[PATCH] D120622: [AArch64] Use correct calling convention for each vararg

[Le Philousophe via Phabricator via llvm-commits]

lephilousophe created this revision.
lephilousophe added reviewers: t.p.northover, mstorsjo.
lephilousophe added a project: LLVM.
Herald added subscribers: hiraditya, kristof.beyls.
lephilousophe requested review of this revision.
Herald added a subscriber: llvm-commits.

While checking is tail call optimization is possible, the calling
convention applied to fixed arguments is not the correct one.
This implies for DarwinPCS that all arguments of a vararg function will
go to the stack although fixed ones can go in registers.

This prevents non-virtual thunks to be tail optimized although they are marked
as musttail.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D120622

Files:
  llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
  llvm/test/CodeGen/AArch64/darwinpcs-musttail.ll


Index: llvm/test/CodeGen/AArch64/darwinpcs-musttail.ll
===================================================================
--- /dev/null
+++ llvm/test/CodeGen/AArch64/darwinpcs-musttail.ll
@@ -0,0 +1,17 @@
+; With Darwin PCS, non-virtual thunks generated are generated with musttail
+; and are expected to build
+; RUN: llc -mtriple=arm64-apple-ios5.0.0 < %s | FileCheck %s
+
+; CHECK-LABEL: __ZThn16_N1C3addEPKcz:
+; CHECK:       b __ZN1C3addEPKcz
+
+%class.C = type { %class.A.base, [4 x i8], %class.B.base, [4 x i8] }
+%class.A.base = type <{ i32 (...)**, i32 }>
+%class.B.base = type <{ i32 (...)**, i32 }>
+
+declare void @_ZN1C3addEPKcz(%class.C*, i8*, ...) unnamed_addr #0 align 2
+
+define void @_ZThn16_N1C3addEPKcz(%class.C* %0, i8* %1, ...) unnamed_addr #0 align 2 {
+  musttail call void (%class.C*, i8*, ...) @_ZN1C3addEPKcz(%class.C* noundef nonnull align 8 dereferenceable(28) undef, i8* noundef %1, ...)
+  ret void
+}
Index: llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
===================================================================
--- llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -5916,6 +5916,7 @@
   MachineFunction &MF = DAG.getMachineFunction();
   const Function &CallerF = MF.getFunction();
   CallingConv::ID CallerCC = CallerF.getCallingConv();
+  bool IsCalleeWin64 = Subtarget->isCallingConvWin64(CalleeCC);
 
   // Functions using the C or Fast calling convention that have an SVE signature
   // preserve more registers and should assume the SVE_VectorCall CC.
@@ -5991,7 +5992,20 @@
     SmallVector<CCValAssign, 16> ArgLocs;
     CCState CCInfo(CalleeCC, isVarArg, MF, ArgLocs, C);
 
-    CCInfo.AnalyzeCallOperands(Outs, CCAssignFnForCall(CalleeCC, true));
+    unsigned NumArgs = Outs.size();
+    for (unsigned i = 0; i != NumArgs; ++i) {
+      MVT ArgVT = Outs[i].VT;
+      ISD::ArgFlagsTy ArgFlags = Outs[i].Flags;
+      bool UseVarArgCC = !Outs[i].IsFixed;
+      // On Windows, the fixed arguments in a vararg call are passed in GPRs
+      // too, so use the vararg CC to force them to integer registers.
+      if (IsCalleeWin64)
+        UseVarArgCC = true;
+      CCAssignFn *AssignFn = CCAssignFnForCall(CalleeCC, UseVarArgCC);
+      bool Res = AssignFn(i, ArgVT, ArgVT, CCValAssign::Full, ArgFlags, CCInfo);
+      assert(!Res && "Call operand has unhandled type");
+      (void)Res;
+    }
     for (const CCValAssign &ArgLoc : ArgLocs)
       if (!ArgLoc.isRegLoc())
         return false;
@@ -6022,7 +6036,20 @@
   SmallVector<CCValAssign, 16> ArgLocs;
   CCState CCInfo(CalleeCC, isVarArg, MF, ArgLocs, C);
 
-  CCInfo.AnalyzeCallOperands(Outs, CCAssignFnForCall(CalleeCC, isVarArg));
+  unsigned NumArgs = Outs.size();
+  for (unsigned i = 0; i != NumArgs; ++i) {
+    MVT ArgVT = Outs[i].VT;
+    ISD::ArgFlagsTy ArgFlags = Outs[i].Flags;
+    bool UseVarArgCC = isVarArg && !Outs[i].IsFixed;
+    // On Windows, the fixed arguments in a vararg call are passed in GPRs
+    // too, so use the vararg CC to force them to integer registers.
+    if (isVarArg && IsCalleeWin64)
+      UseVarArgCC = true;
+    CCAssignFn *AssignFn = CCAssignFnForCall(CalleeCC, UseVarArgCC);
+    bool Res = AssignFn(i, ArgVT, ArgVT, CCValAssign::Full, ArgFlags, CCInfo);
+    assert(!Res && "Call operand has unhandled type");
+    (void)Res;
+  }
 
   const AArch64FunctionInfo *FuncInfo = MF.getInfo<AArch64FunctionInfo>();
 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D120622.411639.patch
Type: text/x-patch
Size: 3431 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220226/aabd4b79/attachment.bin>