Merge branch 'main' into dev/grendel/android-clr-host-local

* main:
  [Android] Run CoreCLR functional tests on Android (dotnet#112283)
  [LoongArch64] Fix some assertion failures for Debug ILC building Debug NativeAOT testcases. (dotnet#112229)
  Fix suspicious code fragments (dotnet#112384)
  `__ComObject` doesn't support dynamic interface map (dotnet#112375)
  Native DLLs: only load imported DLLs from System32 (dotnet#112359)
  [main] Update dependencies from dotnet/roslyn (dotnet#112314)
  Update SVE instructions that writes to GC regs (dotnet#112389)
  Bring up android+coreclr windows build.  (dotnet#112256)
  Never use heap for return buffers (dotnet#112060)
  Wait to complete the test before releasing the agile reference. (dotnet#112387)
  Prevent returning disposed HTTP/1.1 connections to the pool (dotnet#112383)
  Fingerprint dotnet.js if writing import map to html is enabled (dotnet#112407)
  Remove duplicate definition of CORECLR_HOSTING_API_LINKAGE (dotnet#112096)
  Update the exception message to reflect current behavior. (dotnet#112355)
  Use enum for frametype not v table (dotnet#112166)
  Enable AltJits build for LoongArch64 and RiscV64 (dotnet#110282)
  Guard members of MonoType union & fix related bugs (dotnet#111645)
  Add optional hooks for debugging OpenSSL memory allocations (dotnet#111539)
  JIT: Optimize struct parameter register accesses in the backend (dotnet#110819)
  NativeAOT: Cover more opcodes in type preinitializer (dotnet#112073)

Author: grendello

Directory.Build.props

@@ -56,6 +56,7 @@
       Minimum target OS versions, keep in sync with:
         - eng/native/configurecompiler.cmake
         - eng/native/build-commons.sh
+        - eng/native/gen-buildsys.cmd
         - src/native/libs/build-native.sh
         - src/coreclr/tools/aot/ILCompiler.Compiler/Compiler/ObjectWriter/MachObjectWriter.cs
         - src/coreclr/nativeaot/BuildIntegration/Microsoft.NETCore.Native.Unix.targets
@@ -328,7 +329,7 @@
     <!-- Runtime doesn't support Arcade-driven target framework filtering. -->
     <NoTargetFrameworkFiltering>true</NoTargetFrameworkFiltering>
 
-    <NativeBuildPartitionPropertiesToRemove>ClrFullNativeBuild;ClrRuntimeSubset;ClrJitSubset;ClrPalTestsSubset;ClrAllJitsSubset;ClrILToolsSubset;ClrNativeAotSubset;ClrSpmiSubset;ClrCrossComponentsSubset;ClrDebugSubset;HostArchitecture;PgoInstrument;NativeOptimizationDataSupported;CMakeArgs;CxxStandardLibrary;CxxStandardLibraryStatic;CxxAbiLibrary</NativeBuildPartitionPropertiesToRemove>
+    <NativeBuildPartitionPropertiesToRemove>ClrFullNativeBuild;ClrRuntimeSubset;ClrJitSubset;ClrPalTestsSubset;ClrAllJitsSubset;ClrAllJitsCommunitySubset;ClrILToolsSubset;ClrNativeAotSubset;ClrSpmiSubset;ClrCrossComponentsSubset;ClrDebugSubset;HostArchitecture;PgoInstrument;NativeOptimizationDataSupported;CMakeArgs;CxxStandardLibrary;CxxStandardLibraryStatic;CxxAbiLibrary</NativeBuildPartitionPropertiesToRemove>
     <UseSystemZlib Condition="'$(TargetsAppleMobile)' == 'true' or '$(TargetOS)' == 'android' or '$(TargetArchitecture)' == 'armv6'">true</UseSystemZlib>
   </PropertyGroup>
 

docs/design/coreclr/botr/clr-abi.md

@@ -108,9 +108,7 @@ Passing/returning structs according to hardware floating-point calling conventio
 
 ## Return buffers
 
-The same applies to some return buffers. See `MethodTable::IsStructRequiringStackAllocRetBuf()`. When that returns `false`, the return buffer might be on the heap, either due to reflection/remoting code paths mentioned previously or due to a JIT optimization where a call with a return buffer that then assigns to a field (on the GC heap) are changed into passing the field reference as the return buffer. Conversely, when it returns true, the JIT does not need to use a write barrier when storing to the return buffer, but it is still not guaranteed to be a compiler temp, and as such the JIT should not introduce spurious writes to the return buffer.
-
-NOTE: This optimization is now disabled for all platforms (`IsStructRequiringStackAllocRetBuf()` always returns `false`).
+Since .NET 10, return buffers must always be allocated on the stack by the caller. After the call, the caller is responsible for copying the return buffer to the final destination using write barriers if necessary. The JIT can assume that the return buffer is always on the stack and may optimize accordingly, such as by omitting write barriers when writing GC pointers to the return buffer. In addition, the buffer is allowed to be used for temporary storage within the method since its content must not be aliased or cross-thread visible.
 
 ARM64-only: When a method returns a structure that is larger than 16 bytes the caller reserves a return buffer of sufficient size and alignment to hold the result. The address of the buffer is passed as an argument to the method in `R8` (defined in the JIT as `REG_ARG_RET_BUFF`). The callee isn't required to preserve the value stored in `R8`.
 
@@ -778,9 +776,7 @@ The return value is handled as follows:
 1. Floating-point values are returned on the top of the hardware FP stack.
 2. Integers up to 32 bits long are returned in EAX.
 3. 64-bit integers are passed with EAX holding the least significant 32 bits and EDX holding the most significant 32 bits.
-4. All other cases require the use of a return buffer, through which the value is returned.
-
-In addition, there is a guarantee that if a return buffer is used a value is stored there only upon ordinary exit from the method. The buffer is not allowed to be used for temporary storage within the method and its contents will be unaltered if an exception occurs while executing the method.
+4. All other cases require the use of a return buffer, through which the value is returned. See [Return buffers](#return-buffers).
 
 # Control Flow Guard (CFG) support on Windows
 

docs/design/features/tailcalls-with-helpers.md

@@ -259,9 +259,9 @@ stack, even on ARM architectures, due to its return address hijacking mechanism.
 
 When the result is returned by value the JIT will introduce a local and pass a
 pointer to it in the second argument. For ret bufs the JIT will typically
-directly pass along its own return buffer parameter to DispatchTailCalls. It is
-possible that this return buffer is pointing into GC heap, so the result is
-always tracked as a byref in the mechanism.
+directly pass along its own return buffer parameter to DispatchTailCalls. The
+return buffer is always located on the stack, so the result does not need to be
+tracked as a byref.
 
 In certain cases the target function pointer is also stored. For some targets
 this might require the JIT to perform the equivalent of `ldvirtftn` or `ldftn`

eng/Subsets.props

@@ -60,7 +60,7 @@
   <PropertyGroup>
     <DefaultSubsets>clr+mono+libs+tools+host+packs</DefaultSubsets>
     <DefaultSubsets Condition="'$(TargetsMobile)' == 'true'">mono+libs+packs</DefaultSubsets>
-    <DefaultSubsets Condition="'$(TargetsAndroid)' == 'true' and '$(CoreCLRSupported)' == 'true'">clr.runtime+clr.alljits+clr.corelib+clr.nativecorelib+clr.tools+clr.packages+mono+libs+host+packs</DefaultSubsets>
+    <DefaultSubsets Condition="'$(TargetsAndroid)' == 'true' and '$(CoreCLRSupported)' == 'true'">clr+mono+libs+host+packs</DefaultSubsets>
     <DefaultSubsets Condition="'$(TargetsAppleMobile)' == 'true'">clr.nativeaotruntime+clr.nativeaotlibs+mono+libs+packs</DefaultSubsets>
     <DefaultSubsets Condition="'$(TargetsLinuxBionic)' == 'true' and '$(MonoSupported)' == 'true'">clr.nativeaotruntime+clr.nativeaotlibs+mono+libs+host+packs</DefaultSubsets>
     <DefaultSubsets Condition="'$(TargetsLinuxBionic)' == 'true' and '$(MonoSupported)' != 'true'">clr.nativeaotruntime+clr.nativeaotlibs+libs+packs</DefaultSubsets>
@@ -100,6 +100,7 @@
   <PropertyGroup>
     <DefaultCoreClrSubsets>clr.native+clr.corelib+clr.tools+clr.nativecorelib+clr.packages+clr.nativeaotlibs+clr.crossarchtools+host.native</DefaultCoreClrSubsets>
     <DefaultCoreClrSubsets Condition="'$(PackageRID)' == 'linux-armel'">clr.native+clr.corelib+clr.tools+clr.nativecorelib+clr.packages+clr.nativeaotlibs+clr.crossarchtools</DefaultCoreClrSubsets>
+    <DefaultCoreClrSubsets Condition="'$(TargetsAndroid)' == 'true'">clr.native+clr.corelib+clr.tools+clr.nativecorelib+clr.packages+clr.nativeaotlibs+clr.crossarchtools</DefaultCoreClrSubsets>
     <!-- Even on platforms that do not support the CoreCLR runtime, we still want to build ilasm/ildasm. -->
     <DefaultCoreClrSubsets Condition="'$(RuntimeFlavor)' != 'CoreCLR'">clr.iltools+clr.packages</DefaultCoreClrSubsets>
 
@@ -191,6 +192,7 @@
     <SubsetName Include="Clr.Hosts" Description="The CoreCLR corerun test host." />
     <SubsetName Include="Clr.Jit" Description="The JIT for the CoreCLR .NET runtime." />
     <SubsetName Include="Clr.AllJits" Description="All of the cross-targeting JIT compilers for the CoreCLR .NET runtime." />
+    <SubsetName Include="Clr.AllJitsCommunity" Description="All of the cross-targeting JIT compilers for the CoreCLR .NET runtime, including community targets." />
     <SubsetName Include="Clr.Spmi" Description="SuperPMI, a tool for CoreCLR JIT testing." />
     <SubsetName Include="Clr.CoreLib" Description="The managed System.Private.CoreLib library for CoreCLR." />
     <SubsetName Include="Clr.NativeCoreLib" Description="Run crossgen on System.Private.CoreLib library for CoreCLR." />
@@ -291,6 +293,10 @@
     <ClrRuntimeBuildSubsets>$(ClrRuntimeBuildSubsets);ClrAllJitsSubset=true</ClrRuntimeBuildSubsets>
   </PropertyGroup>
 
+  <PropertyGroup Condition="$(_subset.Contains('+clr.alljitscommunity+'))">
+    <ClrRuntimeBuildSubsets>$(ClrRuntimeBuildSubsets);ClrAllJitsCommunitySubset=true</ClrRuntimeBuildSubsets>
+  </PropertyGroup>
+
   <PropertyGroup Condition="$(_subset.Contains('+clr.iltools+'))">
     <ClrRuntimeBuildSubsets>$(ClrRuntimeBuildSubsets);ClrILToolsSubset=true</ClrRuntimeBuildSubsets>
   </PropertyGroup>

eng/Version.Details.xml

@@ -376,17 +376,17 @@
       <Uri>https://github.com/dotnet/runtime-assets</Uri>
       <Sha>3faec1bca5963b4ca54741e12939c77b951df604</Sha>
     </Dependency>
-    <Dependency Name="Microsoft.Net.Compilers.Toolset" Version="4.14.0-2.25107.8">
+    <Dependency Name="Microsoft.Net.Compilers.Toolset" Version="4.14.0-2.25110.8">
       <Uri>https://github.com/dotnet/roslyn</Uri>
-      <Sha>651ee91d035cdf8e659c00a384d5d6671aadbe46</Sha>
+      <Sha>d3972005ce081b597a9dab72761d281e57acdc5a</Sha>
     </Dependency>
-    <Dependency Name="Microsoft.CodeAnalysis" Version="4.14.0-2.25107.8">
+    <Dependency Name="Microsoft.CodeAnalysis" Version="4.14.0-2.25110.8">
       <Uri>https://github.com/dotnet/roslyn</Uri>
-      <Sha>651ee91d035cdf8e659c00a384d5d6671aadbe46</Sha>
+      <Sha>d3972005ce081b597a9dab72761d281e57acdc5a</Sha>
     </Dependency>
-    <Dependency Name="Microsoft.CodeAnalysis.CSharp" Version="4.14.0-2.25107.8">
+    <Dependency Name="Microsoft.CodeAnalysis.CSharp" Version="4.14.0-2.25110.8">
       <Uri>https://github.com/dotnet/roslyn</Uri>
-      <Sha>651ee91d035cdf8e659c00a384d5d6671aadbe46</Sha>
+      <Sha>d3972005ce081b597a9dab72761d281e57acdc5a</Sha>
     </Dependency>
     <Dependency Name="Microsoft.CodeAnalysis.Analyzers" Version="3.12.0-beta1.25081.1">
       <Uri>https://github.com/dotnet/roslyn-analyzers</Uri>
@@ -397,9 +397,9 @@
       <Sha>8fe7aeb135c64e095f43292c427453858d937184</Sha>
     </Dependency>
     <!-- Intermediate is necessary for source build. -->
-    <Dependency Name="Microsoft.SourceBuild.Intermediate.roslyn" Version="4.14.0-2.25107.8">
+    <Dependency Name="Microsoft.SourceBuild.Intermediate.roslyn" Version="4.14.0-2.25110.8">
       <Uri>https://github.com/dotnet/roslyn</Uri>
-      <Sha>651ee91d035cdf8e659c00a384d5d6671aadbe46</Sha>
+      <Sha>d3972005ce081b597a9dab72761d281e57acdc5a</Sha>
       <SourceBuild RepoName="roslyn" ManagedOnly="true" />
     </Dependency>
     <Dependency Name="Microsoft.DotNet.ApiCompat.Task" Version="10.0.100-alpha.1.25077.2">

eng/Versions.props

@@ -44,9 +44,9 @@
       Any tools that contribute to the design-time experience should use the MicrosoftCodeAnalysisVersion_LatestVS property above to ensure
       they do not break the local dev experience.
     -->
-    <MicrosoftCodeAnalysisCSharpVersion>4.14.0-2.25107.8</MicrosoftCodeAnalysisCSharpVersion>
-    <MicrosoftCodeAnalysisVersion>4.14.0-2.25107.8</MicrosoftCodeAnalysisVersion>
-    <MicrosoftNetCompilersToolsetVersion>4.14.0-2.25107.8</MicrosoftNetCompilersToolsetVersion>
+    <MicrosoftCodeAnalysisCSharpVersion>4.14.0-2.25110.8</MicrosoftCodeAnalysisCSharpVersion>
+    <MicrosoftCodeAnalysisVersion>4.14.0-2.25110.8</MicrosoftCodeAnalysisVersion>
+    <MicrosoftNetCompilersToolsetVersion>4.14.0-2.25110.8</MicrosoftNetCompilersToolsetVersion>
   </PropertyGroup>
   <!--
     For source generator support we need to target multiple versions of Roslyn in order to be able to run on older versions of Roslyn.

eng/native/build-commons.sh

@@ -100,10 +100,7 @@ build_native()
             exit 1
         fi
 
-        # cmake cache scripts can't see command line args
-        export ANDROID_BUILD=1
-
-        cmakeArgs="-C $__RepoRootDir/eng/native/tryrun.cmake $cmakeArgs"
+        cmakeArgs="-DANDROID_BUILD=1 -C $__RepoRootDir/eng/native/tryrun.cmake $cmakeArgs"
         cmakeArgs="-DCMAKE_TOOLCHAIN_FILE=$ANDROID_NDK_ROOT/build/cmake/android.toolchain.cmake -DANDROID_PLATFORM=android-${ANDROID_API_LEVEL} -DANDROID_NATIVE_API_LEVEL=${ANDROID_API_LEVEL} $cmakeArgs"
 
         # Don't try to set CC/CXX in init-compiler.sh - it's handled in android.toolchain.cmake already

eng/native/configurecompiler.cmake

@@ -70,6 +70,9 @@ if (MSVC)
   add_compile_options($<$<COMPILE_LANGUAGE:CXX>:$<TARGET_PROPERTY:CLR_EH_OPTION>>)
   add_link_options($<$<BOOL:$<TARGET_PROPERTY:CLR_CONTROL_FLOW_GUARD>>:/guard:cf>)
 
+  # Load all imported DLLs from the System32 directory.
+  add_linker_flag(/DEPENDENTLOADFLAG:0x800)
+
   # Linker flags
   #
   set (WINDOWS_SUBSYSTEM_VERSION 6.01)

eng/native/configureplatform.cmake

@@ -368,6 +368,11 @@ if(CLR_CMAKE_HOST_LINUX_MUSL OR CLR_CMAKE_TARGET_OS STREQUAL alpine)
     set(CLR_CMAKE_TARGET_LINUX_MUSL 1)
 endif(CLR_CMAKE_HOST_LINUX_MUSL OR CLR_CMAKE_TARGET_OS STREQUAL alpine)
 
+macro(set_cache_value)
+  set(${ARGV0} ${ARGV1} CACHE STRING "Result from TRY_RUN" FORCE)
+  set(${ARGV0}__TRYRUN_OUTPUT "dummy output" CACHE STRING "Output from TRY_RUN" FORCE)
+endmacro()
+
 if(CLR_CMAKE_TARGET_OS STREQUAL android)
     set(CLR_CMAKE_TARGET_UNIX 1)
     set(CLR_CMAKE_TARGET_LINUX 1)
@@ -471,7 +476,7 @@ if(CLR_CMAKE_TARGET_OS STREQUAL windows)
 endif()
 
 # check if host & target os/arch combination are valid
-if (NOT (CLR_CMAKE_TARGET_OS STREQUAL CLR_CMAKE_HOST_OS) AND NOT CLR_CMAKE_TARGET_WASI)
+if (NOT (CLR_CMAKE_TARGET_OS STREQUAL CLR_CMAKE_HOST_OS) AND NOT CLR_CMAKE_TARGET_WASI AND NOT CLR_CMAKE_TARGET_ANDROID)
     if(NOT (CLR_CMAKE_HOST_OS STREQUAL windows))
         message(FATAL_ERROR "Invalid host and target os/arch combination. Host OS: ${CLR_CMAKE_HOST_OS}")
     endif()

eng/native/configuretools.cmake

@@ -77,10 +77,17 @@ endif()
 
 if (NOT CLR_CMAKE_HOST_WIN32)
   # detect linker
-  execute_process(COMMAND sh -c "${CMAKE_C_COMPILER} ${CMAKE_SHARED_LINKER_FLAGS} -Wl,--version | head -1"
-    ERROR_QUIET
-    OUTPUT_VARIABLE ldVersionOutput
-    OUTPUT_STRIP_TRAILING_WHITESPACE)
+  if (CMAKE_HOST_SYSTEM_NAME STREQUAL "Windows")
+    execute_process(COMMAND ${CMAKE_C_COMPILER} -Wl,--version
+      ERROR_QUIET
+      OUTPUT_VARIABLE ldVersionOutput
+      OUTPUT_STRIP_TRAILING_WHITESPACE)
+  else()
+    execute_process(COMMAND sh -c "${CMAKE_C_COMPILER} ${CMAKE_SHARED_LINKER_FLAGS} -Wl,--version | head -1"
+      ERROR_QUIET
+      OUTPUT_VARIABLE ldVersionOutput
+      OUTPUT_STRIP_TRAILING_WHITESPACE)
+  endif()
 
   if("${ldVersionOutput}" MATCHES "LLD")
     set(LD_LLVM 1)