From db3c49d9733f49783095a72d1edab4bd390e526a Mon Sep 17 00:00:00 2001
From: linfeng-yuan <1102311262@qq.com>
Date: Tue, 16 Sep 2025 21:05:50 +0800
Subject: [PATCH] [bugfix] fix shared expert dp with hybrid kvcache

Signed-off-by: linfeng-yuan <1102311262@qq.com>
---
 vllm_ascend/torchair/models/torchair_deepseek_v2.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/vllm_ascend/torchair/models/torchair_deepseek_v2.py b/vllm_ascend/torchair/models/torchair_deepseek_v2.py
index ec48b56202..6fef115aac 100644
--- a/vllm_ascend/torchair/models/torchair_deepseek_v2.py
+++ b/vllm_ascend/torchair/models/torchair_deepseek_v2.py
@@ -813,6 +813,8 @@ def forward(
             residual = get_tp_group().all_gather(residual, 0)
 
             attn_metadata = get_forward_context().attn_metadata
+            if attn_metadata is not None and isinstance(attn_metadata, dict):
+                attn_metadata = attn_metadata['model.layers.0.self_attn.attn']
             if attn_metadata is not None:
                 num_tokens = attn_metadata.num_actual_tokens
             else: