1
-        max_num_reqs = vllm_config.scheduler_config.max_num_seqs
-        self.actual_seq_lengths_query = torch.zeros(max_num_reqs + 1, dtype=torch.int32, device=device)
-        self.actual_seq_lengths_key = torch.empty_like(self.actual_seq_lengths_query)
-        self.spec_actual_seq_lengths_query: list[torch.Tensor] | None = None
-        self.spec_actual_seq_lengths_key: list[torch.Tensor] | None = None
         if self.speculative_config:
             spec_token_num = self.speculative_config.num_speculative_tokens
             self.decode_threshold += spec_token_num
@@ -213,20 +208,15 @@ def __init__(
                 npu_fused_infer_attention_score TND layout's limit of 16, \
                 got {self.decode_threshold}"
             )
-            self.spec_actual_seq_lengths_query = [
-                torch.zeros(max_num_reqs * (spec_token_num + 1) + 1, dtype=torch.int32, device=device)
-                for _ in range(spec_token_num)
-            ]
-            self.spec_actual_seq_lengths_key = [
-                torch.zeros(max_num_reqs * (spec_token_num + 1) + 1, dtype=torch.int32, device=device)
-                for _ in range(spec_token_num)
-            ]
-
         self.reorder_batch_threshold = self.decode_threshold
         self.attn_mask_builder = AttentionMaskBuilder(self.device)
         self.rope_dim = self.model_config.hf_text_config.qk_rope_head_dim
         self.enable_dsa_cp = enable_dsa_cp()
 
+        max_num_reqs = vllm_config.scheduler_config.max_num_seqs
+        self.actual_seq_lengths_query = torch.zeros(max_num_reqs + 1, dtype=torch.int32, device=device)
+        self.actual_seq_lengths_key = torch.empty_like(self.actual_seq_lengths_query)
+
     @staticmethod
     def determine_chunked_prefill_workspace_size(vllm_config: VllmConfig) -> int:
         return ascend_chunked_prefill_workspace_size(vllm_config)
@@ -250,22 +240,6 @@ def build(
         common_prefix_len: int,
         common_attn_metadata: AscendCommonAttentionMetadata,
         fast_build: bool = False,
-    ) -> AscendSFAMetadata:
-        # common_prefix_len / fast_build are unused; kept for API compatibility.
-        return self._build(common_attn_metadata, draft_step=None)
-
-    def build_for_drafting(
-        self,
-        draft_step: int,
-        common_attn_metadata: AscendCommonAttentionMetadata,
-        **kwargs,
-    ) -> AscendSFAMetadata:
-        return self._build(common_attn_metadata, draft_step=draft_step)
-
-    def _build(
-        self,
-        common_attn_metadata: AscendCommonAttentionMetadata,
-        draft_step: int | None = None,
     ) -> AscendSFAMetadata:
         num_reqs = common_attn_metadata.num_reqs
         num_actual_tokens = common_attn_metadata.num_actual_tokens
@@ -291,7 +265,7 @@ def _build(
         else:
             seq_lens_cpu = common_attn_metadata.seq_lens[:num_reqs].to("cpu")
 
-        cos, sin = get_cos_and_sin_mla(input_positions, use_cache=(draft_step is None))
+        cos, sin = get_cos_and_sin_mla(input_positions, True)
 
         dsa_cp_context = None
         if self.enable_dsa_cp:
@@ -333,16 +307,8 @@ def _build(
                     got {slot_mapping.shape[0]} and {num_tokens_pad}"
             )
 
-            if draft_step is not None:
-                assert self.spec_actual_seq_lengths_query is not None
-                assert self.spec_actual_seq_lengths_key is not None
-                # Per-draft-step buffers: independent, graph-stable storage so
-                # later draft steps don't clobber earlier ones' metadata.
-                actual_seq_lengths_query = self.spec_actual_seq_lengths_query[draft_step - 1]
-                actual_seq_lengths_key = self.spec_actual_seq_lengths_key[draft_step - 1]
-            else:
-                actual_seq_lengths_query = self.actual_seq_lengths_query
-                actual_seq_lengths_key = self.actual_seq_lengths_key
+            actual_seq_lengths_query = self.actual_seq_lengths_query
+            actual_seq_lengths_key = self.actual_seq_lengths_key
 
             num_segs = cum_query_lens.shape[0]
             last_token = 0

vllm_ascend/spec_decode/llm_base_proposer.py

@@ -1652,13 +1652,6 @@ def attn_update_stack_num_spec_norm(
                 common_attn_metadata,
                 **extra_attn_metadata_args,
             )
-        elif hasattr(attn_metadata_builder, "build_for_drafting"):
-            # e.g. SFA (dsa_cp): route draft steps through a draft-aware build so
-            # per-draft-step buffers are used and cross-step aliasing is avoided.
-            attn_metadata = attn_metadata_builder.build_for_drafting(
-                draft_step,
-                common_attn_metadata,
-            )
         else:
             attn_metadata = attn_metadata_builder.build(
                 0,