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: 'e2e test'

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

concurrency:
  group: pr-${{ github.event.pull_request.number }}
  cancel-in-progress: true

jobs:
  test:
    strategy:
      max-parallel: 2
      matrix:
        os: [linux-arm64-npu-1, linux-arm64-npu-4]
        vllm_verison: [main, v0.8.5.post1]
    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_verison, 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: ${{ secrets.HF_TOKEN }}
    steps:
      - name: Check npu and CANN info
        run: |
          npu-smi info
          cat /usr/local/Ascend/ascend-toolkit/latest/"$(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_verison }}
          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
            pytest -sv tests/singlecard/test_offline_inference.py
            pytest -sv tests/singlecard/test_ilama_lora.py
            pytest -sv tests/ops
            pytest -sv tests/compile
          else
            pytest -sv -k "QwQ" tests/multicard/test_offline_inference_distributed.py
            pytest -sv tests/multicard/test_ilama_lora_tp2.py
            pytest -sv tests/ops
            pytest -sv tests/compile
          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
            pytest -sv tests/singlecard/test_ilama_lora.py
            pytest -sv tests/singlecard/test_offline_inference.py
            pytest -sv tests/ops
          else
            pytest -sv tests/multicard/test_ilama_lora_tp2.py
            pytest -sv -k "QwQ" tests/multicard/test_offline_inference_distributed.py
            pytest -sv -k "DeepSeek" tests/multicard/test_offline_inference_distributed.py
            pytest -sv tests/ops
          fi

      # only run test on spec decode when the related code changed
      - name: Check for changes in Speculative Decode
        if: github.event_name != 'schedule'
        id: filter_spec_decode
        uses: dorny/paths-filter@v3
        with:
          filters: |
            speculative_tests_changed:
              - ".github/workflows/vllm_ascend_test.yaml"
              - "tests/singlecard/spec_decode/**"
              - "tests/multicard/spec_decode_e2e/**"
              - "vllm_ascend/worker/worker.py"
              - "vllm_ascend/worker/model_runner.py"
              - "vllm_ascend/worker/multi_step_runner.py"
              - "vllm_ascend/worker/multi_step_worker.py"
              - "vllm_ascend/worker/draft_model_runner.py"
              - "vllm_ascend/patch/worker/patch_common/patch_metrics.py"
              - "vllm_ascend/patch/worker/patch_common/patch_spec_decode_worker.py"
              - "vllm_ascend/patch/worker/patch_common/patch_multi_step_worker.py"

      - name: Run vllm-project/vllm-ascend Speculative Decode test
        if: steps.filter_spec_decode.outputs.speculative_tests_changed == 'true' || github.event_name == 'schedule'
        run: |
          if [[ "${{ matrix.os }}" == "linux-arm64-npu-1" ]]; then
            VLLM_USE_MODELSCOPE=true pytest -sv tests/singlecard/spec_decode/e2e/test_v1_spec_decode.py
            pytest -sv tests/singlecard/spec_decode/e2e/test_mtp_correctness.py  # it needs a clean process
            pytest -sv tests/singlecard/spec_decode --ignore=tests/singlecard/spec_decode/e2e/test_mtp_correctness.py --ignore=tests/singlecard/spec_decode/e2e/test_v1_spec_decode.py
          fi

      - name: Run vllm-project/vllm test for V0 Engine
        env:
          VLLM_USE_V1: 0
          PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
        run: |
          pytest -sv