docker-build.yml

name: "Docker Build & Deploy"
on:
  pull_request:
  push:
    branches:
      - "inference"
  workflow_dispatch:

# Cancel outdated workflows if they are still running
concurrency:
  group: docker-build-${{ github.head_ref || github.run_id }}
  cancel-in-progress: true

jobs:
  docker-build-rocm:
    name: Build and Install FlexFlow in a Docker Container (ROCm backend)
    runs-on: ubuntu-22.04
    if: ${{ ( github.event_name != 'push' && github.event_name != 'schedule' && github.event_name != 'workflow_dispatch' ) || github.ref_name != 'inference' }}
    env:
      FF_GPU_BACKEND: "hip_rocm"
      hip_version: 5.6
    steps:
      - name: Checkout Git Repository
        uses: actions/checkout@v3
        with:
          submodules: recursive

      - name: Free additional space on runner
        run: .github/workflows/helpers/free_space_on_runner.sh

      - name: Build Docker container
        run: FF_HIP_ARCH="gfx1100,gfx1036" ./docker/build.sh flexflow

      - name: Check availability of flexflow modules in Python
        run: docker run --entrypoint /bin/bash flexflow-${FF_GPU_BACKEND}-${hip_version}:latest -c "python -c 'import flexflow.core; import flexflow.serve as ff; exit()'"

  docker-build-and-publish-rocm:
    name: Build and Deploy FlexFlow Docker Containers (ROCm backend)
    runs-on: "runs-on=${{ github.run_id }}/runner=rocm-builder"
    if: ${{ ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' }}
    strategy:
      matrix:
        hip_version: ["5.6"]
      fail-fast: false
    env:
      FF_GPU_BACKEND: "hip_rocm"
      hip_version: ${{ matrix.hip_version }}
    steps:
      - name: Checkout Git Repository
        uses: actions/checkout@v3
        with:
          submodules: recursive

      - name: Build Docker container
        # On push to inference, build for all compatible architectures, so that we can publish 
        # a pre-built general-purpose image. On all other cases, only build for one architecture
        # to save time.
        run: FF_HIP_ARCH=all ./docker/build.sh flexflow

      - name: Check availability of flexflow modules in Python
        run: docker run --entrypoint /bin/bash flexflow-${FF_GPU_BACKEND}-${hip_version}:latest -c "python -c 'import flexflow.core; import flexflow.serve as ff; exit()'"

      - name: Publish Docker environment image (on push to inference)
        env:
          FLEXFLOW_CONTAINER_TOKEN: ${{ secrets.FLEXFLOW_CONTAINER_TOKEN }}
        run: |
          ./docker/publish.sh flexflow-environment
          ./docker/publish.sh flexflow
  
  docker-build-cuda:
    name: Build and Install FlexFlow in a Docker Container (CUDA backend)
    runs-on: ubuntu-22.04
    strategy:
      matrix:
        cuda_version: ["11.8", "12.0", "12.1", "12.2"]
      fail-fast: false
    env:
      FF_GPU_BACKEND: "cuda"
      cuda_version: ${{ matrix.cuda_version }}
    steps:
      - name: Checkout Git Repository
        if: ${{ ( ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' ) || matrix.cuda_version == '12.0' }}
        uses: actions/checkout@v3
        with:
          submodules: recursive

      - name: Free additional space on runner
        if: ${{ ( ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' ) || matrix.cuda_version == '12.0' }}
        run: .github/workflows/helpers/free_space_on_runner.sh

      - name: Build Docker container
        if: ${{ ( ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' ) || matrix.cuda_version == '12.0' }}
        env:
          deploy_needed: ${{ ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' }}
          build_needed: ${{ matrix.cuda_version == '12.0' }}
        run: |
          # On push to inference, build for all compatible architectures, so that we can publish 
          # a pre-built general-purpose image. On all other cases, only build for one architecture
          # to save time.
          if [[ $deploy_needed == "true" ]] ; then
            export FF_CUDA_ARCH=all
            ./docker/build.sh flexflow
          elif [[ $build_needed == "true" ]]; then
            export FF_CUDA_ARCH=86
            ./docker/build.sh flexflow
          fi

      - name: Check availability of flexflow modules in Python
        if: ${{ ( ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' ) || matrix.cuda_version == '12.0' }}
        run: docker run --entrypoint /bin/bash flexflow-${FF_GPU_BACKEND}-${cuda_version}:latest -c "export LD_LIBRARY_PATH=/usr/local/cuda/lib64/stubs:$LD_LIBRARY_PATH; sudo ln -s /usr/local/cuda/lib64/stubs/libcuda.so /usr/local/cuda/lib64/stubs/libcuda.so.1; python -c 'import flexflow.core; import flexflow.serve as ff; exit()'"

      - name: Publish Docker environment image (on push to inference)
        if: ${{ github.repository_owner == 'flexflow' && ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' }}
        env:
          FLEXFLOW_CONTAINER_TOKEN: ${{ secrets.FLEXFLOW_CONTAINER_TOKEN }}
        run: |
          ./docker/publish.sh flexflow-environment
          ./docker/publish.sh flexflow

  notify-slack:
    name: Notify Slack in case of failure
    runs-on: ubuntu-22.04
    needs: [docker-build-cuda, docker-build-and-publish-rocm]
    if: ${{ failure() && github.repository_owner == 'flexflow' && ( github.event_name == 'push' || github.event_name == 'workflow_dispatch' ) && github.ref_name == 'inference' }} 
    steps:
      - name: Send Slack message
        env:
          SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
        run: |
          curl -X POST -H 'Content-type: application/json' --data "{\"text\":\"flexflow-serve Docker images build failed! <https://github.com/flexflow/flexflow-serve/actions/runs/$GITHUB_RUN_ID|(See here).> :x: \"}" $SLACK_WEBHOOK