[BugFix]converting pa get_workspace back to capturing (#5833)

### What this PR does / why we need it?

This helps to fix a bug in for pa get_workspace. In earlier
implementation, we use `_npu_paged_attention_get_workspace` in
`_update_pa_attn_params`. However, this might cause some potential
memory problems as it dynamically allocate new memory for workspace when
calling this api. Therefor, we move this back to capturing, and use a
fixed `SEQ_LEN_WITH_MAX_PA_WORKSPACE` to get max workspace.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef

Signed-off-by: Angazenn <supperccell@163.com>

vllm_ascend/compilation/acl_graph.py

@@ -229,25 +229,6 @@ def _update_attn_pa_params(update_stream, forward_context, runtime_shape):
             ) = param
             seq_lens = forward_context.attn_metadata[key].seq_lens
 
-            # When using FULL_DECODE_ONLY, there are some rare bugs for FULL_DECODE_ONLY
-            # mode with GQA. This is triggered by getting workspace for _npu_paged_attention
-            # in torch_npu. On some rare cases, _npu_paged_attention with smaller seq_lens
-            # might encounter a bigger workspace, while currently we use max_model_len to
-            # calculate max workspace in capturing. So additional get_workspace is added
-            # here to avoid such bugs.
-            # TODO(Angazenn): we will remove this once _npu_paged_attention is fully
-            # replaced by npu_fused_infer_attention_score which does not contain such bugs.
-            workspace = torch_npu._npu_paged_attention_get_workspace(
-                query=query,
-                key_cache=key_cache,
-                value_cache=value_cache,
-                num_kv_heads=num_kv_heads,
-                num_heads=num_heads,
-                scale_value=scale,
-                block_table=block_table,
-                context_lens=seq_lens,
-                out=output,
-            )
             torch.npu.graph_task_update_begin(update_stream, handle)
             torch_npu._npu_paged_attention(
                 query=query,
@@ -259,7 +240,7 @@ def _update_attn_pa_params(update_stream, forward_context, runtime_shape):
                 block_table=block_table,
                 context_lens=seq_lens,
                 out=output,
-                workspace=workspace,
+                workspace=graph_params.workspaces.get(runtime_shape),
             )
             torch.npu.graph_task_update_end(update_stream)
 

vllm_ascend/worker/model_runner_v1.py

@@ -76,7 +76,7 @@ from vllm.v1.worker.utils import AttentionGroup
 
 from vllm_ascend.ascend_config import get_ascend_config
 from vllm_ascend.attention.attention_v1 import AscendAttentionState
-from vllm_ascend.attention.utils import AscendCommonAttentionMetadata
+from vllm_ascend.attention.utils import AscendCommonAttentionMetadata, using_paged_attention
 # yapf conflicts with isort for this block
 # yapf: disable
 from vllm_ascend.compilation.acl_graph import (ACLGraphWrapper,
@@ -133,6 +133,9 @@ if get_ascend_device_type() == AscendDeviceType._310P:
     torch_npu.npu.set_compile_mode(jit_compile=False)
 
 
+SEQ_LEN_WITH_MAX_PA_WORKSPACE = 6144
+
+
 @dataclass
 class GraphCaptureContext:
     stream: torch.npu.Stream
@@ -1919,6 +1922,7 @@ class NPUModelRunner(GPUModelRunner):
         num_scheduled_tokens: np.ndarray,
         aclgraph_runtime_mode: Optional[CUDAGraphMode] = None,
         force_attention: bool = False,
+        is_graph_capturing: bool = False,
     ) -> Optional[dict[str, Any]]:
         attn_metadata: Optional[dict[str, Any]] = None
 
@@ -1928,7 +1932,12 @@ class NPUModelRunner(GPUModelRunner):
 
             attn_metadata = {}
 
-            seq_lens = max_query_len
+            # The reason why we use a fixed seq_len rather than max_query_len is that
+            # _npu_paged_attention_get_workspace only returns max workspace with specific
+            # seq_lens. We use this seq_len only when capturing graph, and still use max_query_len
+            # in inference. This will be removed once npu_fused_infer_attention_score
+            # outperforms _npu_paged_attention on all cases.
+            seq_lens = SEQ_LEN_WITH_MAX_PA_WORKSPACE if is_graph_capturing and using_paged_attention(num_tokens, self.vllm_config) else max_query_len
             self.seq_lens.np[:num_reqs] = seq_lens
             self.seq_lens.np[num_reqs:] = 0
             self.seq_lens.copy_to_gpu()
@@ -2177,6 +2186,7 @@ class NPUModelRunner(GPUModelRunner):
             max_query_len=max_query_len,
             aclgraph_runtime_mode=cudagraph_runtime_mode,
             force_attention=force_attention,
+            is_graph_capturing=is_graph_capturing,
             num_scheduled_tokens=num_scheduled_tokens,
         )