vllm-ascend 0.11.0rc0__tar.gz → 0.11.0rc2__tar.gz

vllm_ascend-0.11.0rc2/.github/Dockerfile.buildwheel

@@ -0,0 +1,45 @@
+#
+# Copyright (c) 2025 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.
+# This file is a part of the vllm-ascend project.
+#
+ARG PY_VERSION=3.11
+FROM quay.io/ascend/manylinux:8.2.rc1-910b-manylinux_2_28-py${PY_VERSION}
+
+ARG COMPILE_CUSTOM_KERNELS=1
+
+# Define environments
+ENV DEBIAN_FRONTEND=noninteractive
+ENV COMPILE_CUSTOM_KERNELS=${COMPILE_CUSTOM_KERNELS}
+RUN yum update -y && \
+    yum install -y python3-pip git vim wget net-tools gcc gcc-c++ make cmake numactl-devel && \
+    rm -rf /var/cache/yum
+
+WORKDIR /workspace
+
+COPY . /workspace/vllm-ascend/
+
+# Install req
+RUN python3 -m pip install -r vllm-ascend/requirements.txt --extra-index https://download.pytorch.org/whl/cpu/ && \
+    python3 -m pip install twine
+
+# Install vllm-ascend
+RUN source /usr/local/Ascend/ascend-toolkit/set_env.sh && \
+    source /usr/local/Ascend/nnal/atb/set_env.sh && \
+    export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \
+    cd vllm-ascend && \
+    python3 setup.py bdist_wheel && \
+    ls -l dist 
+
+CMD ["/bin/bash"]

vllm_ascend-0.11.0rc2/.github/actionlint.yaml

@@ -0,0 +1,21 @@
+self-hosted-runner:
+  # Labels of self-hosted runner in array of strings.
+  labels:
+    - linux-aarch64-a2-0
+    - linux-aarch64-a2-1
+    - linux-aarch64-a2-2
+    - linux-aarch64-a2-4
+    - linux-aarch64-a2-8
+    - linux-arm64-npu-static-8
+    - linux-aarch64-310p-1
+    - linux-aarch64-310p-2
+    - linux-aarch64-310p-4
+    - ubuntu-24.04-arm
+    - linux-aarch64-a3-1
+    - linux-aarch64-a3-2
+    - linux-aarch64-a3-4
+    - linux-aarch64-a3-8
+    - linux-amd64-cpu-0
+    - linux-amd64-cpu-8
+    - linux-amd64-cpu-16
+    - linux-aarch64-a3-0

vllm_ascend-0.11.0rc2/.github/workflows/_accuracy_test.yaml

@@ -0,0 +1,175 @@
+name: 'accuracy test'
+
+on:
+  workflow_call:
+    inputs:
+      vllm:
+        required: true
+        type: string
+      vllm-ascend:
+        required: false
+        type: string
+        default: main
+      runner:
+        required: true
+        type: string
+      image:
+        required: true
+        type: string
+      model_name:
+        required: true
+        type: string
+      upload:
+        required: false
+        type: boolean
+        default: false
+
+jobs:
+  accuracy_tests:
+
+    runs-on: ${{ inputs.runner }}
+    name: ${{ inputs.model_name }} accuracy
+    container:
+      image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-910b-ubuntu22.04-py3.11
+      env:
+        VLLM_USE_MODELSCOPE: True
+        # 1. If version specified (work_dispatch), do specified branch accuracy test
+        # 2. If no version (labeled PR), do accuracy test by default ref:
+        # The branch, tag or SHA to checkout. When checking out the repository that
+        # triggered a workflow, this defaults to the reference or SHA for that event.
+        # Otherwise, uses the default branch.
+        GHA_VLLM_ASCEND_VERSION: ${{ inputs.vllm-ascend }}
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set model name as output
+        id: set_output
+        run: |
+          echo "model_name=${{ inputs.model_name }}" >> $GITHUB_OUTPUT
+
+      - name: Config mirrors
+        run: |
+          sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
+          pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
+          pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
+          apt-get update -y
+          apt install git -y
+
+      - name: Install system dependencies
+        run: |
+          apt-get -y install `cat packages.txt`
+          apt-get -y install gcc g++ cmake libnuma-dev
+
+      - name: Checkout vllm-project/vllm repo
+        uses: actions/checkout@v4
+        with:
+          repository: vllm-project/vllm
+          ref: ${{ inputs.vllm }}
+          path: ./vllm-empty
+
+      - name: Install vllm-project/vllm from source
+        working-directory: ./vllm-empty
+        run: |
+          VLLM_TARGET_DEVICE=empty pip install -e .
+
+      - name: Resolve vllm-ascend version
+        run: |
+          VERSION_INPUT="${{ inputs.vllm-ascend }}"
+          
+          if [[ "$VERSION_INPUT" == "latest" ]]; then
+            TAGS=$(git ls-remote --tags --sort=-v:refname https://github.com/vllm-project/vllm-ascend "v*" | cut -f2 | sed 's|refs/tags/||')
+            LATEST_TAG=$(echo "$TAGS" | head -n1)
+            if [[ -z "$LATEST_TAG" ]]; then
+              RESOLVED_VERSION="main"
+            else
+              RESOLVED_VERSION="$LATEST_TAG"
+            fi
+          else
+            RESOLVED_VERSION="$VERSION_INPUT"
+          fi
+          echo "GHA_VLLM_ASCEND_VERSION=$RESOLVED_VERSION" >> $GITHUB_ENV
+
+      - name: Checkout vllm-project/vllm-ascend repo
+        uses: actions/checkout@v4
+        with:
+          repository: vllm-project/vllm-ascend
+          path: ./vllm-ascend
+          ref: ${{ env.GHA_VLLM_ASCEND_VERSION }}
+
+      - name: Install vllm-project/vllm-ascend
+        working-directory: ./vllm-ascend
+        env:
+          PIP_EXTRA_INDEX_URL: https://mirrors.huaweicloud.com/ascend/repos/pypi
+        run: |
+          pip install -r requirements-dev.txt
+          pip install -v -e .
+
+      - name: Get vLLM commit hash and URL
+        working-directory: ./vllm-empty
+        run: |
+          VLLM_COMMIT=$(git rev-parse --short=7 HEAD)
+          echo "VLLM_COMMIT=$VLLM_COMMIT" >> $GITHUB_ENV
+
+      - name: Get vLLM-Ascend commit hash and URL
+        working-directory: ./vllm-ascend
+        run: |
+          VLLM_ASCEND_COMMIT=$(git rev-parse --short=7 HEAD)
+          echo "VLLM_ASCEND_COMMIT=$VLLM_ASCEND_COMMIT" >> $GITHUB_ENV
+
+      - name: Collect version info
+        run: |
+          for dir in /usr/local/Ascend/ascend-toolkit/*; do
+            dname=$(basename "$dir")
+            if [ "$dname" != "latest" ]; then
+              TOOLKIT_DIR="$dname"
+              break
+            fi
+          done
+          INFO_FILE="/usr/local/Ascend/ascend-toolkit/${TOOLKIT_DIR}/$(uname -i)-linux/ascend_toolkit_install.info"
+          GHA_CANN_VERSION=$(grep "version=" "$INFO_FILE" \
+                           | head -n1 \
+                           | cut -d'=' -f2 \
+                           | tr -d '"')
+          {
+            echo "GHA_CANN_VERSION=$GHA_CANN_VERSION"
+            pip show torch | grep "Version:" | awk '{print "GHA_TORCH_VERSION="$2}'
+            pip show torch_npu | grep "Version:" | awk '{print "GHA_TORCH_NPU_VERSION="$2}'
+            pip show vllm | grep "Version:" | awk '{print "GHA_VLLM_VERSION="$2}' | sed 's/+.*//'
+          } >> "$GITHUB_ENV"
+
+      - name: Run accuracy test
+        id: report
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_USE_MODELSCOPE: True
+          VLLM_VERSION: ${{ env.GHA_VLLM_VERSION }}
+          VLLM_COMMIT: ${{ env.VLLM_COMMIT }}
+          VLLM_ASCEND_VERSION: ${{ env.GHA_VLLM_ASCEND_VERSION || github.ref }}
+          VLLM_ASCEND_COMMIT: ${{ env.VLLM_ASCEND_COMMIT }}
+          CANN_VERSION: ${{ env.GHA_CANN_VERSION }}
+          TORCH_VERSION: ${{ env.GHA_TORCH_VERSION }}
+          TORCH_NPU_VERSION: ${{ env.GHA_TORCH_NPU_VERSION }}
+        run: |
+          model_base_name=$(basename ${{ inputs.model_name }})
+          markdown_name="${model_base_name}"
+          echo "markdown_name=$markdown_name" >> $GITHUB_OUTPUT
+          mkdir -p ./benchmarks/accuracy
+          pytest -sv ./tests/e2e/models/test_lm_eval_correctness.py \
+          --config ./tests/e2e/models/configs/${{ inputs.model_name }}.yaml
+
+      - name: Generate step summary
+        if: ${{ always() }}
+        run: |
+          cat ./benchmarks/accuracy/${{ steps.report.outputs.markdown_name }}.md >> $GITHUB_STEP_SUMMARY
+
+      - name: Upload Report
+        if: ${{ inputs.upload == true }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: "report-${{ env.GHA_VLLM_ASCEND_VERSION }}-${{ steps.report.outputs.markdown_name }}"
+          path: ./benchmarks/accuracy/${{ steps.report.outputs.markdown_name }}.md
+          if-no-files-found: warn
+          retention-days: 90
+          overwrite: true

vllm_ascend-0.11.0rc2/.github/workflows/_e2e_test.yaml

@@ -0,0 +1,199 @@
+name: 'e2e test'
+
+on:
+  workflow_call:
+    inputs:
+      vllm:
+        required: true
+        type: string
+      runner:
+        required: true
+        type: string
+      image:
+        required: true
+        type: string
+      type:
+        required: true
+        type: string
+
+jobs:
+  e2e:
+    name: singlecard
+    runs-on: ${{ inputs.runner }}-1
+    container:
+      image: ${{ inputs.image }}
+      env:
+        VLLM_LOGGING_LEVEL: ERROR
+        VLLM_USE_MODELSCOPE: True
+    steps:
+      - name: Check npu and CANN info
+        run: |
+          npu-smi info
+          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
+
+      - name: Config mirrors
+        run: |
+          sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
+          pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
+          pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
+          apt-get update -y
+          apt install git -y
+
+      - name: Checkout vllm-project/vllm-ascend repo
+        uses: actions/checkout@v4
+
+      - name: Install system dependencies
+        run: |
+          apt-get -y install `cat packages.txt`
+          apt-get -y install gcc g++ cmake libnuma-dev
+
+      - name: Checkout vllm-project/vllm repo
+        uses: actions/checkout@v4
+        with:
+          repository: vllm-project/vllm
+          ref: ${{ inputs.vllm }}
+          path: ./vllm-empty
+          fetch-depth: 1
+
+      - name: Install vllm-project/vllm from source
+        working-directory: ./vllm-empty
+        run: |
+          VLLM_TARGET_DEVICE=empty pip install -e .
+
+      - name: Install vllm-project/vllm-ascend
+        env:
+          PIP_EXTRA_INDEX_URL: https://mirrors.huaweicloud.com/ascend/repos/pypi
+        run: |
+          pip install -r requirements-dev.txt
+          pip install -v -e .
+
+      - name: Run vllm-project/vllm-ascend test
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_USE_MODELSCOPE: True
+          PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
+        if: ${{ inputs.type == 'light' }}
+        run: |
+          pytest -sv tests/e2e/singlecard/test_aclgraph.py
+          pytest -sv tests/e2e/singlecard/test_quantization.py
+          pytest -sv tests/e2e/singlecard/test_vlm.py::test_multimodal_vl
+
+      - name: Run e2e test
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_USE_MODELSCOPE: True
+          PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
+        if: ${{ inputs.type == 'full' }}
+        run: |
+          # We found that if running aclgraph tests in batch, it will cause AclmdlRICaptureBegin error. So we run
+          # the test separately.
+
+          pytest -sv tests/e2e/singlecard/test_aclgraph.py
+          pytest -sv tests/e2e/singlecard/test_aclgraph_mem.py
+          pytest -sv tests/e2e/singlecard/test_ascend_scheduler.py
+          pytest -sv tests/e2e/singlecard/test_bge_model.py
+          pytest -sv tests/e2e/singlecard/test_camem.py
+          pytest -sv tests/e2e/singlecard/test_chunked.py
+          pytest -sv tests/e2e/singlecard/test_embedding.py
+          pytest -sv tests/e2e/singlecard/test_embedding_aclgraph.py
+          pytest -sv tests/e2e/singlecard/test_guided_decoding.py
+          pytest -sv tests/e2e/singlecard/test_ilama_lora.py
+          pytest -sv tests/e2e/singlecard/test_profile_execute_duration.py
+          pytest -sv tests/e2e/singlecard/test_quantization.py
+          pytest -sv tests/e2e/singlecard/test_sampler.py
+          pytest -sv tests/e2e/singlecard/test_vlm.py
+
+          # ------------------------------------ v1 spec decode test ------------------------------------ #
+          pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_correctness.py
+          pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_torchair_correctness.py
+          # Fix me: test_eagle_correctness OOM error
+          pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_spec_decode.py
+
+          pytest -sv tests/e2e/singlecard/ops/
+
+  e2e-2-cards:
+    name: multicard
+    runs-on: ${{ inputs.runner }}-2
+    container:
+      image: ${{ inputs.image }}
+      env:
+        VLLM_LOGGING_LEVEL: ERROR
+        VLLM_USE_MODELSCOPE: True
+    steps:
+      - name: Check npu and CANN info
+        run: |
+          npu-smi info
+          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
+
+      - name: Config mirrors
+        run: |
+          sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
+          pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
+          pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
+          apt-get update -y
+          apt install git -y
+
+      - name: Checkout vllm-project/vllm-ascend repo
+        uses: actions/checkout@v4
+
+      - name: Install system dependencies
+        run: |
+          apt-get -y install `cat packages.txt`
+          apt-get -y install gcc g++ cmake libnuma-dev
+
+      - name: Checkout vllm-project/vllm repo
+        uses: actions/checkout@v4
+        with:
+          repository: vllm-project/vllm
+          ref: ${{ inputs.vllm }}
+          path: ./vllm-empty
+          fetch-depth: 1
+
+      - name: Install vllm-project/vllm from source
+        working-directory: ./vllm-empty
+        run: |
+          VLLM_TARGET_DEVICE=empty pip install -e .
+
+      - name: Install vllm-project/vllm-ascend
+        env:
+          PIP_EXTRA_INDEX_URL: https://mirrors.huaweicloud.com/ascend/repos/pypi
+        run: |
+          pip install -r requirements-dev.txt
+          pip install -v -e .
+
+      - name: Run vllm-project/vllm-ascend test (light)
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_USE_MODELSCOPE: True
+        if: ${{ inputs.type == 'light' }}
+        run: |
+          pytest -sv tests/e2e/multicard/test_qwen3_moe.py::test_models_distributed_Qwen3_MOE_TP2_WITH_EP
+
+      - name: Run vllm-project/vllm-ascend test (full)
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_USE_MODELSCOPE: True
+        if: ${{ inputs.type == 'full' }}
+        run: |
+          pytest -sv tests/e2e/multicard/test_data_parallel.py
+          pytest -sv tests/e2e/multicard/test_expert_parallel.py
+          pytest -sv tests/e2e/multicard/test_external_launcher.py
+          pytest -sv tests/e2e/multicard/test_single_request_aclgraph.py
+          pytest -sv tests/e2e/multicard/test_fused_moe_allgather_ep.py
+          pytest -sv tests/e2e/multicard/test_ilama_lora_tp2.py
+
+          # To avoid oom, we need to run the test in a single process.
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_QwQ
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_multistream_moe
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen3_W8A8
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen3_W4A8DYNAMIC_new_version
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen3_W4A8DYNAMIC_old_version
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_W4A8DYNAMIC
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_sp_for_qwen3_moe
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen_Dense_with_flashcomm_v1
+          pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen_Dense_with_prefetch_mlp_weight
+
+          pytest -sv tests/e2e/multicard/test_pipeline_parallel.py
+          pytest -sv tests/e2e/multicard/test_prefix_caching.py
+          pytest -sv tests/e2e/multicard/test_qwen3_moe.py
+          pytest -sv tests/e2e/multicard/test_torchair_graph_mode.py

vllm_ascend-0.11.0rc2/.github/workflows/accuracy_test.yaml

@@ -0,0 +1,72 @@
+#
+# Copyright (c) 2025 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.
+# This file is a part of the vllm-ascend project.
+#
+
+# This test will be triggered:
+# - PR labeled with: 'accuracy-test' & 'ready-for-test'
+name: ascend test / accuracy
+
+on:
+  pull_request:
+    branches:
+      - 'main'
+      - '*-dev'
+    types: [ labeled, synchronize ]
+
+# Bash shells do not use ~/.profile or ~/.bashrc so these shells need to be explicitly
+# declared as "shell: bash -el {0}" on steps that need to be properly activated.
+# It's used to activate ascend-toolkit environment variables.
+defaults:
+  run:
+    shell: bash -el {0}
+
+# only cancel in-progress runs of the same workflow
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  run:
+    name: ""
+    strategy:
+      matrix:
+        # Only top series models should be listed in here
+        include:
+          - runner: a2-1
+            model_name: Qwen3-8B
+          - runner: a2-1
+            model_name: Qwen2.5-VL-7B-Instruct
+          - runner: a2-1
+            model_name: Qwen2-Audio-7B-Instruct
+          - runner: a2-2
+            model_name: Qwen3-30B-A3B
+          - runner: a2-2
+            model_name: Qwen3-VL-30B-A3B-Instruct
+          - runner: a2-2
+            model_name: DeepSeek-V2-Lite
+      fail-fast: false
+    # test will be triggered when tag 'accuracy-test' & 'ready-for-test'
+    if:  >-
+      ${{
+      contains(github.event.pull_request.labels.*.name, 'accuracy-test') &&
+      contains(github.event.pull_request.labels.*.name, 'ready-for-test')
+      }}
+    uses: ./.github/workflows/_accuracy_test.yaml
+    with:
+      vllm: v0.11.0
+      runner:  linux-aarch64-${{ matrix.runner }}
+      image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc2-910b-ubuntu22.04-py3.11
+      model_name: ${{ matrix.model_name }}

vllm_ascend-0.11.0rc2/.github/workflows/format_pr_body.yaml

@@ -0,0 +1,57 @@
+#
+# Copyright (c) 2025 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.
+# This file is a part of the vllm-ascend project.
+#
+
+name: format / pr body
+
+on:
+  # The PR updated when PR opened and push new commits
+  pull_request_target:
+    types: [opened, synchronize]
+    branches:
+      - 'main'
+
+permissions:
+  pull-requests: write
+
+jobs:
+  update-description:
+    name: update vLLM version
+    runs-on: ubuntu-latest
+
+    steps:
+
+      - name: Get vLLM version
+        run: |
+          VLLM_COMMIT=v0.11.0
+          echo "VLLM_COMMIT=https://github.com/vllm-project/vllm/commit/$VLLM_COMMIT" >> $GITHUB_ENV
+
+      - name: Checkout repository
+        uses: actions/checkout@ff7abcd0c3c05ccf6adc123a8cd1fd4fb30fb493 # v4.2.2
+
+      - name: Set up Python
+        uses: actions/setup-python@e797f83bcb11b83ae66e0230d6156d7c80228e7c # v6.0.0
+
+      - name: Get vLLM release version
+        run: |
+          VLLM_VERSION=$(python3 docs/source/conf.py | jq .ci_vllm_version | tr -d '"')
+          echo "VLLM_VERSION=$VLLM_VERSION" >> $GITHUB_ENV
+
+      - name: Update PR description
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          bash .github/format_pr_body.sh "${{ github.event.number }}" "${{ env.VLLM_VERSION }}" "${{ env.VLLM_COMMIT }}"

vllm_ascend-0.11.0rc2/.github/workflows/multi_node_test.yaml

@@ -0,0 +1,118 @@
+name: 'e2e test / multi-dp'
+
+on:
+    schedule:
+      - cron: "0 */4 * * *"
+    workflow_dispatch:
+
+# Bash shells do not use ~/.profile or ~/.bashrc so these shells need to be explicitly
+# declared as "shell: bash -el {0}" on steps that need to be properly activated.
+# It's used to activate ascend-toolkit environment variables.
+defaults:
+  run:
+    shell: bash -el {0}
+
+# only cancel in-progress runs of the same workflow
+# and ignore the lint / 8 cards test type
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  e2e:
+    # This is a runner with no NPU for k8s controller
+    runs-on: linux-aarch64-a3-0
+    container:
+      image: m.daocloud.io/quay.io/ascend/cann:8.3.rc2-a3-ubuntu22.04-py3.11
+      env:
+        KUBECONFIG: /tmp/kubeconfig
+        KUBECTL: /root/.cache/.kube/kubectl
+        NAMESPACE: vllm-project
+        LEADER_POD: vllm-0
+    steps:
+        - name: Install system denpendencies
+          run: |
+           # configure apt and pip source
+           sed -i 's|ports.ubuntu.com|mirrors.tuna.tsinghua.edu.cn|g' /etc/apt/sources.list
+           pip config set global.index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
+
+           apt-get update -y && apt-get install -y git curl
+
+           TOKEN=`echo -n "x-access-token:${{ secrets.ADMIN_PTA }}" | base64`
+           git config --global http.https://gh-proxy.test.osinfra.cn/.extraheader "AUTHORIZATION: basic $TOKEN"
+
+        - name: Install kubectl
+          run: |
+            install -o root -g root -m 0755 $KUBECTL /usr/local/bin/kubectl
+
+            # get kubeconfig from secret
+            echo "${{ secrets.KUBECONFIG_B64 }}" | base64 -d > $KUBECONFIG
+
+        - name: Checkout code
+          uses: actions/checkout@v4
+
+        - name: Prepare scripts
+          run: |
+            # prepare for lws entrypoint scripts
+            install -D tests/e2e/multi_node/scripts/run.sh /root/.cache/tests/run.sh
+
+        - name: Launch cluster
+          run: |
+            kubectl apply -f tests/e2e/multi_node/scripts/lws.yaml
+          
+        - name: Waiting for pod ready
+          run: |
+            echo "waiting for Pod [$LEADER_POD] in namespace [$NAMESPACE] to Ready..."
+
+            while true; do
+              # get pod status
+              READY_STATUS=$(kubectl get pod "$LEADER_POD" -n "$NAMESPACE" -o jsonpath='{.status.containerStatuses[*].ready}')
+
+              if [[ "$READY_STATUS" == "true" ]]; then
+                echo "✅ Pod [$LEADER_POD] is Ready!"
+                break
+              else
+                echo "Pod [$LEADER_POD] not ready, waiting..."
+                sleep 3
+              fi
+            done
+
+        - name: Stream logs and monitor pod health
+          run: |
+            set -euo pipefail
+
+            echo "🚀 Start streaming logs for Pod [$LEADER_POD] ..."
+            kubectl logs -f "$LEADER_POD" -n "$NAMESPACE" &
+            LOG_PID=$!
+
+            echo "Start monitoring Pod [$LEADER_POD] status ..."
+            while true; do
+              STATUS=$(kubectl get pod "$LEADER_POD" -n "$NAMESPACE" -o jsonpath='{.status.phase}')
+              if [[ "$STATUS" != "Running" && "$STATUS" != "Succeeded" ]]; then
+                echo "❌ Pod [$LEADER_POD] exited abnormally with status: $STATUS"
+                kubectl describe pod "$LEADER_POD" -n "$NAMESPACE" || true
+                kubectl logs "$LEADER_POD" -n "$NAMESPACE" --previous --all-containers || true
+                kill $LOG_PID || true
+                exit 1
+              fi
+              sleep 5
+            done &
+
+            MONITOR_PID=$!
+            wait $LOG_PID || true
+            kill $MONITOR_PID || true
+
+        - name: Generate summary
+          if: always()
+          run: |
+            if [ -f "/root/.cache/test_summary.md" ]; then
+              cat /root/.cache/test_summary.md >> "$GITHUB_STEP_SUMMARY"
+            else
+              echo "No summary file found." >> "$GITHUB_STEP_SUMMARY"
+            fi
+
+        - name: Post process
+          if: always()
+          run: |
+            kubectl get pods -n $NAMESPACE
+            kubectl delete -f tests/e2e/multi_node/scripts/lws.yaml