From 373f84a19332921940305c23c714ff0752a3f674 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E6=97=A0=E8=84=B8=E7=94=B7?= <244036962@qq.com>
Date: Mon, 29 Sep 2025 11:51:07 +0800
Subject: [PATCH] [Bugfix] Fix the error "cur batch_size is invalid" during
 profile_run in the torchair scenario (#3243)

### What this PR does / why we need it?
Fix the error "cur batch_size is invalid" during profile_run in the
torchair scenario.

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

### How was this patch tested?

- vLLM version: v0.10.2
- vLLM main:
https://github.com/vllm-project/vllm/commit/releases/v0.11.0

Signed-off-by: WithHades <244036962@qq.com>
---
 vllm_ascend/worker/model_runner_v1.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/vllm_ascend/worker/model_runner_v1.py b/vllm_ascend/worker/model_runner_v1.py
index 373a73e..0984e2b 100644
--- a/vllm_ascend/worker/model_runner_v1.py
+++ b/vllm_ascend/worker/model_runner_v1.py
@@ -2513,7 +2513,7 @@ class NPUModelRunner(LoRAModelRunnerMixin):
             if self._select_moe_comm_method(
                     self.mc2_tokens_capacity,
                     with_prefill=True) == MoECommType.MC2:
-                self._dummy_run(self.mc2_tokens_capacity)
+                self._dummy_run(self.mc2_tokens_capacity, with_prefill=True)
 
         output = None
         if get_pp_group().is_last_rank: