[ARM64] [Windows] Fix the local stack size for funclets

[ssijaric]

Summary

The comment was misplaced, and the code didn't do what the comment indicated, namely ignoring the varargs portion when computing the local stack size of a funclet in emitEpilogue. This results in incorrect offset computations within funclets that are contained in varag functions.

[efriedma]

LGTM

lib/Target/AArch64/AArch64FrameLowering.cpp

1292 ↗

(On Diff #175998)

*containing.

llvm/trunk/lib/Target/AArch64/AArch64FrameLowering.cpp

void AArch64FrameLowering::emitEpilogue(MachineFunction &MF,
   //
   // AArch64TargetLowering::LowerCall figures out ArgumentPopSize and keeps
   // it as the 2nd argument of AArch64ISD::TC_RETURN.
 
   auto Cleanup = make_scope_exit([&] { InsertReturnAddressAuth(MF, MBB); });
 
   bool IsWin64 =
       Subtarget.isCallingConvWin64(MF.getFunction().getCallingConv());
-  unsigned FixedObject = IsWin64 ? alignTo(AFI->getVarArgsGPRSize(), 16) : 0;
+  // Var args are accounted for in the containing function, so don't
+  // include them for funclets.
+  unsigned FixedObject =
+      (IsWin64 && !IsFunclet) ? alignTo(AFI->getVarArgsGPRSize(), 16) : 0;
 
   uint64_t AfterCSRPopSize = ArgumentPopSize;
   auto PrologueSaveSize = AFI->getCalleeSavedStackSize() + FixedObject;
-  // Var args are accounted for in the containting function, so don't
-  // include them for funclets.
+  // We cannot rely on the local stack size set in emitPrologue if the function
+  // has funclets, as funclets have different local stack size requirements, and
+  // the current value set in emitPrologue may be that of the containing
+  // function.
   if (MF.hasEHFunclets())
     AFI->setLocalStackSize(NumBytes - PrologueSaveSize);
   bool CombineSPBump = shouldCombineCSRLocalStackBump(MF, NumBytes);
   // Assume we can't combine the last pop with the sp restore.
 
   if (!CombineSPBump && PrologueSaveSize != 0) {
     MachineBasicBlock::iterator Pop = std::prev(MBB.getFirstTerminator());
     while (AArch64InstrInfo::isSEHInstruction(*Pop))

llvm/trunk/test/CodeGen/AArch64/funclet-local-stack-size.ll

+; RUN: llc -o - %s -mtriple=aarch64-windows | FileCheck %s
+; Check that the local stack size is computed correctly for a funclet contained
+; within a varargs function.  The varargs component shouldn't be included in the
+; local stack size computation.
+target datalayout = "e-m:w-p:64:64-i32:32-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64-unknown-windows-msvc19.11.0"
+
+%rtti.TypeDescriptor2 = type { i8**, i8*, [3 x i8] }
+
+$"??_R0H@8" = comdat any
+
+@"??_7type_info@@6B@" = external constant i8*
+@"??_R0H@8" = linkonce_odr global %rtti.TypeDescriptor2 { i8** @"??_7type_info@@6B@", i8* null, [3 x i8] c".H\00" }, comdat
+
+; CHECK-LABEL: ?catch$2@?0??func@@YAHHHZZ@4HA
+; CHECK: stp x29, x30, [sp, #-16]!
+; CHECK: ldp x29, x30, [sp], #16
+; Function Attrs: uwtable
+define dso_local i32 @"?func@@YAHHHZZ"(i32 %a, i32, ...) local_unnamed_addr #0 personality i8* bitcast (i32 (...)* @__CxxFrameHandler3 to i8*) {
+entry:
+  %arr = alloca [10 x i32], align 4
+  %a2 = alloca i32, align 4
+  %1 = bitcast [10 x i32]* %arr to i8*
+  %arraydecay = getelementptr inbounds [10 x i32], [10 x i32]* %arr, i64 0, i64 0
+  %call = call i32 @"?init@@YAHPEAH@Z"(i32* nonnull %arraydecay)
+  %call1 = invoke i32 @"?func2@@YAHXZ"()
+          to label %cleanup unwind label %catch.dispatch
+
+catch.dispatch:                                   ; preds = %entry
+  %2 = catchswitch within none [label %catch] unwind to caller
+
+catch:                                            ; preds = %catch.dispatch
+  %3 = catchpad within %2 [%rtti.TypeDescriptor2* @"??_R0H@8", i32 0, i32* %a2]
+  %4 = load i32, i32* %a2, align 4
+  %add = add nsw i32 %4, 1
+  catchret from %3 to label %cleanup
+
+cleanup:                                          ; preds = %entry, %catch
+  %retval.0 = phi i32 [ %add, %catch ], [ %call1, %entry ]
+  ret i32 %retval.0
+}
+
+declare dso_local i32 @"?init@@YAHPEAH@Z"(i32*)
+
+declare dso_local i32 @"?func2@@YAHXZ"()
+
+declare dso_local i32 @__CxxFrameHandler3(...)
+
+attributes #0 = { uwtable }
+
+!llvm.module.flags = !{!0}
+
+!0 = !{i32 1, !"wchar_size", i32 2}