Adding the build test for the neutron backend library.

Author: jirioc

.github/workflows/trunk.yml

@@ -271,6 +271,36 @@ jobs:
           exit 1
         fi
 
+  nxp-build-test:
+    name: nxp-build-test
+    uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
+    permissions:
+      id-token: write
+      contents: read
+    with:
+      runner: linux.2xlarge
+      docker-image: executorch-ubuntu-22.04-arm-sdk
+      submodules: 'recursive'
+      ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
+      timeout: 90
+      script: |
+        # The generic Linux job chooses to use base env, not the one setup by the image
+        CONDA_ENV=$(conda env list --json | jq -r ".envs | .[-1]")
+        conda activate "${CONDA_ENV}"
+
+        # Build
+        cmake -DEXECUTORCH_BUILD_NXP_NEUTRON=ON -Bcmake-out .
+        cmake --build cmake-out --target executorch_delegate_neutron --config Release
+
+        # Build check for the neutron backend library
+        lib_neutron="cmake-out/backends/nxp/libexecutorch_delegate_neutron.a"
+        if [ -f $lib_neutron ]; then
+            echo "Neutron backend library built."
+        else
+            echo "Neutron backend library not found!"
+            exit 1
+        fi
+
   test-coreml-delegate:
     name: test-coreml-delegate
     uses: pytorch/test-infra/.github/workflows/macos_job.yml@main

backends/nxp/CMakeLists.txt

@@ -6,7 +6,9 @@
 set(
   _common_include_directories 
   ${CMAKE_CURRENT_SOURCE_DIR}/../../..
+  ${CMAKE_CURRENT_SOURCE_DIR}/../../runtime/core/portable_type/c10
 )
+add_compile_definitions(C10_USING_CUSTOM_GENERATED_MACROS)
 
 set(_neutron_sources ${CMAKE_CURRENT_SOURCE_DIR}/runtime/NeutronBackend.cpp )
 

backends/nxp/runtime/NeutronBackend.cpp

@@ -236,8 +236,8 @@ class NeutronBackend final : public PyTorchBackendInterface {
     //    cfg->mcfg.kernels
     const uint8_t* transpositionFlags =
         static_cast<const uint8_t*>(processed->data());
-    uint32_t numInputs = transpositionFlags[INPUT_TENSOR_FORMAT_LEN_POS];
-    uint32_t numOutputs = transpositionFlags[OUTPUT_TENSOR_FORMAT_LEN_POS];
+    int numInputs = transpositionFlags[INPUT_TENSOR_FORMAT_LEN_POS];
+    int numOutputs = transpositionFlags[OUTPUT_TENSOR_FORMAT_LEN_POS];
     cfg->inputTranspositionFlags =
         INPUT_TENSOR_FORMAT_ARRAY_ADDR(transpositionFlags);
     cfg->outputTranspositionFlags =