Sync with Microsoft ONNX Runtime - [18/08/2025]

.github/workflows/linux_cuda_ci.yml

@@ -63,7 +63,7 @@ jobs:
 
       # --- Download Build Artifact to Runner Temp Directory ---
       - name: Download Build Artifact
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v5
         with:
           name: build-output-x64-Release # Must match the upload name
           path: ${{ runner.temp }}/Release # Download contents into temp dir structure

.github/workflows/linux_minimal_build.yml

@@ -582,7 +582,7 @@ jobs:
         with:
           node-version: 20
       - name: Download Test Data Artifact
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v5
         with:
           name: test_data
           path: ${{ runner.temp }}/.test_data/

.github/workflows/linux_tensorrt_ci.yml

@@ -65,7 +65,7 @@ jobs:
 
       # --- Download Build Artifact to Runner Temp Directory ---
       - name: Download Build Artifact
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v5
         with:
           name: build-output-x64-Release # Must match the upload name
           path: ${{ runner.temp }}/Release # Download contents into temp dir structure

.github/workflows/mac.yml

@@ -63,7 +63,7 @@ jobs:
     runs-on: macos-15
 
     env:
-      xcode_version: 16
+      xcode_version: 16.4
 
     strategy:
       matrix:

.github/workflows/windows-web-ci-workflow.yml

@@ -67,7 +67,7 @@ jobs:
           node-version: "20.x"
 
       - name: Download WebAssembly artifacts
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v5
         with:
           name: ${{ inputs.build_config }}_wasm
           path: ${{ github.workspace }}/artifacts_wasm

.github/workflows/windows_cuda.yml

@@ -158,7 +158,7 @@ jobs:
           submodules: 'none'
 
       - name: Download build artifacts
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v5
         with:
           name: build-artifacts
           path: ${{ runner.temp }}\build

.github/workflows/windows_tensorrt.yml

@@ -163,7 +163,7 @@ jobs:
           submodules: 'none'
 
       - name: Download build artifacts
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v5
         with:
           name: build-artifacts
           path: ${{ runner.temp }}\build

cmake/CMakeLists.txt

@@ -1791,3 +1791,12 @@ else()
     include("${CMAKE_CURRENT_SOURCE_DIR}/arm64x.cmake")
   endif()
 endif()
+
+if(onnxruntime_BUILD_UNIT_TESTS)
+  include("${CMAKE_CURRENT_SOURCE_DIR}/onnxruntime_test_pch.cmake")
+endif()
+
+# Include precompiled header configuration for providers
+if(TARGET onnxruntime_providers)
+  include("${CMAKE_CURRENT_SOURCE_DIR}/onnxruntime_providers_pch.cmake")
+endif()

cmake/onnxruntime_providers_pch.cmake

@@ -0,0 +1,8 @@
+# Precompiled header configuration for onnxruntime_providers
+
+if(CMAKE_CXX_COMPILER_ID MATCHES "MSVC")
+  # Visual Studio PCH
+  target_precompile_headers(onnxruntime_providers PRIVATE
+    "${CMAKE_CURRENT_SOURCE_DIR}/providers_pch.h"
+  )
+endif()

cmake/onnxruntime_test_pch.cmake

@@ -0,0 +1,20 @@
+# Precompiled header configuration for onnxruntime_test_all
+
+if(CMAKE_CXX_COMPILER_ID MATCHES "MSVC")
+  # Visual Studio PCH
+  target_precompile_headers(onnxruntime_test_all PRIVATE
+    "${CMAKE_CURRENT_SOURCE_DIR}/test_pch.h"
+  )
+  endif()
+
+# Exclude certain files that might conflict with PCH
+set(PCH_EXCLUDE_FILES
+  # Add any problematic source files here
+  "${TEST_SRC_DIR}/framework/tensor_shape_test.cc"
+)
+
+foreach(file ${PCH_EXCLUDE_FILES})
+  set_source_files_properties(${file} PROPERTIES
+    SKIP_PRECOMPILE_HEADERS ON
+  )
+endforeach()

cmake/providers_pch.h

@@ -0,0 +1,25 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#pragma once
+
+// Core framework headers (highest compilation time impact)
+#include "core/framework/op_kernel.h"
+#include "core/framework/op_kernel_info.h"
+#include "core/framework/execution_provider.h"
+#include "core/framework/op_node_proto_helper.h"
+#include "core/framework/data_types.h"
+#include "core/framework/tensor.h"
+
+// Graph-related headers
+#include "core/graph/graph_viewer.h"
+#include "core/graph/graph.h"
+#include "core/graph/onnx_protobuf.h"
+
+// ONNX schema definitions
+#include "onnx/defs/schema.h"
+
+// Windows-specific headers (if applicable)
+#ifdef _WIN32
+#include <windows.h>
+#endif

cmake/test_pch.h

@@ -0,0 +1,26 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+#pragma once
+
+// Test framework headers (highest compilation time impact)
+#include "gtest/gtest.h"
+#include "gtest/gtest-assertion-result.h"
+#include "gtest/gtest-message.h"
+#include "gtest/internal/gtest-port.h"
+
+// Core test utilities (most frequently used in tests)
+#include "test/providers/provider_test_utils.h"
+#include "test/providers/checkers.h"
+
+// ONNX and Protocol Buffer headers
+#include "core/graph/onnx_protobuf.h"
+#include "onnx/defs/schema.h"
+
+// Data types and framework headers
+#include "core/framework/data_types.h"
+
+// Windows-specific headers (if applicable)
+#ifdef _WIN32
+#include <windows.h>
+#endif

include/onnxruntime/core/session/environment.h

@@ -199,23 +199,6 @@ class Environment {
 
   using OrtAllocatorUniquePtr = std::unique_ptr<OrtAllocator, std::function<void(OrtAllocator*)>>;
 
-  // if the user calls CreateSharedAllocator and wraps the plugin EP's allocator with an arena we end up with
-  // OrtAllocator from EP -> wrapped in IAllocatorImplWrappingOrtAllocator -> inside a BFCArena IAllocator.
-  // we can put that in shared_allocators_ for sessions to use, but to have an OrtAllocator available in
-  // shared_ort_allocators_ that can be used outside of a session we need to additionally wrap that in an
-  // OrtAllocatorImplWrappingIAllocator. way too many levels of indirection but that is what it is currently.
-  // we need something to own that final OrtAllocator, so we add it to arena_ort_allocators_.
-  //
-  // TODO: we could split out the BFCArena implementation so it can be plugged into either an IAllocator
-  // or an OrtAllocator instance to reduce the indirection a little.
-  // with that we get an OrtAllocator from the EP, wrap it with an OrtAllocator based BFCArena, and wrap that with the
-  // IAllocatorImplWrappingOrtAllocator which takes ownership of the OrtAllocator and is in shared_allocators_.
-  //
-  // Alternatively we can disable wrapping an EP's allocator with a BFCArena and say the EP should provide the arena
-  // implementation directly. They're free to copy BFCArena as it came from TF originally. Or we could provide a
-  // cut-and-paste BFCArena implementation that works using the EP API that can be included in the EP source.
-  std::unordered_map<const OrtMemoryInfo*, std::unique_ptr<OrtAllocatorImplWrappingIAllocator>> arena_ort_allocators_;
-
 #if !defined(ORT_MINIMAL_BUILD)
   // register EPs that are built into the ORT binary so they can take part in AutoEP selection
   // added to ep_libraries