[ci] Add workflows to validate TE QA test cases #5
Workflow file for this run
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: QA Pytorch Wheel | |
| on: | |
| push: | |
| branches: | |
| - main | |
| pull_request: | |
| branches: | |
| - main | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-${{ github.actor }} | |
| cancel-in-progress: true | |
| jobs: | |
| qa-l0-pytorch-wheel: | |
| runs-on: [ self-hosted, Linux, X64, nvidia, gpu-8 ] | |
| defaults: | |
| run: | |
| shell: bash | |
| container: | |
| image: harbor.baai.ac.cn/flagscale/cuda12.8.1-torch2.7.1-python3.10-te2.9:20260209 | |
| ports: | |
| - 80:80 | |
| options: >- | |
| --gpus all | |
| --shm-size=500g | |
| --privileged | |
| --ipc=host | |
| --ulimit memlock=-1 | |
| --ulimit stack=67108864 | |
| --ulimit nofile=65535:65535 | |
| --user root | |
| --pull always | |
| steps: | |
| - name: Checkout Code | |
| uses: actions/checkout@v6.0.1 | |
| with: | |
| repository: ${{ github.event.pull_request.head.repo.full_name }} | |
| ref: ${{ github.event.pull_request.head.ref }} | |
| ssh-strict: true | |
| ssh-user: git | |
| persist-credentials: true | |
| clean: true | |
| sparse-checkout-cone-mode: true | |
| fetch-tags: false | |
| show-progress: true | |
| lfs: false | |
| submodules: recursive | |
| set-safe-directory: true | |
| # - name: L0 Pytorch Wheel | |
| # id: L0_pytoech_wheel | |
| # # timeout-minutes: 50 | |
| # env: | |
| # TE_PATH: . | |
| # RUN_LOG: /logs/pytorch/wheel | |
| # run: | | |
| # echo "TE_PATH: ${TE_PATH}" | |
| # sed -i "s/^cd transformer_engine\/pytorch\s*$/pushd transformer_engine\/pytorch/" qa/L0_pytorch_wheel/test.sh | |
| # sed -i '44 s/^cd \s*\$TE_PATH\s*$/popd/' qa/L0_pytorch_wheel/test.sh | |
| # cat qa/L0_pytorch_wheel/test.sh | |
| # source /opt/miniconda3/etc/profile.d/conda.sh | |
| # conda activate flagscale-train | |
| # pip uninstall -y transformer_engine | |
| # bash qa/L0_pytorch_wheel/test.sh | tee ${RUN_LOG}/pytorch_wheel-${{ github.run_id }}.log | |
| # - name: Upload Installation Logs | |
| # if: always() && steps.L0_pytoech_wheel.outcome == 'failure' | |
| # uses: actions/upload-artifact@v4 | |
| # with: | |
| # name: L0-pytorch-logs-${{ github.run_id }} | |
| # path: /logs/pytorch/wheel | |
| # retention-days: 7 | |
| # if-no-files-found: warn |