ci: debug2 #2
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 CPP Distributed | |
| # 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-cpp_distributed: | |
| # runs-on: self-hosted | |
| # defaults: | |
| # run: | |
| # shell: bash | |
| # container: | |
| # image: localhost:5000/flagscale:cuda12.8.1-cudnn9.10.2-torch2.9.1-py3.10-transformerengine_fl | |
| # ports: | |
| # - 80:80 | |
| # options: >- | |
| # --gpus all | |
| # --shm-size=500g | |
| # --privileged | |
| # --ipc=host | |
| # --ulimit memlock=-1 | |
| # --ulimit stack=67108864 | |
| # --ulimit nofile=65535:65535 | |
| # --user root | |
| # steps: | |
| # - name: Checkout Code | |
| # uses: actions/checkout@v6.0.1 | |
| # with: | |
| # repository: ${{ github.event.pull_request.head.repo.full_name }} | |
| # ref: ${{ github.event.pull_request.head.ref }} | |
| # ssh-strict: true | |
| # ssh-user: git | |
| # persist-credentials: true | |
| # clean: true | |
| # sparse-checkout-cone-mode: true | |
| # fetch-tags: false | |
| # show-progress: true | |
| # lfs: false | |
| # submodules: recursive | |
| # set-safe-directory: true | |
| # - name: Pre-Check Environment | |
| # timeout-minutes: 5 | |
| # run: | | |
| # set -euo pipefail | |
| # echo "===== System Info =====" | |
| # uname -a | |
| # python3 --version || (echo "❌ python3 not found"; exit 1) | |
| # pip3 --version || (echo "❌ pip3 not found"; exit 1) | |
| # echo "===== GPU Info =====" | |
| # if command -v nvidia-smi &>/dev/null; then | |
| # nvidia-smi --query-gpu=index,name,memory.total,utilization.gpu --format=csv | |
| # else | |
| # echo "⚠️ GPU check skipped. nvidia-smi is not available in the current environment." | |
| # exit 1 | |
| # fi | |
| # nvcc --version || echo "⚠️ nvcc not found (non-fatal for python-only install)" | |
| # echo "===== Disk Space =====" | |
| # df -h | |
| # echo "===== Available memory =====" | |
| # free -h | |
| # - name: Install dependencies and build transformer_engine | |
| # # timeout-minutes: 30 | |
| # env: | |
| # NVTE_FRAMEWORK: pytorch | |
| # TE_WITH_NCCL: 1 | |
| # run: | | |
| # # =============== Install MPI =============== | |
| # apt update | |
| # apt install -y libopenmpi-dev openmpi-bin openmpi-common | |
| # apt install -y libmpich-dev mpich | |
| # # Verify the MPI header file | |
| # mpicxx -show | awk '{for(i=1;i<=NF;i++) if($i ~ /-I/) print substr($i,3)}' | |
| # # Verify whether the MPI C++ environment is ready | |
| # # 1. Verify whether the MPI C++ compiler (mpicxx) exists | |
| # mpicxx --version | |
| # # 2. Verify if the MPI library file exists | |
| # ls /usr/lib/x86_64-linux-gnu/libmpi_cxx.so | |
| # echo "=============== Install transformer_engine ===============" | |
| # pip install --no-build-isolation -vvv . --no-deps | |
| # cp -rf transformer_engine/plugins/ /usr/local/lib/python3.10/dist-packages/transformer_engine/plugins/ | |
| # - name: Verify installation | |
| # shell: bash | |
| # run: | | |
| # python3 tests/pytorch/test_sanity_import.py | |
| # - name: GPU Usage Check / Verification | |
| # run: | | |
| # source .github/workflows/scripts/gpu_check.sh | |
| # wait_for_gpu | |
| # - name: L0 CPP Distributed | |
| # id: L0_cpp_distributed | |
| # continue-on-error: true | |
| # # timeout-minutes: 10 | |
| # env: | |
| # TE_PATH: . | |
| # run: | | |
| # TE_LIB_PATH=$(pip3 show transformer-engine | grep -E "Location:|Editable project location:" | tail -n 1 | awk '{print $NF}') | |
| # TE_CPP_LIB_PATH="${TE_LIB_PATH}/transformer_engine" | |
| # export CMAKE_PREFIX_PATH="${TE_CPP_LIB_PATH}:${CMAKE_PREFIX_PATH}" | |
| # export LD_LIBRARY_PATH="${TE_CPP_LIB_PATH}:${LD_LIBRARY_PATH}" | |
| # cd $TE_PATH/tests/cpp_distributed | |
| # cmake -GNinja -S. -Bbuild |