[main2main] upgrade vllm main 0202 (#6560)

### What this PR does / why we need it?
1. Fix `TypeError: FusedMoEParallelConfig.__init__() missing 1 required
positional argument: 'is_sequence_parallel'` due to
vllm-project/vllm#32567
2. Fix ` TypeError: '>' not supported between instances of 'MagicMock'
and 'int'` due to vllm-project/vllm#33035
3. Fix `TypeError: Can't instantiate abstract class AscendMLAImpl with
abstract methods forward_mha, forward_mqa` and AttributeError: 'bool'
object has no attribute 'process_weights_after_loading' due to
vllm-project/vllm#33284
4. Fix `'AscendSharedFusedMoE' object has no attribute
'_routed_input_transform'`due to
vllm-project/vllm#32790
5. Fix `NPUModelRunner._dummy_run() got an unexpected keyword argument
'num_active_loras'` due to
vllm-project/vllm#32005
6. Fix the problem caused by` 'tuple' object has no attribute 'job_id'`
due to vllm-project/vllm#27492
7. Fix the problem that all_moe_layers is not equal to vllm.moe_forward,
vllm.moe_forward_shared due to
vllm-project/vllm#33184
8. Add patch to fix the problem "got multiple values for keyword
argument 'add_special_tokens'" due to
vllm-project/vllm#32863
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.15.0
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.15.0

---------

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
Co-authored-by: hfadzxy <starmoon_zhang@163.com>

Author: 3 people

.github/workflows/_pre_commit.yml

@@ -38,6 +38,7 @@ jobs:
         repository: vllm-project/vllm
         path: ./vllm-empty
         ref: ${{ inputs.vllm }}
+
     - uses: dorny/paths-filter@v3
       id: filter
       with:
@@ -62,10 +63,12 @@ jobs:
       run: |
         git config --global --add safe.directory /__w/vllm-ascend/vllm-ascend
         pre-commit run --all-files --hook-stage manual --show-diff-on-failure
+
     - name: Run mypy
       run: |
         PYTHONPATH="$PYTHONPATH:$(pwd)/vllm-empty"
         export PYTHONPATH
+        env
         git config --global --add safe.directory /__w/vllm-ascend/vllm-ascend
         # Run mypy for Python 3.10, 3.11, 3.12 manually
         # Note: We are now separating mypy from pre-commit hooks for performance reasons.

.github/workflows/bot_pr_create.yaml

@@ -37,7 +37,7 @@ jobs:
     steps:
       - name: Get vLLM version
         run: |
-          VLLM_COMMIT=v0.15.0
+          VLLM_COMMIT=d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a
           echo "VLLM_COMMIT=https://github.com/vllm-project/vllm/commit/$VLLM_COMMIT" >> "$GITHUB_ENV"
 
       - name: Checkout repository

.github/workflows/dockerfiles/Dockerfile.lint

@@ -27,7 +27,7 @@ RUN apt-get update -y && \
 
 ARG VLLM_REPO=https://github.com/vllm-project/vllm.git
 # For lint purpose, actually we need make a main2main matching.
-ARG VLLM_COMMIT=v0.15.0
+ARG VLLM_COMMIT=d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a
 RUN git clone $VLLM_REPO /vllm-workspace/vllm && \
     cd /vllm-workspace/vllm && \
     git checkout $VLLM_COMMIT

.github/workflows/pr_test_full.yaml

@@ -75,7 +75,7 @@ jobs:
     name: e2e-full
     strategy:
       matrix:
-        vllm_version: [v0.15.0]
+        vllm_version: [d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a, v0.15.0]
     needs: [changes]
     if: ${{ needs.changes.outputs.e2e_tracker == 'true' || needs.changes.outputs.e2e_tracker == true }}
     uses: ./.github/workflows/_e2e_test.yaml

.github/workflows/pr_test_light.yaml

@@ -41,7 +41,7 @@ jobs:
   lint:
     uses: ./.github/workflows/_pre_commit.yml
     with:
-      vllm: v0.15.0
+      vllm: d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a
   changes:
     runs-on: linux-aarch64-a2-0
     outputs:
@@ -87,7 +87,7 @@ jobs:
     if: ${{ needs.lint.result == 'success' && (needs.changes.outputs.e2e_tracker == 'true' || needs.changes.outputs.ut_tracker == 'true') }}
     strategy:
       matrix:
-        vllm_version: [v0.15.0]
+        vllm_version: [d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a, v0.15.0]
     uses: ./.github/workflows/_unit_test.yaml
     with:
       vllm: ${{ matrix.vllm_version }}
@@ -99,7 +99,7 @@ jobs:
     name: e2e-light
     strategy:
       matrix:
-        vllm_version: [v0.15.0]
+        vllm_version: [d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a, v0.15.0]
     # Note (yikun): If CI resource are limited we can split job into two chain jobs
     needs: [lint, changes]
     # only trigger e2e test after lint passed and the change is e2e related with pull request.

.github/workflows/schedule_codecov_refresh.yaml

@@ -33,7 +33,7 @@ jobs:
     name: refresh codecov
     strategy:
       matrix:
-        vllm_version: [v0.15.0]
+        vllm_version: [d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a]
     uses: ./.github/workflows/_unit_test.yaml
     with:
       vllm: ${{ matrix.vllm_version }}

docs/source/community/versioning_policy.md

@@ -55,7 +55,7 @@ For main branch of vLLM Ascend, we usually make it compatible with the latest vL
 
 | vLLM Ascend | vLLM         | Python           | Stable CANN | PyTorch/torch_npu  |
 |-------------|--------------|------------------|-------------|--------------------|
-|     main    | v0.15.0 tag | >= 3.10, < 3.12   | 8.5.0 | 2.9.0 / 2.9.0 |
+|     main    | d7e17aaacd5ed1b4b4be6bcfef3a1b7cbc84fc9a, v0.15.0 tag | >= 3.10, < 3.12   | 8.5.0 | 2.9.0 / 2.9.0 |
 
 ## Release cadence
 

tests/e2e/conftest.py

@@ -922,4 +922,7 @@ def hunyuan_prompt(questions: list[str]) -> list[str]:
 
 @pytest.fixture(params=PROMPT_CONFIGS.keys())
 def vl_config(request):
-    return PROMPT_CONFIGS[request.param]
+    config = PROMPT_CONFIGS[request.param]
+    if "skip" in config:
+        pytest.skip(config["skip"])
+    return config

tests/ut/eplb/core/test_eplb_utils.py

@@ -9,6 +9,7 @@
 
 from vllm_ascend.ascend_config import init_ascend_config
 from vllm_ascend.eplb.core.eplb_utils import init_eplb_config
+from vllm_ascend.utils import vllm_version_is
 # isort: on
 
 
@@ -21,7 +22,13 @@ def setUp(self, mock_fix_incompatible_config):
             "eplb_config": {"dynamic_eplb": True, "num_redundant_experts": 2},
         }
         from vllm.model_executor.layers.fused_moe.config import RoutingMethodType
-        moe_parallel_config = FusedMoEParallelConfig(2, 0, 1, 2, 1, 1, 1, 1, True, "hccl", enable_eplb=True)
+        if vllm_version_is("0.15.0"):
+            moe_parallel_config = FusedMoEParallelConfig(
+                2, 0, 1, 2, 1, 1, 1, 1, True, "hccl", enable_eplb=True)
+        else:
+            moe_parallel_config = FusedMoEParallelConfig(
+                2, 0, 1, 2, 1, 1, 1, 1, True, "hccl",
+                is_sequence_parallel=False, enable_eplb=True)
         moe_config = FusedMoEConfig(
             num_experts=8,
             experts_per_token=8,

tests/ut/ops/test_mla.py

@@ -82,8 +82,13 @@ def setUp(self):
     @patch("vllm_ascend.ops.mla.get_tensor_model_parallel_world_size")
     def test_initialization(self, mock_tp_size, mock_ascend_config,
                             mock_get_vllm_config):
+        # Create a proper mock for MLAAttention that has the required attributes
+        mock_mla_attn = MagicMock()
+        mock_mla_attn.process_weights_after_loading = MagicMock()
+        mock_mla_attn.impl = MagicMock()
+        mock_mla_attn.impl.process_weights_after_loading = MagicMock()
 
-        with patch("vllm_ascend.ops.mla.MLAAttention", return_value=True):
+        with patch("vllm_ascend.ops.mla.MLAAttention", return_value=mock_mla_attn):
             mock_tp_size.return_value = 2
             mock_ascend_config.return_value.enable_shared_expert_dp = True
             mock_vllm_config = MagicMock(spec=VllmConfig)
@@ -126,7 +131,14 @@ def test_forward(self, mock_get_forward_context, mock_tp_size,
             num_hidden_layers=32, first_k_dense_replace=False)
         mock_get_vllm_config.return_value = mock_vllm_config
         mock_vllm_config.compilation_config = CompilationConfig()
-        with patch("vllm_ascend.ops.mla.MLAAttention", return_value=True):
+
+        # Create a proper mock for MLAAttention that has the required attributes
+        mock_mla_attn = MagicMock()
+        mock_mla_attn.process_weights_after_loading = MagicMock()
+        mock_mla_attn.impl = MagicMock()
+        mock_mla_attn.impl.process_weights_after_loading = MagicMock()
+
+        with patch("vllm_ascend.ops.mla.MLAAttention", return_value=mock_mla_attn):
             attn = AscendMultiHeadLatentAttention(
                 hidden_size=self.hidden_size,
                 num_heads=self.num_heads,