gh-130039: Tailcall for windows builds

.github/workflows/tail-call.yml

@@ -6,12 +6,14 @@ on:
       - 'Python/ceval.c'
       - 'Python/ceval_macros.h'
       - 'Python/generated_cases.c.h'
+      - '.github/workflows/tail-call.yml'
   push:
     paths:
       - 'Python/bytecodes.c'
       - 'Python/ceval.c'
       - 'Python/ceval_macros.h'
       - 'Python/generated_cases.c.h'
+      - '.github/workflows/tail-call.yml'
   workflow_dispatch:
 
 permissions:
@@ -35,7 +37,7 @@ jobs:
         target:
 # Un-comment as we add support for more platforms for tail-calling interpreters.
 #          - i686-pc-windows-msvc/msvc
-#          - x86_64-pc-windows-msvc/msvc
+          - x86_64-pc-windows-msvc/msvc
 #          - aarch64-pc-windows-msvc/msvc
           - x86_64-apple-darwin/clang
           - aarch64-apple-darwin/clang
@@ -48,9 +50,9 @@ jobs:
 #          - target: i686-pc-windows-msvc/msvc
 #            architecture: Win32
 #            runner: windows-latest
-#          - target: x86_64-pc-windows-msvc/msvc
-#            architecture: x64
-#            runner: windows-latest
+          - target: x86_64-pc-windows-msvc/msvc
+            architecture: x64
+            runner: windows-latest
 #          - target: aarch64-pc-windows-msvc/msvc
 #            architecture: ARM64
 #            runner: windows-latest
@@ -79,23 +81,31 @@ jobs:
 
       - name: Native Windows (debug)
         if: runner.os == 'Windows' && matrix.architecture != 'ARM64'
+        shell: cmd
         run: |
-          choco install llvm --allow-downgrade --no-progress --version ${{ matrix.llvm }}.1.0
+          choco install llvm --allow-downgrade --no-progress --version ${{ matrix.llvm }}.1.5
+          set PlatformToolset=clangcl
+          set LLVMToolsVersion=19.1.5
+          set LLVMInstallDir=C:\Program Files\LLVM
           ./PCbuild/build.bat --tail-call-interp -d -p ${{ matrix.architecture }}
           ./PCbuild/rt.bat -d -p ${{ matrix.architecture }} -q --multiprocess 0 --timeout 4500 --verbose2 --verbose3
 
       # No tests (yet):
       - name: Emulated Windows (release)
         if: runner.os == 'Windows' && matrix.architecture == 'ARM64'
+        shell: cmd
         run: |
-          choco install llvm --allow-downgrade --no-progress --version ${{ matrix.llvm }}.1.0
+          choco install llvm --allow-downgrade --no-progress --version ${{ matrix.llvm }}.1.5
+          set PlatformToolset=clangcl
+          set LLVMToolsVersion=19.1.5
+          set LLVMInstallDir=C:\Program Files\LLVM
           ./PCbuild/build.bat --tail-call-interp -p ${{ matrix.architecture }}
 
         # The `find` line is required as a result of https://github.com/actions/runner-images/issues/9966.
         # This is a bug in the macOS runner image where the pre-installed Python is installed in the same
         # directory as the Homebrew Python, which causes the build to fail for macos-13. This line removes
         # the symlink to the pre-installed Python so that the Homebrew Python is used instead.
-      - name: Native macOS (debug)
+      - name: Native macOS (release)
         if: runner.os == 'macOS'
         run: |
           brew update
@@ -104,16 +114,16 @@ jobs:
           export SDKROOT="$(xcrun --show-sdk-path)"
           export PATH="/opt/homebrew/opt/llvm/bin:$PATH"
           export PATH="/usr/local/opt/llvm/bin:$PATH"
-          CC=clang-19 ./configure --with-tail-call-interp --with-pydebug
+          CC=clang-19 ./configure --with-tail-call-interp
           make all --jobs 4
           ./python.exe -m test --multiprocess 0 --timeout 4500 --verbose2 --verbose3
 
-      - name: Native Linux (release)
+      - name: Native Linux (debug)
         if: runner.os == 'Linux' && matrix.target != 'free-threading'
         run: |
           sudo bash -c "$(wget -O - https://apt.llvm.org/llvm.sh)" ./llvm.sh ${{ matrix.llvm }}
           export PATH="$(llvm-config-${{ matrix.llvm }} --bindir):$PATH"
-          CC=clang-19 ./configure --with-tail-call-interp
+          CC=clang-19 ./configure --with-tail-call-interp --with-pydebug
           make all --jobs 4
           ./python -m test --multiprocess 0 --timeout 4500 --verbose2 --verbose3
 

Include/internal/pycore_debug_offsets.h

@@ -23,7 +23,7 @@ extern "C" {
    declaration                                      \
    _GENERATE_DEBUG_SECTION_LINUX(name)
 
-#if defined(MS_WINDOWS)
+#if defined(MS_WINDOWS) && !defined(__clang__)
 #define _GENERATE_DEBUG_SECTION_WINDOWS(name)                       \
    _Pragma(Py_STRINGIFY(section(Py_STRINGIFY(name), read, write))) \
    __declspec(allocate(Py_STRINGIFY(name)))

PCbuild/build.bat

@@ -40,6 +40,7 @@ echo.  --experimental-jit          Enable the experimental just-in-time compiler
 echo.  --experimental-jit-off      Ditto but off by default (PYTHON_JIT=1 enables).
 echo.  --experimental-jit-interpreter  Enable the experimental Tier 2 interpreter.
 echo.  --pystats      Enable PyStats collection.
+echo.  --tail-call-interp  Enable tail-calling interpreter.
 echo.
 echo.Available flags to avoid building certain modules.
 echo.These flags have no effect if '-e' is not given:
@@ -95,6 +96,7 @@ if "%~1"=="--experimental-jit-off" (set UseJIT=true) & (set UseTIER2=3) & shift
 if "%~1"=="--experimental-jit-interpreter" (set UseTIER2=4) & shift & goto CheckOpts
 if "%~1"=="--experimental-jit-interpreter-off" (set UseTIER2=6) & shift & goto CheckOpts
 if "%~1"=="--pystats" (set PyStats=1) & shift & goto CheckOpts
+if "%~1"=="--tail-call-interp" (set UseTailCallInterp=true) & shift & goto CheckOpts
 rem These use the actual property names used by MSBuild.  We could just let
 rem them in through the environment, but we specify them on the command line
 rem anyway for visibility so set defaults after this
@@ -189,6 +191,7 @@ echo on
  /p:UseJIT=%UseJIT%^
  /p:UseTIER2=%UseTIER2%^
  /p:PyStats=%PyStats%^
+ /p:UseTailCallInterp=%UseTailCallInterp%^
  %1 %2 %3 %4 %5 %6 %7 %8 %9
 
 @echo off

PCbuild/pythoncore.vcxproj

@@ -106,6 +106,7 @@
       <PreprocessorDefinitions Condition="$(IncludeExternals)">_Py_HAVE_ZLIB;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <PreprocessorDefinitions Condition="'$(UseJIT)' == 'true'">_Py_JIT;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <PreprocessorDefinitions Condition="'$(UseTIER2)' != '' and '$(UseTIER2)' != '0'">_Py_TIER2=$(UseTIER2);%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions Condition="'$(UseTailCallInterp)' == 'true'">Py_TAIL_CALL_INTERP=1;%(PreprocessorDefinitions)</PreprocessorDefinitions>
     </ClCompile>
     <Link>
       <AdditionalDependencies>version.lib;ws2_32.lib;pathcch.lib;bcrypt.lib;%(AdditionalDependencies)</AdditionalDependencies>