[ci] build aarch64 wheel builds on a real aarch64 machine (#6843)

.ci/setup.sh

@@ -3,7 +3,6 @@
 set -e -E -u -o pipefail
 
 # defaults
-AZURE=${AZURE:-"false"}
 IN_UBUNTU_BASE_CONTAINER=${IN_UBUNTU_BASE_CONTAINER:-"false"}
 SETUP_CONDA=${SETUP_CONDA:-"true"}
 

.github/workflows/python_package.yml

@@ -18,7 +18,54 @@ env:
   SKBUILD_STRICT_CONFIG: true
 
 jobs:
-  test:
+  test-linux-aarch64:
+    name: bdist wheel (ubuntu-24.04-arm)
+    runs-on: ubuntu-24.04-arm
+    timeout-minutes: 60
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 5
+          submodules: true
+      - name: Setup and run tests
+        shell: bash
+        # this uses 'docker run' instead of just setting 'container:'
+        # because actions/checkout requires GLIBC 2.28 and that is too
+        # new for manylinux2014
+        env:
+          BUILD_DIRECTORY: /LightGBM
+        run: |
+          cat > ./docker-script.sh <<EOF
+          mkdir -p \$BUILD_ARTIFACTSTAGINGDIRECTORY
+          export CONDA=\$HOME/miniforge
+          export PATH=\$CONDA/bin:\$PATH
+          ${{ env.BUILD_DIRECTORY }}/.ci/setup.sh || exit 1
+          ${{ env.BUILD_DIRECTORY }}/.ci/test.sh || exit 1
+          EOF
+          IMAGE_URI="lightgbm/vsts-agent:manylinux2014_aarch64"
+          docker pull "${IMAGE_URI}" || exit 1
+          docker run \
+            --platform "${PLATFORM}" \
+            --rm \
+            --env BUILD_DIRECTORY=${{ env.BUILD_DIRECTORY }} \
+            --env BUILD_ARTIFACTSTAGINGDIRECTORY=${{ env.BUILD_DIRECTORY }}/artifacts/ \
+            --env COMPILER=gcc \
+            --env METHOD=wheel \
+            --env OS_NAME=linux \
+            --env PRODUCES_ARTIFACTS=true \
+            --env PYTHON_VERSION="3.13" \
+            --env TASK=bdist \
+            -v "${PWD}":"${{ env.BUILD_DIRECTORY }}" \
+            -w ${{ env.BUILD_DIRECTORY }} \
+            "${IMAGE_URI}" \
+            /bin/bash ./docker-script.sh
+      - name: upload wheels
+        uses: actions/upload-artifact@v4
+        with:
+          name: linux-aarch64-wheel
+          path: artifacts/*.whl
+  test-macos:
     name: ${{ matrix.task }} ${{ matrix.method }} (${{ matrix.os }}, Python ${{ matrix.python_version }})
     runs-on: ${{ matrix.os }}
     timeout-minutes: 60
@@ -65,17 +112,13 @@ jobs:
         run: |
           export TASK="${{ matrix.task }}"
           export METHOD="${{ matrix.method }}"
+          export OS_NAME="macos"
           export PYTHON_VERSION="${{ matrix.python_version }}"
           if [[ "${{ matrix.os }}" == "macos-14" ]]; then
               # use clang when creating macOS release artifacts
               export COMPILER="clang"
-              export OS_NAME="macos"
-          elif [[ "${{ matrix.os }}" == "macos-13" ]]; then
+          else
               export COMPILER="gcc"
-              export OS_NAME="macos"
-          elif [[ "${{ matrix.os }}" == "ubuntu-latest" ]]; then
-              export COMPILER="clang"
-              export OS_NAME="linux"
           fi
           export BUILD_DIRECTORY="$GITHUB_WORKSPACE"
           export CONDA=${HOME}/miniforge
@@ -152,7 +195,11 @@ jobs:
   all-python-package-jobs-successful:
     if: always()
     runs-on: ubuntu-latest
-    needs: [test, test-latest-versions, test-old-versions]
+    needs:
+      - test-latest-versions
+      - test-macos
+      - test-linux-aarch64
+      - test-old-versions
     steps:
       - name: Note that all tests succeeded
         uses: re-actors/alls-green@v1.2.2

.vsts-ci.yml

@@ -223,81 +223,6 @@ jobs:
         inputs:
           filePath: $(Build.SourcesDirectory)/.ci/test.sh
           targetType: 'filePath'
-  ##################
-  # QEMU_multiarch #
-  ##################
-  - job: QEMU_multiarch
-    variables:
-      BUILD_DIRECTORY: /LightGBM
-      COMPILER: gcc
-      PRODUCES_ARTIFACTS: 'true'
-    pool:
-      vmImage: ubuntu-22.04
-    timeoutInMinutes: 180
-    strategy:
-      matrix:
-        bdist:
-          TASK: bdist
-          ARCH: aarch64
-    steps:
-      - script: |
-          sudo apt-get update
-          sudo apt-get install --no-install-recommends -y \
-            binfmt-support \
-            qemu \
-            qemu-user \
-            qemu-user-static
-        displayName: 'Install QEMU'
-      - script: |
-          docker run --rm --privileged multiarch/qemu-user-static --reset -p yes
-        displayName: 'Enable Docker multi-architecture support'
-      - script: |
-          git clean -d -f -x
-        displayName: 'Clean source directory'
-      # LGBM_SKIP_DASK_TESTS=true is set below only because running the tests under emulation is so slow...
-      # in theory, 'lightgbm.dask' should work without issue on aarch64 Linux systems.
-      # That could probably be removed as part of https://github.com/microsoft/LightGBM/issues/6788
-      - script: |
-          cat > docker-script.sh <<EOF
-          export CONDA=\$HOME/miniforge
-          export PATH=\$CONDA/bin:/opt/rh/llvm-toolset-7.0/root/usr/bin:\$PATH
-          export LD_LIBRARY_PATH=/opt/rh/llvm-toolset-7.0/root/usr/lib64:\$LD_LIBRARY_PATH
-          \$BUILD_DIRECTORY/.ci/setup.sh || exit 1
-          \$BUILD_DIRECTORY/.ci/test.sh || exit 1
-          EOF
-          IMAGE_URI="lightgbm/vsts-agent:manylinux2014_aarch64"
-          docker pull "${IMAGE_URI}" || exit 1
-          PLATFORM=$(docker inspect --format='{{.Os}}/{{.Architecture}}' "${IMAGE_URI}") || exit 1
-          echo "detected image platform: ${PLATFORM}"
-          docker run \
-            --platform "${PLATFORM}" \
-            --rm \
-            --env AZURE=true \
-            --env BUILD_ARTIFACTSTAGINGDIRECTORY=$BUILD_ARTIFACTSTAGINGDIRECTORY \
-            --env BUILD_DIRECTORY=$BUILD_DIRECTORY \
-            --env COMPILER=$COMPILER \
-            --env LGBM_SKIP_DASK_TESTS=true \
-            --env METHOD=$METHOD \
-            --env OS_NAME=linux \
-            --env PRODUCES_ARTIFACTS=$PRODUCES_ARTIFACTS \
-            --env PYTHON_VERSION=$PYTHON_VERSION \
-            --env TASK=$TASK \
-            -v "$(Build.SourcesDirectory)":"$BUILD_DIRECTORY" \
-            -v "$(Build.ArtifactStagingDirectory)":"$(Build.ArtifactStagingDirectory)" \
-            "${IMAGE_URI}" \
-            /bin/bash $BUILD_DIRECTORY/docker-script.sh
-        displayName: 'Setup and run tests'
-      - task: PublishBuildArtifacts@1
-        condition: >
-          and(
-          succeeded(),
-          in(variables['TASK'], 'bdist'),
-          not(startsWith(variables['Build.SourceBranch'], 'refs/pull/'))
-          )
-        inputs:
-          pathtoPublish: '$(Build.ArtifactStagingDirectory)'
-          artifactName: PackageAssets
-          artifactType: container
   #########
   # macOS #
   #########
@@ -441,7 +366,6 @@ jobs:
     dependsOn:
       - Linux
       - Linux_latest
-      - QEMU_multiarch
       - macOS
       - Windows
       - R_artifact

tests/python_package_test/test_dask.py

@@ -55,10 +55,6 @@ pytestmark = [
     pytest.mark.skipif(getenv("TASK", "") == "mpi", reason="Fails to run with MPI interface"),
     pytest.mark.skipif(getenv("TASK", "") == "gpu", reason="Fails to run with GPU interface"),
     pytest.mark.skipif(getenv("TASK", "") == "cuda", reason="Fails to run with CUDA interface"),
-    pytest.mark.skipif(
-        getenv("LGBM_SKIP_DASK_TESTS", "") == "true",
-        reason="Skipping lightgbm.dask tests (found env variable LGBM_SKIP_DASK_TESTS=true)",
-    ),
 ]