[Feature][310P] support shared experts path in fused MoE for qwen3.5 (#7674)

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

310P originally supported only the Qwen3 series. Recent adaptation work
for Qwen3.5 introduced the new shared-experts structure, which had not
been considered on the 310P path, so this fix was made. The fix aligns
the 310P execution flow with the A2/A3 implementation path.

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

NO

### How was this patch tested?

local e2e test

- vLLM version: v0.18.0
- vLLM main:
vllm-project/vllm@35141a7

---------

Signed-off-by: Tflowers-0129 <2906339855@qq.com>

Author: Tflowers-0129

tests/e2e/310p/multicard/test_moe_model_multicard.py

@@ -63,3 +63,18 @@ def test_qwen3_moe_tp2_w8a8():
         max_model_len=16384,
     ) as vllm_model:
         vllm_model.generate_greedy(example_prompts, max_tokens)
+
+
+def test_qwen3_5_moe_tp4_fp16():
+    example_prompts = [
+        "Hello, my name is",
+    ]
+    max_tokens = 5
+    with VllmRunner(
+        "Qwen/Qwen3.5-35B-A3B",
+        tensor_parallel_size=4,
+        enforce_eager=True,
+        dtype="float16",
+        max_model_len=16384,
+    ) as vllm_model:
+        vllm_model.generate_greedy(example_prompts, max_tokens)

tests/e2e/310p/singlecard/test_dense_model_singlecard.py

@@ -47,3 +47,18 @@ def test_qwen3_dense_tp1_w8a8():
         max_model_len=16384,
     ) as vllm_model:
         vllm_model.generate_greedy(example_prompts, max_tokens)
+
+
+def test_qwen3_5_dense_tp1_fp16():
+    example_prompts = [
+        "Hello, my name is",
+    ]
+    max_tokens = 5
+    with VllmRunner(
+        "Qwen/Qwen3.5-4B",
+        tensor_parallel_size=1,
+        enforce_eager=True,
+        dtype="float16",
+        max_model_len=16384,
+    ) as vllm_model:
+        vllm_model.generate_greedy(example_prompts, max_tokens)

tests/ut/_310p/fused_moe/test_shared_fused_moe_310.py

@@ -0,0 +1,110 @@
+#
+# Copyright (c) 2026 Huawei Technologies Co., Ltd. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from unittest.mock import patch
+
+import torch
+import torch.nn.functional as F
+
+from vllm_ascend._310p.fused_moe.fused_moe import (
+    AscendFusedMoE310,
+    AscendSharedFusedMoE310,
+)
+
+
+class _DummyGate(torch.nn.Module):
+    def forward(self, hidden_states: torch.Tensor):
+        # Keep gate output deterministic: sigmoid(0)=0.5.
+        return torch.zeros(
+            hidden_states.shape[0],
+            1,
+            dtype=hidden_states.dtype,
+            device=hidden_states.device,
+        ), None
+
+
+class _DummySharedExperts(torch.nn.Module):
+    def __init__(self, with_gate: bool):
+        super().__init__()
+        self.expert_gate = _DummyGate() if with_gate else None
+
+    def forward(self, hidden_states: torch.Tensor):
+        out = hidden_states * 2.0 + 1.0
+        if self.expert_gate is not None:
+            gate_out, _ = self.expert_gate(hidden_states)
+            out = F.sigmoid(gate_out) * out
+        return out
+
+
+def _build_layer(shared_experts: torch.nn.Module | None) -> AscendSharedFusedMoE310:
+    layer = AscendSharedFusedMoE310.__new__(AscendSharedFusedMoE310)
+    # The test bypasses full layer init with __new__, so we must initialize
+    # nn.Module internals before assigning child modules.
+    torch.nn.Module.__init__(layer)
+    layer._shared_experts = shared_experts
+    return layer
+
+
+def test_forward_shared_experts_without_gate_310():
+    layer = _build_layer(_DummySharedExperts(with_gate=False))
+    hidden_states = torch.randn(4, 8)
+    output = layer._forward_shared_experts(hidden_states)
+    expected = hidden_states * 2.0 + 1.0
+    torch.testing.assert_close(output, expected)
+
+
+def test_forward_shared_experts_with_gate_310():
+    layer = _build_layer(_DummySharedExperts(with_gate=True))
+    hidden_states = torch.randn(4, 8)
+    output = layer._forward_shared_experts(hidden_states)
+    expected = 0.5 * (hidden_states * 2.0 + 1.0)
+    torch.testing.assert_close(output, expected)
+
+
+def test_forward_impl_with_shared_experts_returns_tuple_310():
+    layer = _build_layer(_DummySharedExperts(with_gate=True))
+    hidden_states = torch.randn(3, 8)
+    router_logits = torch.randn(3, 8)
+    routed_out = torch.randn(3, 8)
+
+    with patch.object(AscendFusedMoE310, "forward_impl", return_value=routed_out):
+        shared_out, routed = layer.forward_impl(hidden_states, router_logits)
+
+    expected_shared = 0.5 * (hidden_states * 2.0 + 1.0)
+    torch.testing.assert_close(shared_out, expected_shared)
+    torch.testing.assert_close(routed, routed_out)
+
+
+def test_forward_impl_without_shared_experts_integration_310():
+    layer = _build_layer(None)
+    hidden_states = torch.randn(3, 8)
+    assert layer._forward_shared_experts(hidden_states) is None
+
+
+def test_forward_impl_without_shared_experts_returns_routed_only_310():
+    layer = _build_layer(None)
+    hidden_states = torch.randn(3, 8)
+    router_logits = torch.randn(3, 8)
+    routed_out = torch.randn(3, 8)
+
+    with patch.object(AscendFusedMoE310, "forward_impl", return_value=routed_out):
+        output = layer.forward_impl(hidden_states, router_logits)
+
+    torch.testing.assert_close(output, routed_out)
+
+
+def test_is_internal_router_is_false_310():
+    layer = _build_layer(_DummySharedExperts(with_gate=True))
+    assert layer.is_internal_router is False

vllm_ascend/_310p/fused_moe/fused_moe.py

@@ -274,6 +274,14 @@ def __init__(
         self.use_overlapped = use_overlapped
         self.shared_expert_stream = None
         self._gate = gate
+        # Recreate runner after shared_experts/gate are set so custom op dispatch
+        # goes through moe_forward_shared.
+        self.runner = self._init_runner()
+
+    @property
+    def is_internal_router(self) -> bool:
+        # 310P Ascend path expects router logits from the model forward path.
+        return False
 
     def forward(
         self,
@@ -298,9 +306,7 @@ def forward(
     def _forward_shared_experts(self, hidden_states: torch.Tensor):
         if self._shared_experts is None:
             return None
-        part1_out = self._shared_experts_part1(hidden_states)
-        shared_out = self._shared_experts_part2(hidden_states, part1_out)
-        return shared_out
+        return self._shared_experts(hidden_states)
 
     def forward_impl(  # type: ignore[override]
         self, hidden_states: torch.Tensor, router_logits: torch.Tensor

vllm_ascend/patch/worker/patch_qwen3_5_310.py

@@ -28,7 +28,7 @@
 from vllm_ascend._310p.ops.fla.fused_gdn_gating import fused_gdn_gating_pytorch
 from vllm_ascend._310p.ops.fla.fused_recurrent_gated_delta_rule import fused_recurrent_gated_delta_rule_pytorch
 from vllm_ascend.attention.utils import maybe_save_kv_layer_to_connector
-from vllm_ascend.utils import enable_sp
+from vllm_ascend.utils import enable_sp, vllm_version_is
 
 
 class Ascend310Qwen3_5GatedDeltaNet(Qwen3_5GatedDeltaNet):
@@ -65,7 +65,7 @@ def _forward_core(
         non_spec_token_indx = attn_metadata.non_spec_token_indx
         spec_state_indices_tensor = attn_metadata.spec_state_indices_tensor  # noqa: E501
         non_spec_state_indices_tensor = attn_metadata.non_spec_state_indices_tensor  # noqa: E501
-        self_kv_cache = self.kv_cache[forward_context.virtual_engine]
+        self_kv_cache = self.kv_cache[forward_context.virtual_engine if vllm_version_is("0.18.0") else 0]
         conv_state = self_kv_cache[0].transpose(-1, -2)
         ssm_state = self_kv_cache[1]
         num_actual_tokens = attn_metadata.num_actual_tokens