[ARM64] [Windows] Implement _InterlockedExchangeAdd*_* builtins.
These apparently need to be proper builtins to handle the Windows
SDK.
Differential Revision: https://reviews.llvm.org/D53916
llvm-svn: 345779
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index f058109..4dcf158 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -95,9 +95,9 @@
/// Utility to insert an atomic instruction based on Instrinsic::ID
/// and the expression node.
-static Value *MakeBinaryAtomicValue(CodeGenFunction &CGF,
- llvm::AtomicRMWInst::BinOp Kind,
- const CallExpr *E) {
+static Value *MakeBinaryAtomicValue(
+ CodeGenFunction &CGF, llvm::AtomicRMWInst::BinOp Kind, const CallExpr *E,
+ AtomicOrdering Ordering = AtomicOrdering::SequentiallyConsistent) {
QualType T = E->getType();
assert(E->getArg(0)->getType()->isPointerType());
assert(CGF.getContext().hasSameUnqualifiedType(T,
@@ -119,7 +119,7 @@
Args[1] = EmitToInt(CGF, Args[1], T, IntType);
llvm::Value *Result = CGF.Builder.CreateAtomicRMW(
- Kind, Args[0], Args[1], llvm::AtomicOrdering::SequentiallyConsistent);
+ Kind, Args[0], Args[1], Ordering);
return EmitFromInt(CGF, Result, T, ValueType);
}
@@ -745,6 +745,9 @@
_InterlockedIncrement,
_InterlockedOr,
_InterlockedXor,
+ _InterlockedExchangeAdd_acq,
+ _InterlockedExchangeAdd_rel,
+ _InterlockedExchangeAdd_nf,
__fastfail,
};
@@ -811,6 +814,15 @@
return MakeBinaryAtomicValue(*this, AtomicRMWInst::Or, E);
case MSVCIntrin::_InterlockedXor:
return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xor, E);
+ case MSVCIntrin::_InterlockedExchangeAdd_acq:
+ return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E,
+ AtomicOrdering::Acquire);
+ case MSVCIntrin::_InterlockedExchangeAdd_rel:
+ return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E,
+ AtomicOrdering::Release);
+ case MSVCIntrin::_InterlockedExchangeAdd_nf:
+ return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E,
+ AtomicOrdering::Monotonic);
case MSVCIntrin::_InterlockedDecrement: {
llvm::Type *IntTy = ConvertType(E->getType());
@@ -6109,6 +6121,21 @@
return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedDecrement, E);
case ARM::BI_InterlockedIncrement64:
return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedIncrement, E);
+ case ARM::BI_InterlockedExchangeAdd8_acq:
+ case ARM::BI_InterlockedExchangeAdd16_acq:
+ case ARM::BI_InterlockedExchangeAdd_acq:
+ case ARM::BI_InterlockedExchangeAdd64_acq:
+ return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedExchangeAdd_acq, E);
+ case ARM::BI_InterlockedExchangeAdd8_rel:
+ case ARM::BI_InterlockedExchangeAdd16_rel:
+ case ARM::BI_InterlockedExchangeAdd_rel:
+ case ARM::BI_InterlockedExchangeAdd64_rel:
+ return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedExchangeAdd_rel, E);
+ case ARM::BI_InterlockedExchangeAdd8_nf:
+ case ARM::BI_InterlockedExchangeAdd16_nf:
+ case ARM::BI_InterlockedExchangeAdd_nf:
+ case ARM::BI_InterlockedExchangeAdd64_nf:
+ return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedExchangeAdd_nf, E);
}
// Get the last argument, which specifies the vector type.
@@ -8590,6 +8617,21 @@
return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedDecrement, E);
case AArch64::BI_InterlockedIncrement64:
return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedIncrement, E);
+ case AArch64::BI_InterlockedExchangeAdd8_acq:
+ case AArch64::BI_InterlockedExchangeAdd16_acq:
+ case AArch64::BI_InterlockedExchangeAdd_acq:
+ case AArch64::BI_InterlockedExchangeAdd64_acq:
+ return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedExchangeAdd_acq, E);
+ case AArch64::BI_InterlockedExchangeAdd8_rel:
+ case AArch64::BI_InterlockedExchangeAdd16_rel:
+ case AArch64::BI_InterlockedExchangeAdd_rel:
+ case AArch64::BI_InterlockedExchangeAdd64_rel:
+ return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedExchangeAdd_rel, E);
+ case AArch64::BI_InterlockedExchangeAdd8_nf:
+ case AArch64::BI_InterlockedExchangeAdd16_nf:
+ case AArch64::BI_InterlockedExchangeAdd_nf:
+ case AArch64::BI_InterlockedExchangeAdd64_nf:
+ return EmitMSVCBuiltinExpr(MSVCIntrin::_InterlockedExchangeAdd_nf, E);
case AArch64::BI_InterlockedAdd: {
Value *Arg0 = EmitScalarExpr(E->getArg(0));
diff --git a/clang/lib/Headers/intrin.h b/clang/lib/Headers/intrin.h
index 72157fa..7987d43 100644
--- a/clang/lib/Headers/intrin.h
+++ b/clang/lib/Headers/intrin.h
@@ -329,54 +329,18 @@
|* Interlocked Exchange Add
\*----------------------------------------------------------------------------*/
#if defined(__arm__) || defined(__aarch64__)
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd8_acq(char volatile *_Addend, char _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd8_nf(char volatile *_Addend, char _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ char __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd8_rel(char volatile *_Addend, char _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd16_acq(short volatile *_Addend, short _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd16_nf(short volatile *_Addend, short _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ short __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd16_rel(short volatile *_Addend, short _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELEASE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd_acq(long volatile *_Addend, long _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd_nf(long volatile *_Addend, long _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ long __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd_rel(long volatile *_Addend, long _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELEASE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd64_acq(__int64 volatile *_Addend, __int64 _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_ACQUIRE);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd64_nf(__int64 volatile *_Addend, __int64 _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELAXED);
-}
-static __inline__ __int64 __DEFAULT_FN_ATTRS
-_InterlockedExchangeAdd64_rel(__int64 volatile *_Addend, __int64 _Value) {
- return __atomic_fetch_add(_Addend, _Value, __ATOMIC_RELEASE);
-}
+char _InterlockedExchangeAdd8_acq(char volatile *_Addend, char _Value);
+char _InterlockedExchangeAdd8_nf(char volatile *_Addend, char _Value);
+char _InterlockedExchangeAdd8_rel(char volatile *_Addend, char _Value);
+short _InterlockedExchangeAdd16_acq(short volatile *_Addend, short _Value);
+short _InterlockedExchangeAdd16_nf(short volatile *_Addend, short _Value);
+short _InterlockedExchangeAdd16_rel(short volatile *_Addend, short _Value);
+long _InterlockedExchangeAdd_acq(long volatile *_Addend, long _Value);
+long _InterlockedExchangeAdd_nf(long volatile *_Addend, long _Value);
+long _InterlockedExchangeAdd_rel(long volatile *_Addend, long _Value);
+__int64 _InterlockedExchangeAdd64_acq(__int64 volatile *_Addend, __int64 _Value);
+__int64 _InterlockedExchangeAdd64_nf(__int64 volatile *_Addend, __int64 _Value);
+__int64 _InterlockedExchangeAdd64_rel(__int64 volatile *_Addend, __int64 _Value);
#endif
/*----------------------------------------------------------------------------*\
|* Interlocked Increment