Merge branch 'main' into Arm-backend-Do-not-run-model-unit-tests-in-parallel

Author: zingo

.github/workflows/build-presets.yml

@@ -6,6 +6,8 @@ on:
     branches:
       - main
       - release/*
+    paths:
+      - .github/workflows/build-presets.yml
   workflow_dispatch:
 
 concurrency:
@@ -16,15 +18,51 @@ jobs:
   apple:
     uses: pytorch/test-infra/.github/workflows/macos_job.yml@main
     strategy:
+      fail-fast: false
       matrix:
-        preset: [macos-arm64]
+        preset: [macos-arm64, pybind]
     with:
       job-name: build
+      ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
       runner: macos-latest-xlarge
       python-version: 3.12
       submodules: recursive
+      timeout: 90
       script: |
         set -eux
         ${CONDA_RUN} ./install_requirements.sh > /dev/null
         ${CONDA_RUN} cmake --preset ${{ matrix.preset }}
         ${CONDA_RUN} cmake --build cmake-out --parallel
+
+  linux:
+    uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
+    strategy:
+      fail-fast: false
+      matrix:
+        preset: [pybind]
+        runner: [linux.2xlarge, linux.arm64.2xlarge]
+        docker-image: [executorch-ubuntu-22.04-clang12, executorch-ubuntu-22.04-gcc11-aarch64]
+        # Excluding specific runner + docker image combinations that don't make sense:
+        #   - Excluding the ARM64 gcc image on the x86 runner (linux.2xlarge)
+        #   - Excluding the x86 clang image on the ARM64 runner (linux.arm64.2xlarge)
+        exclude:
+          - runner: linux.2xlarge
+            docker-image: executorch-ubuntu-22.04-gcc11-aarch64
+          - runner: linux.arm64.2xlarge
+            docker-image: executorch-ubuntu-22.04-clang12
+    with:
+      job-name: build
+      ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
+      runner: ${{ matrix.runner }}
+      docker-image: ${{ matrix.docker-image }}
+      submodules: recursive
+      timeout: 90
+      script: |
+        set -eux
+        # The generic Linux job chooses to use base env, not the one setup by the image
+        CONDA_ENV=$(conda env list --json | jq -r ".envs | .[-1]")
+        conda activate "${CONDA_ENV}"
+
+        ./install_requirements.sh > /dev/null
+        cmake --preset ${{ matrix.preset }}
+        cmake --build cmake-out --parallel

examples/demo-apps/android/LlamaDemo/README.md

@@ -1,6 +1,6 @@
 # ExecuTorch Llama Android Demo App
 
-**[UPDATE - 10/24]** We have added support for running quantized Llama 3.2 1B/3B models in demo apps on the [XNNPACK backend](https://github.com/pytorch/executorch/blob/main/examples/demo-apps/android/LlamaDemo/docs/delegates/xnnpack_README.md). We currently support inference with SpinQuant and QAT+LoRA quantization methods.
+**[UPDATE - 2025-05-15]** We have added support for running Qwen3 0.6B and 4B model. Please see [this tutorial](https://github.com/pytorch/executorch/tree/main/examples/models/qwen3#summary) for export. Loading and running Qwen3 with this app is the same as Llama, as in this doc.
 
 We’re excited to share that the newly revamped Android demo app is live and includes many new updates to provide a more intuitive and smoother user experience with a chat use case! The primary goal of this app is to showcase how easily ExecuTorch can be integrated into an Android demo app and how to exercise the many features ExecuTorch and Llama models have to offer.
 
@@ -26,6 +26,7 @@ As a whole, the models that this app supports are (varies by delegate):
 * Llama 3 8B
 * Llama 2 7B
 * LLaVA-1.5 vision model (only XNNPACK)
+* Qwen 3 0.6B, 1.7B, and 4B
 
 
 ## Building the APK
@@ -170,4 +171,4 @@ Go to `examples/demo-apps/android/LlamaDemo`,
 ```
 
 ## Reporting Issues
-If you encountered any bugs or issues following this tutorial please file a bug/issue here on [Github](https://github.com/pytorch/executorch/issues/new).
+If you encountered any bugs or issues following this tutorial please file a bug/issue here on [Github](https://github.com/pytorch/executorch/issues/new), or join our discord [here](https://lnkd.in/gWCM4ViK).

kernels/optimized/CMakeLists.txt

@@ -39,6 +39,7 @@ include(${EXECUTORCH_ROOT}/tools/cmake/Codegen.cmake)
 # Build cpublas.
 list(TRANSFORM _optimized_cpublas__srcs PREPEND "${EXECUTORCH_ROOT}/")
 add_library(cpublas STATIC ${_optimized_cpublas__srcs})
+target_include_directories(cpublas PRIVATE ${TORCH_INCLUDE_DIRS})
 target_link_libraries(
   cpublas PUBLIC executorch_core eigen_blas extension_threadpool
 )