Backmerging with Msft commits

.github/actions/setup-android-ndk/action.yml

@@ -0,0 +1,98 @@
+# .github/actions/setup-android-ndk/action.yml
+name: 'Setup Android NDK'
+description: 'Installs and configures a specific version of the Android NDK'
+inputs:
+  ndk-version:
+    description: 'The version of the Android NDK to install (e.g., 27.2.12479018)'
+    required: true
+    default: '27.2.12479018'
+  android-sdk-root:
+    description: 'The root directory of the Android SDK'
+    required: true
+    default: '/usr/local/lib/android/sdk'
+
+runs:
+  using: "composite"  # Use a composite action for multiple shell commands
+  steps:
+    - name: Install coreutils and ninja
+      shell: bash
+      run: sudo apt-get update -y && sudo apt-get install -y coreutils ninja-build
+
+    - name: Install Android NDK
+      shell: bash
+      run: |
+        set -e
+        "${{ inputs.android-sdk-root }}/cmdline-tools/latest/bin/sdkmanager" --install "ndk;${{ inputs.ndk-version }}"
+
+        NDK_PATH="${{ inputs.android-sdk-root }}/ndk/${{ inputs.ndk-version }}"
+        if [[ ! -d "${NDK_PATH}" ]]; then
+          echo "NDK directory is not in expected location: ${NDK_PATH}"
+          exit 1
+        fi
+
+        # Use standard environment variable setting in bash and add to GITHUB_ENV
+        echo "ANDROID_NDK_HOME=${NDK_PATH}" >> $GITHUB_ENV
+        echo "ANDROID_NDK_ROOT=${NDK_PATH}" >> $GITHUB_ENV
+        echo "ANDROID_NDK_HOME: ${NDK_PATH}"
+        echo "ANDROID_NDK_ROOT: ${NDK_PATH}"
+
+    - name: Check if emulator are installed and add to PATH
+      shell: bash
+      run: |
+          if [[ ":$PATH:" == *":${ANDROID_SDK_ROOT}/emulator:"* ]]; then
+            echo "${ANDROID_SDK_ROOT}/emulator is in PATH"
+          else
+            ${ANDROID_SDK_ROOT}/cmdline-tools/latest/bin/sdkmanager --install "emulator"
+            echo "${ANDROID_SDK_ROOT}/emulator" >> $GITHUB_PATH
+          fi
+
+    - name: Check if platform tools are installed and add to PATH
+      shell: bash
+      run: |
+          if [[ ":$PATH:" == *":${ANDROID_SDK_ROOT}/platform-tools:"* ]]; then
+            echo "${ANDROID_SDK_ROOT}/platform-tools is in PATH"
+          else
+            ${ANDROID_SDK_ROOT}/cmdline-tools/latest/bin/sdkmanager --install "platform-tools"
+            echo "${ANDROID_SDK_ROOT}/platform-tools" >> $GITHUB_PATH
+          fi
+          ls -R "${ANDROID_SDK_ROOT}/platform-tools"
+
+    - name: Create Android Emulator
+      shell: bash
+      env:
+        ANDROID_AVD_HOME: ${{ runner.temp }}/android-avd
+      run: |
+          python3 tools/python/run_android_emulator.py \
+            --android-sdk-root "${ANDROID_SDK_ROOT}" \
+            --create-avd --system-image "system-images;android-31;default;x86_64"
+
+    - name: List Android AVDs
+      shell: bash
+      env:
+        ANDROID_AVD_HOME: ${{ runner.temp }}/android-avd
+      run: |
+          "${ANDROID_SDK_ROOT}/cmdline-tools/latest/bin/avdmanager" list avd
+
+    - name: Check emulator.pid does not exist
+      shell: bash
+      run: |
+          if test -f ./emulator.pid; then
+            echo "Emulator PID file was not expected to exist but does and has pid: `cat ./emulator.pid`"
+            exit 1
+          fi
+
+    - name: Start Android Emulator
+      shell: bash
+      env:
+        ANDROID_AVD_HOME: ${{ runner.temp }}/android-avd
+      run: |
+          set -e -x
+          python3 tools/python/run_android_emulator.py \
+            --android-sdk-root "${ANDROID_SDK_ROOT}" \
+            --start --emulator-extra-args="-partition-size 2047" \
+            --emulator-pid-file ./emulator.pid
+          echo "Emulator PID: `cat ./emulator.pid`"
+
+    - name: View Android ENVs
+      shell: bash
+      run: env | grep ANDROID

.github/workflows/android.yml

@@ -0,0 +1,147 @@
+name: Android CI
+# This workflow is used to build and test on Android Emulator on Linux
+
+on:
+  push:
+    branches:
+      - main
+      - rel-*
+  pull_request:
+    branches:
+      - main
+      - rel-*
+  workflow_dispatch:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}-${{ github.event_name == 'workflow_dispatch' }}
+  cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}
+
+jobs:
+  android_nnapi_ep:
+    runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-Ubuntu2204-AMD-CPU"]
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Use jdk 17
+        uses: actions/setup-java@v4
+        with:
+          distribution: 'temurin'
+          java-version: '17'
+          architecture: x64
+
+      - name: Setup Android NDK
+        uses: ./.github/actions/setup-android-ndk
+        with:
+          ndk-version: 27.2.12479018
+
+      - name: Export GitHub Actions cache environment variables
+        uses: actions/github-script@v7
+        with:
+          script: |
+            core.exportVariable('ACTIONS_CACHE_URL', process.env.ACTIONS_CACHE_URL || '');
+            core.exportVariable('ACTIONS_RUNTIME_TOKEN', process.env.ACTIONS_RUNTIME_TOKEN || '');
+
+      - name: NNAPI EP, Build, Test on Android Emulator
+        run: >-
+          python3 tools/ci_build/build.py
+          --enable_lto
+          --android
+          --build_dir build_nnapi
+          --android_sdk_path "$ANDROID_HOME"
+          --android_ndk_path "$ANDROID_NDK_HOME"
+          --android_abi=x86_64
+          --android_api=29
+          --skip_submodule_sync
+          --parallel --use_vcpkg --use_vcpkg_ms_internal_asset_cache
+          --use_nnapi
+          --build_shared_lib
+          --cmake_generator=Ninja
+          --build_java
+        shell: bash
+
+
+      - name: Build Minimal ORT with NNAPI and run tests
+        run:  tools/ci_build/github/linux/ort_minimal/nnapi_minimal_build_minimal_ort_and_run_tests.sh "$(pwd)"
+        shell: bash
+
+      - name: Install psutil for emulator shutdown by run_android_emulator.py
+        if: always()
+        run: python3 -m pip install psutil
+        shell: bash
+
+      - name: Stop Android Emulator
+        if: always()
+        run: |
+          env | grep ANDROID
+          if test -f ${{ github.workspace }}/emulator.pid; then
+            echo "Emulator PID:"`cat ${{ github.workspace }}/emulator.pid`
+            python3 tools/python/run_android_emulator.py \
+              --android-sdk-root "${ANDROID_SDK_ROOT}" \
+              --stop \
+              --emulator-pid-file ${{ github.workspace }}/emulator.pid
+            rm ${{ github.workspace }}/emulator.pid
+          else
+            echo "Emulator PID file was expected to exist but does not."
+          fi
+        shell: bash
+
+  android_cpu_ep:
+    name: Android CI Pipeline
+    runs-on: ["self-hosted", "1ES.Pool=onnxruntime-github-Ubuntu2204-AMD-CPU"]
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Use jdk 17
+        uses: actions/setup-java@v4
+        with:
+          distribution: 'temurin'
+          java-version: '17'
+          architecture: x64
+
+      - name: Setup Android NDK
+        uses: ./.github/actions/setup-android-ndk
+        with:
+          ndk-version: 27.2.12479018
+
+      - name: Export GitHub Actions cache environment variables
+        uses: actions/github-script@v7
+        with:
+          script: |
+            core.exportVariable('ACTIONS_CACHE_URL', process.env.ACTIONS_CACHE_URL || '');
+            core.exportVariable('ACTIONS_RUNTIME_TOKEN', process.env.ACTIONS_RUNTIME_TOKEN || '');
+
+      - name: CPU EP, Build and Test
+        run: >-
+          python3 tools/ci_build/build.py
+          --enable_lto
+          --android
+          --build_dir build
+          --android_sdk_path $ANDROID_HOME
+          --android_ndk_path $ANDROID_NDK_HOME
+          --android_abi=x86_64
+          --android_api=30
+          --skip_submodule_sync
+          --parallel --use_vcpkg --use_vcpkg_ms_internal_asset_cache
+          --cmake_generator=Ninja
+          --build_java
+        shell: bash
+
+      - name: Install psutil for emulator shutdown by run_android_emulator.py
+        if: always()
+        run: python3 -m pip install psutil
+        shell: bash
+
+      - name: Stop Android Emulator
+        if: always()
+        run: |
+          if test -f ${{ github.workspace }}/emulator.pid; then
+           echo "Emulator PID:"`cat ${{ github.workspace }}/emulator.pid`
+           python3 tools/python/run_android_emulator.py \
+           --android-sdk-root "${ANDROID_SDK_ROOT}" \
+           --stop \
+           --emulator-pid-file ${{ github.workspace }}/emulator.pid
+           rm ${{ github.workspace }}/emulator.pid
+          else
+           echo "Emulator PID file was expected to exist but does not."
+          fi
+        shell: bash

.github/workflows/mac.yml

@@ -154,6 +154,46 @@ jobs:
           --use_xnnpack \
           --use_binskim_compliant_compile_flags
 
+  ARM64-Xcode16-webgpu:
+    runs-on: macos-15
+
+    env:
+      xcode_version: 16
+
+    timeout-minutes: 60
+
+    steps:
+    - uses: actions/setup-python@v5
+      with:
+        python-version: ${{ env.python_version }}
+
+    - name: Verify ARM64 machine
+      shell: python
+      run: |
+        import platform
+        assert platform.machine() == "arm64", "This job expects to be run on an ARM64 machine."
+
+    - name: Use Xcode ${{ env.xcode_version }}
+      shell: bash
+      run: |
+        XCODE_DEVELOPER_DIR="/Applications/Xcode_${{ env.xcode_version }}.app/Contents/Developer"
+        sudo xcode-select --switch "${XCODE_DEVELOPER_DIR}"
+
+    - uses: actions/checkout@v4
+
+    - name: Build and test
+      shell: bash
+      run: |
+        python ./tools/ci_build/build.py \
+          --build_dir ./build \
+          --update \
+          --build --parallel \
+          --test \
+          --build_shared_lib \
+          --build_nodejs \
+          --use_webgpu \
+          --use_binskim_compliant_compile_flags
+
   ARM64-Xcode16-targeting-iphonesimulator:
     runs-on: macos-15
 
@@ -164,7 +204,7 @@ jobs:
       matrix:
         target_arch: [x86_64, arm64]
 
-    timeout-minutes: 75
+    timeout-minutes: 90
 
     steps:
     - uses: actions/setup-python@v5

cmake/CMakePresets.json

@@ -109,6 +109,29 @@
                 "rhs": "Darwin"
             }
         },
+        {
+            "name": "arm64-osx",
+            "inherits": [
+                "unit-test"
+            ],
+            "generator": "Xcode",
+            "binaryDir": "${sourceParentDir}/cmake_build/arm64-osx",
+            "installDir": "${sourceParentDir}/cmake_build/out",
+            "cacheVariables": {
+                "CMAKE_OSX_ARCHITECTURES": "arm64",
+                "onnxruntime_BUILD_SHARED_LIB": true,
+                "onnxruntime_USE_XNNPACK": false,
+                "onnxruntime_USE_COREML": true,
+                "onnxruntime_BUILD_OBJC": true,
+                "onnxruntime_BUILD_APPLE_FRAMEWORK": true,
+                "CMAKE_CONFIGURATION_TYPES": "Debug;Release"
+            },
+            "condition": {
+                "type": "equals",
+                "lhs": "${hostSystemName}",
+                "rhs": "Darwin"
+            }
+        },
         {
             "name": "x64-osx-vcpkg",
             "inherits": [

docs/ContribOperators.md

@@ -1625,7 +1625,7 @@ This version of the operator has been available since version 1 of the 'com.micr
 #### Type Constraints
 
 <dl>
-<dt><tt>T</tt> : tensor(int8), tensor(int16), tensor(int32), tensor(int64), tensor(uint8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(float16), tensor(float), tensor(double), tensor(bfloat16)</dt>
+<dt><tt>T</tt> : tensor(bool), tensor(int8), tensor(int16), tensor(int32), tensor(int64), tensor(uint8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(float16), tensor(float), tensor(double), tensor(bfloat16)</dt>
 <dd>Constrain input and output types.</dd>
 </dl>
 

docs/OperatorKernels.md

@@ -582,7 +582,7 @@ Do not modify directly.*
 | Op Name | Parameters | OpSet Version | Types Supported |
 |---------|------------|---------------|-----------------|
 |**Operator Domain:** *ai.onnx*||||
-|Abs|*in* X:**T**<br> *out* Y:**T**|13+|**T** = tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
+|Abs|*in* X:**T**<br> *out* Y:**T**|13+|**T** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |||[6, 12]|**T** = tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |Add|*in* A:**T**<br> *in* B:**T**<br> *out* C:**T**|14+|**T** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |||13|**T** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64)|
@@ -839,7 +839,7 @@ Do not modify directly.*
 |Shrink|*in* input:**T**<br> *out* output:**T**|9+|**T** = tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |Sigmoid|*in* X:**T**<br> *out* Y:**T**|13+|**T** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16)|
 |||[6, 12]|**T** = tensor(double), tensor(float), tensor(float16)|
-|Sign|*in* input:**T**<br> *out* output:**T**|13+|**T** = tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
+|Sign|*in* input:**T**<br> *out* output:**T**|13+|**T** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |SimplifiedLayerNormalization|*in* X:**T**<br> *in* scale:**V**<br> *out* Y:**V**<br> *out* inv_std_var:**U**|1+|**T** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16)<br/> **U** = tensor(double), tensor(float)<br/> **V** = tensor(bfloat16), tensor(double), tensor(float), tensor(float16)|
 |Sin|*in* input:**T**<br> *out* output:**T**|7+|**T** = tensor(double), tensor(float), tensor(float16)|
 |Size|*in* data:**T**<br> *out* size:**T1**|13+|**T** = tensor(bfloat16), tensor(bool), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)<br/> **T1** = tensor(int64)|

js/react_native/android/src/main/AndroidManifest.xml

@@ -1,3 +1,2 @@
-<manifest xmlns:android="http://schemas.android.com/apk/res/android"
-          package="ai.onnxruntime.reactnative">
+<manifest xmlns:android="http://schemas.android.com/apk/res/android">
 </manifest>

js/react_native/e2e/.detoxrc.js

@@ -38,8 +38,8 @@ module.exports = {
     simulator: {
       type: 'ios.simulator',
       device: {
-        type: 'iPhone 14',
-        os: 'iOS 16.4',
+        type: 'iPhone 15',
+        os: 'iOS 17.4',
       },
     },
     attached: {