xc-llm-ascend/.github/workflows/vllm_ascend_test.yaml

#
# 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: 'test'

on:
  schedule:
    - cron: '0 23 * * *'
  pull_request:
    branches:
      - 'main'
      - '*-dev'
    paths:
      - '*.txt'
      - '**/*.py'
      - '.github/workflows/vllm_ascend_test.yaml'
      - '!docs/**'
      - 'pytest.ini'
      - '!benchmarks/**'
      - 'tools/mypy.sh'
      - 'mypy.ini'

# 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}

jobs:
  lint:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        python-version: ["3.10"]
    steps:
      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
      - name: Set up Python ${{ matrix.python-version }}
        uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0
        with:
          python-version: ${{ matrix.python-version }}
      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -r requirements-lint.txt
      - name: Run codespell check
        run: |
          CODESPELL_EXCLUDES=('--skip' 'tests/prompts/**,./benchmarks/sonnet.txt,*tests/lora/data/**,build/**,./vllm_ascend.egg-info/**')
          CODESPELL_IGNORE_WORDS=('-L' 'CANN,cann,NNAL,nnal,ASCEND,ascend,EnQue,CopyIn')

          codespell --toml pyproject.toml "${CODESPELL_EXCLUDES[@]}" "${CODESPELL_IGNORE_WORDS[@]}"
      - name: Analysing the code with ruff
        run: |
          echo "::add-matcher::.github/workflows/matchers/ruff.json"
          ruff check --output-format github .
      - name: Run isort
        run: |
          isort . --check-only
      - name: Running yapf
        run: |
          python -m pip install --upgrade pip
          pip install toml
          pip install yapf==0.32.0
          yapf --diff --recursive .

      - name: Install dependencies
        run: |
          pip install -r requirements-dev.txt --extra-index-url https://download.pytorch.org/whl/cpu

      - name: Checkout vllm-project/vllm repo
        uses: actions/checkout@v4
        with:
          repository: vllm-project/vllm
          path: vllm-empty

      - name: Install vllm-project/vllm from source
        working-directory: vllm-empty
        run: |
          pip install -r requirements/build.txt --extra-index-url https://download.pytorch.org/whl/cpu
          VLLM_TARGET_DEVICE=empty pip install .

      - name: Mypy Check
        run: |
          echo "::add-matcher::.github/workflows/matchers/mypy.json"
          tools/mypy.sh 1 ${{ matrix.python-version }}

  e2e:
    needs: [lint]
    if: ${{ needs.lint.result == 'success' }}
    strategy:
      max-parallel: 2
      matrix:
        os: [linux-arm64-npu-1, linux-arm64-npu-4]
        vllm_version: [main, v0.9.0]
    concurrency:
      group: >
        ${{
        matrix.os == 'linux-arm64-npu-4'
          && github.event.pull_request.number
          && format('pr-{0}-limit-npu-4', github.event.pull_request.number)
        || format('job-{0}-{1}-{2}', matrix.os, matrix.vllm_version, github.event.pull_request.number)
        }}
      cancel-in-progress: false
    name: vLLM Ascend test
    runs-on: ${{ matrix.os }}
    container:
      # TODO(yikun): Remove m.daocloud.io prefix when infra proxy ready
      image: m.daocloud.io/quay.io/ascend/cann:8.1.rc1-910b-ubuntu22.04-py3.10
      env:
        HF_ENDPOINT: https://hf-mirror.com
        HF_TOKEN: ${{ secrets.HF_TOKEN }}
        VLLM_LOGGING_LEVEL: ERROR
    steps:
      - name: Check npu and CANN info
        run: |
          npu-smi info
          cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info

      - name: Config mirrors
        run: |
          sed -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 install git -y
          git config --global url."https://gh-proxy.test.osinfra.cn/https://github.com/".insteadOf https://github.com/

      - 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: ${{ matrix.vllm_version }}
          path: ./vllm-empty

      - 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
        run: |
          pip install -r requirements-dev.txt
          pip install -v -e .

      - name: Run vllm-project/vllm-ascend test for V1 Engine
        env:
          VLLM_USE_V1: 1
          VLLM_WORKER_MULTIPROC_METHOD: spawn
        run: |
          if [[ "${{ matrix.os }}" == "linux-arm64-npu-1" ]]; then
            VLLM_USE_MODELSCOPE=True pytest -sv tests/singlecard/test_offline_inference.py
            pytest -sv tests/singlecard/test_scheduler.py
            # guided decoding doesn't work, fix it later
            # pytest -sv tests/singlecard/test_guided_decoding.py.py
            # test_ascend_config.py should be ran separately because it will regenerate the global config many times.
            pytest -sv tests/singlecard/test_ascend_config.py
            pytest -sv tests/singlecard/test_camem.py
            pytest -sv tests/singlecard/ \
            --ignore=tests/singlecard/test_offline_inference.py \
            --ignore=tests/singlecard/test_scheduler.py \
            --ignore=tests/singlecard/test_guided_decoding.py \
            --ignore=tests/singlecard/test_ascend_config.py \
            --ignore=tests/singlecard/test_camem.py
          else
            pytest -sv tests/multicard/test_ilama_lora_tp2.py
            # To avoid oom, we need to run the test in a single process.
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_QwQ
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_topk
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_W8A8
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/ --ignore=tests/multicard/test_ilama_lora_tp2.py --ignore=tests/multicard/test_offline_inference_distributed.py
          fi

      - name: Run vllm-project/vllm-ascend test on V0 engine
        env:
          VLLM_USE_V1: 0
        run: |
          if [[ "${{ matrix.os }}" == "linux-arm64-npu-1" ]]; then
            VLLM_USE_MODELSCOPE=True  pytest -sv tests/singlecard/test_offline_inference.py
            pytest -sv tests/singlecard/test_scheduler.py
            # guided decoding doesn't work, fix it later
            # pytest -sv tests/singlecard/test_guided_decoding.py.py
            pytest -sv tests/singlecard/test_camem.py
            # test_ascend_config.py should be ran separately because it will regenerate the global config many times.
            pytest -sv tests/singlecard/test_ascend_config.py
            pytest -sv tests/singlecard/test_prompt_embedding.py
            pytest -sv tests/singlecard/ \
              --ignore=tests/singlecard/test_offline_inference.py \
              --ignore=tests/singlecard/test_scheduler.py \
              --ignore=tests/singlecard/test_guided_decoding.py \
              --ignore=tests/singlecard/test_camem.py \
              --ignore=tests/singlecard/test_ascend_config.py \
              --ignore=tests/singlecard/test_prompt_embedding.py
          else
            pytest -sv tests/multicard/test_ilama_lora_tp2.py
            # Fixme: run VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py will raise error.
            # To avoid oom, we need to run the test in a single process.
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_QwQ
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_topk
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_W8A8
            VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/ --ignore=tests/multicard/test_ilama_lora_tp2.py --ignore=tests/multicard/test_offline_inference_distributed.py
          fi