diff --git a/.github/workflows/release-docker-amd-rocm720-nightly.yml b/.github/workflows/release-docker-amd-rocm720-nightly.yml index d801c77cc287..b42b543562b2 100644 --- a/.github/workflows/release-docker-amd-rocm720-nightly.yml +++ b/.github/workflows/release-docker-amd-rocm720-nightly.yml @@ -1,6 +1,16 @@ name: Release Docker Images Nightly ROCm7.2 (AMD) on: workflow_dispatch: + inputs: + job_select: + description: 'Select which release job to run' + required: false + type: choice + default: 'all' + options: + - 'all' + - publish + - publish_dsv4 schedule: - cron: '0 12 * * *' @@ -14,7 +24,7 @@ concurrency: jobs: publish: - if: github.repository == 'sgl-project/sglang' + if: github.repository == 'sgl-project/sglang' && (github.event_name != 'workflow_dispatch' || inputs.job_select == 'all' || inputs.job_select == 'publish') runs-on: amd-docker-scale environment: 'prod' strategy: @@ -123,7 +133,7 @@ jobs: # of publish failed; legs without an artifact will fail at download and be # the only ones marked red. push_local_registry: - if: ${{ !cancelled() && github.repository == 'sgl-project/sglang' }} + if: ${{ !cancelled() && github.repository == 'sgl-project/sglang' && (github.event_name != 'workflow_dispatch' || inputs.job_select == 'all' || inputs.job_select == 'publish') }} runs-on: linux-mi300-1gpu-sglang environment: 'prod' needs: publish @@ -160,3 +170,90 @@ jobs: docker pull "${src}" docker tag "${src}" "${dst}" docker push "${dst}" + + publish_dsv4: + if: github.repository == 'sgl-project/sglang' && (github.event_name != 'workflow_dispatch' || inputs.job_select == 'all' || inputs.job_select == 'publish_dsv4') + runs-on: amd-docker-scale + environment: 'prod' + strategy: + fail-fast: false + matrix: + gpu_arch: ['gfx942-rocm720', 'gfx950-rocm720'] + build_type: ['all'] + steps: + - name: Checkout repository + uses: actions/checkout@v4 + with: + ref: amd/deepseek_v4 + fetch-depth: 0 # Required for git describe to find tags + + - name: Set up Python + uses: actions/setup-python@v5 + with: + python-version: "3.10" + + - name: "Set Date" + run: | + echo "DATE=$(date +%Y%m%d)" >> $GITHUB_ENV + + - name: Get version from latest tag + id: version + run: | + # Use the shared helper so stable/post releases sort above rc tags. + VERSION=$(python3 python/tools/get_version_tag.py --tag-only | sed 's/^v//') + + if [ -z "$VERSION" ]; then + echo "::error::Could not determine version from git tags" + exit 1 + fi + + # Get short commit hash of current HEAD + COMMIT_SHA=$(git rev-parse HEAD) + COMMIT_HASH=${COMMIT_SHA:0:7} + + # Compose pretend version for setuptools_scm: e.g., 0.5.8.post1.dev20260211+g1a2b3c4 + PRETEND_VERSION="${VERSION}.dev${{ env.DATE }}+g${COMMIT_HASH}" + + echo "commit_sha=${COMMIT_SHA}" >> "$GITHUB_OUTPUT" + echo "commit_hash=${COMMIT_HASH}" >> "$GITHUB_OUTPUT" + echo "version=${VERSION}" >> "$GITHUB_OUTPUT" + echo "pretend_version=${PRETEND_VERSION}" >> "$GITHUB_OUTPUT" + echo "DeepSeek V4 commit: ${COMMIT_SHA}" + echo "Detected version: ${VERSION}" + echo "Pretend version for pip: ${PRETEND_VERSION}" + + - name: Login to Docker Hub + uses: docker/login-action@v2 + with: + username: ${{ secrets.DOCKERHUB_AMD_USERNAME }} + password: ${{ secrets.DOCKERHUB_AMD_TOKEN }} + + - name: Build and Push DSv4 image to rocm/sgl-dev + run: | + version=${{ steps.version.outputs.version }} + pretend_version=${{ steps.version.outputs.pretend_version }} + echo "Version: ${version}" + echo "Pretend version: ${pretend_version}" + + if [ "${{ matrix.gpu_arch }}" = "gfx942-rocm720" ]; then + rocm_tag="rocm720-mi30x" + elif [ "${{ matrix.gpu_arch }}" = "gfx950-rocm720" ]; then + rocm_tag="rocm720-mi35x" + else + echo "Unsupported gfx arch" + exit 1 + fi + + image_tag="${rocm_tag}-${{ steps.version.outputs.commit_hash }}-${{ env.DATE }}-DSv4" + echo "IMAGE_TAG=${image_tag}" >> "$GITHUB_ENV" + echo "Building rocm/sgl-dev:${image_tag} from amd/deepseek_v4 @ ${{ steps.version.outputs.commit_sha }}" + + docker build . -f docker/rocm.Dockerfile \ + --build-arg SGL_BRANCH=${{ steps.version.outputs.commit_sha }} \ + --build-arg BUILD_TYPE=${{ matrix.build_type }} \ + --build-arg GPU_ARCH=${{ matrix.gpu_arch }} \ + --build-arg ENABLE_MORI=1 \ + --build-arg SETUPTOOLS_SCM_PRETEND_VERSION=${pretend_version} \ + -t rocm/sgl-dev:${image_tag} \ + --no-cache + docker push rocm/sgl-dev:${image_tag}