GH-36469: [Java][Packaging] Distribute linux aarch64 libs with mavencentral jars

ci/docker/java-jni-manylinux-201x.dockerfile

@@ -34,8 +34,12 @@ RUN vcpkg install \
 
 # Install Java
 ARG java=1.8.0
-RUN yum install -y java-$java-openjdk-devel rh-maven35 && yum clean all
-ENV JAVA_HOME=/usr/lib/jvm/java-$java-openjdk/
+ARG maven=3.9.3
+RUN yum install -y java-$java-openjdk-devel && \
+      yum clean all && \
+      curl https://dlcdn.apache.org/maven/maven-3/${maven}/binaries/apache-maven-${maven}-bin.tar.gz | \
+        tar xfz - -C /usr/local && \
+      ln -s /usr/local/apache-maven-${maven}/bin/mvn /usr/local/bin
 
 # Install the gcs testbench
 COPY ci/scripts/install_gcs_testbench.sh /arrow/ci/scripts/

ci/scripts/java_jni_macos_build.sh

@@ -99,6 +99,7 @@ if [ "${ARROW_BUILD_TESTS}" == "ON" ]; then
   # MinIO is required
   exclude_tests="arrow-s3fs-test"
   # unstable
+  exclude_tests="${exclude_tests}|arrow-acero-asof-join-node-test"
   exclude_tests="${exclude_tests}|arrow-acero-hash-join-node-test"
   ctest \
     --exclude-regex "${exclude_tests}" \

ci/scripts/java_jni_manylinux_build.sh

@@ -21,8 +21,14 @@ set -ex
 
 arrow_dir=${1}
 build_dir=${2}
+normalized_arch=$(arch)
+case ${normalized_arch} in
+  aarch64)
+    normalized_arch=aarch_64
+    ;;
+esac
 # The directory where the final binaries will be stored when scripts finish
-dist_dir=${3}/$(arch)
+dist_dir=${3}/${normalized_arch}
 
 echo "=== Clear output directories and leftovers ==="
 # Clear output directories and leftovers
@@ -103,9 +109,18 @@ ninja install
 if [ "${ARROW_BUILD_TESTS}" = "ON" ]; then
   # MinIO is required
   exclude_tests="arrow-s3fs-test"
+  case $(arch) in
+    aarch64)
+      # GCS testbench is crashed on aarch64:
+      # ImportError: ../grpc/_cython/cygrpc.cpython-38-aarch64-linux-gnu.so:
+      # undefined symbol: vtable for std::__cxx11::basic_ostringstream<
+      #   char, std::char_traits<char>, std::allocator<char> >
+      exclude_tests="${exclude_tests}|arrow-gcsfs-test"
+      ;;
+  esac
   # unstable
-  exclude_tests="${exclude_tests}|arrow-compute-hash-join-node-test"
-  exclude_tests="${exclude_tests}|arrow-dataset-scanner-test"
+  exclude_tests="${exclude_tests}|arrow-acero-asof-join-node-test"
+  exclude_tests="${exclude_tests}|arrow-acero-hash-join-node-test"
   # strptime
   exclude_tests="${exclude_tests}|arrow-utility-test"
   ctest \
@@ -138,6 +153,7 @@ fi
 echo "=== Checking shared dependencies for libraries ==="
 pushd ${dist_dir}
 archery linking check-dependencies \
+  --allow ld-linux-aarch64 \
   --allow ld-linux-x86-64 \
   --allow libc \
   --allow libdl \

cpp/src/arrow/compute/kernels/scalar_arithmetic_test.cc

@@ -911,11 +911,11 @@ TYPED_TEST(TestBinaryArithmeticFloating, Power) {
     this->AssertBinop(Power, "[null, 1, 3.3, null, 2]", "[1, 4, 2, 5, 0.1]",
                       "[null, 1, 10.89, null, 1.07177346]");
     // Scalar exponentiated by array
-    this->AssertBinop(Power, 10.0F, "[null, 1, 2.5, null, 2, 5]",
-                      "[null, 10, 316.227766017, null, 100, 100000]");
+    this->AssertBinop(Power, 4.0F, "[null, 1, 0.5, null, 2, 5]",
+                      "[null, 4, 2.0, null, 16, 1024]");
     // Array exponentiated by scalar
-    this->AssertBinop(Power, "[null, 1, 2.5, null, 2, 5]", 10.0F,
-                      "[null, 1, 9536.74316406, null, 1024, 9765625]");
+    this->AssertBinop(Power, "[null, 1, 0.5, null, 2, 5]", 4.0F,
+                      "[null, 1, 0.0625, null, 16, 625]");
     // Array with infinity
     this->AssertBinop(Power, "[3.4, Inf, -Inf, 1.1, 100000]", "[1, 2, 3, Inf, 100000]",
                       "[3.4, Inf, -Inf, Inf, Inf]");
@@ -925,7 +925,7 @@ TYPED_TEST(TestBinaryArithmeticFloating, Power) {
     this->AssertBinop(Power, 21.0F, 3.0F, 9261.0F);
     // Divide by zero
     this->AssertBinop(Power, "[0.0, 0.0]", "[-1.0, -3.0]", "[Inf, Inf]");
-    // Check overflow behaviour
+    // Check overflow behavior
     this->AssertBinop(Power, max, 10, INFINITY);
   }
 

dev/tasks/java-jars/github.yml

@@ -22,8 +22,28 @@
 jobs:
 
   build-cpp-ubuntu:
-    name: Build C++ libraries Ubuntu
-    runs-on: ubuntu-latest
+    {% set arch = '${{ matrix.platform.arch }}' %}
+    name: Build C++ libraries Ubuntu {{ arch }}
+    runs-on: {{ '${{ matrix.platform.runs_on }}' }}
+    env:
+      # architecture name used for archery build
+      ARCH: {{ '${{ matrix.platform.archery_arch }}' }}
+      ARCH_ALIAS: {{ '${{ matrix.platform.archery_arch_alias }}' }}
+      ARCH_SHORT: {{ '${{ matrix.platform.archery_arch_short }}' }}
+    strategy:
+      fail-fast: false
+      matrix:
+        platform:
+          - runs_on: ["ubuntu-latest"]
+            arch: "x86_64"
+            archery_arch: "amd64"
+            archery_arch_alias: "x86_64"
+            archery_arch_short: "amd64"
+          - runs_on: ["self-hosted", "Linux", "arm64"]
+            arch: "aarch_64"
+            archery_arch: "arm64v8"
+            archery_arch_alias: "aarch64"
+            archery_arch_short: "arm64"
     steps:
       {{ macros.github_checkout_arrow()|indent }}
       {{ macros.github_install_archery()|indent }}
@@ -36,12 +56,12 @@ jobs:
             -e ARROW_JAVA_TEST=OFF \
             java-jni-manylinux-2014
       - name: Compress into single artifact to keep directory structure
-        run: tar -cvzf arrow-shared-libs-linux.tar.gz arrow/java-dist/
+        run: tar -cvzf arrow-shared-libs-linux-{{ arch }}.tar.gz arrow/java-dist/
       - name: Upload artifacts
         uses: actions/upload-artifact@v2
         with:
-          name: ubuntu-shared-lib
-          path: arrow-shared-libs-linux.tar.gz
+          name: ubuntu-shared-lib-{{ arch }}
+          path: arrow-shared-libs-linux-{{ arch }}.tar.gz
     {% if arrow.is_default_branch() %}
       {{ macros.github_login_dockerhub()|indent }}
       - name: Push Docker image
@@ -178,7 +198,8 @@ jobs:
       - name: Decompress artifacts
         run: |
           mv artifacts/*/*.tar.gz .
-          tar -xvzf arrow-shared-libs-linux.tar.gz
+          tar -xvzf arrow-shared-libs-linux-x86_64.tar.gz
+          tar -xvzf arrow-shared-libs-linux-aarch_64.tar.gz
           tar -xvzf arrow-shared-libs-macos-x86_64.tar.gz
           tar -xvzf arrow-shared-libs-macos-aarch_64.tar.gz
           tar -xvzf arrow-shared-libs-windows.tar.gz
@@ -191,6 +212,11 @@ jobs:
           test -f arrow/java-dist/x86_64/libarrow_orc_jni.so
           test -f arrow/java-dist/x86_64/libgandiva_jni.so
 
+          test -f arrow/java-dist/aarch_64/libarrow_cdata_jni.so
+          test -f arrow/java-dist/aarch_64/libarrow_dataset_jni.so
+          test -f arrow/java-dist/aarch_64/libarrow_orc_jni.so
+          test -f arrow/java-dist/aarch_64/libgandiva_jni.so
+
           test -f arrow/java-dist/x86_64/libarrow_cdata_jni.dylib
           test -f arrow/java-dist/x86_64/libarrow_dataset_jni.dylib
           test -f arrow/java-dist/x86_64/libarrow_orc_jni.dylib

docker-compose.yml

@@ -1147,7 +1147,6 @@ services:
     command:
       ["pip install -e /arrow/dev/archery && \
         /arrow/ci/scripts/java_jni_manylinux_build.sh /arrow /build /arrow/java-dist && \
-        source /opt/rh/rh-maven35/enable && \
         /arrow/ci/scripts/java_build.sh /arrow /build /arrow/java-dist && \
         /arrow/ci/scripts/java_test.sh /arrow /build /arrow/java-dist"]