pytorch/.github/workflows/docker-builds.yml

name: docker-builds

on:
  workflow_dispatch:
  pull_request:
    paths:
      - .ci/docker/**
      - .github/workflows/docker-builds.yml
      - .lintrunner.toml
  push:
    branches:
      - main
      - release/*
      - landchecks/*
    paths:
      - .ci/docker/**
      - .github/workflows/docker-builds.yml
      - .lintrunner.toml
  schedule:
    - cron: 1 3 * * 3

concurrency:
  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.sha }}-${{ github.event_name == 'workflow_dispatch' }}
  cancel-in-progress: true

env:
  ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine
  AWS_DEFAULT_REGION: us-east-1

permissions: read-all

jobs:
  get-label-type:
    if: github.repository_owner == 'pytorch'
    name: get-label-type
    uses: pytorch/pytorch/.github/workflows/_runner-determinator.yml@main
    with:
      triggering_actor: ${{ github.triggering_actor }}
      issue_owner: ${{ github.event.pull_request.user.login || github.event.issue.user.login }}
      curr_branch: ${{ github.head_ref || github.ref_name }}
      curr_ref_type: ${{ github.ref_type }}

  docker-build:
    environment: ${{ (github.ref == 'refs/heads/main' || startsWith(github.event.ref, 'refs/tags/v')) && 'docker-build' || '' }}
    needs: get-label-type
    timeout-minutes: 240
    strategy:
      fail-fast: false
      matrix:
        runner: [linux.12xlarge]
        docker-image-name: [
          pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks,
          pytorch-linux-focal-cuda12.4-cudnn9-py3.12-gcc9-inductor-benchmarks,
          pytorch-linux-focal-cuda12.4-cudnn9-py3.13-gcc9-inductor-benchmarks,
          pytorch-linux-focal-cuda12.6-cudnn9-py3-gcc11,
          pytorch-linux-focal-cuda12.6-cudnn9-py3-gcc9-inductor-benchmarks,
          pytorch-linux-focal-cuda12.6-cudnn9-py3.12-gcc9-inductor-benchmarks,
          pytorch-linux-focal-cuda12.6-cudnn9-py3.13-gcc9-inductor-benchmarks,
          pytorch-linux-focal-cuda11.8-cudnn9-py3-gcc9,
          pytorch-linux-focal-py3.9-clang10,
          pytorch-linux-focal-py3.11-clang10,
          pytorch-linux-focal-py3.12-clang10,
          pytorch-linux-focal-py3.13-clang10,
          pytorch-linux-jammy-rocm-n-1-py3,
          pytorch-linux-jammy-rocm-n-py3,
          pytorch-linux-jammy-cuda11.8-cudnn9-py3.9-clang12,
          pytorch-linux-jammy-py3.9-gcc11,
          pytorch-linux-jammy-py3.9-gcc11-inductor-benchmarks,
          pytorch-linux-jammy-py3.12-halide,
          pytorch-linux-jammy-xpu-2024.0-py3,
          pytorch-linux-jammy-xpu-2025.0-py3,
          pytorch-linux-jammy-py3-clang15-asan,
          pytorch-linux-jammy-py3-clang18-asan,
          pytorch-linux-focal-py3-clang10-onnx,
          pytorch-linux-focal-linter,
          pytorch-linux-jammy-cuda11.8-cudnn9-py3.9-linter,
          pytorch-linux-jammy-py3-clang12-executorch,
          pytorch-linux-jammy-py3.12-triton-cpu
        ]
        include:
          - docker-image-name: pytorch-linux-jammy-aarch64-py3.10-gcc11
            runner: linux.arm64.m7g.4xlarge
          - docker-image-name: pytorch-linux-jammy-aarch64-py3.10-gcc11-inductor-benchmarks
            runner: linux.arm64.m7g.4xlarge
            timeout-minutes: 600
    # Docker uploads fail from LF runners, see https://github.com/pytorch/pytorch/pull/137358
    # runs-on: "${{ needs.get-label-type.outputs.label-type }}${{ matrix.runner }}"
    runs-on: "${{ matrix.runner }}"
    steps:
      - name: Clean workspace
        shell: bash
        run: |
          echo "${GITHUB_WORKSPACE}"
          sudo rm -rf "${GITHUB_WORKSPACE}"
          mkdir "${GITHUB_WORKSPACE}"

      # [see note: pytorch repo ref]
      # deep clone (fetch-depth 0) required for git merge-base
      - name: Checkout PyTorch
        uses: pytorch/pytorch/.github/actions/checkout-pytorch@main

      - name: Setup Linux
        uses: ./.github/actions/setup-linux

      - name: Build docker image
        id: build-docker-image
        uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
        with:
          docker-image-name: ci-image:${{ matrix.docker-image-name }}
          always-rebuild: true
          push: true

      - name: Push docker image to old name
        shell: bash
        env:
          ECR_DOCKER_IMAGE: ${{ steps.build-docker-image.outputs.docker-image }}
        run: |
          # This can be deleted after people have rebased their PRs/the new name has been in main for a while
          set -euox pipefail
          docker_image_name="308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/${{ matrix.docker-image-name }}"
          foldersha=${ECR_DOCKER_IMAGE##*-}
          docker tag "${ECR_DOCKER_IMAGE}" "${docker_image_name}:${foldersha}"
          docker push "${docker_image_name}:${foldersha}"

      - name: Pull docker image
        uses: pytorch/test-infra/.github/actions/pull-docker-image@main
        with:
          docker-image: ${{ steps.build-docker-image.outputs.docker-image }}

      - uses: nick-fields/retry@7152eba30c6575329ac0576536151aca5a72780e # v3.0.0
        name: Push to https://ghcr.io/
        id: push-to-ghcr-io
        if: ${{ github.event_name == 'push' }}
        env:
          ECR_DOCKER_IMAGE: ${{ steps.build-docker-image.outputs.docker-image }}
          GHCR_PAT: ${{ secrets.GHCR_PAT }}
        with:
          shell: bash
          timeout_minutes: 30
          max_attempts: 5
          retry_wait_seconds: 90
          command: |
            ghcr_image="ghcr.io/pytorch/ci-image"
            tag=${ECR_DOCKER_IMAGE##*:}
            # Push docker image to the ghcr.io
            echo $GHCR_PAT | docker login ghcr.io -u pytorch --password-stdin
            docker tag "${ECR_DOCKER_IMAGE}" "${ghcr_image}:${tag}"
            docker push "${ghcr_image}:${tag}"

      - name: Chown workspace
        uses: ./.github/actions/chown-workspace
        with:
          ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/${{ contains(matrix.runner, 'arm64') && 'arm64v8' || 'tool' }}/alpine
        if: always()

      - name: Teardown Linux
        uses: pytorch/test-infra/.github/actions/teardown-linux@main
        if: always()