Fix TransformerEngine and Activation checkpointing interaction for thunderFX #5033
Workflow file for this run
This file contains 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: CI testing | |
| # see: https://help.github.com/en/actions/reference/events-that-trigger-workflows | |
| on: # Trigger the workflow on push or pull request, but only for the main branch | |
| push: | |
| branches: [main] | |
| pull_request: {} | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref }} | |
| cancel-in-progress: ${{ github.ref != 'refs/heads/main' }} | |
| defaults: | |
| run: | |
| shell: bash | |
| env: | |
| #CI: "true" | |
| TORCH_URL_RC: "https://download.pytorch.org/whl/test/cpu/torch_test.html" | |
| TORCH_URL_NIGHTLY: "https://download.pytorch.org/whl/nightly/cpu" | |
| TORCH_URL_STABLE: "https://download.pytorch.org/whl/cpu" | |
| jobs: | |
| pytester: | |
| runs-on: ${{ matrix.os }} | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| os: ["ubuntu-22.04", "macOS-14", "windows-2022"] | |
| python-version: ["3.10"] | |
| requires: ["latest", "nightly"] # , 'oldest' | |
| suite: ["core", "ops"] | |
| include: | |
| - { os: "ubuntu-22.04", python-version: "3.11", requires: "latest" } | |
| - { os: "ubuntu-22.04", python-version: "3.12", requires: "latest" } | |
| exclude: | |
| - { os: "windows-2022", suite: "ops" } | |
| # Timeout: https://stackoverflow.com/a/59076067/4521646 | |
| timeout-minutes: 35 | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Set up Python ${{ matrix.python-version }} | |
| uses: actions/setup-python@v5 | |
| with: | |
| python-version: ${{ matrix.python-version }} | |
| - name: Setup Ubuntu | |
| if: runner.os == 'ubuntu' | |
| run: | | |
| sudo apt-get install -y graphviz | |
| - name: Set min. dependencies | |
| if: matrix.requires == 'oldest' | |
| run: | | |
| for fpath in ('requirements/base.txt', 'requirements/test.txt'): | |
| req = open(fpath).read().replace('>=', '==') | |
| open(fpath, 'w').write(req) | |
| shell: python | |
| - name: Get pip cache dir | |
| id: pip-cache | |
| run: echo "dir=$(pip cache dir)" >> $GITHUB_OUTPUT | |
| - name: Cache pip | |
| uses: actions/cache@v4 | |
| with: | |
| path: ${{ steps.pip-cache.outputs.dir }} | |
| key: ${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}-pip-${{ hashFiles('requirements/*.txt') }} | |
| restore-keys: | | |
| ${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}-pip- | |
| - name: switch Torch source | |
| run: | | |
| if [[ "${{ matrix.requires }}" == "nightly" ]]; then | |
| echo "TORCH_URL=$TORCH_URL_NIGHTLY" >> $GITHUB_ENV | |
| echo "PIP_EXTRA_FLAG=--pre" >> $GITHUB_ENV | |
| else | |
| echo "TORCH_URL=$TORCH_URL_STABLE" >> $GITHUB_ENV | |
| fi | |
| - name: Install package & dependencies | |
| run: | | |
| if [[ "${{ runner.os }}" == "macOS" ]]; then | |
| brew install libomp | |
| fi | |
| pip --version | |
| pip install -U \ | |
| ${PIP_EXTRA_FLAG} torch torchvision torchaudio \ | |
| --index-url=${TORCH_URL} | |
| pip install -e . -U \ | |
| ${PIP_EXTRA_FLAG} -r requirements/test.txt | |
| pip list | |
| shell: bash | |
| - name: Testing Local | |
| if: matrix.python-version == '3.10' && matrix.suite == 'core' | |
| run: | | |
| coverage run --source thunder -m \ | |
| pytest thunder/tests/ \ | |
| --ignore=thunder/tests/distributed \ | |
| --ignore=thunder/tests/test_ops.py \ | |
| --ignore=thunder/tests/test_grad.py \ | |
| -v --datefmt="%Y%m%d-%H:%M:%S.%f" \ | |
| --random-order-seed=$GITHUB_RUN_ID \ | |
| -n 4 --durations=250 | |
| - name: Testing Distributed | |
| # run all found tests in given past as standalone | |
| if: matrix.python-version == '3.10' && runner.os == 'Linux' && matrix.suite == 'core' | |
| run: | | |
| pytest thunder/tests/distributed/ \ | |
| -v --datefmt="%Y%m%d-%H:%M:%S.%f" \ | |
| --random-order-seed=$GITHUB_RUN_ID \ | |
| --durations=250 | |
| - name: Testing Ops | |
| if: matrix.python-version == '3.10' && matrix.suite == 'ops' && runner.os != 'Windows' | |
| run: | | |
| coverage run --source thunder -m \ | |
| pytest thunder/tests/test_ops.py thunder/tests/test_grad.py \ | |
| -v --datefmt="%Y%m%d-%H:%M:%S.%f" \ | |
| --random-order-seed=$GITHUB_RUN_ID \ | |
| -n 4 --durations=250 | |
| - name: Testing interpreter | |
| if: matrix.python-version == '3.11' || matrix.python-version == '3.12' | |
| #continue-on-error: true | |
| run: | | |
| python -m pytest \ | |
| thunder/tests/test_interpreter.py \ | |
| -v --durations=50 --cov=thunder | |
| python -m pytest thunder/tests/test_jit_general.py -v --durations=50 --cov=thunder | |
| - name: Statistics | |
| run: | | |
| coverage report | |
| coverage xml | |
| - name: Upload coverage to Codecov | |
| uses: codecov/codecov-action@v4 | |
| with: | |
| token: ${{ secrets.CODECOV_TOKEN }} | |
| file: ./coverage.xml | |
| flags: unittests | |
| env_vars: OS,PYTHON | |
| name: codecov-umbrella | |
| fail_ci_if_error: false | |
| testing-guardian: | |
| runs-on: ubuntu-latest | |
| needs: pytester | |
| if: always() | |
| steps: | |
| - run: echo "${{ needs.pytester.result }}" | |
| - name: failing... | |
| if: needs.pytester.result == 'failure' | |
| run: exit 1 | |
| - name: cancelled or skipped... | |
| if: contains(fromJSON('["cancelled", "skipped"]'), needs.pytester.result) | |
| timeout-minutes: 1 | |
| run: sleep 90 |