Vendor import of llvm-project main llvmorg-14-init-13186-g0c553cc1af2e.vendor/llvm-project/llvmorg-14-init-13186-g0c553cc1af2e

210 files changed, 6399 insertions, 3171 deletions

diff --git a/clang/lib/ARCMigrate/ARCMT.cpp b/clang/lib/ARCMigrate/ARCMT.cpp
index 4851c434d765..68ee7c59270e 100644
--- a/clang/lib/ARCMigrate/ARCMT.cpp
+++ b/clang/lib/ARCMigrate/ARCMT.cpp
@@ -162,9 +162,7 @@ static bool HasARCRuntime(CompilerInvocation &origCI) {
     return triple.getOSMajorVersion() >= 11;
 
   if (triple.getOS() == llvm::Triple::MacOSX) {
-    unsigned Major, Minor, Micro;
-    triple.getOSVersion(Major, Minor, Micro);
-    return Major > 10 || (Major == 10 && Minor >= 7);
+    return triple.getOSVersion() >= VersionTuple(10, 7);
   }
 
   return false;
diff --git a/clang/lib/AST/ASTContext.cpp b/clang/lib/AST/ASTContext.cpp
index 2d85d72e5b8a..008b703d4c1a 100644
--- a/clang/lib/AST/ASTContext.cpp
+++ b/clang/lib/AST/ASTContext.cpp
@@ -2286,8 +2286,8 @@ TypeInfo ASTContext::getTypeInfoImpl(const Type *T) const {
     Align = toBits(Layout.getAlignment());
     break;
   }
-  case Type::ExtInt: {
-    const auto *EIT = cast<ExtIntType>(T);
+  case Type::BitInt: {
+    const auto *EIT = cast<BitIntType>(T);
     Align =
         std::min(static_cast<unsigned>(std::max(
                      getCharWidth(), llvm::PowerOf2Ceil(EIT->getNumBits()))),
@@ -2349,6 +2349,9 @@ TypeInfo ASTContext::getTypeInfoImpl(const Type *T) const {
   case Type::ObjCTypeParam:
     return getTypeInfo(cast<ObjCTypeParamType>(T)->desugar().getTypePtr());
 
+  case Type::Using:
+    return getTypeInfo(cast<UsingType>(T)->desugar().getTypePtr());
+
   case Type::Typedef: {
     const TypedefNameDecl *Typedef = cast<TypedefType>(T)->getDecl();
     TypeInfo Info = getTypeInfo(Typedef->getUnderlyingType().getTypePtr());
@@ -3569,8 +3572,8 @@ QualType ASTContext::getVariableArrayDecayedType(QualType type) const {
   case Type::Auto:
   case Type::DeducedTemplateSpecialization:
   case Type::PackExpansion:
-  case Type::ExtInt:
-  case Type::DependentExtInt:
+  case Type::BitInt:
+  case Type::DependentBitInt:
     llvm_unreachable("type should never be variably-modified");
 
   // These types can be variably-modified but should never need to
@@ -4482,34 +4485,34 @@ QualType ASTContext::getWritePipeType(QualType T) const {
   return getPipeType(T, false);
 }
 
-QualType ASTContext::getExtIntType(bool IsUnsigned, unsigned NumBits) const {
+QualType ASTContext::getBitIntType(bool IsUnsigned, unsigned NumBits) const {
   llvm::FoldingSetNodeID ID;
-  ExtIntType::Profile(ID, IsUnsigned, NumBits);
+  BitIntType::Profile(ID, IsUnsigned, NumBits);
 
   void *InsertPos = nullptr;
-  if (ExtIntType *EIT = ExtIntTypes.FindNodeOrInsertPos(ID, InsertPos))
+  if (BitIntType *EIT = BitIntTypes.FindNodeOrInsertPos(ID, InsertPos))
     return QualType(EIT, 0);
 
-  auto *New = new (*this, TypeAlignment) ExtIntType(IsUnsigned, NumBits);
-  ExtIntTypes.InsertNode(New, InsertPos);
+  auto *New = new (*this, TypeAlignment) BitIntType(IsUnsigned, NumBits);
+  BitIntTypes.InsertNode(New, InsertPos);
   Types.push_back(New);
   return QualType(New, 0);
 }
 
-QualType ASTContext::getDependentExtIntType(bool IsUnsigned,
+QualType ASTContext::getDependentBitIntType(bool IsUnsigned,
                                             Expr *NumBitsExpr) const {
   assert(NumBitsExpr->isInstantiationDependent() && "Only good for dependent");
   llvm::FoldingSetNodeID ID;
-  DependentExtIntType::Profile(ID, *this, IsUnsigned, NumBitsExpr);
+  DependentBitIntType::Profile(ID, *this, IsUnsigned, NumBitsExpr);
 
   void *InsertPos = nullptr;
-  if (DependentExtIntType *Existing =
-          DependentExtIntTypes.FindNodeOrInsertPos(ID, InsertPos))
+  if (DependentBitIntType *Existing =
+          DependentBitIntTypes.FindNodeOrInsertPos(ID, InsertPos))
     return QualType(Existing, 0);
 
   auto *New = new (*this, TypeAlignment)
-      DependentExtIntType(*this, IsUnsigned, NumBitsExpr);
-  DependentExtIntTypes.InsertNode(New, InsertPos);
+      DependentBitIntType(*this, IsUnsigned, NumBitsExpr);
+  DependentBitIntTypes.InsertNode(New, InsertPos);
 
   Types.push_back(New);
   return QualType(New, 0);
@@ -4568,9 +4571,7 @@ QualType ASTContext::getTypeDeclTypeSlow(const TypeDecl *Decl) const {
     assert(Enum->isFirstDecl() && "enum has previous declaration");
     return getEnumType(Enum);
   } else if (const auto *Using = dyn_cast<UnresolvedUsingTypenameDecl>(Decl)) {
-    Type *newType = new (*this, TypeAlignment) UnresolvedUsingType(Using);
-    Decl->TypeForDecl = newType;
-    Types.push_back(newType);
+    return getUnresolvedUsingType(Using);
   } else
     llvm_unreachable("TypeDecl without a type?");
 
@@ -4593,6 +4594,27 @@ QualType ASTContext::getTypedefType(const TypedefNameDecl *Decl,
   return QualType(newType, 0);
 }
 
+QualType ASTContext::getUsingType(const UsingShadowDecl *Found,
+                                  QualType Underlying) const {
+  llvm::FoldingSetNodeID ID;
+  UsingType::Profile(ID, Found);
+
+  void *InsertPos = nullptr;
+  UsingType *T = UsingTypes.FindNodeOrInsertPos(ID, InsertPos);
+  if (T)
+    return QualType(T, 0);
+
+  assert(!Underlying.hasLocalQualifiers());
+  assert(Underlying == getTypeDeclType(cast<TypeDecl>(Found->getTargetDecl())));
+  QualType Canon = Underlying.getCanonicalType();
+
+  UsingType *NewType =
+      new (*this, TypeAlignment) UsingType(Found, Underlying, Canon);
+  Types.push_back(NewType);
+  UsingTypes.InsertNode(NewType, InsertPos);
+  return QualType(NewType, 0);
+}
+
 QualType ASTContext::getRecordType(const RecordDecl *Decl) const {
   if (Decl->TypeForDecl) return QualType(Decl->TypeForDecl, 0);
 
@@ -4619,6 +4641,22 @@ QualType ASTContext::getEnumType(const EnumDecl *Decl) const {
   return QualType(newType, 0);
 }
 
+QualType ASTContext::getUnresolvedUsingType(
+    const UnresolvedUsingTypenameDecl *Decl) const {
+  if (Decl->TypeForDecl)
+    return QualType(Decl->TypeForDecl, 0);
+
+  if (const UnresolvedUsingTypenameDecl *CanonicalDecl =
+          Decl->getCanonicalDecl())
+    if (CanonicalDecl->TypeForDecl)
+      return QualType(Decl->TypeForDecl = CanonicalDecl->TypeForDecl, 0);
+
+  Type *newType = new (*this, TypeAlignment) UnresolvedUsingType(Decl);
+  Decl->TypeForDecl = newType;
+  Types.push_back(newType);
+  return QualType(newType, 0);
+}
+
 QualType ASTContext::getAttributedType(attr::Kind attrKind,
                                        QualType modifiedType,
                                        QualType equivalentType) {
@@ -6444,7 +6482,7 @@ unsigned ASTContext::getIntegerRank(const Type *T) const {
 
   // Results in this 'losing' to any type of the same size, but winning if
   // larger.
-  if (const auto *EIT = dyn_cast<ExtIntType>(T))
+  if (const auto *EIT = dyn_cast<BitIntType>(T))
     return 0 + (EIT->getNumBits() << 3);
 
   switch (cast<BuiltinType>(T)->getKind()) {
@@ -7885,7 +7923,7 @@ void ASTContext::getObjCEncodingForTypeImpl(QualType T, std::string &S,
     return;
 
   case Type::Pipe:
-  case Type::ExtInt:
+  case Type::BitInt:
 #define ABSTRACT_TYPE(KIND, BASE)
 #define TYPE(KIND, BASE)
 #define DEPENDENT_TYPE(KIND, BASE) \
@@ -9234,7 +9272,7 @@ void getIntersectionOfProtocols(ASTContext &Context,
   // Remove any implied protocols from the list of inherited protocols.
   if (!ImpliedProtocols.empty()) {
     llvm::erase_if(IntersectionSet, [&](ObjCProtocolDecl *proto) -> bool {
-      return ImpliedProtocols.count(proto) > 0;
+      return ImpliedProtocols.contains(proto);
     });
   }
 
@@ -10101,12 +10139,12 @@ QualType ASTContext::mergeTypes(QualType LHS, QualType RHS,
     assert(LHS != RHS &&
            "Equivalent pipe types should have already been handled!");
     return {};
-  case Type::ExtInt: {
-    // Merge two ext-int types, while trying to preserve typedef info.
-    bool LHSUnsigned  = LHS->castAs<ExtIntType>()->isUnsigned();
-    bool RHSUnsigned = RHS->castAs<ExtIntType>()->isUnsigned();
-    unsigned LHSBits = LHS->castAs<ExtIntType>()->getNumBits();
-    unsigned RHSBits = RHS->castAs<ExtIntType>()->getNumBits();
+  case Type::BitInt: {
+    // Merge two bit-precise int types, while trying to preserve typedef info.
+    bool LHSUnsigned = LHS->castAs<BitIntType>()->isUnsigned();
+    bool RHSUnsigned = RHS->castAs<BitIntType>()->isUnsigned();
+    unsigned LHSBits = LHS->castAs<BitIntType>()->getNumBits();
+    unsigned RHSBits = RHS->castAs<BitIntType>()->getNumBits();
 
     // Like unsigned/int, shouldn't have a type if they don't match.
     if (LHSUnsigned != RHSUnsigned)
@@ -10256,7 +10294,7 @@ unsigned ASTContext::getIntWidth(QualType T) const {
     T = ET->getDecl()->getIntegerType();
   if (T->isBooleanType())
     return 1;
-  if(const auto *EIT = T->getAs<ExtIntType>())
+  if (const auto *EIT = T->getAs<BitIntType>())
     return EIT->getNumBits();
   // For builtin types, just use the standard type sizing method
   return (unsigned)getTypeSize(T);
@@ -10271,9 +10309,9 @@ QualType ASTContext::getCorrespondingUnsignedType(QualType T) const {
     return getVectorType(getCorrespondingUnsignedType(VTy->getElementType()),
                          VTy->getNumElements(), VTy->getVectorKind());
 
-  // For _ExtInt, return an unsigned _ExtInt with same width.
-  if (const auto *EITy = T->getAs<ExtIntType>())
-    return getExtIntType(/*IsUnsigned=*/true, EITy->getNumBits());
+  // For _BitInt, return an unsigned _BitInt with same width.
+  if (const auto *EITy = T->getAs<BitIntType>())
+    return getBitIntType(/*IsUnsigned=*/true, EITy->getNumBits());
 
   // For enums, get the underlying integer type of the enum, and let the general
   // integer type signchanging code handle it.
@@ -10339,9 +10377,9 @@ QualType ASTContext::getCorrespondingSignedType(QualType T) const {
     return getVectorType(getCorrespondingSignedType(VTy->getElementType()),
                          VTy->getNumElements(), VTy->getVectorKind());
 
-  // For _ExtInt, return a signed _ExtInt with same width.
-  if (const auto *EITy = T->getAs<ExtIntType>())
-    return getExtIntType(/*IsUnsigned=*/false, EITy->getNumBits());
+  // For _BitInt, return a signed _BitInt with same width.
+  if (const auto *EITy = T->getAs<BitIntType>())
+    return getBitIntType(/*IsUnsigned=*/false, EITy->getNumBits());
 
   // For enums, get the underlying integer type of the enum, and let the general
   // integer type signchanging code handle it.
diff --git a/clang/lib/AST/ASTDiagnostic.cpp b/clang/lib/AST/ASTDiagnostic.cpp
index 7e435e8b35b8..724ede272fbf 100644
--- a/clang/lib/AST/ASTDiagnostic.cpp
+++ b/clang/lib/AST/ASTDiagnostic.cpp
@@ -26,7 +26,8 @@ using namespace clang;
 
 // Returns a desugared version of the QualType, and marks ShouldAKA as true
 // whenever we remove significant sugar from the type.
-static QualType Desugar(ASTContext &Context, QualType QT, bool &ShouldAKA) {
+QualType clang::desugarForDiagnostic(ASTContext &Context, QualType QT,
+                                     bool &ShouldAKA) {
   QualifierCollector QC;
 
   while (true) {
@@ -37,6 +38,11 @@ static QualType Desugar(ASTContext &Context, QualType QT, bool &ShouldAKA) {
       QT = ET->desugar();
       continue;
     }
+    // ... or a using type ...
+    if (const UsingType *UT = dyn_cast<UsingType>(Ty)) {
+      QT = UT->desugar();
+      continue;
+    }
     // ... or a paren type ...
     if (const ParenType *PT = dyn_cast<ParenType>(Ty)) {
       QT = PT->desugar();
@@ -76,7 +82,7 @@ static QualType Desugar(ASTContext &Context, QualType QT, bool &ShouldAKA) {
     if (const FunctionType *FT = dyn_cast<FunctionType>(Ty)) {
       bool DesugarReturn = false;
       QualType SugarRT = FT->getReturnType();
-      QualType RT = Desugar(Context, SugarRT, DesugarReturn);
+      QualType RT = desugarForDiagnostic(Context, SugarRT, DesugarReturn);
       if (auto nullability = AttributedType::stripOuterNullability(SugarRT)) {
         RT = Context.getAttributedType(
             AttributedType::getNullabilityAttrKind(*nullability), RT, RT);
@@ -87,7 +93,7 @@ static QualType Desugar(ASTContext &Context, QualType QT, bool &ShouldAKA) {
       const FunctionProtoType *FPT = dyn_cast<FunctionProtoType>(FT);
       if (FPT) {
         for (QualType SugarPT : FPT->param_types()) {
-          QualType PT = Desugar(Context, SugarPT, DesugarArgument);
+          QualType PT = desugarForDiagnostic(Context, SugarPT, DesugarArgument);
           if (auto nullability =
                   AttributedType::stripOuterNullability(SugarPT)) {
             PT = Context.getAttributedType(
@@ -115,7 +121,8 @@ static QualType Desugar(ASTContext &Context, QualType QT, bool &ShouldAKA) {
         for (unsigned I = 0, N = TST->getNumArgs(); I != N; ++I) {
           const TemplateArgument &Arg = TST->getArg(I);
           if (Arg.getKind() == TemplateArgument::Type)
-            Args.push_back(Desugar(Context, Arg.getAsType(), DesugarArgument));
+            Args.push_back(desugarForDiagnostic(Context, Arg.getAsType(),
+                                                DesugarArgument));
           else
             Args.push_back(Arg);
         }
@@ -129,6 +136,29 @@ static QualType Desugar(ASTContext &Context, QualType QT, bool &ShouldAKA) {
       }
     }
 
+    if (const auto *AT = dyn_cast<ArrayType>(Ty)) {
+      QualType ElementTy =
+          desugarForDiagnostic(Context, AT->getElementType(), ShouldAKA);
+      if (const auto *CAT = dyn_cast<ConstantArrayType>(AT))
+        QT = Context.getConstantArrayType(
+            ElementTy, CAT->getSize(), CAT->getSizeExpr(),
+            CAT->getSizeModifier(), CAT->getIndexTypeCVRQualifiers());
+      else if (const auto *VAT = dyn_cast<VariableArrayType>(AT))
+        QT = Context.getVariableArrayType(
+            ElementTy, VAT->getSizeExpr(), VAT->getSizeModifier(),
+            VAT->getIndexTypeCVRQualifiers(), VAT->getBracketsRange());
+      else if (const auto *DSAT = dyn_cast<DependentSizedArrayType>(AT))
+        QT = Context.getDependentSizedArrayType(
+            ElementTy, DSAT->getSizeExpr(), DSAT->getSizeModifier(),
+            DSAT->getIndexTypeCVRQualifiers(), DSAT->getBracketsRange());
+      else if (const auto *IAT = dyn_cast<IncompleteArrayType>(AT))
+        QT = Context.getIncompleteArrayType(ElementTy, IAT->getSizeModifier(),
+                                            IAT->getIndexTypeCVRQualifiers());
+      else
+        llvm_unreachable("Unhandled array type");
+      break;
+    }
+
     // Don't desugar magic Objective-C types.
     if (QualType(Ty,0) == Context.getObjCIdType() ||
         QualType(Ty,0) == Context.getObjCClassType() ||
@@ -181,24 +211,25 @@ break; \
   // If we have a pointer-like type, desugar the pointee as well.
   // FIXME: Handle other pointer-like types.
   if (const PointerType *Ty = QT->getAs<PointerType>()) {
-    QT = Context.getPointerType(Desugar(Context, Ty->getPointeeType(),
-                                        ShouldAKA));
+    QT = Context.getPointerType(
+        desugarForDiagnostic(Context, Ty->getPointeeType(), ShouldAKA));
   } else if (const auto *Ty = QT->getAs<ObjCObjectPointerType>()) {
-    QT = Context.getObjCObjectPointerType(Desugar(Context, Ty->getPointeeType(),
-                                                  ShouldAKA));
+    QT = Context.getObjCObjectPointerType(
+        desugarForDiagnostic(Context, Ty->getPointeeType(), ShouldAKA));
   } else if (const LValueReferenceType *Ty = QT->getAs<LValueReferenceType>()) {
-    QT = Context.getLValueReferenceType(Desugar(Context, Ty->getPointeeType(),
-                                                ShouldAKA));
+    QT = Context.getLValueReferenceType(
+        desugarForDiagnostic(Context, Ty->getPointeeType(), ShouldAKA));
   } else if (const RValueReferenceType *Ty = QT->getAs<RValueReferenceType>()) {
-    QT = Context.getRValueReferenceType(Desugar(Context, Ty->getPointeeType(),
-                                                ShouldAKA));
+    QT = Context.getRValueReferenceType(
+        desugarForDiagnostic(Context, Ty->getPointeeType(), ShouldAKA));
   } else if (const auto *Ty = QT->getAs<ObjCObjectType>()) {
     if (Ty->getBaseType().getTypePtr() != Ty && !ShouldAKA) {
-      QualType BaseType = Desugar(Context, Ty->getBaseType(), ShouldAKA);
-      QT = Context.getObjCObjectType(BaseType, Ty->getTypeArgsAsWritten(),
-                                     llvm::makeArrayRef(Ty->qual_begin(),
-                                                        Ty->getNumProtocols()),
-                                     Ty->isKindOfTypeAsWritten());
+      QualType BaseType =
+          desugarForDiagnostic(Context, Ty->getBaseType(), ShouldAKA);
+      QT = Context.getObjCObjectType(
+          BaseType, Ty->getTypeArgsAsWritten(),
+          llvm::makeArrayRef(Ty->qual_begin(), Ty->getNumProtocols()),
+          Ty->isKindOfTypeAsWritten());
     }
   }
 
@@ -251,7 +282,8 @@ ConvertTypeToDiagnosticString(ASTContext &Context, QualType Ty,
       continue;  // Same canonical types
     std::string CompareS = CompareTy.getAsString(Context.getPrintingPolicy());
     bool ShouldAKA = false;
-    QualType CompareDesugar = Desugar(Context, CompareTy, ShouldAKA);
+    QualType CompareDesugar =
+        desugarForDiagnostic(Context, CompareTy, ShouldAKA);
     std::string CompareDesugarStr =
         CompareDesugar.getAsString(Context.getPrintingPolicy());
     if (CompareS != S && CompareDesugarStr != S)
@@ -286,7 +318,7 @@ ConvertTypeToDiagnosticString(ASTContext &Context, QualType Ty,
   // sugar gives us something "significantly different".
   if (!Repeated) {
     bool ShouldAKA = false;
-    QualType DesugaredTy = Desugar(Context, Ty, ShouldAKA);
+    QualType DesugaredTy = desugarForDiagnostic(Context, Ty, ShouldAKA);
     if (ShouldAKA || ForceAKA) {
       if (DesugaredTy == Ty) {
         DesugaredTy = Ty.getCanonicalType();
@@ -308,7 +340,7 @@ ConvertTypeToDiagnosticString(ASTContext &Context, QualType Ty,
       OS << "'" << S << "' (vector of " << VTy->getNumElements() << " '"
          << VTy->getElementType().getAsString(Context.getPrintingPolicy())
          << "' " << Values << ")";
-      return OS.str();
+      return DecoratedString;
     }
   }
 
diff --git a/clang/lib/AST/ASTImporter.cpp b/clang/lib/AST/ASTImporter.cpp
index 710e40bbb4b7..7f78da10e0b3 100644
--- a/clang/lib/AST/ASTImporter.cpp
+++ b/clang/lib/AST/ASTImporter.cpp
@@ -354,6 +354,7 @@ namespace clang {
     ExpectedType VisitTypeOfExprType(const TypeOfExprType *T);
     // FIXME: DependentTypeOfExprType
     ExpectedType VisitTypeOfType(const TypeOfType *T);
+    ExpectedType VisitUsingType(const UsingType *T);
     ExpectedType VisitDecltypeType(const DecltypeType *T);
     ExpectedType VisitUnaryTransformType(const UnaryTransformType *T);
     ExpectedType VisitAutoType(const AutoType *T);
@@ -1340,6 +1341,17 @@ ExpectedType ASTNodeImporter::VisitTypeOfType(const TypeOfType *T) {
   return Importer.getToContext().getTypeOfType(*ToUnderlyingTypeOrErr);
 }
 
+ExpectedType ASTNodeImporter::VisitUsingType(const UsingType *T) {
+  Expected<UsingShadowDecl *> FoundOrErr = import(T->getFoundDecl());
+  if (!FoundOrErr)
+    return FoundOrErr.takeError();
+  Expected<QualType> UnderlyingOrErr = import(T->getUnderlyingType());
+  if (!UnderlyingOrErr)
+    return UnderlyingOrErr.takeError();
+
+  return Importer.getToContext().getUsingType(*FoundOrErr, *UnderlyingOrErr);
+}
+
 ExpectedType ASTNodeImporter::VisitDecltypeType(const DecltypeType *T) {
   // FIXME: Make sure that the "to" context supports C++0x!
   ExpectedExpr ToExprOrErr = import(T->getUnderlyingExpr());
@@ -6066,20 +6078,24 @@ ASTNodeImporter::VisitFunctionTemplateDecl(FunctionTemplateDecl *D) {
   if (Error Err = importInto(TemplatedFD, D->getTemplatedDecl()))
     return std::move(Err);
 
-  // Template parameters of the ClassTemplateDecl and FunctionTemplateDecl are
-  // shared, if the FunctionTemplateDecl is a deduction guide for the class.
-  // At import the ClassTemplateDecl object is always created first (FIXME: is
-  // this really true?) because the dependency, then the FunctionTemplateDecl.
-  // The DeclContext of the template parameters is changed when the
-  // FunctionTemplateDecl is created, but was set already when the class
-  // template was created. So here it is not the TU (default value) any more.
-  // FIXME: The DeclContext of the parameters is now set finally to the
-  // CXXDeductionGuideDecl object that was imported later. This may not be the
-  // same that is in the original AST, specially if there are multiple deduction
-  // guides.
-  DeclContext *OldParamDC = nullptr;
-  if (Params->size() > 0)
-    OldParamDC = Params->getParam(0)->getDeclContext();
+  // At creation of the template the template parameters are "adopted"
+  // (DeclContext is changed). After this possible change the lookup table
+  // must be updated.
+  // At deduction guides the DeclContext of the template parameters may be
+  // different from what we would expect, it may be the class template, or a
+  // probably different CXXDeductionGuideDecl. This may come from the fact that
+  // the template parameter objects may be shared between deduction guides or
+  // the class template, and at creation of multiple FunctionTemplateDecl
+  // objects (for deduction guides) the same parameters are re-used. The
+  // "adoption" happens multiple times with different parent, even recursively
+  // for TemplateTemplateParmDecl. The same happens at import when the
+  // FunctionTemplateDecl objects are created, but in different order.
+  // In this way the DeclContext of these template parameters is not necessarily
+  // the same as in the "from" context.
+  SmallVector<DeclContext *, 2> OldParamDC;
+  OldParamDC.reserve(Params->size());
+  llvm::transform(*Params, std::back_inserter(OldParamDC),
+                  [](NamedDecl *ND) { return ND->getDeclContext(); });
 
   FunctionTemplateDecl *ToFunc;
   if (GetImportedOrCreateDecl(ToFunc, D, Importer.getToContext(), DC, Loc, Name,
@@ -6091,7 +6107,12 @@ ASTNodeImporter::VisitFunctionTemplateDecl(FunctionTemplateDecl *D) {
   ToFunc->setAccess(D->getAccess());
   ToFunc->setLexicalDeclContext(LexicalDC);
   LexicalDC->addDeclInternal(ToFunc);
-  updateLookupTableForTemplateParameters(*Params, OldParamDC);
+
+  ASTImporterLookupTable *LT = Importer.SharedState->getLookupTable();
+  if (LT && !OldParamDC.empty()) {
+    for (unsigned int I = 0; I < OldParamDC.size(); ++I)
+      LT->updateForced(Params->getParam(I), OldParamDC[I]);
+  }
 
   if (FoundByLookup) {
     auto *Recent =
diff --git a/clang/lib/AST/ASTImporterLookupTable.cpp b/clang/lib/AST/ASTImporterLookupTable.cpp
index ef42561c6f94..b7d17a5e92d0 100644
--- a/clang/lib/AST/ASTImporterLookupTable.cpp
+++ b/clang/lib/AST/ASTImporterLookupTable.cpp
@@ -140,6 +140,11 @@ void ASTImporterLookupTable::update(NamedDecl *ND, DeclContext *OldDC) {
   add(ND);
 }
 
+void ASTImporterLookupTable::updateForced(NamedDecl *ND, DeclContext *OldDC) {
+  LookupTable[OldDC][ND->getDeclName()].remove(ND);
+  add(ND);
+}
+
 ASTImporterLookupTable::LookupResult
 ASTImporterLookupTable::lookup(DeclContext *DC, DeclarationName Name) const {
   auto DCI = LookupTable.find(DC->getPrimaryContext());
diff --git a/clang/lib/AST/ASTStructuralEquivalence.cpp b/clang/lib/AST/ASTStructuralEquivalence.cpp
index 7fd24e2aa9ad..0813a5204a5e 100644
--- a/clang/lib/AST/ASTStructuralEquivalence.cpp
+++ b/clang/lib/AST/ASTStructuralEquivalence.cpp
@@ -945,6 +945,12 @@ static bool IsStructurallyEquivalent(StructuralEquivalenceContext &Context,
       return false;
     break;
 
+  case Type::Using:
+    if (!IsStructurallyEquivalent(Context, cast<UsingType>(T1)->getFoundDecl(),
+                                  cast<UsingType>(T2)->getFoundDecl()))
+      return false;
+    break;
+
   case Type::Typedef:
     if (!IsStructurallyEquivalent(Context, cast<TypedefType>(T1)->getDecl(),
                                   cast<TypedefType>(T2)->getDecl()))
@@ -1205,18 +1211,18 @@ static bool IsStructurallyEquivalent(StructuralEquivalenceContext &Context,
                                   cast<PipeType>(T2)->getElementType()))
       return false;
     break;
-  case Type::ExtInt: {
-    const auto *Int1 = cast<ExtIntType>(T1);
-    const auto *Int2 = cast<ExtIntType>(T2);
+  case Type::BitInt: {
+    const auto *Int1 = cast<BitIntType>(T1);
+    const auto *Int2 = cast<BitIntType>(T2);
 
     if (Int1->isUnsigned() != Int2->isUnsigned() ||
         Int1->getNumBits() != Int2->getNumBits())
       return false;
     break;
   }
-  case Type::DependentExtInt: {
-    const auto *Int1 = cast<DependentExtIntType>(T1);
-    const auto *Int2 = cast<DependentExtIntType>(T2);
+  case Type::DependentBitInt: {
+    const auto *Int1 = cast<DependentBitIntType>(T1);
+    const auto *Int2 = cast<DependentBitIntType>(T2);
 
     if (Int1->isUnsigned() != Int2->isUnsigned() ||
         !IsStructurallyEquivalent(Context, Int1->getNumBitsExpr(),
diff --git a/clang/lib/AST/AttrImpl.cpp b/clang/lib/AST/AttrImpl.cpp
index a3b46752c511..c2f13cf63830 100644
--- a/clang/lib/AST/AttrImpl.cpp
+++ b/clang/lib/AST/AttrImpl.cpp
@@ -60,7 +60,7 @@ std::string LoopHintAttr::getValueString(const PrintingPolicy &Policy) const {
   else
     OS << "disable";
   OS << ")";
-  return OS.str();
+  return ValueName;
 }
 
 // Return a string suitable for identifying this attribute in diagnostics.
diff --git a/clang/lib/AST/Comment.cpp b/clang/lib/AST/Comment.cpp
index fae3640d5ff7..43820fc566e4 100644
--- a/clang/lib/AST/Comment.cpp
+++ b/clang/lib/AST/Comment.cpp
@@ -108,12 +108,7 @@ Comment::child_iterator Comment::child_end() const {
 }
 
 bool TextComment::isWhitespaceNoCache() const {
-  for (StringRef::const_iterator I = Text.begin(), E = Text.end();
-       I != E; ++I) {
-    if (!clang::isWhitespace(*I))
-      return false;
-  }
-  return true;
+  return llvm::all_of(Text, clang::isWhitespace);
 }
 
 bool ParagraphComment::isWhitespaceNoCache() const {
diff --git a/clang/lib/AST/CommentBriefParser.cpp b/clang/lib/AST/CommentBriefParser.cpp
index 2a5f7452b776..bf9e17993497 100644
--- a/clang/lib/AST/CommentBriefParser.cpp
+++ b/clang/lib/AST/CommentBriefParser.cpp
@@ -8,15 +8,12 @@
 
 #include "clang/AST/CommentBriefParser.h"
 #include "clang/AST/CommentCommandTraits.h"
+#include "clang/Basic/CharInfo.h"
 
 namespace clang {
 namespace comments {
 
 namespace {
-inline bool isWhitespace(char C) {
-  return C == ' ' || C == '\n' || C == '\r' ||
-         C == '\t' || C == '\f' || C == '\v';
-}
 
 /// Convert all whitespace into spaces, remove leading and trailing spaces,
 /// compress multiple spaces into one.
@@ -26,12 +23,11 @@ void cleanupBrief(std::string &S) {
   for (std::string::iterator I = S.begin(), E = S.end();
        I != E; ++I) {
     const char C = *I;
-    if (isWhitespace(C)) {
+    if (clang::isWhitespace(C)) {
       if (!PrevWasSpace) {
         *O++ = ' ';
         PrevWasSpace = true;
       }
-      continue;
     } else {
       *O++ = C;
       PrevWasSpace = false;
@@ -44,12 +40,7 @@ void cleanupBrief(std::string &S) {
 }
 
 bool isWhitespace(StringRef Text) {
-  for (StringRef::const_iterator I = Text.begin(), E = Text.end();
-       I != E; ++I) {
-    if (!isWhitespace(*I))
-      return false;
-  }
-  return true;
+  return llvm::all_of(Text, clang::isWhitespace);
 }
 } // unnamed namespace
 
diff --git a/clang/lib/AST/Decl.cpp b/clang/lib/AST/Decl.cpp
index 68dfef248f65..e63560f1b6fe 100644
--- a/clang/lib/AST/Decl.cpp
+++ b/clang/lib/AST/Decl.cpp
@@ -604,8 +604,14 @@ static LinkageInfo getExternalLinkageFor(const NamedDecl *D) {
   //   - A name declared at namespace scope that does not have internal linkage
   //     by the previous rules and that is introduced by a non-exported
   //     declaration has module linkage.
-  if (isInModulePurview(D) && !isExportedFromModuleInterfaceUnit(
-                                  cast<NamedDecl>(D->getCanonicalDecl())))
+  //
+  // [basic.namespace.general]/p2
+  //   A namespace is never attached to a named module and never has a name with
+  //   module linkage.
+  if (isInModulePurview(D) &&
+      !isExportedFromModuleInterfaceUnit(
+          cast<NamedDecl>(D->getCanonicalDecl())) &&
+      !isa<NamespaceDecl>(D))
     return LinkageInfo(ModuleLinkage, DefaultVisibility, false);
 
   return LinkageInfo::external();
@@ -1583,7 +1589,7 @@ std::string NamedDecl::getQualifiedNameAsString() const {
   std::string QualName;
   llvm::raw_string_ostream OS(QualName);
   printQualifiedName(OS, getASTContext().getPrintingPolicy());
-  return OS.str();
+  return QualName;
 }
 
 void NamedDecl::printQualifiedName(raw_ostream &OS) const {
diff --git a/clang/lib/AST/DeclarationName.cpp b/clang/lib/AST/DeclarationName.cpp
index 56cf4b457a48..b2232ddfced3 100644
--- a/clang/lib/AST/DeclarationName.cpp
+++ b/clang/lib/AST/DeclarationName.cpp
@@ -236,7 +236,7 @@ std::string DeclarationName::getAsString() const {
   std::string Result;
   llvm::raw_string_ostream OS(Result);
   OS << *this;
-  return OS.str();
+  return Result;
 }
 
 void *DeclarationName::getFETokenInfoSlow() const {
@@ -460,7 +460,7 @@ std::string DeclarationNameInfo::getAsString() const {
   std::string Result;
   llvm::raw_string_ostream OS(Result);
   OS << *this;
-  return OS.str();
+  return Result;
 }
 
 raw_ostream &clang::operator<<(raw_ostream &OS, DeclarationNameInfo DNInfo) {
diff --git a/clang/lib/AST/Expr.cpp b/clang/lib/AST/Expr.cpp
index d3cb2ff3734c..2530beb89d17 100644
--- a/clang/lib/AST/Expr.cpp
+++ b/clang/lib/AST/Expr.cpp
@@ -202,6 +202,23 @@ bool Expr::isKnownToHaveBooleanValue(bool Semantic) const {
   return false;
 }
 
+const ValueDecl *
+Expr::getAsBuiltinConstantDeclRef(const ASTContext &Context) const {
+  Expr::EvalResult Eval;
+
+  if (EvaluateAsConstantExpr(Eval, Context)) {
+    APValue &Value = Eval.Val;
+
+    if (Value.isMemberPointer())
+      return Value.getMemberPointerDecl();
+
+    if (Value.isLValue() && Value.getLValueOffset().isZero())
+      return Value.getLValueBase().dyn_cast<const ValueDecl *>();
+  }
+
+  return nullptr;
+}
+
 // Amusing macro metaprogramming hack: check whether a class provides
 // a more specific implementation of getExprLoc().
 //
diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp
index 99babd58b027..469339e8cd62 100644
--- a/clang/lib/AST/ExprConstant.cpp
+++ b/clang/lib/AST/ExprConstant.cpp
@@ -1954,11 +1954,12 @@ static bool EvaluateIgnoredValue(EvalInfo &Info, const Expr *E) {
   return true;
 }
 
-/// Should this call expression be treated as a string literal?
-static bool IsStringLiteralCall(const CallExpr *E) {
+/// Should this call expression be treated as a constant?
+static bool IsConstantCall(const CallExpr *E) {
   unsigned Builtin = E->getBuiltinCallee();
   return (Builtin == Builtin::BI__builtin___CFStringMakeConstantString ||
-          Builtin == Builtin::BI__builtin___NSStringMakeConstantString);
+          Builtin == Builtin::BI__builtin___NSStringMakeConstantString ||
+          Builtin == Builtin::BI__builtin_function_start);
 }
 
 static bool IsGlobalLValue(APValue::LValueBase B) {
@@ -2004,7 +2005,7 @@ static bool IsGlobalLValue(APValue::LValueBase B) {
   case Expr::ObjCBoxedExprClass:
     return cast<ObjCBoxedExpr>(E)->isExpressibleAsConstantInitializer();
   case Expr::CallExprClass:
-    return IsStringLiteralCall(cast<CallExpr>(E));
+    return IsConstantCall(cast<CallExpr>(E));
   // For GCC compatibility, &&label has static storage duration.
   case Expr::AddrLabelExprClass:
     return true;
@@ -2931,6 +2932,11 @@ handleCompareOpForVectorHelper(const APTy &LHSValue, BinaryOperatorKind Opcode,
     break;
   }
 
+  // The boolean operations on these vector types use an instruction that
+  // results in a mask of '-1' for the 'truth' value.  Ensure that we negate 1
+  // to -1 to make sure that we produce the correct value.
+  Result.negate();
+
   return true;
 }
 
@@ -8962,7 +8968,7 @@ bool PointerExprEvaluator::visitNonBuiltinCallExpr(const CallExpr *E) {
 }
 
 bool PointerExprEvaluator::VisitCallExpr(const CallExpr *E) {
-  if (IsStringLiteralCall(E))
+  if (IsConstantCall(E))
     return Success(E);
 
   if (unsigned BuiltinOp = E->getBuiltinCallee())
@@ -10179,7 +10185,8 @@ namespace {
     bool VisitInitListExpr(const InitListExpr *E);
     bool VisitUnaryImag(const UnaryOperator *E);
     bool VisitBinaryOperator(const BinaryOperator *E);
-    // FIXME: Missing: unary -, unary ~, conditional operator (for GNU
+    bool VisitUnaryOperator(const UnaryOperator *E);
+    // FIXME: Missing: conditional operator (for GNU
     //                 conditional select), shufflevector, ExtVectorElementExpr
   };
 } // end anonymous namespace
@@ -10364,6 +10371,83 @@ bool VectorExprEvaluator::VisitBinaryOperator(const BinaryOperator *E) {
   return Success(LHSValue, E);
 }
 
+static llvm::Optional<APValue> handleVectorUnaryOperator(ASTContext &Ctx,
+                                                         QualType ResultTy,
+                                                         UnaryOperatorKind Op,
+                                                         APValue Elt) {
+  switch (Op) {
+  case UO_Plus:
+    // Nothing to do here.
+    return Elt;
+  case UO_Minus:
+    if (Elt.getKind() == APValue::Int) {
+      Elt.getInt().negate();
+    } else {
+      assert(Elt.getKind() == APValue::Float &&
+             "Vector can only be int or float type");
+      Elt.getFloat().changeSign();
+    }
+    return Elt;
+  case UO_Not:
+    // This is only valid for integral types anyway, so we don't have to handle
+    // float here.
+    assert(Elt.getKind() == APValue::Int &&
+           "Vector operator ~ can only be int");
+    Elt.getInt().flipAllBits();
+    return Elt;
+  case UO_LNot: {
+    if (Elt.getKind() == APValue::Int) {
+      Elt.getInt() = !Elt.getInt();
+      // operator ! on vectors returns -1 for 'truth', so negate it.
+      Elt.getInt().negate();
+      return Elt;
+    }
+    assert(Elt.getKind() == APValue::Float &&
+           "Vector can only be int or float type");
+    // Float types result in an int of the same size, but -1 for true, or 0 for
+    // false.
+    APSInt EltResult{Ctx.getIntWidth(ResultTy),
+                     ResultTy->isUnsignedIntegerType()};
+    if (Elt.getFloat().isZero())
+      EltResult.setAllBits();
+    else
+      EltResult.clearAllBits();
+
+    return APValue{EltResult};
+  }
+  default:
+    // FIXME: Implement the rest of the unary operators.
+    return llvm::None;
+  }
+}
+
+bool VectorExprEvaluator::VisitUnaryOperator(const UnaryOperator *E) {
+  Expr *SubExpr = E->getSubExpr();
+  const auto *VD = SubExpr->getType()->castAs<VectorType>();
+  // This result element type differs in the case of negating a floating point
+  // vector, since the result type is the a vector of the equivilant sized
+  // integer.
+  const QualType ResultEltTy = VD->getElementType();
+  UnaryOperatorKind Op = E->getOpcode();
+
+  APValue SubExprValue;
+  if (!Evaluate(SubExprValue, Info, SubExpr))
+    return false;
+
+  assert(SubExprValue.getVectorLength() == VD->getNumElements() &&
+         "Vector length doesn't match type?");
+
+  SmallVector<APValue, 4> ResultElements;
+  for (unsigned EltNum = 0; EltNum < VD->getNumElements(); ++EltNum) {
+    llvm::Optional<APValue> Elt = handleVectorUnaryOperator(
+        Info.Ctx, ResultEltTy, Op, SubExprValue.getVectorElt(EltNum));
+    if (!Elt)
+      return false;
+    ResultElements.push_back(*Elt);
+  }
+  return Success(APValue(ResultElements.data(), ResultElements.size()), E);
+}
+
 //===----------------------------------------------------------------------===//
 // Array Evaluation
 //===----------------------------------------------------------------------===//
@@ -11077,7 +11161,7 @@ EvaluateBuiltinClassifyType(QualType T, const LangOptions &LangOpts) {
   case Type::ObjCInterface:
   case Type::ObjCObjectPointer:
   case Type::Pipe:
-  case Type::ExtInt:
+  case Type::BitInt:
     // GCC classifies vectors as None. We follow its lead and classify all
     // other types that don't fit into the regular classification the same way.
     return GCCTypeClass::None;
diff --git a/clang/lib/AST/ItaniumMangle.cpp b/clang/lib/AST/ItaniumMangle.cpp
index 07579d04e275..7afc1250a36f 100644
--- a/clang/lib/AST/ItaniumMangle.cpp
+++ b/clang/lib/AST/ItaniumMangle.cpp
@@ -2263,8 +2263,8 @@ bool CXXNameMangler::mangleUnresolvedTypeOrSimpleId(QualType Ty,
   case Type::Atomic:
   case Type::Pipe:
   case Type::MacroQualified:
-  case Type::ExtInt:
-  case Type::DependentExtInt:
+  case Type::BitInt:
+  case Type::DependentBitInt:
     llvm_unreachable("type is illegal as a nested name specifier");
 
   case Type::SubstTemplateTypeParmPack:
@@ -2380,6 +2380,9 @@ bool CXXNameMangler::mangleUnresolvedTypeOrSimpleId(QualType Ty,
     break;
   }
 
+  case Type::Using:
+    return mangleUnresolvedTypeOrSimpleId(cast<UsingType>(Ty)->desugar(),
+                                          Prefix);
   case Type::Elaborated:
     return mangleUnresolvedTypeOrSimpleId(
         cast<ElaboratedType>(Ty)->getNamedType(), Prefix);
@@ -3967,26 +3970,20 @@ void CXXNameMangler::mangleType(const PipeType *T) {
   Out << "8ocl_pipe";
 }
 
-void CXXNameMangler::mangleType(const ExtIntType *T) {
-  Out << "U7_ExtInt";
-  llvm::APSInt BW(32, true);
-  BW = T->getNumBits();
-  TemplateArgument TA(Context.getASTContext(), BW, getASTContext().IntTy);
-  mangleTemplateArgs(TemplateName(), &TA, 1);
-  if (T->isUnsigned())
-    Out << "j";
-  else
-    Out << "i";
+void CXXNameMangler::mangleType(const BitIntType *T) {
+  // 5.1.5.2 Builtin types
+  // <type> ::= DB <number | instantiation-dependent expression> _
+  //        ::= DU <number | instantiation-dependent expression> _
+  Out << "D" << (T->isUnsigned() ? "U" : "B") << T->getNumBits() << "_";
 }
 
-void CXXNameMangler::mangleType(const DependentExtIntType *T) {
-  Out << "U7_ExtInt";
-  TemplateArgument TA(T->getNumBitsExpr());
-  mangleTemplateArgs(TemplateName(), &TA, 1);
-  if (T->isUnsigned())
-    Out << "j";
-  else
-    Out << "i";
+void CXXNameMangler::mangleType(const DependentBitIntType *T) {
+  // 5.1.5.2 Builtin types
+  // <type> ::= DB <number | instantiation-dependent expression> _
+  //        ::= DU <number | instantiation-dependent expression> _
+  Out << "D" << (T->isUnsigned() ? "U" : "B");
+  mangleExpression(T->getNumBitsExpr());
+  Out << "_";
 }
 
 void CXXNameMangler::mangleIntegerLiteral(QualType T,
diff --git a/clang/lib/AST/JSONNodeDumper.cpp b/clang/lib/AST/JSONNodeDumper.cpp
index 86879b8c3533..ae585def4d07 100644
--- a/clang/lib/AST/JSONNodeDumper.cpp
+++ b/clang/lib/AST/JSONNodeDumper.cpp
@@ -744,11 +744,18 @@ void JSONNodeDumper::VisitNamedDecl(const NamedDecl *ND) {
     JOS.attribute("name", ND->getNameAsString());
     // FIXME: There are likely other contexts in which it makes no sense to ask
     // for a mangled name.
-    if (!isa<RequiresExprBodyDecl>(ND->getDeclContext())) {
-      std::string MangledName = ASTNameGen.getName(ND);
-      if (!MangledName.empty())
-        JOS.attribute("mangledName", MangledName);
-    }
+    if (isa<RequiresExprBodyDecl>(ND->getDeclContext()))
+      return;
+
+    // Mangled names are not meaningful for locals, and may not be well-defined
+    // in the case of VLAs.
+    auto *VD = dyn_cast<VarDecl>(ND);
+    if (VD && VD->hasLocalStorage())
+      return;
+
+    std::string MangledName = ASTNameGen.getName(ND);
+    if (!MangledName.empty())
+      JOS.attribute("mangledName", MangledName);
   }
 }
 
diff --git a/clang/lib/AST/MicrosoftMangle.cpp b/clang/lib/AST/MicrosoftMangle.cpp
index 79a448a2435c..8802b6e500a6 100644
--- a/clang/lib/AST/MicrosoftMangle.cpp
+++ b/clang/lib/AST/MicrosoftMangle.cpp
@@ -3365,26 +3365,26 @@ void MicrosoftMangleContextImpl::mangleCXXName(GlobalDecl GD,
   return Mangler.mangle(GD);
 }
 
-void MicrosoftCXXNameMangler::mangleType(const ExtIntType *T, Qualifiers,
+void MicrosoftCXXNameMangler::mangleType(const BitIntType *T, Qualifiers,
                                          SourceRange Range) {
   llvm::SmallString<64> TemplateMangling;
   llvm::raw_svector_ostream Stream(TemplateMangling);
   MicrosoftCXXNameMangler Extra(Context, Stream);
   Stream << "?$";
   if (T->isUnsigned())
-    Extra.mangleSourceName("_UExtInt");
+    Extra.mangleSourceName("_UBitInt");
   else
-    Extra.mangleSourceName("_ExtInt");
+    Extra.mangleSourceName("_BitInt");
   Extra.mangleIntegerLiteral(llvm::APSInt::getUnsigned(T->getNumBits()));
 
   mangleArtificialTagType(TTK_Struct, TemplateMangling, {"__clang"});
 }
 
-void MicrosoftCXXNameMangler::mangleType(const DependentExtIntType *T,
+void MicrosoftCXXNameMangler::mangleType(const DependentBitIntType *T,
                                          Qualifiers, SourceRange Range) {
   DiagnosticsEngine &Diags = Context.getDiags();
   unsigned DiagID = Diags.getCustomDiagID(
-      DiagnosticsEngine::Error, "cannot mangle this DependentExtInt type yet");
+      DiagnosticsEngine::Error, "cannot mangle this DependentBitInt type yet");
   Diags.Report(Range.getBegin(), DiagID) << Range;
 }
 
diff --git a/clang/lib/AST/OpenMPClause.cpp b/clang/lib/AST/OpenMPClause.cpp
index f721e56f7fdd..1bd049b88005 100644
--- a/clang/lib/AST/OpenMPClause.cpp
+++ b/clang/lib/AST/OpenMPClause.cpp
@@ -126,6 +126,7 @@ const OMPClauseWithPreInit *OMPClauseWithPreInit::get(const OMPClause *C) {
   case OMPC_write:
   case OMPC_update:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -217,6 +218,7 @@ const OMPClauseWithPostUpdate *OMPClauseWithPostUpdate::get(const OMPClause *C)
   case OMPC_write:
   case OMPC_update:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -1792,6 +1794,10 @@ void OMPClausePrinter::VisitOMPCaptureClause(OMPCaptureClause *) {
   OS << "capture";
 }
 
+void OMPClausePrinter::VisitOMPCompareClause(OMPCompareClause *) {
+  OS << "compare";
+}
+
 void OMPClausePrinter::VisitOMPSeqCstClause(OMPSeqCstClause *) {
   OS << "seq_cst";
 }
@@ -2454,7 +2460,7 @@ std::string OMPTraitInfo::getMangledName() const {
                                                 Property.RawString);
     }
   }
-  return OS.str();
+  return MangledName;
 }
 
 OMPTraitInfo::OMPTraitInfo(StringRef MangledName) {
diff --git a/clang/lib/AST/ParentMap.cpp b/clang/lib/AST/ParentMap.cpp
index 2ff5c9d8aeb5..da21e573c320 100644
--- a/clang/lib/AST/ParentMap.cpp
+++ b/clang/lib/AST/ParentMap.cpp
@@ -133,8 +133,7 @@ void ParentMap::setParent(const Stmt *S, const Stmt *Parent) {
 
 Stmt* ParentMap::getParent(Stmt* S) const {
   MapTy* M = (MapTy*) Impl;
-  MapTy::iterator I = M->find(S);
-  return I == M->end() ? nullptr : I->second;
+  return M->lookup(S);
 }
 
 Stmt *ParentMap::getParentIgnoreParens(Stmt *S) const {
diff --git a/clang/lib/AST/QualTypeNames.cpp b/clang/lib/AST/QualTypeNames.cpp
index 673821078345..561757b1ba64 100644
--- a/clang/lib/AST/QualTypeNames.cpp
+++ b/clang/lib/AST/QualTypeNames.cpp
@@ -418,6 +418,13 @@ QualType getFullyQualifiedType(QualType QT, const ASTContext &Ctx,
     return QT;
   }
 
+  // We don't consider the alias introduced by `using a::X` as a new type.
+  // The qualified name is still a::X.
+  if (isa<UsingType>(QT.getTypePtr())) {
+    return getFullyQualifiedType(QT.getSingleStepDesugaredType(Ctx), Ctx,
+                                 WithGlobalNsPrefix);
+  }
+
   // Remove the part of the type related to the type being a template
   // parameter (we won't report it as part of the 'type name' and it
   // is actually make the code below to be more complex (to handle
diff --git a/clang/lib/AST/StmtProfile.cpp b/clang/lib/AST/StmtProfile.cpp
index 4339c249e027..09853e0f0e49 100644
--- a/clang/lib/AST/StmtProfile.cpp
+++ b/clang/lib/AST/StmtProfile.cpp
@@ -551,6 +551,8 @@ void OMPClauseProfiler::VisitOMPUpdateClause(const OMPUpdateClause *) {}
 
 void OMPClauseProfiler::VisitOMPCaptureClause(const OMPCaptureClause *) {}
 
+void OMPClauseProfiler::VisitOMPCompareClause(const OMPCompareClause *) {}
+
 void OMPClauseProfiler::VisitOMPSeqCstClause(const OMPSeqCstClause *) {}
 
 void OMPClauseProfiler::VisitOMPAcqRelClause(const OMPAcqRelClause *) {}
diff --git a/clang/lib/AST/TextNodeDumper.cpp b/clang/lib/AST/TextNodeDumper.cpp
index b21e806e307c..67c934847c7f 100644
--- a/clang/lib/AST/TextNodeDumper.cpp
+++ b/clang/lib/AST/TextNodeDumper.cpp
@@ -1534,6 +1534,10 @@ void TextNodeDumper::VisitUnresolvedUsingType(const UnresolvedUsingType *T) {
   dumpDeclRef(T->getDecl());
 }
 
+void TextNodeDumper::VisitUsingType(const UsingType *T) {
+  dumpDeclRef(T->getFoundDecl());
+}
+
 void TextNodeDumper::VisitTypedefType(const TypedefType *T) {
   dumpDeclRef(T->getDecl());
 }
diff --git a/clang/lib/AST/Type.cpp b/clang/lib/AST/Type.cpp
index e0ac3f5b1351..c771fe264b0c 100644
--- a/clang/lib/AST/Type.cpp
+++ b/clang/lib/AST/Type.cpp
@@ -338,25 +338,25 @@ VectorType::VectorType(TypeClass tc, QualType vecType, unsigned nElements,
   VectorTypeBits.NumElements = nElements;
 }
 
-ExtIntType::ExtIntType(bool IsUnsigned, unsigned NumBits)
-    : Type(ExtInt, QualType{}, TypeDependence::None), IsUnsigned(IsUnsigned),
+BitIntType::BitIntType(bool IsUnsigned, unsigned NumBits)
+    : Type(BitInt, QualType{}, TypeDependence::None), IsUnsigned(IsUnsigned),
       NumBits(NumBits) {}
 
-DependentExtIntType::DependentExtIntType(const ASTContext &Context,
+DependentBitIntType::DependentBitIntType(const ASTContext &Context,
                                          bool IsUnsigned, Expr *NumBitsExpr)
-    : Type(DependentExtInt, QualType{},
+    : Type(DependentBitInt, QualType{},
            toTypeDependence(NumBitsExpr->getDependence())),
       Context(Context), ExprAndUnsigned(NumBitsExpr, IsUnsigned) {}
 
-bool DependentExtIntType::isUnsigned() const {
+bool DependentBitIntType::isUnsigned() const {
   return ExprAndUnsigned.getInt();
 }
 
-clang::Expr *DependentExtIntType::getNumBitsExpr() const {
+clang::Expr *DependentBitIntType::getNumBitsExpr() const {
   return ExprAndUnsigned.getPointer();
 }
 
-void DependentExtIntType::Profile(llvm::FoldingSetNodeID &ID,
+void DependentBitIntType::Profile(llvm::FoldingSetNodeID &ID,
                                   const ASTContext &Context, bool IsUnsigned,
                                   Expr *NumBitsExpr) {
   ID.AddBoolean(IsUnsigned);
@@ -1932,7 +1932,7 @@ bool Type::isIntegralType(const ASTContext &Ctx) const {
     if (const auto *ET = dyn_cast<EnumType>(CanonicalType))
       return ET->getDecl()->isComplete();
 
-  return isExtIntType();
+  return isBitIntType();
 }
 
 bool Type::isIntegralOrUnscopedEnumerationType() const {
@@ -1940,7 +1940,7 @@ bool Type::isIntegralOrUnscopedEnumerationType() const {
     return BT->getKind() >= BuiltinType::Bool &&
            BT->getKind() <= BuiltinType::Int128;
 
-  if (isExtIntType())
+  if (isBitIntType())
     return true;
 
   return isUnscopedEnumerationType();
@@ -2023,7 +2023,9 @@ bool Type::isSignedIntegerType() const {
       return ET->getDecl()->getIntegerType()->isSignedIntegerType();
   }
 
-  if (const ExtIntType *IT = dyn_cast<ExtIntType>(CanonicalType))
+  if (const auto *IT = dyn_cast<BitIntType>(CanonicalType))
+    return IT->isSigned();
+  if (const auto *IT = dyn_cast<DependentBitIntType>(CanonicalType))
     return IT->isSigned();
 
   return false;
@@ -2040,9 +2042,10 @@ bool Type::isSignedIntegerOrEnumerationType() const {
       return ET->getDecl()->getIntegerType()->isSignedIntegerType();
   }
 
-  if (const ExtIntType *IT = dyn_cast<ExtIntType>(CanonicalType))
+  if (const auto *IT = dyn_cast<BitIntType>(CanonicalType))
+    return IT->isSigned();
+  if (const auto *IT = dyn_cast<DependentBitIntType>(CanonicalType))
     return IT->isSigned();
-
 
   return false;
 }
@@ -2070,7 +2073,9 @@ bool Type::isUnsignedIntegerType() const {
       return ET->getDecl()->getIntegerType()->isUnsignedIntegerType();
   }
 
-  if (const ExtIntType *IT = dyn_cast<ExtIntType>(CanonicalType))
+  if (const auto *IT = dyn_cast<BitIntType>(CanonicalType))
+    return IT->isUnsigned();
+  if (const auto *IT = dyn_cast<DependentBitIntType>(CanonicalType))
     return IT->isUnsigned();
 
   return false;
@@ -2087,7 +2092,9 @@ bool Type::isUnsignedIntegerOrEnumerationType() const {
       return ET->getDecl()->getIntegerType()->isUnsignedIntegerType();
   }
 
-  if (const ExtIntType *IT = dyn_cast<ExtIntType>(CanonicalType))
+  if (const auto *IT = dyn_cast<BitIntType>(CanonicalType))
+    return IT->isUnsigned();
+  if (const auto *IT = dyn_cast<DependentBitIntType>(CanonicalType))
     return IT->isUnsigned();
 
   return false;
@@ -2129,7 +2136,7 @@ bool Type::isRealType() const {
            BT->getKind() <= BuiltinType::Ibm128;
   if (const auto *ET = dyn_cast<EnumType>(CanonicalType))
       return ET->getDecl()->isComplete() && !ET->getDecl()->isScoped();
-  return isExtIntType();
+  return isBitIntType();
 }
 
 bool Type::isArithmeticType() const {
@@ -2145,7 +2152,7 @@ bool Type::isArithmeticType() const {
     // false for scoped enumerations since that will disable any
     // unwanted implicit conversions.
     return !ET->getDecl()->isScoped() && ET->getDecl()->isComplete();
-  return isa<ComplexType>(CanonicalType) || isExtIntType();
+  return isa<ComplexType>(CanonicalType) || isBitIntType();
 }
 
 Type::ScalarTypeKind Type::getScalarTypeKind() const {
@@ -2174,7 +2181,7 @@ Type::ScalarTypeKind Type::getScalarTypeKind() const {
     if (CT->getElementType()->isRealFloatingType())
       return STK_FloatingComplex;
     return STK_IntegralComplex;
-  } else if (isExtIntType()) {
+  } else if (isBitIntType()) {
     return STK_Integral;
   }
 
@@ -2381,7 +2388,7 @@ bool QualType::isCXX98PODType(const ASTContext &Context) const {
   case Type::MemberPointer:
   case Type::Vector:
   case Type::ExtVector:
-  case Type::ExtInt:
+  case Type::BitInt:
     return true;
 
   case Type::Enum:
@@ -3400,6 +3407,17 @@ QualType TypedefType::desugar() const {
   return getDecl()->getUnderlyingType();
 }
 
+UsingType::UsingType(const UsingShadowDecl *Found, QualType Underlying,
+                     QualType Canon)
+    : Type(Using, Canon, Underlying->getDependence()),
+      Found(const_cast<UsingShadowDecl *>(Found)) {
+  assert(Underlying == getUnderlyingType());
+}
+
+QualType UsingType::getUnderlyingType() const {
+  return QualType(cast<TypeDecl>(Found->getTargetDecl())->getTypeForDecl(), 0);
+}
+
 QualType MacroQualifiedType::desugar() const { return getUnderlyingType(); }
 
 QualType MacroQualifiedType::getModifiedType() const {
@@ -3849,7 +3867,7 @@ static CachedProperties computeCachedProperties(const Type *T) {
     // here in error recovery.
     return CachedProperties(ExternalLinkage, false);
 
-  case Type::ExtInt:
+  case Type::BitInt:
   case Type::Builtin:
     // C++ [basic.link]p8:
     //   A type is said to have linkage if and only if:
@@ -3949,7 +3967,7 @@ LinkageInfo LinkageComputer::computeTypeLinkageInfo(const Type *T) {
     assert(T->isInstantiationDependentType());
     return LinkageInfo::external();
 
-  case Type::ExtInt:
+  case Type::BitInt:
   case Type::Builtin:
     return LinkageInfo::external();
 
@@ -4169,8 +4187,8 @@ bool Type::canHaveNullability(bool ResultIfUnknown) const {
   case Type::ObjCInterface:
   case Type::Atomic:
   case Type::Pipe:
-  case Type::ExtInt:
-  case Type::DependentExtInt:
+  case Type::BitInt:
+  case Type::DependentBitInt:
     return false;
   }
   llvm_unreachable("bad type kind!");
diff --git a/clang/lib/AST/TypePrinter.cpp b/clang/lib/AST/TypePrinter.cpp
index eca9af3e5f36..2a33a69f288d 100644
--- a/clang/lib/AST/TypePrinter.cpp
+++ b/clang/lib/AST/TypePrinter.cpp
@@ -212,6 +212,7 @@ bool TypePrinter::canPrefixQualifiers(const Type *T,
     case Type::Builtin:
     case Type::Complex:
     case Type::UnresolvedUsing:
+    case Type::Using:
     case Type::Typedef:
     case Type::TypeOfExpr:
     case Type::TypeOf:
@@ -232,8 +233,8 @@ bool TypePrinter::canPrefixQualifiers(const Type *T,
     case Type::ObjCInterface:
     case Type::Atomic:
     case Type::Pipe:
-    case Type::ExtInt:
-    case Type::DependentExtInt:
+    case Type::BitInt:
+    case Type::DependentBitInt:
       CanPrefixQualifiers = true;
       break;
 
@@ -1046,6 +1047,21 @@ void TypePrinter::printUnresolvedUsingBefore(const UnresolvedUsingType *T,
 void TypePrinter::printUnresolvedUsingAfter(const UnresolvedUsingType *T,
                                             raw_ostream &OS) {}
 
+void TypePrinter::printUsingBefore(const UsingType *T, raw_ostream &OS) {
+  // After `namespace b { using a::X }`, is the type X within B a::X or b::X?
+  //
+  // - b::X is more formally correct given the UsingType model
+  // - b::X makes sense if "re-exporting" a symbol in a new namespace
+  // - a::X makes sense if "importing" a symbol for convenience
+  //
+  // The "importing" use seems much more common, so we print a::X.
+  // This could be a policy option, but the right choice seems to rest more
+  // with the intent of the code than the caller.
+  printTypeSpec(T->getFoundDecl()->getUnderlyingDecl(), OS);
+}
+
+void TypePrinter::printUsingAfter(const UsingType *T, raw_ostream &OS) {}
+
 void TypePrinter::printTypedefBefore(const TypedefType *T, raw_ostream &OS) {
   printTypeSpec(T->getDecl(), OS);
 }
@@ -1200,26 +1216,26 @@ void TypePrinter::printPipeBefore(const PipeType *T, raw_ostream &OS) {
 
 void TypePrinter::printPipeAfter(const PipeType *T, raw_ostream &OS) {}
 
-void TypePrinter::printExtIntBefore(const ExtIntType *T, raw_ostream &OS) {
+void TypePrinter::printBitIntBefore(const BitIntType *T, raw_ostream &OS) {
   if (T->isUnsigned())
     OS << "unsigned ";
-  OS << "_ExtInt(" << T->getNumBits() << ")";
+  OS << "_BitInt(" << T->getNumBits() << ")";
   spaceBeforePlaceHolder(OS);
 }
 
-void TypePrinter::printExtIntAfter(const ExtIntType *T, raw_ostream &OS) {}
+void TypePrinter::printBitIntAfter(const BitIntType *T, raw_ostream &OS) {}
 
-void TypePrinter::printDependentExtIntBefore(const DependentExtIntType *T,
+void TypePrinter::printDependentBitIntBefore(const DependentBitIntType *T,
                                              raw_ostream &OS) {
   if (T->isUnsigned())
     OS << "unsigned ";
-  OS << "_ExtInt(";
+  OS << "_BitInt(";
   T->getNumBitsExpr()->printPretty(OS, nullptr, Policy);
   OS << ")";
   spaceBeforePlaceHolder(OS);
 }
 
-void TypePrinter::printDependentExtIntAfter(const DependentExtIntType *T,
+void TypePrinter::printDependentBitIntAfter(const DependentBitIntType *T,
                                             raw_ostream &OS) {}
 
 /// Appends the given scope to the end of a string.
diff --git a/clang/lib/ASTMatchers/ASTMatchersInternal.cpp b/clang/lib/ASTMatchers/ASTMatchersInternal.cpp
index 7680eb38283e..815058f32de4 100644
--- a/clang/lib/ASTMatchers/ASTMatchersInternal.cpp
+++ b/clang/lib/ASTMatchers/ASTMatchersInternal.cpp
@@ -1059,6 +1059,7 @@ const AstTypeMatcher<UnaryTransformType> unaryTransformType;
 const AstTypeMatcher<RecordType> recordType;
 const AstTypeMatcher<TagType> tagType;
 const AstTypeMatcher<ElaboratedType> elaboratedType;
+const AstTypeMatcher<UsingType> usingType;
 const AstTypeMatcher<SubstTemplateTypeParmType> substTemplateTypeParmType;
 const AstTypeMatcher<TemplateTypeParmType> templateTypeParmType;
 const AstTypeMatcher<InjectedClassNameType> injectedClassNameType;
diff --git a/clang/lib/ASTMatchers/Dynamic/Diagnostics.cpp b/clang/lib/ASTMatchers/Dynamic/Diagnostics.cpp
index ba2f49e6b623..41ab0ed70fda 100644
--- a/clang/lib/ASTMatchers/Dynamic/Diagnostics.cpp
+++ b/clang/lib/ASTMatchers/Dynamic/Diagnostics.cpp
@@ -204,7 +204,7 @@ std::string Diagnostics::toString() const {
   std::string S;
   llvm::raw_string_ostream OS(S);
   printToStream(OS);
-  return OS.str();
+  return S;
 }
 
 void Diagnostics::printToStreamFull(llvm::raw_ostream &OS) const {
@@ -223,7 +223,7 @@ std::string Diagnostics::toStringFull() const {
   std::string S;
   llvm::raw_string_ostream OS(S);
   printToStreamFull(OS);
-  return OS.str();
+  return S;
 }
 
 }  // namespace dynamic
diff --git a/clang/lib/ASTMatchers/Dynamic/Marshallers.h b/clang/lib/ASTMatchers/Dynamic/Marshallers.h
index 783fb203c408..fa9d42247e24 100644
--- a/clang/lib/ASTMatchers/Dynamic/Marshallers.h
+++ b/clang/lib/ASTMatchers/Dynamic/Marshallers.h
@@ -1035,7 +1035,6 @@ public:
   void getArgKinds(ASTNodeKind ThisKind, unsigned,
                    std::vector<ArgKind> &ArgKinds) const override {
     ArgKinds.push_back(ArgKind::MakeNodeArg(ThisKind));
-    return;
   }
   bool isConvertibleTo(ASTNodeKind Kind, unsigned *Specificity = nullptr,
                        ASTNodeKind *LeastDerivedKind = nullptr) const override {
diff --git a/clang/lib/ASTMatchers/Dynamic/Parser.cpp b/clang/lib/ASTMatchers/Dynamic/Parser.cpp
index c6a77bb6c2e0..cab1476acf94 100644
--- a/clang/lib/ASTMatchers/Dynamic/Parser.cpp
+++ b/clang/lib/ASTMatchers/Dynamic/Parser.cpp
@@ -645,7 +645,7 @@ bool Parser::parseMatcherExpressionImpl(const TokenInfo &NameToken,
   Tokenizer->SkipNewlines();
 
   {
-    ScopedContextEntry SCE(this, Ctor ? *Ctor : nullptr);
+    ScopedContextEntry SCE(this, Ctor.getValueOr(nullptr));
 
     while (Tokenizer->nextTokenKind() != TokenInfo::TK_Eof) {
       if (Tokenizer->nextTokenKind() == TokenInfo::TK_CloseParen) {
diff --git a/clang/lib/ASTMatchers/Dynamic/Registry.cpp b/clang/lib/ASTMatchers/Dynamic/Registry.cpp
index 878547923d27..4f3efdb0a663 100644
--- a/clang/lib/ASTMatchers/Dynamic/Registry.cpp
+++ b/clang/lib/ASTMatchers/Dynamic/Registry.cpp
@@ -228,6 +228,7 @@ RegistryMaps::RegistryMaps() {
   REGISTER_MATCHER(eachOf);
   REGISTER_MATCHER(elaboratedType);
   REGISTER_MATCHER(elaboratedTypeLoc);
+  REGISTER_MATCHER(usingType);
   REGISTER_MATCHER(enumConstantDecl);
   REGISTER_MATCHER(enumDecl);
   REGISTER_MATCHER(enumType);
diff --git a/clang/lib/Analysis/AnalysisDeclContext.cpp b/clang/lib/Analysis/AnalysisDeclContext.cpp
index d8466ac34a3d..06f1f813aeed 100644
--- a/clang/lib/Analysis/AnalysisDeclContext.cpp
+++ b/clang/lib/Analysis/AnalysisDeclContext.cpp
@@ -387,7 +387,7 @@ std::string AnalysisDeclContext::getFunctionName(const Decl *D) {
     OS << ' ' << OMD->getSelector().getAsString() << ']';
   }
 
-  return OS.str();
+  return Str;
 }
 
 LocationContextManager &AnalysisDeclContext::getLocationContextManager() {
diff --git a/clang/lib/Analysis/CFG.cpp b/clang/lib/Analysis/CFG.cpp
index abf65e3efce9..9ef3b5b6277a 100644
--- a/clang/lib/Analysis/CFG.cpp
+++ b/clang/lib/Analysis/CFG.cpp
@@ -1820,8 +1820,6 @@ void CFGBuilder::addScopesEnd(LocalScope::const_iterator B,
 
   for (VarDecl *VD : llvm::reverse(DeclsWithEndedScope))
     appendScopeEnd(Block, VD, S);
-
-  return;
 }
 
 /// addAutomaticObjDtors - Add to current block automatic objects destructors
diff --git a/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp b/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp
index bb7eb9971068..413e8d14bf0a 100644
--- a/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp
+++ b/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp
@@ -11,15 +11,82 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include <utility>
 #include <vector>
 
+#include "clang/Analysis/Analyses/PostOrderCFGView.h"
 #include "clang/Analysis/CFG.h"
 #include "clang/Analysis/FlowSensitive/DataflowEnvironment.h"
+#include "clang/Analysis/FlowSensitive/DataflowWorklist.h"
 #include "clang/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.h"
+#include "llvm/ADT/None.h"
 #include "llvm/ADT/Optional.h"
+#include "llvm/Support/raw_ostream.h"
 
-using namespace clang;
-using namespace dataflow;
+namespace clang {
+namespace dataflow {
+
+/// Computes the input state for a given basic block by joining the output
+/// states of its predecessors.
+///
+/// Requirements:
+///
+///   All predecessors of `Block` except those with loop back edges must have
+///   already been transferred. States in `BlockStates` that are set to
+///   `llvm::None` represent basic blocks that are not evaluated yet.
+static TypeErasedDataflowAnalysisState computeBlockInputState(
+    std::vector<llvm::Optional<TypeErasedDataflowAnalysisState>> &BlockStates,
+    const CFGBlock &Block, const Environment &InitEnv,
+    TypeErasedDataflowAnalysis &Analysis) {
+  // FIXME: Consider passing `Block` to `Analysis.typeErasedInitialElement()`
+  // to enable building analyses like computation of dominators that initialize
+  // the state of each basic block differently.
+  TypeErasedDataflowAnalysisState State = {Analysis.typeErasedInitialElement(),
+                                           InitEnv};
+  for (const CFGBlock *Pred : Block.preds()) {
+    // Skip if the `Block` is unreachable or control flow cannot get past it.
+    if (!Pred || Pred->hasNoReturnElement())
+      continue;
+
+    // Skip if `Pred` was not evaluated yet. This could happen if `Pred` has a
+    // loop back edge to `Block`.
+    const llvm::Optional<TypeErasedDataflowAnalysisState> &MaybePredState =
+        BlockStates[Pred->getBlockID()];
+    if (!MaybePredState.hasValue())
+      continue;
+
+    const TypeErasedDataflowAnalysisState &PredState =
+        MaybePredState.getValue();
+    Analysis.joinTypeErased(State.Lattice, PredState.Lattice);
+    State.Env.join(PredState.Env);
+  }
+  return State;
+}
+
+TypeErasedDataflowAnalysisState transferBlock(
+    std::vector<llvm::Optional<TypeErasedDataflowAnalysisState>> &BlockStates,
+    const CFGBlock &Block, const Environment &InitEnv,
+    TypeErasedDataflowAnalysis &Analysis,
+    std::function<void(const CFGStmt &,
+                       const TypeErasedDataflowAnalysisState &)>
+        HandleTransferredStmt) {
+  TypeErasedDataflowAnalysisState State =
+      computeBlockInputState(BlockStates, Block, InitEnv, Analysis);
+  for (const CFGElement &Element : Block) {
+    // FIXME: Evaluate other kinds of `CFGElement`.
+    const llvm::Optional<CFGStmt> Stmt = Element.getAs<CFGStmt>();
+    if (!Stmt.hasValue())
+      continue;
+
+    // FIXME: Evaluate the statement contained in `Stmt`.
+
+    State.Lattice = Analysis.transferTypeErased(Stmt.getValue().getStmt(),
+                                                State.Lattice, State.Env);
+    if (HandleTransferredStmt != nullptr)
+      HandleTransferredStmt(Stmt.getValue(), State);
+  }
+  return State;
+}
 
 std::vector<llvm::Optional<TypeErasedDataflowAnalysisState>>
 runTypeErasedDataflowAnalysis(const CFG &Cfg,
@@ -29,7 +96,59 @@ runTypeErasedDataflowAnalysis(const CFG &Cfg,
   // are specified in the header. This could be done by remembering
   // what options were used to build `Cfg` and asserting on them here.
 
-  // FIXME: Implement work list-based algorithm to compute the fixed
-  // point of `Analysis::transform` for every basic block in `Cfg`.
-  return {};
+  PostOrderCFGView POV(&Cfg);
+  ForwardDataflowWorklist Worklist(Cfg, &POV);
+
+  std::vector<llvm::Optional<TypeErasedDataflowAnalysisState>> BlockStates;
+  BlockStates.resize(Cfg.size(), llvm::None);
+
+  // The entry basic block doesn't contain statements so it can be skipped.
+  const CFGBlock &Entry = Cfg.getEntry();
+  BlockStates[Entry.getBlockID()] = {Analysis.typeErasedInitialElement(),
+                                     InitEnv};
+  Worklist.enqueueSuccessors(&Entry);
+
+  // Bugs in lattices and transfer functions can prevent the analysis from
+  // converging. To limit the damage (infinite loops) that these bugs can cause,
+  // limit the number of iterations.
+  // FIXME: Consider making the maximum number of iterations configurable.
+  // FIXME: Set up statistics (see llvm/ADT/Statistic.h) to count average number
+  // of iterations, number of functions that time out, etc.
+  unsigned Iterations = 0;
+  static constexpr unsigned MaxIterations = 1 << 16;
+  while (const CFGBlock *Block = Worklist.dequeue()) {
+    if (++Iterations > MaxIterations) {
+      llvm::errs() << "Maximum number of iterations reached, giving up.\n";
+      break;
+    }
+
+    const llvm::Optional<TypeErasedDataflowAnalysisState> &OldBlockState =
+        BlockStates[Block->getBlockID()];
+    TypeErasedDataflowAnalysisState NewBlockState =
+        transferBlock(BlockStates, *Block, InitEnv, Analysis);
+
+    if (OldBlockState.hasValue() &&
+        Analysis.isEqualTypeErased(OldBlockState.getValue().Lattice,
+                                   NewBlockState.Lattice) &&
+        OldBlockState->Env == NewBlockState.Env) {
+      // The state of `Block` didn't change after transfer so there's no need to
+      // revisit its successors.
+      continue;
+    }
+
+    BlockStates[Block->getBlockID()] = std::move(NewBlockState);
+
+    // Do not add unreachable successor blocks to `Worklist`.
+    if (Block->hasNoReturnElement())
+      continue;
+
+    Worklist.enqueueSuccessors(Block);
+  }
+  // FIXME: Consider evaluating unreachable basic blocks (those that have a
+  // state set to `llvm::None` at this point) to also analyze dead code.
+
+  return BlockStates;
 }
+
+} // namespace dataflow
+} // namespace clang
diff --git a/clang/lib/Analysis/ThreadSafety.cpp b/clang/lib/Analysis/ThreadSafety.cpp
index b196ffa73cbf..9cc990bd35a3 100644
--- a/clang/lib/Analysis/ThreadSafety.cpp
+++ b/clang/lib/Analysis/ThreadSafety.cpp
@@ -418,7 +418,6 @@ public:
 private:
   Context::Factory ContextFactory;
   std::vector<VarDefinition> VarDefinitions;
-  std::vector<unsigned> CtxIndices;
   std::vector<std::pair<const Stmt *, Context>> SavedContexts;
 
 public:
@@ -731,8 +730,6 @@ void LocalVariableMap::traverseCFG(CFG *CFGraph,
                                    std::vector<CFGBlockInfo> &BlockInfo) {
   PostOrderCFGView::CFGBlockSet VisitedBlocks(CFGraph);
 
-  CtxIndices.resize(CFGraph->getNumBlockIDs());
-
   for (const auto *CurrBlock : *SortedGraph) {
     unsigned CurrBlockID = CurrBlock->getBlockID();
     CFGBlockInfo *CurrBlockInfo = &BlockInfo[CurrBlockID];
diff --git a/clang/lib/Analysis/UninitializedValues.cpp b/clang/lib/Analysis/UninitializedValues.cpp
index 67cd39728c35..a38ae34f4b81 100644
--- a/clang/lib/Analysis/UninitializedValues.cpp
+++ b/clang/lib/Analysis/UninitializedValues.cpp
@@ -591,8 +591,8 @@ public:
 
         if (AtPredExit == MayUninitialized) {
           // If the predecessor's terminator is an "asm goto" that initializes
-          // the variable, then it won't be counted as "initialized" on the
-          // non-fallthrough paths.
+          // the variable, then don't count it as "initialized" on the indirect
+          // paths.
           CFGTerminator term = Pred->getTerminator();
           if (const auto *as = dyn_cast_or_null<GCCAsmStmt>(term.getStmt())) {
             const CFGBlock *fallthrough = *Pred->succ_begin();
@@ -810,13 +810,22 @@ void TransferFunctions::VisitGCCAsmStmt(GCCAsmStmt *as) {
   if (!as->isAsmGoto())
     return;
 
-  for (const Expr *o : as->outputs())
-    if (const VarDecl *VD = findVar(o).getDecl())
+  ASTContext &C = ac.getASTContext();
+  for (const Expr *O : as->outputs()) {
+    const Expr *Ex = stripCasts(C, O);
+
+    // Strip away any unary operators. Invalid l-values are reported by other
+    // semantic analysis passes.
+    while (const auto *UO = dyn_cast<UnaryOperator>(Ex))
+      Ex = stripCasts(C, UO->getSubExpr());
+
+    if (const VarDecl *VD = findVar(Ex).getDecl())
       if (vals[VD] != Initialized)
         // If the variable isn't initialized by the time we get here, then we
         // mark it as potentially uninitialized for those cases where it's used
         // on an indirect path, where it's not guaranteed to be defined.
         vals[VD] = MayUninitialized;
+  }
 }
 
 void TransferFunctions::VisitObjCMessageExpr(ObjCMessageExpr *ME) {
diff --git a/clang/lib/Basic/Cuda.cpp b/clang/lib/Basic/Cuda.cpp
index e82a3a705e70..2d75578b3de0 100644
--- a/clang/lib/Basic/Cuda.cpp
+++ b/clang/lib/Basic/Cuda.cpp
@@ -123,6 +123,7 @@ static const CudaArchToStringMap arch_names[] = {
     GFX(1033), // gfx1033
     GFX(1034), // gfx1034
     GFX(1035), // gfx1035
+    {CudaArch::Generic, "generic", ""},
     // clang-format on
 };
 #undef SM
diff --git a/clang/lib/Basic/OpenMPKinds.cpp b/clang/lib/Basic/OpenMPKinds.cpp
index 9e74e05bd863..1761c6d3d89b 100644
--- a/clang/lib/Basic/OpenMPKinds.cpp
+++ b/clang/lib/Basic/OpenMPKinds.cpp
@@ -163,6 +163,7 @@ unsigned clang::getOpenMPSimpleClauseType(OpenMPClauseKind Kind, StringRef Str,
   case OMPC_read:
   case OMPC_write:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -428,6 +429,7 @@ const char *clang::getOpenMPSimpleClauseTypeName(OpenMPClauseKind Kind,
   case OMPC_read:
   case OMPC_write:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
diff --git a/clang/lib/Basic/SourceLocation.cpp b/clang/lib/Basic/SourceLocation.cpp
index 6986fcd322f2..6e5e55fb09ce 100644
--- a/clang/lib/Basic/SourceLocation.cpp
+++ b/clang/lib/Basic/SourceLocation.cpp
@@ -90,7 +90,7 @@ SourceLocation::printToString(const SourceManager &SM) const {
   std::string S;
   llvm::raw_string_ostream OS(S);
   print(OS, SM);
-  return OS.str();
+  return S;
 }
 
 LLVM_DUMP_METHOD void SourceLocation::dump(const SourceManager &SM) const {
@@ -149,7 +149,7 @@ SourceRange::printToString(const SourceManager &SM) const {
   std::string S;
   llvm::raw_string_ostream OS(S);
   print(OS, SM);
-  return OS.str();
+  return S;
 }
 
 //===----------------------------------------------------------------------===//
diff --git a/clang/lib/Basic/Targets/AArch64.cpp b/clang/lib/Basic/Targets/AArch64.cpp
index 4d403ae1809d..4089a393b762 100644
--- a/clang/lib/Basic/Targets/AArch64.cpp
+++ b/clang/lib/Basic/Targets/AArch64.cpp
@@ -307,6 +307,9 @@ void AArch64TargetInfo::getTargetDefines(const LangOptions &Opts,
   if (FPU & SveMode)
     Builder.defineMacro("__ARM_FEATURE_SVE", "1");
 
+  if ((FPU & NeonMode) && (FPU & SveMode))
+    Builder.defineMacro("__ARM_NEON_SVE_BRIDGE", "1");
+
   if (HasSVE2)
     Builder.defineMacro("__ARM_FEATURE_SVE2", "1");
 
@@ -474,10 +477,12 @@ ArrayRef<Builtin::Info> AArch64TargetInfo::getTargetBuiltins() const {
 Optional<std::pair<unsigned, unsigned>>
 AArch64TargetInfo::getVScaleRange(const LangOptions &LangOpts) const {
   if (LangOpts.VScaleMin || LangOpts.VScaleMax)
-    return std::pair<unsigned, unsigned>(LangOpts.VScaleMin,
-                                         LangOpts.VScaleMax);
+    return std::pair<unsigned, unsigned>(
+        LangOpts.VScaleMin ? LangOpts.VScaleMin : 1, LangOpts.VScaleMax);
+
   if (hasFeature("sve"))
-    return std::pair<unsigned, unsigned>(0, 16);
+    return std::pair<unsigned, unsigned>(1, 16);
+
   return None;
 }
 
diff --git a/clang/lib/Basic/Targets/AArch64.h b/clang/lib/Basic/Targets/AArch64.h
index dea59a9b015d..74745df3be8d 100644
--- a/clang/lib/Basic/Targets/AArch64.h
+++ b/clang/lib/Basic/Targets/AArch64.h
@@ -150,7 +150,7 @@ public:
   const char *getBFloat16Mangling() const override { return "u6__bf16"; };
   bool hasInt128Type() const override;
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 
 class LLVM_LIBRARY_VISIBILITY AArch64leTargetInfo : public AArch64TargetInfo {
diff --git a/clang/lib/Basic/Targets/AMDGPU.h b/clang/lib/Basic/Targets/AMDGPU.h
index 8b9d7ce79c16..974922191488 100644
--- a/clang/lib/Basic/Targets/AMDGPU.h
+++ b/clang/lib/Basic/Targets/AMDGPU.h
@@ -426,7 +426,7 @@ public:
 
   void setAuxTarget(const TargetInfo *Aux) override;
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 
   // Record offload arch features since they are needed for defining the
   // pre-defined macros.
diff --git a/clang/lib/Basic/Targets/ARC.h b/clang/lib/Basic/Targets/ARC.h
index 3c0c5f6df2f4..5411cd2cd869 100644
--- a/clang/lib/Basic/Targets/ARC.h
+++ b/clang/lib/Basic/Targets/ARC.h
@@ -66,7 +66,7 @@ public:
     return false;
   }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 
   bool isCLZForZeroUndef() const override { return false; }
 };
diff --git a/clang/lib/Basic/Targets/ARM.cpp b/clang/lib/Basic/Targets/ARM.cpp
index f330780300f2..c619d6cde41d 100644
--- a/clang/lib/Basic/Targets/ARM.cpp
+++ b/clang/lib/Basic/Targets/ARM.cpp
@@ -465,6 +465,8 @@ bool ARMTargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
   HWDiv = 0;
   DotProd = 0;
   HasMatMul = 0;
+  HasPAC = 0;
+  HasBTI = 0;
   HasFloat16 = true;
   ARMCDECoprocMask = 0;
   HasBFloat16 = false;
@@ -547,6 +549,9 @@ bool ARMTargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
       HasBFloat16 = true;
     } else if (Feature == "-fpregs") {
       FPRegsDisabled = true;
+    } else if (Feature == "+pacbti") {
+      HasPAC = 1;
+      HasBTI = 1;
     }
   }
 
@@ -890,6 +895,12 @@ void ARMTargetInfo::getTargetDefines(const LangOptions &Opts,
   if (HasMatMul)
     Builder.defineMacro("__ARM_FEATURE_MATMUL_INT8", "1");
 
+  if (HasPAC)
+    Builder.defineMacro("__ARM_FEATURE_PAUTH", "1");
+
+  if (HasBTI)
+    Builder.defineMacro("__ARM_FEATURE_BTI", "1");
+
   if (HasBFloat16) {
     Builder.defineMacro("__ARM_FEATURE_BF16", "1");
     Builder.defineMacro("__ARM_FEATURE_BF16_VECTOR_ARITHMETIC", "1");
@@ -900,7 +911,7 @@ void ARMTargetInfo::getTargetDefines(const LangOptions &Opts,
     Builder.defineMacro("__ARM_FEATURE_BTI_DEFAULT", "1");
 
   if (Opts.hasSignReturnAddress()) {
-    unsigned Value = Opts.isSignReturnAddressWithAKey() ? 1 : 2;
+    unsigned Value = 1;
     if (Opts.isSignReturnAddressScopeAll())
       Value |= 1 << 2;
     Builder.defineMacro("__ARM_FEATURE_PAC_DEFAULT", Twine(Value));
diff --git a/clang/lib/Basic/Targets/ARM.h b/clang/lib/Basic/Targets/ARM.h
index 7d0011d134ea..40c658f3f40e 100644
--- a/clang/lib/Basic/Targets/ARM.h
+++ b/clang/lib/Basic/Targets/ARM.h
@@ -79,6 +79,8 @@ class LLVM_LIBRARY_VISIBILITY ARMTargetInfo : public TargetInfo {
   unsigned DotProd : 1;
   unsigned HasMatMul : 1;
   unsigned FPRegsDisabled : 1;
+  unsigned HasPAC : 1;
+  unsigned HasBTI : 1;
 
   enum {
     LDREX_B = (1 << 0), /// byte (8-bit)
@@ -191,8 +193,8 @@ public:
 
   bool hasSjLjLowering() const override;
 
-  bool hasExtIntType() const override { return true; }
-  
+  bool hasBitIntType() const override { return true; }
+
   const char *getBFloat16Mangling() const override { return "u6__bf16"; };
 };
 
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp b/clang/lib/Basic/Targets/Hexagon.cpp
index 9c37dee7e89a..161369242926 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -68,6 +68,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions &Opts,
   } else if (CPU == "hexagonv68") {
     Builder.defineMacro("__HEXAGON_V68__");
     Builder.defineMacro("__HEXAGON_ARCH__", "68");
+  } else if (CPU == "hexagonv69") {
+    Builder.defineMacro("__HEXAGON_V69__");
+    Builder.defineMacro("__HEXAGON_ARCH__", "69");
   }
 
   if (hasFeature("hvx-length64b")) {
@@ -128,6 +131,10 @@ bool HexagonTargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
     else if (F == "+audio")
       HasAudio = true;
   }
+  if (CPU.compare("hexagonv68") >= 0) {
+    HasLegalHalfType = true;
+    HasFloat16 = true;
+  }
   return true;
 }
 
@@ -214,7 +221,7 @@ static constexpr CPUSuffix Suffixes[] = {
     {{"hexagonv60"}, {"60"}}, {{"hexagonv62"},  {"62"}},
     {{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
     {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
-    {{"hexagonv68"}, {"68"}},
+    {{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
 };
 
 const char *HexagonTargetInfo::getHexagonCPUSuffix(StringRef Name) {
diff --git a/clang/lib/Basic/Targets/Hexagon.h b/clang/lib/Basic/Targets/Hexagon.h
index d6c7da5f1e40..94441998f355 100644
--- a/clang/lib/Basic/Targets/Hexagon.h
+++ b/clang/lib/Basic/Targets/Hexagon.h
@@ -139,7 +139,7 @@ public:
     return CPU.find('t') != std::string::npos;
   }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/Lanai.h b/clang/lib/Basic/Targets/Lanai.h
index 9af5427b81c4..56c6cced938a 100644
--- a/clang/lib/Basic/Targets/Lanai.h
+++ b/clang/lib/Basic/Targets/Lanai.h
@@ -87,7 +87,7 @@ public:
 
   const char *getClobbers() const override { return ""; }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/Mips.h b/clang/lib/Basic/Targets/Mips.h
index b475c03889a1..b54d36e1c95f 100644
--- a/clang/lib/Basic/Targets/Mips.h
+++ b/clang/lib/Basic/Targets/Mips.h
@@ -406,7 +406,7 @@ public:
   unsigned getUnwindWordWidth() const override;
 
   bool validateTarget(DiagnosticsEngine &Diags) const override;
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/NVPTX.cpp b/clang/lib/Basic/Targets/NVPTX.cpp
index 3561b22677bc..75e82d819900 100644
--- a/clang/lib/Basic/Targets/NVPTX.cpp
+++ b/clang/lib/Basic/Targets/NVPTX.cpp
@@ -215,6 +215,7 @@ void NVPTXTargetInfo::getTargetDefines(const LangOptions &Opts,
       case CudaArch::GFX1033:
       case CudaArch::GFX1034:
       case CudaArch::GFX1035:
+      case CudaArch::Generic:
       case CudaArch::LAST:
         break;
       case CudaArch::UNUSED:
diff --git a/clang/lib/Basic/Targets/NVPTX.h b/clang/lib/Basic/Targets/NVPTX.h
index ef751b8e1a8d..589f24f4bb03 100644
--- a/clang/lib/Basic/Targets/NVPTX.h
+++ b/clang/lib/Basic/Targets/NVPTX.h
@@ -121,7 +121,7 @@ public:
 
   void fillValidCPUList(SmallVectorImpl<StringRef> &Values) const override {
     for (int i = static_cast<int>(CudaArch::SM_20);
-         i < static_cast<int>(CudaArch::LAST); ++i)
+         i < static_cast<int>(CudaArch::Generic); ++i)
       Values.emplace_back(CudaArchToString(static_cast<CudaArch>(i)));
   }
 
@@ -175,7 +175,7 @@ public:
     return CCCR_Warning;
   }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/OSTargets.cpp b/clang/lib/Basic/Targets/OSTargets.cpp
index 53748bf067cd..f8f12daaa072 100644
--- a/clang/lib/Basic/Targets/OSTargets.cpp
+++ b/clang/lib/Basic/Targets/OSTargets.cpp
@@ -48,12 +48,12 @@ void getDarwinDefines(MacroBuilder &Builder, const LangOptions &Opts,
     Builder.defineMacro("_REENTRANT");
 
   // Get the platform type and version number from the triple.
-  unsigned Maj, Min, Rev;
+  VersionTuple OsVersion;
   if (Triple.isMacOSX()) {
-    Triple.getMacOSXVersion(Maj, Min, Rev);
+    Triple.getMacOSXVersion(OsVersion);
     PlatformName = "macos";
   } else {
-    Triple.getOSVersion(Maj, Min, Rev);
+    OsVersion = Triple.getOSVersion();
     PlatformName = llvm::Triple::getOSTypeName(Triple.getOS());
     if (PlatformName == "ios" && Triple.isMacCatalystEnvironment())
       PlatformName = "maccatalyst";
@@ -63,29 +63,29 @@ void getDarwinDefines(MacroBuilder &Builder, const LangOptions &Opts,
   // generating code for Win32 ABI. No need to emit
   // __ENVIRONMENT_XX_OS_VERSION_MIN_REQUIRED__.
   if (PlatformName == "win32") {
-    PlatformMinVersion = VersionTuple(Maj, Min, Rev);
+    PlatformMinVersion = OsVersion;
     return;
   }
 
   // Set the appropriate OS version define.
   if (Triple.isiOS()) {
-    assert(Maj < 100 && Min < 100 && Rev < 100 && "Invalid version!");
+    assert(OsVersion < VersionTuple(100) && "Invalid version!");
     char Str[7];
-    if (Maj < 10) {
-      Str[0] = '0' + Maj;
-      Str[1] = '0' + (Min / 10);
-      Str[2] = '0' + (Min % 10);
-      Str[3] = '0' + (Rev / 10);
-      Str[4] = '0' + (Rev % 10);
+    if (OsVersion.getMajor() < 10) {
+      Str[0] = '0' + OsVersion.getMajor();
+      Str[1] = '0' + (OsVersion.getMinor().getValueOr(0) / 10);
+      Str[2] = '0' + (OsVersion.getMinor().getValueOr(0) % 10);
+      Str[3] = '0' + (OsVersion.getSubminor().getValueOr(0) / 10);
+      Str[4] = '0' + (OsVersion.getSubminor().getValueOr(0) % 10);
       Str[5] = '\0';
     } else {
       // Handle versions >= 10.
-      Str[0] = '0' + (Maj / 10);
-      Str[1] = '0' + (Maj % 10);
-      Str[2] = '0' + (Min / 10);
-      Str[3] = '0' + (Min % 10);
-      Str[4] = '0' + (Rev / 10);
-      Str[5] = '0' + (Rev % 10);
+      Str[0] = '0' + (OsVersion.getMajor() / 10);
+      Str[1] = '0' + (OsVersion.getMajor() % 10);
+      Str[2] = '0' + (OsVersion.getMinor().getValueOr(0) / 10);
+      Str[3] = '0' + (OsVersion.getMinor().getValueOr(0) % 10);
+      Str[4] = '0' + (OsVersion.getSubminor().getValueOr(0) / 10);
+      Str[5] = '0' + (OsVersion.getSubminor().getValueOr(0) % 10);
       Str[6] = '\0';
     }
     if (Triple.isTvOS())
@@ -95,13 +95,13 @@ void getDarwinDefines(MacroBuilder &Builder, const LangOptions &Opts,
                           Str);
 
   } else if (Triple.isWatchOS()) {
-    assert(Maj < 10 && Min < 100 && Rev < 100 && "Invalid version!");
+    assert(OsVersion < VersionTuple(10) && "Invalid version!");
     char Str[6];
-    Str[0] = '0' + Maj;
-    Str[1] = '0' + (Min / 10);
-    Str[2] = '0' + (Min % 10);
-    Str[3] = '0' + (Rev / 10);
-    Str[4] = '0' + (Rev % 10);
+    Str[0] = '0' + OsVersion.getMajor();
+    Str[1] = '0' + (OsVersion.getMinor().getValueOr(0) / 10);
+    Str[2] = '0' + (OsVersion.getMinor().getValueOr(0) % 10);
+    Str[3] = '0' + (OsVersion.getSubminor().getValueOr(0) / 10);
+    Str[4] = '0' + (OsVersion.getSubminor().getValueOr(0) % 10);
     Str[5] = '\0';
     Builder.defineMacro("__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__", Str);
   } else if (Triple.isMacOSX()) {
@@ -109,22 +109,22 @@ void getDarwinDefines(MacroBuilder &Builder, const LangOptions &Opts,
     // define (because we only get a single digit for the minor and micro
     // revision numbers). So, we limit them to the maximum representable
     // version.
-    assert(Maj < 100 && Min < 100 && Rev < 100 && "Invalid version!");
+    assert(OsVersion < VersionTuple(100) && "Invalid version!");
     char Str[7];
-    if (Maj < 10 || (Maj == 10 && Min < 10)) {
-      Str[0] = '0' + (Maj / 10);
-      Str[1] = '0' + (Maj % 10);
-      Str[2] = '0' + std::min(Min, 9U);
-      Str[3] = '0' + std::min(Rev, 9U);
+    if (OsVersion < VersionTuple(10, 10)) {
+      Str[0] = '0' + (OsVersion.getMajor() / 10);
+      Str[1] = '0' + (OsVersion.getMajor() % 10);
+      Str[2] = '0' + std::min(OsVersion.getMinor().getValueOr(0), 9U);
+      Str[3] = '0' + std::min(OsVersion.getSubminor().getValueOr(0), 9U);
       Str[4] = '\0';
     } else {
       // Handle versions > 10.9.
-      Str[0] = '0' + (Maj / 10);
-      Str[1] = '0' + (Maj % 10);
-      Str[2] = '0' + (Min / 10);
-      Str[3] = '0' + (Min % 10);
-      Str[4] = '0' + (Rev / 10);
-      Str[5] = '0' + (Rev % 10);
+      Str[0] = '0' + (OsVersion.getMajor() / 10);
+      Str[1] = '0' + (OsVersion.getMajor() % 10);
+      Str[2] = '0' + (OsVersion.getMinor().getValueOr(0) / 10);
+      Str[3] = '0' + (OsVersion.getMinor().getValueOr(0) % 10);
+      Str[4] = '0' + (OsVersion.getSubminor().getValueOr(0) / 10);
+      Str[5] = '0' + (OsVersion.getSubminor().getValueOr(0) % 10);
       Str[6] = '\0';
     }
     Builder.defineMacro("__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__", Str);
@@ -134,7 +134,7 @@ void getDarwinDefines(MacroBuilder &Builder, const LangOptions &Opts,
   if (Triple.isOSDarwin())
     Builder.defineMacro("__MACH__");
 
-  PlatformMinVersion = VersionTuple(Maj, Min, Rev);
+  PlatformMinVersion = OsVersion;
 }
 
 static void addMinGWDefines(const llvm::Triple &Triple, const LangOptions &Opts,
@@ -203,6 +203,7 @@ static void addVisualCDefines(const LangOptions &Opts, MacroBuilder &Builder) {
   }
 
   Builder.defineMacro("_INTEGRAL_MAX_BITS", "64");
+  Builder.defineMacro("__STDC_NO_THREADS__");
 
   // Starting with VS 2022 17.1, MSVC predefines the below macro to inform
   // users of the execution character set defined at compile time.
diff --git a/clang/lib/Basic/Targets/OSTargets.h b/clang/lib/Basic/Targets/OSTargets.h
index 7fbe2cbc5653..3c1830d5f8e8 100644
--- a/clang/lib/Basic/Targets/OSTargets.h
+++ b/clang/lib/Basic/Targets/OSTargets.h
@@ -148,9 +148,7 @@ public:
       return 64;
     }
 
-    unsigned Major, Minor, Micro;
-    T.getOSVersion(Major, Minor, Micro);
-    if (llvm::VersionTuple(Major, Minor, Micro) < MinVersion)
+    if (T.getOSVersion() < MinVersion)
       return 64;
     return OSTargetInfo<Target>::getExnObjectAlignment();
   }
@@ -294,7 +292,7 @@ protected:
     Builder.defineMacro("__HAIKU__");
     Builder.defineMacro("__ELF__");
     DefineStd(Builder, "unix", Opts);
-    if (this->HasFloat128) 
+    if (this->HasFloat128)
       Builder.defineMacro("__FLOAT128__");
   }
 
@@ -376,10 +374,9 @@ protected:
     Builder.defineMacro("__ELF__");
     if (Triple.isAndroid()) {
       Builder.defineMacro("__ANDROID__", "1");
-      unsigned Maj, Min, Rev;
-      Triple.getEnvironmentVersion(Maj, Min, Rev);
       this->PlatformName = "android";
-      this->PlatformMinVersion = VersionTuple(Maj, Min, Rev);
+      this->PlatformMinVersion = Triple.getEnvironmentVersion();
+      const unsigned Maj = this->PlatformMinVersion.getMajor();
       if (Maj) {
         Builder.defineMacro("__ANDROID_MIN_SDK_VERSION__", Twine(Maj));
         // This historical but ambiguous name for the minSdkVersion macro. Keep
@@ -693,23 +690,32 @@ protected:
     if (Opts.EnableAIXExtendedAltivecABI)
       Builder.defineMacro("__EXTABI__");
 
-    unsigned Major, Minor, Micro;
-    Triple.getOSVersion(Major, Minor, Micro);
+    VersionTuple OsVersion = Triple.getOSVersion();
 
     // Define AIX OS-Version Macros.
     // Includes logic for legacy versions of AIX; no specific intent to support.
-    std::pair<int, int> OsVersion = {Major, Minor};
-    if (OsVersion >= std::make_pair(3, 2)) Builder.defineMacro("_AIX32");
-    if (OsVersion >= std::make_pair(4, 1)) Builder.defineMacro("_AIX41");
-    if (OsVersion >= std::make_pair(4, 3)) Builder.defineMacro("_AIX43");
-    if (OsVersion >= std::make_pair(5, 0)) Builder.defineMacro("_AIX50");
-    if (OsVersion >= std::make_pair(5, 1)) Builder.defineMacro("_AIX51");
-    if (OsVersion >= std::make_pair(5, 2)) Builder.defineMacro("_AIX52");
-    if (OsVersion >= std::make_pair(5, 3)) Builder.defineMacro("_AIX53");
-    if (OsVersion >= std::make_pair(6, 1)) Builder.defineMacro("_AIX61");
-    if (OsVersion >= std::make_pair(7, 1)) Builder.defineMacro("_AIX71");
-    if (OsVersion >= std::make_pair(7, 2)) Builder.defineMacro("_AIX72");
-    if (OsVersion >= std::make_pair(7, 3)) Builder.defineMacro("_AIX73");
+    if (OsVersion >= VersionTuple(3, 2))
+      Builder.defineMacro("_AIX32");
+    if (OsVersion >= VersionTuple(4, 1))
+      Builder.defineMacro("_AIX41");
+    if (OsVersion >= VersionTuple(4, 3))
+      Builder.defineMacro("_AIX43");
+    if (OsVersion >= VersionTuple(5, 0))
+      Builder.defineMacro("_AIX50");
+    if (OsVersion >= VersionTuple(5, 1))
+      Builder.defineMacro("_AIX51");
+    if (OsVersion >= VersionTuple(5, 2))
+      Builder.defineMacro("_AIX52");
+    if (OsVersion >= VersionTuple(5, 3))
+      Builder.defineMacro("_AIX53");
+    if (OsVersion >= VersionTuple(6, 1))
+      Builder.defineMacro("_AIX61");
+    if (OsVersion >= VersionTuple(7, 1))
+      Builder.defineMacro("_AIX71");
+    if (OsVersion >= VersionTuple(7, 2))
+      Builder.defineMacro("_AIX72");
+    if (OsVersion >= VersionTuple(7, 3))
+      Builder.defineMacro("_AIX73");
 
     // FIXME: Do not define _LONG_LONG when -fno-long-long is specified.
     Builder.defineMacro("_LONG_LONG");
diff --git a/clang/lib/Basic/Targets/PNaCl.h b/clang/lib/Basic/Targets/PNaCl.h
index d5bfc369583f..b5cf73d73e95 100644
--- a/clang/lib/Basic/Targets/PNaCl.h
+++ b/clang/lib/Basic/Targets/PNaCl.h
@@ -69,7 +69,7 @@ public:
 
   const char *getClobbers() const override { return ""; }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 
 // We attempt to use PNaCl (le32) frontend and Mips32EL backend.
diff --git a/clang/lib/Basic/Targets/PPC.cpp b/clang/lib/Basic/Targets/PPC.cpp
index c3c61ed443ca..7f7b44b658eb 100644
--- a/clang/lib/Basic/Targets/PPC.cpp
+++ b/clang/lib/Basic/Targets/PPC.cpp
@@ -623,14 +623,11 @@ void PPCTargetInfo::addP10SpecificFeatures(
   Features["pcrelative-memops"] = true;
   Features["prefix-instrs"] = true;
   Features["isa-v31-instructions"] = true;
-  return;
 }
 
 // Add features specific to the "Future" CPU.
 void PPCTargetInfo::addFutureSpecificFeatures(
-    llvm::StringMap<bool> &Features) const {
-  return;
-}
+    llvm::StringMap<bool> &Features) const {}
 
 bool PPCTargetInfo::hasFeature(StringRef Feature) const {
   return llvm::StringSwitch<bool>(Feature)
diff --git a/clang/lib/Basic/Targets/PPC.h b/clang/lib/Basic/Targets/PPC.h
index f19d3ebbc066..60701072ac4b 100644
--- a/clang/lib/Basic/Targets/PPC.h
+++ b/clang/lib/Basic/Targets/PPC.h
@@ -351,7 +351,7 @@ public:
   const char *getFloat128Mangling() const override { return "u9__ieee128"; }
   const char *getIbm128Mangling() const override { return "g"; }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 
   bool isSPRegName(StringRef RegName) const override {
     return RegName.equals("r1") || RegName.equals("x1");
diff --git a/clang/lib/Basic/Targets/RISCV.cpp b/clang/lib/Basic/Targets/RISCV.cpp
index 93562dde2f54..770d37a1c1be 100644
--- a/clang/lib/Basic/Targets/RISCV.cpp
+++ b/clang/lib/Basic/Targets/RISCV.cpp
@@ -246,6 +246,9 @@ bool RISCVTargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
     ISAInfo = std::move(*ParseResult);
   }
 
+  if (ABI.empty())
+    ABI = llvm::RISCV::computeDefaultABIFromArch(*ISAInfo).str();
+
   return true;
 }
 
diff --git a/clang/lib/Basic/Targets/RISCV.h b/clang/lib/Basic/Targets/RISCV.h
index f7ffe9febcd0..5331ed4a50ae 100644
--- a/clang/lib/Basic/Targets/RISCV.h
+++ b/clang/lib/Basic/Targets/RISCV.h
@@ -95,7 +95,7 @@ public:
   bool handleTargetFeatures(std::vector<std::string> &Features,
                             DiagnosticsEngine &Diags) override;
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 class LLVM_LIBRARY_VISIBILITY RISCV32TargetInfo : public RISCVTargetInfo {
 public:
diff --git a/clang/lib/Basic/Targets/SPIR.h b/clang/lib/Basic/Targets/SPIR.h
index 8cf18b6c20f1..a40d4b3ca27e 100644
--- a/clang/lib/Basic/Targets/SPIR.h
+++ b/clang/lib/Basic/Targets/SPIR.h
@@ -162,7 +162,7 @@ public:
     supportAllOpenCLOpts();
   }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 
   bool hasInt128Type() const override { return false; }
 };
diff --git a/clang/lib/Basic/Targets/Sparc.h b/clang/lib/Basic/Targets/Sparc.h
index 22a1621fcb9f..177a117520da 100644
--- a/clang/lib/Basic/Targets/Sparc.h
+++ b/clang/lib/Basic/Targets/Sparc.h
@@ -48,8 +48,6 @@ public:
 
   bool hasFeature(StringRef Feature) const override;
 
-  bool hasSjLjLowering() const override { return true; }
-
   ArrayRef<Builtin::Info> getTargetBuiltins() const override {
     // FIXME: Implement!
     return None;
@@ -178,8 +176,7 @@ public:
   void getTargetDefines(const LangOptions &Opts,
                         MacroBuilder &Builder) const override;
 
-  bool hasSjLjLowering() const override { return true; }
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 
 // SPARCV8el is the 32-bit little-endian mode selected by Triple::sparcel.
@@ -232,7 +229,7 @@ public:
     return getCPUGeneration(CPU) == CG_V9;
   }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/SystemZ.h b/clang/lib/Basic/Targets/SystemZ.h
index d3e3ed50dd47..92cefeea5d26 100644
--- a/clang/lib/Basic/Targets/SystemZ.h
+++ b/clang/lib/Basic/Targets/SystemZ.h
@@ -170,7 +170,7 @@ public:
 
   const char *getLongDoubleMangling() const override { return "g"; }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 
   int getEHDataRegisterNumber(unsigned RegNo) const override {
     return RegNo < 4 ? 6 + RegNo : -1;
diff --git a/clang/lib/Basic/Targets/WebAssembly.h b/clang/lib/Basic/Targets/WebAssembly.h
index 16534d3ef99b..075486990558 100644
--- a/clang/lib/Basic/Targets/WebAssembly.h
+++ b/clang/lib/Basic/Targets/WebAssembly.h
@@ -137,7 +137,7 @@ private:
     }
   }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 
   bool hasProtectedVisibility() const override { return false; }
 
diff --git a/clang/lib/Basic/Targets/X86.h b/clang/lib/Basic/Targets/X86.h
index b9b2ac79815b..c952b8c9a336 100644
--- a/clang/lib/Basic/Targets/X86.h
+++ b/clang/lib/Basic/Targets/X86.h
@@ -462,7 +462,7 @@ public:
 
   ArrayRef<Builtin::Info> getTargetBuiltins() const override;
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 
 class LLVM_LIBRARY_VISIBILITY NetBSDI386TargetInfo
@@ -472,10 +472,9 @@ public:
       : NetBSDTargetInfo<X86_32TargetInfo>(Triple, Opts) {}
 
   unsigned getFloatEvalMethod() const override {
-    unsigned Major, Minor, Micro;
-    getTriple().getOSVersion(Major, Minor, Micro);
+    VersionTuple OsVersion = getTriple().getOSVersion();
     // New NetBSD uses the default rounding mode.
-    if (Major >= 7 || (Major == 6 && Minor == 99 && Micro >= 26) || Major == 0)
+    if (OsVersion >= VersionTuple(6, 99, 26) || OsVersion.getMajor() == 0)
       return X86_32TargetInfo::getFloatEvalMethod();
     // NetBSD before 6.99.26 defaults to "double" rounding.
     return 1;
@@ -769,7 +768,7 @@ public:
 
   ArrayRef<Builtin::Info> getTargetBuiltins() const override;
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 
 // x86-64 Windows target
diff --git a/clang/lib/Basic/Targets/XCore.h b/clang/lib/Basic/Targets/XCore.h
index c33766751aa1..25f20581839d 100644
--- a/clang/lib/Basic/Targets/XCore.h
+++ b/clang/lib/Basic/Targets/XCore.h
@@ -76,7 +76,7 @@ public:
 
   bool allowsLargerPreferedTypeAlignment() const override { return false; }
 
-  bool hasExtIntType() const override { return true; }
+  bool hasBitIntType() const override { return true; }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Version.cpp b/clang/lib/Basic/Version.cpp
index af3118b0f6da..e205da7adec1 100644
--- a/clang/lib/Basic/Version.cpp
+++ b/clang/lib/Basic/Version.cpp
@@ -82,7 +82,7 @@ std::string getClangFullRepositoryVersion() {
       OS << LLVMRepo << ' ';
     OS << LLVMRev << ')';
   }
-  return OS.str();
+  return buf;
 }
 
 std::string getClangFullVersion() {
@@ -102,7 +102,7 @@ std::string getClangToolFullVersion(StringRef ToolName) {
     OS << " " << repo;
   }
 
-  return OS.str();
+  return buf;
 }
 
 std::string getClangFullCPPVersion() {
@@ -120,7 +120,7 @@ std::string getClangFullCPPVersion() {
     OS << " " << repo;
   }
 
-  return OS.str();
+  return buf;
 }
 
 } // end namespace clang
diff --git a/clang/lib/CodeGen/ABIInfo.h b/clang/lib/CodeGen/ABIInfo.h
index 56f0dd4322d2..0d12183055e1 100644
--- a/clang/lib/CodeGen/ABIInfo.h
+++ b/clang/lib/CodeGen/ABIInfo.h
@@ -105,7 +105,7 @@ namespace swiftcall {
                                 uint64_t &Members) const;
 
     // Implement the Type::IsPromotableIntegerType for ABI specific needs. The
-    // only difference is that this considers _ExtInt as well.
+    // only difference is that this considers bit-precise integer types as well.
     bool isPromotableIntegerTypeForABI(QualType Ty) const;
 
     /// A convenience method to return an indirect ABIArgInfo with an
diff --git a/clang/lib/CodeGen/Address.h b/clang/lib/CodeGen/Address.h
index 6a8e57f8db33..37c20291c0e8 100644
--- a/clang/lib/CodeGen/Address.h
+++ b/clang/lib/CodeGen/Address.h
@@ -23,15 +23,29 @@ namespace CodeGen {
 /// An aligned address.
 class Address {
   llvm::Value *Pointer;
+  llvm::Type *ElementType;
   CharUnits Alignment;
+
+protected:
+  Address(std::nullptr_t) : Pointer(nullptr), ElementType(nullptr) {}
+
 public:
-  Address(llvm::Value *pointer, CharUnits alignment)
-      : Pointer(pointer), Alignment(alignment) {
-    assert((!alignment.isZero() || pointer == nullptr) &&
-           "creating valid address with invalid alignment");
+  Address(llvm::Value *pointer, llvm::Type *elementType, CharUnits alignment)
+      : Pointer(pointer), ElementType(elementType), Alignment(alignment) {
+    assert(pointer != nullptr && "Pointer cannot be null");
+    assert(elementType != nullptr && "Element type cannot be null");
+    assert(llvm::cast<llvm::PointerType>(pointer->getType())
+               ->isOpaqueOrPointeeTypeMatches(elementType) &&
+           "Incorrect pointer element type");
+    assert(!alignment.isZero() && "Alignment cannot be zero");
   }
 
-  static Address invalid() { return Address(nullptr, CharUnits()); }
+  // Deprecated: Use constructor with explicit element type instead.
+  Address(llvm::Value *Pointer, CharUnits Alignment)
+      : Address(Pointer, Pointer->getType()->getPointerElementType(),
+                Alignment) {}
+
+  static Address invalid() { return Address(nullptr); }
   bool isValid() const { return Pointer != nullptr; }
 
   llvm::Value *getPointer() const {
@@ -45,11 +59,9 @@ public:
   }
 
   /// Return the type of the values stored in this address.
-  ///
-  /// When IR pointer types lose their element type, we should simply
-  /// store it in Address instead for the convenience of writing code.
   llvm::Type *getElementType() const {
-    return getType()->getElementType();
+    assert(isValid());
+    return ElementType;
   }
 
   /// Return the address space that this address resides in.
@@ -67,30 +79,42 @@ public:
     assert(isValid());
     return Alignment;
   }
+
+  /// Return address with different pointer, but same element type and
+  /// alignment.
+  Address withPointer(llvm::Value *NewPointer) const {
+    return Address(NewPointer, ElementType, Alignment);
+  }
+
+  /// Return address with different alignment, but same pointer and element
+  /// type.
+  Address withAlignment(CharUnits NewAlignment) const {
+    return Address(Pointer, ElementType, NewAlignment);
+  }
 };
 
 /// A specialization of Address that requires the address to be an
 /// LLVM Constant.
 class ConstantAddress : public Address {
+  ConstantAddress(std::nullptr_t) : Address(nullptr) {}
+
 public:
-  ConstantAddress(llvm::Constant *pointer, CharUnits alignment)
-    : Address(pointer, alignment) {}
+  ConstantAddress(llvm::Constant *pointer, llvm::Type *elementType,
+                  CharUnits alignment)
+      : Address(pointer, elementType, alignment) {}
 
   static ConstantAddress invalid() {
-    return ConstantAddress(nullptr, CharUnits());
+    return ConstantAddress(nullptr);
   }
 
   llvm::Constant *getPointer() const {
     return llvm::cast<llvm::Constant>(Address::getPointer());
   }
 
-  ConstantAddress getBitCast(llvm::Type *ty) const {
-    return ConstantAddress(llvm::ConstantExpr::getBitCast(getPointer(), ty),
-                           getAlignment());
-  }
-
-  ConstantAddress getElementBitCast(llvm::Type *ty) const {
-    return getBitCast(ty->getPointerTo(getAddressSpace()));
+  ConstantAddress getElementBitCast(llvm::Type *ElemTy) const {
+    llvm::Constant *BitCast = llvm::ConstantExpr::getBitCast(
+        getPointer(), ElemTy->getPointerTo(getAddressSpace()));
+    return ConstantAddress(BitCast, ElemTy, getAlignment());
   }
 
   static bool isaImpl(Address addr) {
@@ -98,7 +122,7 @@ public:
   }
   static ConstantAddress castImpl(Address addr) {
     return ConstantAddress(llvm::cast<llvm::Constant>(addr.getPointer()),
-                           addr.getAlignment());
+                           addr.getElementType(), addr.getAlignment());
   }
 };
 
diff --git a/clang/lib/CodeGen/BackendUtil.cpp b/clang/lib/CodeGen/BackendUtil.cpp
index 510f3911939c..bacac0a20d4d 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -94,10 +94,16 @@ using namespace llvm;
   llvm::PassPluginLibraryInfo get##Ext##PluginInfo();
 #include "llvm/Support/Extension.def"
 
+namespace llvm {
+extern cl::opt<bool> DebugInfoCorrelate;
+}
+
 namespace {
 
 // Default filename used for profile generation.
-static constexpr StringLiteral DefaultProfileGenName = "default_%m.profraw";
+std::string getDefaultProfileGenName() {
+  return DebugInfoCorrelate ? "default_%p.proflite" : "default_%m.profraw";
+}
 
 class EmitAssemblyHelper {
   DiagnosticsEngine &Diags;
@@ -597,8 +603,6 @@ static bool initTargetOptions(DiagnosticsEngine &Diags,
   Options.ForceDwarfFrameSection = CodeGenOpts.ForceDwarfFrameSection;
   Options.EmitCallSiteInfo = CodeGenOpts.EmitCallSiteInfo;
   Options.EnableAIXExtendedAltivecABI = CodeGenOpts.EnableAIXExtendedAltivecABI;
-  Options.ValueTrackingVariableLocations =
-      CodeGenOpts.ValueTrackingVariableLocations;
   Options.XRayOmitFunctionIndex = CodeGenOpts.XRayOmitFunctionIndex;
   Options.LoopAlignment = CodeGenOpts.LoopAlignment;
 
@@ -640,6 +644,7 @@ static bool initTargetOptions(DiagnosticsEngine &Diags,
   Options.MCOptions.Argv0 = CodeGenOpts.Argv0;
   Options.MCOptions.CommandLineArgs = CodeGenOpts.CommandLineArgs;
   Options.DebugStrictDwarf = CodeGenOpts.DebugStrictDwarf;
+  Options.ObjectFilenameForDebug = CodeGenOpts.ObjectFilenameForDebug;
 
   return true;
 }
@@ -886,7 +891,7 @@ void EmitAssemblyHelper::CreatePasses(legacy::PassManager &MPM,
     if (!CodeGenOpts.InstrProfileOutput.empty())
       PMBuilder.PGOInstrGen = CodeGenOpts.InstrProfileOutput;
     else
-      PMBuilder.PGOInstrGen = std::string(DefaultProfileGenName);
+      PMBuilder.PGOInstrGen = getDefaultProfileGenName();
   }
   if (CodeGenOpts.hasProfileIRUse()) {
     PMBuilder.PGOInstrUse = CodeGenOpts.ProfileInstrumentUsePath;
@@ -1231,7 +1236,7 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
   if (CodeGenOpts.hasProfileIRInstr())
     // -fprofile-generate.
     PGOOpt = PGOOptions(CodeGenOpts.InstrProfileOutput.empty()
-                            ? std::string(DefaultProfileGenName)
+                            ? getDefaultProfileGenName()
                             : CodeGenOpts.InstrProfileOutput,
                         "", "", PGOOptions::IRInstr, PGOOptions::NoCSAction,
                         CodeGenOpts.DebugInfoForProfiling);
@@ -1269,13 +1274,13 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
              "Cannot run CSProfileGen pass with ProfileGen or SampleUse "
              " pass");
       PGOOpt->CSProfileGenFile = CodeGenOpts.InstrProfileOutput.empty()
-                                     ? std::string(DefaultProfileGenName)
+                                     ? getDefaultProfileGenName()
                                      : CodeGenOpts.InstrProfileOutput;
       PGOOpt->CSAction = PGOOptions::CSIRInstr;
     } else
       PGOOpt = PGOOptions("",
                           CodeGenOpts.InstrProfileOutput.empty()
-                              ? std::string(DefaultProfileGenName)
+                              ? getDefaultProfileGenName()
                               : CodeGenOpts.InstrProfileOutput,
                           "", PGOOptions::NoAction, PGOOptions::CSIRInstr,
                           CodeGenOpts.DebugInfoForProfiling);
@@ -1577,7 +1582,8 @@ static void runThinLTOBackend(
     return;
 
   auto AddStream = [&](size_t Task) {
-    return std::make_unique<CachedFileStream>(std::move(OS));
+    return std::make_unique<CachedFileStream>(std::move(OS),
+                                              CGOpts.ObjectFilenameForDebug);
   };
   lto::Config Conf;
   if (CGOpts.SaveTempsFilePrefix != "") {
diff --git a/clang/lib/CodeGen/CGAtomic.cpp b/clang/lib/CodeGen/CGAtomic.cpp
index b68e6328acdf..e81c5ba5055c 100644
--- a/clang/lib/CodeGen/CGAtomic.cpp
+++ b/clang/lib/CodeGen/CGAtomic.cpp
@@ -1079,8 +1079,8 @@ RValue CodeGenFunction::EmitAtomicExpr(AtomicExpr *E) {
       if (AS == LangAS::opencl_generic)
         return V;
       auto DestAS = getContext().getTargetAddressSpace(LangAS::opencl_generic);
-      auto T = V->getType();
-      auto *DestType = T->getPointerElementType()->getPointerTo(DestAS);
+      auto T = llvm::cast<llvm::PointerType>(V->getType());
+      auto *DestType = llvm::PointerType::getWithSamePointeeType(T, DestAS);
 
       return getTargetHooks().performAddrSpaceCast(
           *this, V, AS, LangAS::opencl_generic, DestType, false);
@@ -1321,15 +1321,14 @@ RValue CodeGenFunction::EmitAtomicExpr(AtomicExpr *E) {
         ResVal = Builder.CreateNot(ResVal);
 
       Builder.CreateStore(
-          ResVal,
-          Builder.CreateBitCast(Dest, ResVal->getType()->getPointerTo()));
+          ResVal, Builder.CreateElementBitCast(Dest, ResVal->getType()));
     }
 
     if (RValTy->isVoidType())
       return RValue::get(nullptr);
 
     return convertTempToRValue(
-        Builder.CreateBitCast(Dest, ConvertTypeForMem(RValTy)->getPointerTo()),
+        Builder.CreateElementBitCast(Dest, ConvertTypeForMem(RValTy)),
         RValTy, E->getExprLoc());
   }
 
@@ -1382,8 +1381,7 @@ RValue CodeGenFunction::EmitAtomicExpr(AtomicExpr *E) {
       return RValue::get(nullptr);
 
     return convertTempToRValue(
-        Builder.CreateBitCast(Dest, ConvertTypeForMem(RValTy)->getPointerTo(
-                                        Dest.getAddressSpace())),
+        Builder.CreateElementBitCast(Dest, ConvertTypeForMem(RValTy)),
         RValTy, E->getExprLoc());
   }
 
@@ -1455,17 +1453,14 @@ RValue CodeGenFunction::EmitAtomicExpr(AtomicExpr *E) {
 
   assert(Atomics.getValueSizeInBits() <= Atomics.getAtomicSizeInBits());
   return convertTempToRValue(
-      Builder.CreateBitCast(Dest, ConvertTypeForMem(RValTy)->getPointerTo(
-                                      Dest.getAddressSpace())),
+      Builder.CreateElementBitCast(Dest, ConvertTypeForMem(RValTy)),
       RValTy, E->getExprLoc());
 }
 
 Address AtomicInfo::emitCastToAtomicIntPointer(Address addr) const {
-  unsigned addrspace =
-    cast<llvm::PointerType>(addr.getPointer()->getType())->getAddressSpace();
   llvm::IntegerType *ty =
     llvm::IntegerType::get(CGF.getLLVMContext(), AtomicSizeInBits);
-  return CGF.Builder.CreateBitCast(addr, ty->getPointerTo(addrspace));
+  return CGF.Builder.CreateElementBitCast(addr, ty);
 }
 
 Address AtomicInfo::convertToAtomicIntPointer(Address Addr) const {
diff --git a/clang/lib/CodeGen/CGBlocks.cpp b/clang/lib/CodeGen/CGBlocks.cpp
index 2da2014345d8..7bb6dbb8a8ac 100644
--- a/clang/lib/CodeGen/CGBlocks.cpp
+++ b/clang/lib/CodeGen/CGBlocks.cpp
@@ -2721,8 +2721,7 @@ void CodeGenFunction::emitByrefStructureInit(const AutoVarEmission &emission) {
   Address addr = emission.Addr;
 
   // That's an alloca of the byref structure type.
-  llvm::StructType *byrefType = cast<llvm::StructType>(
-    cast<llvm::PointerType>(addr.getPointer()->getType())->getElementType());
+  llvm::StructType *byrefType = cast<llvm::StructType>(addr.getElementType());
 
   unsigned nextHeaderIndex = 0;
   CharUnits nextHeaderOffset;
diff --git a/clang/lib/CodeGen/CGBuilder.h b/clang/lib/CodeGen/CGBuilder.h
index 4fad44a105cd..7c9f41e84eaf 100644
--- a/clang/lib/CodeGen/CGBuilder.h
+++ b/clang/lib/CodeGen/CGBuilder.h
@@ -86,7 +86,8 @@ public:
   llvm::LoadInst *CreateAlignedLoad(llvm::Type *Ty, llvm::Value *Addr,
                                     CharUnits Align,
                                     const llvm::Twine &Name = "") {
-    assert(Addr->getType()->getPointerElementType() == Ty);
+    assert(llvm::cast<llvm::PointerType>(Addr->getType())
+               ->isOpaqueOrPointeeTypeMatches(Ty));
     return CreateAlignedLoad(Ty, Addr, Align.getAsAlign(), Name);
   }
 
@@ -115,13 +116,15 @@ public:
   /// Emit a load from an i1 flag variable.
   llvm::LoadInst *CreateFlagLoad(llvm::Value *Addr,
                                  const llvm::Twine &Name = "") {
-    assert(Addr->getType()->getPointerElementType() == getInt1Ty());
+    assert(llvm::cast<llvm::PointerType>(Addr->getType())
+               ->isOpaqueOrPointeeTypeMatches(getInt1Ty()));
     return CreateAlignedLoad(getInt1Ty(), Addr, CharUnits::One(), Name);
   }
 
   /// Emit a store to an i1 flag variable.
   llvm::StoreInst *CreateFlagStore(bool Value, llvm::Value *Addr) {
-    assert(Addr->getType()->getPointerElementType() == getInt1Ty());
+    assert(llvm::cast<llvm::PointerType>(Addr->getType())
+               ->isOpaqueOrPointeeTypeMatches(getInt1Ty()));
     return CreateAlignedStore(getInt1(Value), Addr, CharUnits::One());
   }
 
@@ -165,8 +168,9 @@ public:
   /// preserving information like the alignment and address space.
   Address CreateElementBitCast(Address Addr, llvm::Type *Ty,
                                const llvm::Twine &Name = "") {
-    auto PtrTy = Ty->getPointerTo(Addr.getAddressSpace());
-    return CreateBitCast(Addr, PtrTy, Name);
+    auto *PtrTy = Ty->getPointerTo(Addr.getAddressSpace());
+    return Address(CreateBitCast(Addr.getPointer(), PtrTy, Name),
+                   Ty, Addr.getAlignment());
   }
 
   using CGBuilderBaseTy::CreatePointerBitCastOrAddrSpaceCast;
@@ -194,6 +198,7 @@ public:
 
     return Address(CreateStructGEP(Addr.getElementType(),
                                    Addr.getPointer(), Index, Name),
+                   ElTy->getElementType(Index),
                    Addr.getAlignment().alignmentAtOffset(Offset));
   }
 
@@ -215,6 +220,7 @@ public:
     return Address(
         CreateInBoundsGEP(Addr.getElementType(), Addr.getPointer(),
                           {getSize(CharUnits::Zero()), getSize(Index)}, Name),
+        ElTy->getElementType(),
         Addr.getAlignment().alignmentAtOffset(Index * EltSize));
   }
 
@@ -231,6 +237,7 @@ public:
 
     return Address(CreateInBoundsGEP(Addr.getElementType(), Addr.getPointer(),
                                      getSize(Index), Name),
+                   ElTy,
                    Addr.getAlignment().alignmentAtOffset(Index * EltSize));
   }
 
@@ -247,15 +254,32 @@ public:
 
     return Address(CreateGEP(Addr.getElementType(), Addr.getPointer(),
                              getSize(Index), Name),
+                   Addr.getElementType(),
                    Addr.getAlignment().alignmentAtOffset(Index * EltSize));
   }
 
+  /// Create GEP with single dynamic index. The address alignment is reduced
+  /// according to the element size.
+  using CGBuilderBaseTy::CreateGEP;
+  Address CreateGEP(Address Addr, llvm::Value *Index,
+                    const llvm::Twine &Name = "") {
+    const llvm::DataLayout &DL = BB->getParent()->getParent()->getDataLayout();
+    CharUnits EltSize =
+        CharUnits::fromQuantity(DL.getTypeAllocSize(Addr.getElementType()));
+
+    return Address(CreateGEP(Addr.getElementType(), Addr.getPointer(), Index,
+                             Name),
+                   Addr.getElementType(),
+                   Addr.getAlignment().alignmentOfArrayElement(EltSize));
+  }
+
   /// Given a pointer to i8, adjust it by a given constant offset.
   Address CreateConstInBoundsByteGEP(Address Addr, CharUnits Offset,
                                      const llvm::Twine &Name = "") {
     assert(Addr.getElementType() == TypeCache.Int8Ty);
     return Address(CreateInBoundsGEP(Addr.getElementType(), Addr.getPointer(),
                                      getSize(Offset), Name),
+                   Addr.getElementType(),
                    Addr.getAlignment().alignmentAtOffset(Offset));
   }
   Address CreateConstByteGEP(Address Addr, CharUnits Offset,
@@ -263,6 +287,7 @@ public:
     assert(Addr.getElementType() == TypeCache.Int8Ty);
     return Address(CreateGEP(Addr.getElementType(), Addr.getPointer(),
                              getSize(Offset), Name),
+                   Addr.getElementType(),
                    Addr.getAlignment().alignmentAtOffset(Offset));
   }
 
@@ -278,8 +303,9 @@ public:
         /*isSigned=*/true);
     if (!GEP->accumulateConstantOffset(DL, Offset))
       llvm_unreachable("offset of GEP with constants is always computable");
-    return Address(GEP, Addr.getAlignment().alignmentAtOffset(
-                            CharUnits::fromQuantity(Offset.getSExtValue())));
+    return Address(GEP, GEP->getResultElementType(),
+                   Addr.getAlignment().alignmentAtOffset(
+                       CharUnits::fromQuantity(Offset.getSExtValue())));
   }
 
   using CGBuilderBaseTy::CreateMemCpy;
@@ -330,8 +356,14 @@ public:
 
     return Address(CreatePreserveStructAccessIndex(ElTy, Addr.getPointer(),
                                                    Index, FieldIndex, DbgInfo),
+                   ElTy->getElementType(Index),
                    Addr.getAlignment().alignmentAtOffset(Offset));
   }
+
+  using CGBuilderBaseTy::CreateLaunderInvariantGroup;
+  Address CreateLaunderInvariantGroup(Address Addr) {
+    return Addr.withPointer(CreateLaunderInvariantGroup(Addr.getPointer()));
+  }
 };
 
 }  // end namespace CodeGen
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index 5d6df59cc405..1982b40ff667 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -96,13 +96,33 @@ llvm::Constant *CodeGenModule::getBuiltinLibFunction(const FunctionDecl *FD,
   StringRef Name;
   GlobalDecl D(FD);
 
+  // TODO: This list should be expanded or refactored after all GCC-compatible
+  // std libcall builtins are implemented.
+  static SmallDenseMap<unsigned, StringRef, 8> F128Builtins{
+      {Builtin::BI__builtin_printf, "__printfieee128"},
+      {Builtin::BI__builtin_vsnprintf, "__vsnprintfieee128"},
+      {Builtin::BI__builtin_vsprintf, "__vsprintfieee128"},
+      {Builtin::BI__builtin_sprintf, "__sprintfieee128"},
+      {Builtin::BI__builtin_snprintf, "__snprintfieee128"},
+      {Builtin::BI__builtin_fprintf, "__fprintfieee128"},
+      {Builtin::BI__builtin_nexttowardf128, "__nexttowardieee128"},
+  };
+
   // If the builtin has been declared explicitly with an assembler label,
   // use the mangled name. This differs from the plain label on platforms
   // that prefix labels.
   if (FD->hasAttr<AsmLabelAttr>())
     Name = getMangledName(D);
-  else
-    Name = Context.BuiltinInfo.getName(BuiltinID) + 10;
+  else {
+    // TODO: This mutation should also be applied to other targets other than
+    // PPC, after backend supports IEEE 128-bit style libcalls.
+    if (getTriple().isPPC64() &&
+        &getTarget().getLongDoubleFormat() == &llvm::APFloat::IEEEquad() &&
+        F128Builtins.find(BuiltinID) != F128Builtins.end())
+      Name = F128Builtins[BuiltinID];
+    else
+      Name = Context.BuiltinInfo.getName(BuiltinID) + 10;
+  }
 
   llvm::FunctionType *Ty =
     cast<llvm::FunctionType>(getTypes().ConvertType(FD->getType()));
@@ -667,7 +687,7 @@ getIntegerWidthAndSignedness(const clang::ASTContext &context,
                              const clang::QualType Type) {
   assert(Type->isIntegerType() && "Given type is not an integer.");
   unsigned Width = Type->isBooleanType()  ? 1
-                   : Type->isExtIntType() ? context.getIntWidth(Type)
+                   : Type->isBitIntType() ? context.getIntWidth(Type)
                                           : context.getTypeInfo(Type).Width;
   bool Signed = Type->isSignedIntegerType();
   return {Width, Signed};
@@ -1482,8 +1502,7 @@ Value *CodeGenFunction::EmitMSVCBuiltinExpr(MSVCIntrin BuiltinID,
     Value *ArgValue = EmitScalarExpr(E->getArg(1));
 
     llvm::Type *ArgType = ArgValue->getType();
-    llvm::Type *IndexType =
-        IndexAddress.getPointer()->getType()->getPointerElementType();
+    llvm::Type *IndexType = IndexAddress.getElementType();
     llvm::Type *ResultType = ConvertType(E->getType());
 
     Value *ArgZero = llvm::Constant::getNullValue(ArgType);
@@ -3113,6 +3132,14 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
                                             "elt.abs");
     return RValue::get(Result);
   }
+
+  case Builtin::BI__builtin_elementwise_ceil: {
+    Value *Op0 = EmitScalarExpr(E->getArg(0));
+    Value *Result = Builder.CreateUnaryIntrinsic(llvm::Intrinsic::ceil, Op0,
+                                                 nullptr, "elt.ceil");
+    return RValue::get(Result);
+  }
+
   case Builtin::BI__builtin_elementwise_max: {
     Value *Op0 = EmitScalarExpr(E->getArg(0));
     Value *Op1 = EmitScalarExpr(E->getArg(1));
@@ -3184,6 +3211,13 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     return RValue::get(Result);
   }
 
+  case Builtin::BI__builtin_reduce_xor: {
+    Value *Op0 = EmitScalarExpr(E->getArg(0));
+    Value *Result = Builder.CreateUnaryIntrinsic(
+        llvm::Intrinsic::vector_reduce_xor, Op0, nullptr, "rdx.xor");
+    return RValue::get(Result);
+  }
+
   case Builtin::BI__builtin_matrix_transpose: {
     const auto *MatrixTy = E->getArg(0)->getType()->getAs<ConstantMatrixType>();
     Value *MatValue = EmitScalarExpr(E->getArg(0));
@@ -4478,6 +4512,9 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
   }
   case Builtin::BI__builtin_addressof:
     return RValue::get(EmitLValue(E->getArg(0)).getPointer(*this));
+  case Builtin::BI__builtin_function_start:
+    return RValue::get(CGM.GetFunctionStart(
+        E->getArg(0)->getAsBuiltinConstantDeclRef(CGM.getContext())));
   case Builtin::BI__builtin_operator_new:
     return EmitBuiltinNewDeleteCall(
         E->getCallee()->getType()->castAs<FunctionProtoType>(), E, false);
@@ -4674,8 +4711,6 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     return EmitCoroutineIntrinsic(E, Intrinsic::coro_end);
   case Builtin::BI__builtin_coro_suspend:
     return EmitCoroutineIntrinsic(E, Intrinsic::coro_suspend);
-  case Builtin::BI__builtin_coro_param:
-    return EmitCoroutineIntrinsic(E, Intrinsic::coro_param);
 
   // OpenCL v2.0 s6.13.16.2, Built-in pipe read and write functions
   case Builtin::BIread_pipe:
@@ -5221,9 +5256,9 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     llvm::Type *BPP = Int8PtrPtrTy;
 
     DestAddr = Address(Builder.CreateBitCast(DestAddr.getPointer(), BPP, "cp"),
-                       DestAddr.getAlignment());
+                       Int8PtrTy, DestAddr.getAlignment());
     SrcAddr = Address(Builder.CreateBitCast(SrcAddr.getPointer(), BPP, "ap"),
-                      SrcAddr.getAlignment());
+                      Int8PtrTy, SrcAddr.getAlignment());
 
     Value *ArgPtr = Builder.CreateLoad(SrcAddr, "ap.val");
     return RValue::get(Builder.CreateStore(ArgPtr, DestAddr));
@@ -6385,6 +6420,7 @@ static const ARMVectorIntrinsicInfo AArch64SISDIntrinsicMap[] = {
 static const ARMVectorIntrinsicInfo AArch64SVEIntrinsicMap[] = {
 #define GET_SVE_LLVM_INTRINSIC_MAP
 #include "clang/Basic/arm_sve_builtin_cg.inc"
+#include "clang/Basic/BuiltinsAArch64NeonSVEBridge_cg.def"
 #undef GET_SVE_LLVM_INTRINSIC_MAP
 };
 
@@ -9308,6 +9344,54 @@ Value *CodeGenFunction::EmitAArch64SVEBuiltinExpr(unsigned BuiltinID,
     Function *F = CGM.getIntrinsic(Intrinsic::aarch64_sve_tbl2, VTy);
     return Builder.CreateCall(F, {V0, V1, Ops[1]});
   }
+
+  case SVE::BI__builtin_sve_svset_neonq_s8:
+  case SVE::BI__builtin_sve_svset_neonq_s16:
+  case SVE::BI__builtin_sve_svset_neonq_s32:
+  case SVE::BI__builtin_sve_svset_neonq_s64:
+  case SVE::BI__builtin_sve_svset_neonq_u8:
+  case SVE::BI__builtin_sve_svset_neonq_u16:
+  case SVE::BI__builtin_sve_svset_neonq_u32:
+  case SVE::BI__builtin_sve_svset_neonq_u64:
+  case SVE::BI__builtin_sve_svset_neonq_f16:
+  case SVE::BI__builtin_sve_svset_neonq_f32:
+  case SVE::BI__builtin_sve_svset_neonq_f64:
+  case SVE::BI__builtin_sve_svset_neonq_bf16: {
+    return Builder.CreateInsertVector(Ty, Ops[0], Ops[1], Builder.getInt64(0));
+  }
+
+  case SVE::BI__builtin_sve_svget_neonq_s8:
+  case SVE::BI__builtin_sve_svget_neonq_s16:
+  case SVE::BI__builtin_sve_svget_neonq_s32:
+  case SVE::BI__builtin_sve_svget_neonq_s64:
+  case SVE::BI__builtin_sve_svget_neonq_u8:
+  case SVE::BI__builtin_sve_svget_neonq_u16:
+  case SVE::BI__builtin_sve_svget_neonq_u32:
+  case SVE::BI__builtin_sve_svget_neonq_u64:
+  case SVE::BI__builtin_sve_svget_neonq_f16:
+  case SVE::BI__builtin_sve_svget_neonq_f32:
+  case SVE::BI__builtin_sve_svget_neonq_f64:
+  case SVE::BI__builtin_sve_svget_neonq_bf16: {
+    return Builder.CreateExtractVector(Ty, Ops[0], Builder.getInt64(0));
+  }
+
+  case SVE::BI__builtin_sve_svdup_neonq_s8:
+  case SVE::BI__builtin_sve_svdup_neonq_s16:
+  case SVE::BI__builtin_sve_svdup_neonq_s32:
+  case SVE::BI__builtin_sve_svdup_neonq_s64:
+  case SVE::BI__builtin_sve_svdup_neonq_u8:
+  case SVE::BI__builtin_sve_svdup_neonq_u16:
+  case SVE::BI__builtin_sve_svdup_neonq_u32:
+  case SVE::BI__builtin_sve_svdup_neonq_u64:
+  case SVE::BI__builtin_sve_svdup_neonq_f16:
+  case SVE::BI__builtin_sve_svdup_neonq_f32:
+  case SVE::BI__builtin_sve_svdup_neonq_f64:
+  case SVE::BI__builtin_sve_svdup_neonq_bf16: {
+    Value *Insert = Builder.CreateInsertVector(Ty, UndefValue::get(Ty), Ops[0],
+                                               Builder.getInt64(0));
+    return Builder.CreateIntrinsic(Intrinsic::aarch64_sve_dupq_lane, {Ty},
+                                   {Insert, Builder.getInt64(0)});
+  }
   }
 
   /// Should not happen
@@ -15331,7 +15415,8 @@ Value *CodeGenFunction::EmitPPCBuiltinExpr(unsigned BuiltinID,
     // If the user wants the entire vector, just load the entire vector.
     if (NumBytes == 16) {
       Value *BC = Builder.CreateBitCast(Ops[0], ResTy->getPointerTo());
-      Value *LD = Builder.CreateLoad(Address(BC, CharUnits::fromQuantity(1)));
+      Value *LD =
+          Builder.CreateLoad(Address(BC, ResTy, CharUnits::fromQuantity(1)));
       if (!IsLE)
         return LD;
 
@@ -15392,8 +15477,8 @@ Value *CodeGenFunction::EmitPPCBuiltinExpr(unsigned BuiltinID,
             RevMask.push_back(15 - Idx);
           StVec = Builder.CreateShuffleVector(Ops[2], Ops[2], RevMask);
         }
-        return Builder.CreateStore(StVec,
-                                   Address(BC, CharUnits::fromQuantity(1)));
+        return Builder.CreateStore(
+            StVec, Address(BC, Ops[2]->getType(), CharUnits::fromQuantity(1)));
       }
       auto *ConvTy = Int64Ty;
       unsigned NumElts = 0;
@@ -15427,8 +15512,8 @@ Value *CodeGenFunction::EmitPPCBuiltinExpr(unsigned BuiltinID,
         Function *F = CGM.getIntrinsic(Intrinsic::bswap, ConvTy);
         Elt = Builder.CreateCall(F, Elt);
       }
-      return Builder.CreateStore(Elt,
-                                 Address(PtrBC, CharUnits::fromQuantity(1)));
+      return Builder.CreateStore(
+          Elt, Address(PtrBC, ConvTy, CharUnits::fromQuantity(1)));
     };
     unsigned Stored = 0;
     unsigned RemainingBytes = NumBytes;
@@ -16222,7 +16307,8 @@ Value *EmitAMDGPUWorkGroupSize(CodeGenFunction &CGF, unsigned Index) {
   auto *DstTy =
       CGF.Int16Ty->getPointerTo(GEP->getType()->getPointerAddressSpace());
   auto *Cast = CGF.Builder.CreateBitCast(GEP, DstTy);
-  auto *LD = CGF.Builder.CreateLoad(Address(Cast, CharUnits::fromQuantity(2)));
+  auto *LD = CGF.Builder.CreateLoad(
+      Address(Cast, CGF.Int16Ty, CharUnits::fromQuantity(2)));
   llvm::MDBuilder MDHelper(CGF.getLLVMContext());
   llvm::MDNode *RNode = MDHelper.createRange(APInt(16, 1),
       APInt(16, CGF.getTarget().getMaxOpenCLWorkGroupSize() + 1));
@@ -16242,7 +16328,8 @@ Value *EmitAMDGPUGridSize(CodeGenFunction &CGF, unsigned Index) {
   auto *DstTy =
       CGF.Int32Ty->getPointerTo(GEP->getType()->getPointerAddressSpace());
   auto *Cast = CGF.Builder.CreateBitCast(GEP, DstTy);
-  auto *LD = CGF.Builder.CreateLoad(Address(Cast, CharUnits::fromQuantity(4)));
+  auto *LD = CGF.Builder.CreateLoad(
+      Address(Cast, CGF.Int32Ty, CharUnits::fromQuantity(4)));
   LD->setMetadata(llvm::LLVMContext::MD_invariant_load,
                   llvm::MDNode::get(CGF.getLLVMContext(), None));
   return LD;
@@ -16314,8 +16401,7 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
     llvm::Value *Result = Builder.CreateExtractValue(Tmp, 0);
     llvm::Value *Flag = Builder.CreateExtractValue(Tmp, 1);
 
-    llvm::Type *RealFlagType
-      = FlagOutPtr.getPointer()->getType()->getPointerElementType();
+    llvm::Type *RealFlagType = FlagOutPtr.getElementType();
 
     llvm::Value *FlagExt = Builder.CreateZExt(Flag, RealFlagType);
     Builder.CreateStore(FlagExt, FlagOutPtr);
@@ -16572,6 +16658,15 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
     llvm::Value *RayInverseDir = EmitScalarExpr(E->getArg(4));
     llvm::Value *TextureDescr = EmitScalarExpr(E->getArg(5));
 
+    // The builtins take these arguments as vec4 where the last element is
+    // ignored. The intrinsic takes them as vec3.
+    RayOrigin = Builder.CreateShuffleVector(RayOrigin, RayOrigin,
+                                            ArrayRef<int>{0, 1, 2});
+    RayDir =
+        Builder.CreateShuffleVector(RayDir, RayDir, ArrayRef<int>{0, 1, 2});
+    RayInverseDir = Builder.CreateShuffleVector(RayInverseDir, RayInverseDir,
+                                                ArrayRef<int>{0, 1, 2});
+
     Function *F = CGM.getIntrinsic(Intrinsic::amdgcn_image_bvh_intersect_ray,
                                    {NodePtr->getType(), RayDir->getType()});
     return Builder.CreateCall(F, {NodePtr, RayExtent, RayOrigin, RayDir,
@@ -17938,7 +18033,7 @@ RValue CodeGenFunction::EmitBuiltinAlignTo(const CallExpr *E, bool AlignUp) {
     if (getLangOpts().isSignedOverflowDefined())
       Result = Builder.CreateGEP(Int8Ty, Base, Difference, "aligned_result");
     else
-      Result = EmitCheckedInBoundsGEP(Base, Difference,
+      Result = EmitCheckedInBoundsGEP(Int8Ty, Base, Difference,
                                       /*SignedIndices=*/true,
                                       /*isSubtraction=*/!AlignUp,
                                       E->getExprLoc(), "aligned_result");
@@ -18501,6 +18596,7 @@ getIntrinsicForHexagonNonGCCBuiltin(unsigned BuiltinID) {
     CUSTOM_BUILTIN_MAPPING(S2_storerf_pcr, 0)
     CUSTOM_BUILTIN_MAPPING(S2_storeri_pcr, 0)
     CUSTOM_BUILTIN_MAPPING(S2_storerd_pcr, 0)
+    // Legacy builtins that take a vector in place of a vector predicate.
     CUSTOM_BUILTIN_MAPPING(V6_vmaskedstoreq, 64)
     CUSTOM_BUILTIN_MAPPING(V6_vmaskedstorenq, 64)
     CUSTOM_BUILTIN_MAPPING(V6_vmaskedstorentq, 64)
@@ -18534,8 +18630,8 @@ Value *CodeGenFunction::EmitHexagonBuiltinExpr(unsigned BuiltinID,
   auto MakeCircOp = [this, E](unsigned IntID, bool IsLoad) {
     // The base pointer is passed by address, so it needs to be loaded.
     Address A = EmitPointerWithAlignment(E->getArg(0));
-    Address BP = Address(
-        Builder.CreateBitCast(A.getPointer(), Int8PtrPtrTy), A.getAlignment());
+    Address BP = Address(Builder.CreateBitCast(
+        A.getPointer(), Int8PtrPtrTy), Int8PtrTy, A.getAlignment());
     llvm::Value *Base = Builder.CreateLoad(BP);
     // The treatment of both loads and stores is the same: the arguments for
     // the builtin are the same as the arguments for the intrinsic.
@@ -18579,7 +18675,7 @@ Value *CodeGenFunction::EmitHexagonBuiltinExpr(unsigned BuiltinID,
     // per call.
     Address DestAddr = EmitPointerWithAlignment(E->getArg(1));
     DestAddr = Address(Builder.CreateBitCast(DestAddr.getPointer(), Int8PtrTy),
-                       DestAddr.getAlignment());
+                       Int8Ty, DestAddr.getAlignment());
     llvm::Value *DestAddress = DestAddr.getPointer();
 
     // Operands are Base, Dest, Modifier.
@@ -18626,8 +18722,8 @@ Value *CodeGenFunction::EmitHexagonBuiltinExpr(unsigned BuiltinID,
   case Hexagon::BI__builtin_HEXAGON_V6_vsubcarry_128B: {
     // Get the type from the 0-th argument.
     llvm::Type *VecType = ConvertType(E->getArg(0)->getType());
-    Address PredAddr = Builder.CreateBitCast(
-        EmitPointerWithAlignment(E->getArg(2)), VecType->getPointerTo(0));
+    Address PredAddr = Builder.CreateElementBitCast(
+        EmitPointerWithAlignment(E->getArg(2)), VecType);
     llvm::Value *PredIn = V2Q(Builder.CreateLoad(PredAddr));
     llvm::Value *Result = Builder.CreateCall(CGM.getIntrinsic(ID),
         {EmitScalarExpr(E->getArg(0)), EmitScalarExpr(E->getArg(1)), PredIn});
@@ -18638,6 +18734,27 @@ Value *CodeGenFunction::EmitHexagonBuiltinExpr(unsigned BuiltinID,
     return Builder.CreateExtractValue(Result, 0);
   }
 
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstoreq:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstorenq:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstorentq:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstorentnq:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstoreq_128B:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstorenq_128B:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstorentq_128B:
+  case Hexagon::BI__builtin_HEXAGON_V6_vmaskedstorentnq_128B: {
+    SmallVector<llvm::Value*,4> Ops;
+    const Expr *PredOp = E->getArg(0);
+    // There will be an implicit cast to a boolean vector. Strip it.
+    if (auto *Cast = dyn_cast<ImplicitCastExpr>(PredOp)) {
+      if (Cast->getCastKind() == CK_BitCast)
+        PredOp = Cast->getSubExpr();
+      Ops.push_back(V2Q(EmitScalarExpr(PredOp)));
+    }
+    for (int i = 1, e = E->getNumArgs(); i != e; ++i)
+      Ops.push_back(EmitScalarExpr(E->getArg(i)));
+    return Builder.CreateCall(CGM.getIntrinsic(ID), Ops);
+  }
+
   case Hexagon::BI__builtin_HEXAGON_L2_loadrub_pci:
   case Hexagon::BI__builtin_HEXAGON_L2_loadrb_pci:
   case Hexagon::BI__builtin_HEXAGON_L2_loadruh_pci:
@@ -18674,40 +18791,6 @@ Value *CodeGenFunction::EmitHexagonBuiltinExpr(unsigned BuiltinID,
     return MakeBrevLd(Intrinsic::hexagon_L2_loadri_pbr, Int32Ty);
   case Hexagon::BI__builtin_brev_ldd:
     return MakeBrevLd(Intrinsic::hexagon_L2_loadrd_pbr, Int64Ty);
-
-  default: {
-    if (ID == Intrinsic::not_intrinsic)
-      return nullptr;
-
-    auto IsVectorPredTy = [](llvm::Type *T) {
-      return T->isVectorTy() &&
-             cast<llvm::VectorType>(T)->getElementType()->isIntegerTy(1);
-    };
-
-    llvm::Function *IntrFn = CGM.getIntrinsic(ID);
-    llvm::FunctionType *IntrTy = IntrFn->getFunctionType();
-    SmallVector<llvm::Value*,4> Ops;
-    for (unsigned i = 0, e = IntrTy->getNumParams(); i != e; ++i) {
-      llvm::Type *T = IntrTy->getParamType(i);
-      const Expr *A = E->getArg(i);
-      if (IsVectorPredTy(T)) {
-        // There will be an implicit cast to a boolean vector. Strip it.
-        if (auto *Cast = dyn_cast<ImplicitCastExpr>(A)) {
-          if (Cast->getCastKind() == CK_BitCast)
-            A = Cast->getSubExpr();
-        }
-        Ops.push_back(V2Q(EmitScalarExpr(A)));
-      } else {
-        Ops.push_back(EmitScalarExpr(A));
-      }
-    }
-
-    llvm::Value *Call = Builder.CreateCall(IntrFn, Ops);
-    if (IsVectorPredTy(IntrTy->getReturnType()))
-      Call = Q2V(Call);
-
-    return Call;
-  } // default
   } // switch
 
   return nullptr;
diff --git a/clang/lib/CodeGen/CGCUDANV.cpp b/clang/lib/CodeGen/CGCUDANV.cpp
index a1b4431ca8c4..c4e3f7f54f4f 100644
--- a/clang/lib/CodeGen/CGCUDANV.cpp
+++ b/clang/lib/CodeGen/CGCUDANV.cpp
@@ -814,6 +814,9 @@ llvm::Function *CGNVCUDARuntime::makeModuleCtorFunction() {
         Linkage,
         /*Initializer=*/llvm::ConstantPointerNull::get(VoidPtrPtrTy),
         "__hip_gpubin_handle");
+    if (Linkage == llvm::GlobalValue::LinkOnceAnyLinkage)
+      GpuBinaryHandle->setComdat(
+          CGM.getModule().getOrInsertComdat(GpuBinaryHandle->getName()));
     GpuBinaryHandle->setAlignment(CGM.getPointerAlign().getAsAlign());
     // Prevent the weak symbol in different shared libraries being merged.
     if (Linkage != llvm::GlobalValue::InternalLinkage)
diff --git a/clang/lib/CodeGen/CGCall.cpp b/clang/lib/CodeGen/CGCall.cpp
index d830a7e01709..d70f78fea6b4 100644
--- a/clang/lib/CodeGen/CGCall.cpp
+++ b/clang/lib/CodeGen/CGCall.cpp
@@ -1261,8 +1261,7 @@ static llvm::Value *CreateCoercedLoad(Address Src, llvm::Type *Ty,
     //
     // FIXME: Assert that we aren't truncating non-padding bits when have access
     // to that information.
-    Src = CGF.Builder.CreateBitCast(Src,
-                                    Ty->getPointerTo(Src.getAddressSpace()));
+    Src = CGF.Builder.CreateElementBitCast(Src, Ty);
     return CGF.Builder.CreateLoad(Src);
   }
 
@@ -1832,11 +1831,6 @@ void CodeGenModule::getDefaultFunctionAttributes(StringRef Name,
     if (LangOpts.getFPExceptionMode() == LangOptions::FPE_Ignore)
       FuncAttrs.addAttribute("no-trapping-math", "true");
 
-    // Strict (compliant) code is the default, so only add this attribute to
-    // indicate that we are trying to workaround a problem case.
-    if (!CodeGenOpts.StrictFloatCastOverflow)
-      FuncAttrs.addAttribute("strict-float-cast-overflow", "false");
-
     // TODO: Are these all needed?
     // unsafe/inf/nan/nsz are handled by instruction-level FastMathFlags.
     if (LangOpts.NoHonorInfs)
@@ -1971,7 +1965,7 @@ static bool DetermineNoUndef(QualType QTy, CodeGenTypes &Types,
       // there's no internal padding (typeSizeEqualsStoreSize).
       return false;
   }
-  if (QTy->isExtIntType())
+  if (QTy->isBitIntType())
     return true;
   if (QTy->isReferenceType())
     return true;
@@ -2686,8 +2680,8 @@ void CodeGenFunction::EmitFunctionProlog(const CGFunctionInfo &FI,
     case ABIArgInfo::Indirect:
     case ABIArgInfo::IndirectAliased: {
       assert(NumIRArgs == 1);
-      Address ParamAddr =
-          Address(Fn->getArg(FirstIRArg), ArgI.getIndirectAlign());
+      Address ParamAddr = Address(Fn->getArg(FirstIRArg), ConvertTypeForMem(Ty),
+                                  ArgI.getIndirectAlign());
 
       if (!hasScalarEvaluationKind(Ty)) {
         // Aggregates and complex variables are accessed by reference. All we
@@ -3475,12 +3469,19 @@ void CodeGenFunction::EmitFunctionEpilog(const CGFunctionInfo &FI,
     case TEK_Aggregate:
       // Do nothing; aggregrates get evaluated directly into the destination.
       break;
-    case TEK_Scalar:
-      EmitStoreOfScalar(Builder.CreateLoad(ReturnValue),
-                        MakeNaturalAlignAddrLValue(&*AI, RetTy),
-                        /*isInit*/ true);
+    case TEK_Scalar: {
+      LValueBaseInfo BaseInfo;
+      TBAAAccessInfo TBAAInfo;
+      CharUnits Alignment =
+          CGM.getNaturalTypeAlignment(RetTy, &BaseInfo, &TBAAInfo);
+      Address ArgAddr(&*AI, ConvertType(RetTy), Alignment);
+      LValue ArgVal =
+          LValue::MakeAddr(ArgAddr, RetTy, getContext(), BaseInfo, TBAAInfo);
+      EmitStoreOfScalar(
+          Builder.CreateLoad(ReturnValue), ArgVal, /*isInit*/ true);
       break;
     }
+    }
     break;
   }
 
@@ -4134,8 +4135,7 @@ void CodeGenFunction::EmitCallArgs(
   }
 
   // If we still have any arguments, emit them using the type of the argument.
-  for (auto *A : llvm::make_range(std::next(ArgRange.begin(), ArgTypes.size()),
-                                  ArgRange.end()))
+  for (auto *A : llvm::drop_begin(ArgRange, ArgTypes.size()))
     ArgTypes.push_back(IsVariadic ? getVarArgType(A) : A->getType());
   assert((int)ArgTypes.size() == (ArgRange.end() - ArgRange.begin()));
 
@@ -4308,11 +4308,8 @@ void CodeGenFunction::EmitCallArg(CallArgList &args, const Expr *E,
       type->castAs<RecordType>()->getDecl()->isParamDestroyedInCallee()) {
     // If we're using inalloca, use the argument memory.  Otherwise, use a
     // temporary.
-    AggValueSlot Slot;
-    if (args.isUsingInAlloca())
-      Slot = createPlaceholderSlot(*this, type);
-    else
-      Slot = CreateAggTemp(type, "agg.tmp");
+    AggValueSlot Slot = args.isUsingInAlloca()
+        ? createPlaceholderSlot(*this, type) : CreateAggTemp(type, "agg.tmp");
 
     bool DestroyedInCallee = true, NeedsEHCleanup = true;
     if (const auto *RD = type->getAsCXXRecordDecl())
@@ -4651,13 +4648,13 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo &CallInfo,
     //
     // In other cases, we assert that the types match up (until pointers stop
     // having pointee types).
-    llvm::Type *TypeFromVal;
     if (Callee.isVirtual())
-      TypeFromVal = Callee.getVirtualFunctionType();
-    else
-      TypeFromVal =
-          Callee.getFunctionPointer()->getType()->getPointerElementType();
-    assert(IRFuncTy == TypeFromVal);
+      assert(IRFuncTy == Callee.getVirtualFunctionType());
+    else {
+      llvm::PointerType *PtrTy =
+          llvm::cast<llvm::PointerType>(Callee.getFunctionPointer()->getType());
+      assert(PtrTy->isOpaqueOrPointeeTypeMatches(IRFuncTy));
+    }
   }
 #endif
 
@@ -4872,7 +4869,8 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo &CallInfo,
           I->copyInto(*this, AI);
         } else {
           // Skip the extra memcpy call.
-          auto *T = V->getType()->getPointerElementType()->getPointerTo(
+          auto *T = llvm::PointerType::getWithSamePointeeType(
+              cast<llvm::PointerType>(V->getType()),
               CGM.getDataLayout().getAllocaAddrSpace());
           IRCallArgs[FirstIRArg] = getTargetHooks().performAddrSpaceCast(
               *this, V, LangAS::Default, CGM.getASTAllocaAddressSpace(), T,
@@ -4967,8 +4965,7 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo &CallInfo,
           Builder.CreateMemCpy(TempAlloca, Src, SrcSize);
           Src = TempAlloca;
         } else {
-          Src = Builder.CreateBitCast(Src,
-                                      STy->getPointerTo(Src.getAddressSpace()));
+          Src = Builder.CreateElementBitCast(Src, STy);
         }
 
         assert(NumIRArgs == STy->getNumElements());
diff --git a/clang/lib/CodeGen/CGCall.h b/clang/lib/CodeGen/CGCall.h
index e3d9fec6d363..c8594068c3fc 100644
--- a/clang/lib/CodeGen/CGCall.h
+++ b/clang/lib/CodeGen/CGCall.h
@@ -115,7 +115,8 @@ public:
     AbstractInfo = abstractInfo;
     assert(functionPtr && "configuring callee without function pointer");
     assert(functionPtr->getType()->isPointerTy());
-    assert(functionPtr->getType()->getPointerElementType()->isFunctionTy());
+    assert(functionPtr->getType()->isOpaquePointerTy() ||
+           functionPtr->getType()->getPointerElementType()->isFunctionTy());
   }
 
   static CGCallee forBuiltin(unsigned builtinID,
diff --git a/clang/lib/CodeGen/CGClass.cpp b/clang/lib/CodeGen/CGClass.cpp
index 0df64d4d5d26..8f99ff0d50ff 100644
--- a/clang/lib/CodeGen/CGClass.cpp
+++ b/clang/lib/CodeGen/CGClass.cpp
@@ -127,18 +127,18 @@ CodeGenModule::getDynamicOffsetAlignment(CharUnits actualBaseAlign,
 
 Address CodeGenFunction::LoadCXXThisAddress() {
   assert(CurFuncDecl && "loading 'this' without a func declaration?");
-  assert(isa<CXXMethodDecl>(CurFuncDecl));
+  auto *MD = cast<CXXMethodDecl>(CurFuncDecl);
 
   // Lazily compute CXXThisAlignment.
   if (CXXThisAlignment.isZero()) {
     // Just use the best known alignment for the parent.
     // TODO: if we're currently emitting a complete-object ctor/dtor,
     // we can always use the complete-object alignment.
-    auto RD = cast<CXXMethodDecl>(CurFuncDecl)->getParent();
-    CXXThisAlignment = CGM.getClassPointerAlignment(RD);
+    CXXThisAlignment = CGM.getClassPointerAlignment(MD->getParent());
   }
 
-  return Address(LoadCXXThis(), CXXThisAlignment);
+  llvm::Type *Ty = ConvertType(MD->getThisType()->getPointeeType());
+  return Address(LoadCXXThis(), Ty, CXXThisAlignment);
 }
 
 /// Emit the address of a field using a member data pointer.
@@ -286,7 +286,7 @@ ApplyNonVirtualAndVirtualOffset(CodeGenFunction &CGF, Address addr,
   }
   alignment = alignment.alignmentAtOffset(nonVirtualOffset);
 
-  return Address(ptr, alignment);
+  return Address(ptr, CGF.Int8Ty, alignment);
 }
 
 Address CodeGenFunction::GetAddressOfBaseClass(
@@ -326,9 +326,9 @@ Address CodeGenFunction::GetAddressOfBaseClass(
   }
 
   // Get the base pointer type.
+  llvm::Type *BaseValueTy = ConvertType((PathEnd[-1])->getType());
   llvm::Type *BasePtrTy =
-      ConvertType((PathEnd[-1])->getType())
-          ->getPointerTo(Value.getType()->getPointerAddressSpace());
+      BaseValueTy->getPointerTo(Value.getType()->getPointerAddressSpace());
 
   QualType DerivedTy = getContext().getRecordType(Derived);
   CharUnits DerivedAlign = CGM.getClassPointerAlignment(Derived);
@@ -342,7 +342,7 @@ Address CodeGenFunction::GetAddressOfBaseClass(
       EmitTypeCheck(TCK_Upcast, Loc, Value.getPointer(),
                     DerivedTy, DerivedAlign, SkippedChecks);
     }
-    return Builder.CreateBitCast(Value, BasePtrTy);
+    return Builder.CreateElementBitCast(Value, BaseValueTy);
   }
 
   llvm::BasicBlock *origBB = nullptr;
@@ -379,7 +379,7 @@ Address CodeGenFunction::GetAddressOfBaseClass(
                                           VirtualOffset, Derived, VBase);
 
   // Cast to the destination type.
-  Value = Builder.CreateBitCast(Value, BasePtrTy);
+  Value = Builder.CreateElementBitCast(Value, BaseValueTy);
 
   // Build a phi if we needed a null check.
   if (NullCheckValue) {
@@ -406,16 +406,16 @@ CodeGenFunction::GetAddressOfDerivedClass(Address BaseAddr,
 
   QualType DerivedTy =
     getContext().getCanonicalType(getContext().getTagDeclType(Derived));
-  unsigned AddrSpace =
-    BaseAddr.getPointer()->getType()->getPointerAddressSpace();
-  llvm::Type *DerivedPtrTy = ConvertType(DerivedTy)->getPointerTo(AddrSpace);
+  unsigned AddrSpace = BaseAddr.getAddressSpace();
+  llvm::Type *DerivedValueTy = ConvertType(DerivedTy);
+  llvm::Type *DerivedPtrTy = DerivedValueTy->getPointerTo(AddrSpace);
 
   llvm::Value *NonVirtualOffset =
     CGM.GetNonVirtualBaseClassOffset(Derived, PathBegin, PathEnd);
 
   if (!NonVirtualOffset) {
     // No offset, we can just cast back.
-    return Builder.CreateBitCast(BaseAddr, DerivedPtrTy);
+    return Builder.CreateElementBitCast(BaseAddr, DerivedValueTy);
   }
 
   llvm::BasicBlock *CastNull = nullptr;
@@ -453,7 +453,7 @@ CodeGenFunction::GetAddressOfDerivedClass(Address BaseAddr,
     Value = PHI;
   }
 
-  return Address(Value, CGM.getClassPointerAlignment(Derived));
+  return Address(Value, DerivedValueTy, CGM.getClassPointerAlignment(Derived));
 }
 
 llvm::Value *CodeGenFunction::GetVTTParameter(GlobalDecl GD,
@@ -996,16 +996,8 @@ namespace {
 
   private:
     void emitMemcpyIR(Address DestPtr, Address SrcPtr, CharUnits Size) {
-      llvm::PointerType *DPT = DestPtr.getType();
-      llvm::Type *DBP =
-        llvm::Type::getInt8PtrTy(CGF.getLLVMContext(), DPT->getAddressSpace());
-      DestPtr = CGF.Builder.CreateBitCast(DestPtr, DBP);
-
-      llvm::PointerType *SPT = SrcPtr.getType();
-      llvm::Type *SBP =
-        llvm::Type::getInt8PtrTy(CGF.getLLVMContext(), SPT->getAddressSpace());
-      SrcPtr = CGF.Builder.CreateBitCast(SrcPtr, SBP);
-
+      DestPtr = CGF.Builder.CreateElementBitCast(DestPtr, CGF.Int8Ty);
+      SrcPtr = CGF.Builder.CreateElementBitCast(SrcPtr, CGF.Int8Ty);
       CGF.Builder.CreateMemCpy(DestPtr, SrcPtr, Size.getQuantity());
     }
 
@@ -2068,8 +2060,8 @@ void CodeGenFunction::EmitCXXConstructorCall(const CXXConstructorDecl *D,
 
   if (SlotAS != ThisAS) {
     unsigned TargetThisAS = getContext().getTargetAddressSpace(ThisAS);
-    llvm::Type *NewType =
-        ThisPtr->getType()->getPointerElementType()->getPointerTo(TargetThisAS);
+    llvm::Type *NewType = llvm::PointerType::getWithSamePointeeType(
+        This.getType(), TargetThisAS);
     ThisPtr = getTargetHooks().performAddrSpaceCast(*this, This.getPointer(),
                                                     ThisAS, SlotAS, NewType);
   }
@@ -2507,9 +2499,6 @@ void CodeGenFunction::InitializeVTablePointer(const VPtr &Vptr) {
 
   // Apply the offsets.
   Address VTableField = LoadCXXThisAddress();
-  unsigned ThisAddrSpace =
-      VTableField.getPointer()->getType()->getPointerAddressSpace();
-
   if (!NonVirtualOffset.isZero() || VirtualOffset)
     VTableField = ApplyNonVirtualAndVirtualOffset(
         *this, VTableField, NonVirtualOffset, VirtualOffset, Vptr.VTableClass,
@@ -2525,8 +2514,7 @@ void CodeGenFunction::InitializeVTablePointer(const VPtr &Vptr) {
           ->getPointerTo(GlobalsAS);
   // vtable field is is derived from `this` pointer, therefore they should be in
   // the same addr space. Note that this might not be LLVM address space 0.
-  VTableField = Builder.CreateBitCast(VTableField,
-                                      VTablePtrTy->getPointerTo(ThisAddrSpace));
+  VTableField = Builder.CreateElementBitCast(VTableField, VTablePtrTy);
   VTableAddressPoint = Builder.CreateBitCast(VTableAddressPoint, VTablePtrTy);
 
   llvm::StoreInst *Store = Builder.CreateStore(VTableAddressPoint, VTableField);
diff --git a/clang/lib/CodeGen/CGCleanup.h b/clang/lib/CodeGen/CGCleanup.h
index 1b54c0018d27..76f3a48f32f3 100644
--- a/clang/lib/CodeGen/CGCleanup.h
+++ b/clang/lib/CodeGen/CGCleanup.h
@@ -242,7 +242,7 @@ class alignas(8) EHCleanupScope : public EHScope {
 
   /// An optional i1 variable indicating whether this cleanup has been
   /// activated yet.
-  llvm::AllocaInst *ActiveFlag;
+  Address ActiveFlag;
 
   /// Extra information required for cleanups that have resolved
   /// branches through them.  This has to be allocated on the side
@@ -290,7 +290,8 @@ public:
                  EHScopeStack::stable_iterator enclosingEH)
       : EHScope(EHScope::Cleanup, enclosingEH),
         EnclosingNormal(enclosingNormal), NormalBlock(nullptr),
-        ActiveFlag(nullptr), ExtInfo(nullptr), FixupDepth(fixupDepth) {
+        ActiveFlag(Address::invalid()), ExtInfo(nullptr),
+        FixupDepth(fixupDepth) {
     CleanupBits.IsNormalCleanup = isNormal;
     CleanupBits.IsEHCleanup = isEH;
     CleanupBits.IsActive = true;
@@ -320,13 +321,13 @@ public:
   bool isLifetimeMarker() const { return CleanupBits.IsLifetimeMarker; }
   void setLifetimeMarker() { CleanupBits.IsLifetimeMarker = true; }
 
-  bool hasActiveFlag() const { return ActiveFlag != nullptr; }
+  bool hasActiveFlag() const { return ActiveFlag.isValid(); }
   Address getActiveFlag() const {
-    return Address(ActiveFlag, CharUnits::One());
+    return ActiveFlag;
   }
   void setActiveFlag(Address Var) {
     assert(Var.getAlignment().isOne());
-    ActiveFlag = cast<llvm::AllocaInst>(Var.getPointer());
+    ActiveFlag = Var;
   }
 
   void setTestFlagInNormalCleanup() {
diff --git a/clang/lib/CodeGen/CGCoroutine.cpp b/clang/lib/CodeGen/CGCoroutine.cpp
index ca071d3d2e80..2041d2a5b4c9 100644
--- a/clang/lib/CodeGen/CGCoroutine.cpp
+++ b/clang/lib/CodeGen/CGCoroutine.cpp
@@ -597,6 +597,10 @@ void CodeGenFunction::EmitCoroutineBody(const CoroutineBodyStmt &S) {
       CGM.getIntrinsic(llvm::Intrinsic::coro_begin), {CoroId, Phi});
   CurCoro.Data->CoroBegin = CoroBegin;
 
+  // We need to emit `get_­return_­object` first. According to:
+  // [dcl.fct.def.coroutine]p7
+  // The call to get_­return_­object is sequenced before the call to
+  // initial_­suspend and is invoked at most once.
   GetReturnObjectManager GroManager(*this, S);
   GroManager.EmitGroAlloca();
 
diff --git a/clang/lib/CodeGen/CGDebugInfo.cpp b/clang/lib/CodeGen/CGDebugInfo.cpp
index af651e6f44b7..6e189a61dd20 100644
--- a/clang/lib/CodeGen/CGDebugInfo.cpp
+++ b/clang/lib/CodeGen/CGDebugInfo.cpp
@@ -768,7 +768,7 @@ llvm::DIType *CGDebugInfo::CreateType(const BuiltinType *BT) {
       }
 
       // Element count = (VLENB / SEW) x LMUL
-      SmallVector<int64_t, 9> Expr(
+      SmallVector<int64_t, 12> Expr(
           // The DW_OP_bregx operation has two operands: a register which is
           // specified by an unsigned LEB128 number, followed by a signed LEB128
           // offset.
@@ -782,6 +782,8 @@ llvm::DIType *CGDebugInfo::CreateType(const BuiltinType *BT) {
         Expr.push_back(llvm::dwarf::DW_OP_div);
       else
         Expr.push_back(llvm::dwarf::DW_OP_mul);
+      // Element max index = count - 1
+      Expr.append({llvm::dwarf::DW_OP_constu, 1, llvm::dwarf::DW_OP_minus});
 
       auto *LowerBound =
           llvm::ConstantAsMetadata::get(llvm::ConstantInt::getSigned(
@@ -884,9 +886,9 @@ llvm::DIType *CGDebugInfo::CreateType(const AutoType *Ty) {
   return DBuilder.createUnspecifiedType("auto");
 }
 
-llvm::DIType *CGDebugInfo::CreateType(const ExtIntType *Ty) {
+llvm::DIType *CGDebugInfo::CreateType(const BitIntType *Ty) {
 
-  StringRef Name = Ty->isUnsigned() ? "unsigned _ExtInt" : "_ExtInt";
+  StringRef Name = Ty->isUnsigned() ? "unsigned _BitInt" : "_BitInt";
   llvm::dwarf::TypeKind Encoding = Ty->isUnsigned()
                                        ? llvm::dwarf::DW_ATE_unsigned
                                        : llvm::dwarf::DW_ATE_signed;
@@ -3353,6 +3355,9 @@ static QualType UnwrapTypeForDebugInfo(QualType T, const ASTContext &C) {
     case Type::Elaborated:
       T = cast<ElaboratedType>(T)->getNamedType();
       break;
+    case Type::Using:
+      T = cast<UsingType>(T)->getUnderlyingType();
+      break;
     case Type::Paren:
       T = cast<ParenType>(T)->getInnerType();
       break;
@@ -3531,8 +3536,8 @@ llvm::DIType *CGDebugInfo::CreateTypeNode(QualType Ty, llvm::DIFile *Unit,
   case Type::Atomic:
     return CreateType(cast<AtomicType>(Ty), Unit);
 
-  case Type::ExtInt:
-    return CreateType(cast<ExtIntType>(Ty));
+  case Type::BitInt:
+    return CreateType(cast<BitIntType>(Ty));
   case Type::Pipe:
     return CreateType(cast<PipeType>(Ty), Unit);
 
@@ -3545,6 +3550,7 @@ llvm::DIType *CGDebugInfo::CreateTypeNode(QualType Ty, llvm::DIFile *Unit,
   case Type::Decayed:
   case Type::DeducedTemplateSpecialization:
   case Type::Elaborated:
+  case Type::Using:
   case Type::Paren:
   case Type::MacroQualified:
   case Type::SubstTemplateTypeParm:
@@ -3633,6 +3639,9 @@ llvm::DICompositeType *CGDebugInfo::CreateLimitedType(const RecordType *Ty) {
     // Record exports it symbols to the containing structure.
     if (CXXRD->isAnonymousStructOrUnion())
         Flags |= llvm::DINode::FlagExportSymbols;
+
+    Flags |= getAccessFlag(CXXRD->getAccess(),
+                           dyn_cast<CXXRecordDecl>(CXXRD->getDeclContext()));
   }
 
   llvm::DINodeArray Annotations = CollectBTFDeclTagAnnotations(D);
diff --git a/clang/lib/CodeGen/CGDebugInfo.h b/clang/lib/CodeGen/CGDebugInfo.h
index a7b72fa5f5a6..14ff0eeabd21 100644
--- a/clang/lib/CodeGen/CGDebugInfo.h
+++ b/clang/lib/CodeGen/CGDebugInfo.h
@@ -177,7 +177,7 @@ class CGDebugInfo {
   llvm::DIType *CreateType(const BuiltinType *Ty);
   llvm::DIType *CreateType(const ComplexType *Ty);
   llvm::DIType *CreateType(const AutoType *Ty);
-  llvm::DIType *CreateType(const ExtIntType *Ty);
+  llvm::DIType *CreateType(const BitIntType *Ty);
   llvm::DIType *CreateQualifiedType(QualType Ty, llvm::DIFile *Fg,
                                     TypeLoc TL = TypeLoc());
   llvm::DIType *CreateQualifiedType(const FunctionProtoType *Ty,
diff --git a/clang/lib/CodeGen/CGDecl.cpp b/clang/lib/CodeGen/CGDecl.cpp
index 941671c61482..e09279c1d455 100644
--- a/clang/lib/CodeGen/CGDecl.cpp
+++ b/clang/lib/CodeGen/CGDecl.cpp
@@ -405,7 +405,8 @@ void CodeGenFunction::EmitStaticVarDecl(const VarDecl &D,
 
   // Store into LocalDeclMap before generating initializer to handle
   // circular references.
-  setAddrOfLocalVar(&D, Address(addr, alignment));
+  llvm::Type *elemTy = ConvertTypeForMem(D.getType());
+  setAddrOfLocalVar(&D, Address(addr, elemTy, alignment));
 
   // We can't have a VLA here, but we can have a pointer to a VLA,
   // even though that doesn't really make any sense.
@@ -458,8 +459,7 @@ void CodeGenFunction::EmitStaticVarDecl(const VarDecl &D,
   // RAUW's the GV uses of this constant will be invalid.
   llvm::Constant *castedAddr =
     llvm::ConstantExpr::getPointerBitCastOrAddrSpaceCast(var, expectedType);
-  if (var != castedAddr)
-    LocalDeclMap.find(&D)->second = Address(castedAddr, alignment);
+  LocalDeclMap.find(&D)->second = Address(castedAddr, elemTy, alignment);
   CGM.setStaticLocalDeclAddress(&D, castedAddr);
 
   CGM.getSanitizerMetadata()->reportGlobalToASan(var, D);
@@ -1146,7 +1146,7 @@ Address CodeGenModule::createUnnamedGlobalFrom(const VarDecl &D,
     CacheEntry->setAlignment(Align.getAsAlign());
   }
 
-  return Address(CacheEntry, Align);
+  return Address(CacheEntry, CacheEntry->getValueType(), Align);
 }
 
 static Address createUnnamedGlobalForMemcpyFrom(CodeGenModule &CGM,
@@ -1193,7 +1193,7 @@ static void emitStoresForConstant(CodeGenModule &CGM, const VarDecl &D,
     bool valueAlreadyCorrect =
         constant->isNullValue() || isa<llvm::UndefValue>(constant);
     if (!valueAlreadyCorrect) {
-      Loc = Builder.CreateBitCast(Loc, Ty->getPointerTo(Loc.getAddressSpace()));
+      Loc = Builder.CreateElementBitCast(Loc, Ty);
       emitStoresForInitAfterBZero(CGM, constant, Loc, isVolatile, Builder,
                                   IsAutoInit);
     }
diff --git a/clang/lib/CodeGen/CGDeclCXX.cpp b/clang/lib/CodeGen/CGDeclCXX.cpp
index d22f9dc3b68c..3579761f1429 100644
--- a/clang/lib/CodeGen/CGDeclCXX.cpp
+++ b/clang/lib/CodeGen/CGDeclCXX.cpp
@@ -172,7 +172,7 @@ void CodeGenFunction::EmitInvariantStart(llvm::Constant *Addr, CharUnits Size) {
 }
 
 void CodeGenFunction::EmitCXXGlobalVarDeclInit(const VarDecl &D,
-                                               llvm::Constant *DeclPtr,
+                                               llvm::GlobalVariable *GV,
                                                bool PerformInit) {
 
   const Expr *Init = D.getInit();
@@ -194,14 +194,16 @@ void CodeGenFunction::EmitCXXGlobalVarDeclInit(const VarDecl &D,
   // "shared" address space qualifier, but the constructor of StructWithCtor
   // expects "this" in the "generic" address space.
   unsigned ExpectedAddrSpace = getContext().getTargetAddressSpace(T);
-  unsigned ActualAddrSpace = DeclPtr->getType()->getPointerAddressSpace();
+  unsigned ActualAddrSpace = GV->getAddressSpace();
+  llvm::Constant *DeclPtr = GV;
   if (ActualAddrSpace != ExpectedAddrSpace) {
-    llvm::Type *LTy = CGM.getTypes().ConvertTypeForMem(T);
-    llvm::PointerType *PTy = llvm::PointerType::get(LTy, ExpectedAddrSpace);
+    llvm::PointerType *PTy = llvm::PointerType::getWithSamePointeeType(
+        GV->getType(), ExpectedAddrSpace);
     DeclPtr = llvm::ConstantExpr::getAddrSpaceCast(DeclPtr, PTy);
   }
 
-  ConstantAddress DeclAddr(DeclPtr, getContext().getDeclAlign(&D));
+  ConstantAddress DeclAddr(
+      DeclPtr, GV->getValueType(), getContext().getDeclAlign(&D));
 
   if (!T->isReferenceType()) {
     if (getLangOpts().OpenMP && !getLangOpts().OpenMPSimd &&
diff --git a/clang/lib/CodeGen/CGException.cpp b/clang/lib/CodeGen/CGException.cpp
index aff9c77d53c7..91ecbecc843f 100644
--- a/clang/lib/CodeGen/CGException.cpp
+++ b/clang/lib/CodeGen/CGException.cpp
@@ -400,8 +400,8 @@ void CodeGenFunction::EmitAnyExprToExn(const Expr *e, Address addr) {
 
   // __cxa_allocate_exception returns a void*;  we need to cast this
   // to the appropriate type for the object.
-  llvm::Type *ty = ConvertTypeForMem(e->getType())->getPointerTo();
-  Address typedAddr = Builder.CreateBitCast(addr, ty);
+  llvm::Type *ty = ConvertTypeForMem(e->getType());
+  Address typedAddr = Builder.CreateElementBitCast(addr, ty);
 
   // FIXME: this isn't quite right!  If there's a final unelided call
   // to a copy constructor, then according to [except.terminate]p1 we
@@ -421,13 +421,13 @@ void CodeGenFunction::EmitAnyExprToExn(const Expr *e, Address addr) {
 Address CodeGenFunction::getExceptionSlot() {
   if (!ExceptionSlot)
     ExceptionSlot = CreateTempAlloca(Int8PtrTy, "exn.slot");
-  return Address(ExceptionSlot, getPointerAlign());
+  return Address(ExceptionSlot, Int8PtrTy, getPointerAlign());
 }
 
 Address CodeGenFunction::getEHSelectorSlot() {
   if (!EHSelectorSlot)
     EHSelectorSlot = CreateTempAlloca(Int32Ty, "ehselector.slot");
-  return Address(EHSelectorSlot, CharUnits::fromQuantity(4));
+  return Address(EHSelectorSlot, Int32Ty, CharUnits::fromQuantity(4));
 }
 
 llvm::Value *CodeGenFunction::getExceptionFromSlot() {
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 4332e74dbb24..34b4951a7f72 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -71,7 +71,7 @@ Address CodeGenFunction::CreateTempAllocaWithoutCast(llvm::Type *Ty,
                                                      llvm::Value *ArraySize) {
   auto Alloca = CreateTempAlloca(Ty, Name, ArraySize);
   Alloca->setAlignment(Align.getAsAlign());
-  return Address(Alloca, Align);
+  return Address(Alloca, Ty, Align);
 }
 
 /// CreateTempAlloca - This creates a alloca and inserts it into the entry
@@ -101,7 +101,7 @@ Address CodeGenFunction::CreateTempAlloca(llvm::Type *Ty, CharUnits Align,
         Ty->getPointerTo(DestAddrSpace), /*non-null*/ true);
   }
 
-  return Address(V, Align);
+  return Address(V, Ty, Align);
 }
 
 /// CreateTempAlloca - This creates an alloca and inserts it into the entry
@@ -144,7 +144,7 @@ Address CodeGenFunction::CreateMemTemp(QualType Ty, CharUnits Align,
                                     /*ArraySize=*/nullptr, Alloca);
 
   if (Ty->isConstantMatrixType()) {
-    auto *ArrayTy = cast<llvm::ArrayType>(Result.getType()->getElementType());
+    auto *ArrayTy = cast<llvm::ArrayType>(Result.getElementType());
     auto *VectorTy = llvm::FixedVectorType::get(ArrayTy->getElementType(),
                                                 ArrayTy->getNumElements());
 
@@ -1099,7 +1099,7 @@ Address CodeGenFunction::EmitPointerWithAlignment(const Expr *E,
           if (InnerBaseInfo.getAlignmentSource() != AlignmentSource::Decl) {
             if (BaseInfo)
               BaseInfo->mergeForCast(TargetTypeBaseInfo);
-            Addr = Address(Addr.getPointer(), Align);
+            Addr = Address(Addr.getPointer(), Addr.getElementType(), Align);
           }
         }
 
@@ -1111,10 +1111,12 @@ Address CodeGenFunction::EmitPointerWithAlignment(const Expr *E,
                                       CodeGenFunction::CFITCK_UnrelatedCast,
                                       CE->getBeginLoc());
         }
-        return CE->getCastKind() != CK_AddressSpaceConversion
-                   ? Builder.CreateBitCast(Addr, ConvertType(E->getType()))
-                   : Builder.CreateAddrSpaceCast(Addr,
-                                                 ConvertType(E->getType()));
+
+        if (CE->getCastKind() == CK_AddressSpaceConversion)
+         return Builder.CreateAddrSpaceCast(Addr, ConvertType(E->getType()));
+
+        llvm::Type *ElemTy = ConvertTypeForMem(E->getType()->getPointeeType());
+        return Builder.CreateElementBitCast(Addr, ElemTy);
       }
       break;
 
@@ -1160,7 +1162,8 @@ Address CodeGenFunction::EmitPointerWithAlignment(const Expr *E,
   // Otherwise, use the alignment of the type.
   CharUnits Align =
       CGM.getNaturalPointeeTypeAlignment(E->getType(), BaseInfo, TBAAInfo);
-  return Address(EmitScalarExpr(E), Align);
+  llvm::Type *ElemTy = ConvertTypeForMem(E->getType()->getPointeeType());
+  return Address(EmitScalarExpr(E), ElemTy, Align);
 }
 
 llvm::Value *CodeGenFunction::EmitNonNullRValueCheck(RValue RV, QualType T) {
@@ -1306,7 +1309,8 @@ LValue CodeGenFunction::EmitLValue(const Expr *E) {
     const ConstantExpr *CE = cast<ConstantExpr>(E);
     if (llvm::Value *Result = ConstantEmitter(*this).tryEmitConstantExpr(CE)) {
       QualType RetType = cast<CallExpr>(CE->getSubExpr()->IgnoreImplicit())
-                             ->getCallReturnType(getContext());
+                             ->getCallReturnType(getContext())
+                             ->getPointeeType();
       return MakeNaturalAlignAddrLValue(Result, RetType);
     }
     return EmitLValue(cast<ConstantExpr>(E)->getSubExpr());
@@ -1342,10 +1346,11 @@ LValue CodeGenFunction::EmitLValue(const Expr *E) {
     if (LV.isSimple()) {
       // Defend against branches out of gnu statement expressions surrounded by
       // cleanups.
-      llvm::Value *V = LV.getPointer(*this);
+      Address Addr = LV.getAddress(*this);
+      llvm::Value *V = Addr.getPointer();
       Scope.ForceCleanup({&V});
-      return LValue::MakeAddr(Address(V, LV.getAlignment()), LV.getType(),
-                              getContext(), LV.getBaseInfo(), LV.getTBAAInfo());
+      return LValue::MakeAddr(Addr.withPointer(V), LV.getType(), getContext(),
+                              LV.getBaseInfo(), LV.getTBAAInfo());
     }
     // FIXME: Is it possible to create an ExprWithCleanups that produces a
     // bitfield lvalue or some other non-simple lvalue?
@@ -1777,16 +1782,14 @@ llvm::Value *CodeGenFunction::EmitFromMemory(llvm::Value *Value, QualType Ty) {
 // MatrixType), if it points to a array (the memory type of MatrixType).
 static Address MaybeConvertMatrixAddress(Address Addr, CodeGenFunction &CGF,
                                          bool IsVector = true) {
-  auto *ArrayTy = dyn_cast<llvm::ArrayType>(
-      cast<llvm::PointerType>(Addr.getPointer()->getType())->getElementType());
+  auto *ArrayTy = dyn_cast<llvm::ArrayType>(Addr.getElementType());
   if (ArrayTy && IsVector) {
     auto *VectorTy = llvm::FixedVectorType::get(ArrayTy->getElementType(),
                                                 ArrayTy->getNumElements());
 
     return Address(CGF.Builder.CreateElementBitCast(Addr, VectorTy));
   }
-  auto *VectorTy = dyn_cast<llvm::VectorType>(
-      cast<llvm::PointerType>(Addr.getPointer()->getType())->getElementType());
+  auto *VectorTy = dyn_cast<llvm::VectorType>(Addr.getElementType());
   if (VectorTy && !IsVector) {
     auto *ArrayTy = llvm::ArrayType::get(
         VectorTy->getElementType(),
@@ -2475,10 +2478,11 @@ CodeGenFunction::EmitLoadOfReference(LValue RefLVal,
       Builder.CreateLoad(RefLVal.getAddress(*this), RefLVal.isVolatile());
   CGM.DecorateInstructionWithTBAA(Load, RefLVal.getTBAAInfo());
 
+  QualType PointeeType = RefLVal.getType()->getPointeeType();
   CharUnits Align = CGM.getNaturalTypeAlignment(
-      RefLVal.getType()->getPointeeType(), PointeeBaseInfo, PointeeTBAAInfo,
+      PointeeType, PointeeBaseInfo, PointeeTBAAInfo,
       /* forPointeeType= */ true);
-  return Address(Load, Align);
+  return Address(Load, ConvertTypeForMem(PointeeType), Align);
 }
 
 LValue CodeGenFunction::EmitLoadOfReferenceLValue(LValue RefLVal) {
@@ -2528,7 +2532,7 @@ static LValue EmitGlobalVarDeclLValue(CodeGenFunction &CGF,
   llvm::Type *RealVarTy = CGF.getTypes().ConvertTypeForMem(VD->getType());
   V = EmitBitCastOfLValueToProperType(CGF, V, RealVarTy);
   CharUnits Alignment = CGF.getContext().getDeclAlign(VD);
-  Address Addr(V, Alignment);
+  Address Addr(V, RealVarTy, Alignment);
   // Emit reference to the private copy of the variable if it is an OpenMP
   // threadprivate variable.
   if (CGF.getLangOpts().OpenMP && !CGF.getLangOpts().OpenMPSimd &&
@@ -2610,7 +2614,7 @@ static LValue EmitGlobalNamedRegister(const VarDecl *VD, CodeGenModule &CGM) {
 
   llvm::Value *Ptr =
     llvm::MetadataAsValue::get(CGM.getLLVMContext(), M->getOperand(0));
-  return LValue::MakeGlobalReg(Address(Ptr, Alignment), VD->getType());
+  return LValue::MakeGlobalReg(Ptr, Alignment, VD->getType());
 }
 
 /// Determine whether we can emit a reference to \p VD from the current
@@ -2706,7 +2710,7 @@ LValue CodeGenFunction::EmitDeclRefLValue(const DeclRefExpr *E) {
                                         /* BaseInfo= */ nullptr,
                                         /* TBAAInfo= */ nullptr,
                                         /* forPointeeType= */ true);
-        Addr = Address(Val, Alignment);
+        Addr = Address(Val, ConvertTypeForMem(E->getType()), Alignment);
       }
       return MakeAddrLValue(Addr, T, AlignmentSource::Decl);
     }
@@ -2783,9 +2787,10 @@ LValue CodeGenFunction::EmitDeclRefLValue(const DeclRefExpr *E) {
     // Otherwise, it might be static local we haven't emitted yet for
     // some reason; most likely, because it's in an outer function.
     } else if (VD->isStaticLocal()) {
-      addr = Address(CGM.getOrCreateStaticVarDecl(
-          *VD, CGM.getLLVMLinkageVarDefinition(VD, /*IsConstant=*/false)),
-                     getContext().getDeclAlign(VD));
+      llvm::Constant *var = CGM.getOrCreateStaticVarDecl(
+          *VD, CGM.getLLVMLinkageVarDefinition(VD, /*IsConstant=*/false));
+      addr = Address(
+          var, ConvertTypeForMem(VD->getType()), getContext().getDeclAlign(VD));
 
     // No other cases for now.
     } else {
@@ -3586,7 +3591,7 @@ static llvm::Value *emitArraySubscriptGEP(CodeGenFunction &CGF,
                                           SourceLocation loc,
                                     const llvm::Twine &name = "arrayidx") {
   if (inbounds) {
-    return CGF.EmitCheckedInBoundsGEP(ptr, indices, signedIndices,
+    return CGF.EmitCheckedInBoundsGEP(elemType, ptr, indices, signedIndices,
                                       CodeGenFunction::NotSubtraction, loc,
                                       name);
   } else {
@@ -3698,7 +3703,7 @@ static Address emitArraySubscriptGEP(CodeGenFunction &CGF, Address addr,
                                                         idx, DbgInfo);
   }
 
-  return Address(eltPtr, eltAlign);
+  return Address(eltPtr, CGF.ConvertTypeForMem(eltType), eltAlign);
 }
 
 LValue CodeGenFunction::EmitArraySubscriptExpr(const ArraySubscriptExpr *E,
@@ -4380,8 +4385,7 @@ LValue CodeGenFunction::EmitLValueForField(LValue base,
         hasAnyVptr(FieldType, getContext()))
       // Because unions can easily skip invariant.barriers, we need to add
       // a barrier every time CXXRecord field with vptr is referenced.
-      addr = Address(Builder.CreateLaunderInvariantGroup(addr.getPointer()),
-                     addr.getAlignment());
+      addr = Builder.CreateLaunderInvariantGroup(addr);
 
     if (IsInPreservedAIRegion ||
         (getDebugInfo() && rec->hasAttr<BPFPreserveAccessIndexAttr>())) {
@@ -4539,10 +4543,10 @@ EmitConditionalOperatorLValue(const AbstractConditionalOperator *expr) {
       // because it can't be used.
       if (auto *ThrowExpr = dyn_cast<CXXThrowExpr>(live->IgnoreParens())) {
         EmitCXXThrowExpr(ThrowExpr);
-        llvm::Type *Ty =
-            llvm::PointerType::getUnqual(ConvertType(dead->getType()));
+        llvm::Type *ElemTy = ConvertType(dead->getType());
+        llvm::Type *Ty = llvm::PointerType::getUnqual(ElemTy);
         return MakeAddrLValue(
-            Address(llvm::UndefValue::get(Ty), CharUnits::One()),
+            Address(llvm::UndefValue::get(Ty), ElemTy, CharUnits::One()),
             dead->getType());
       }
       return EmitLValue(live);
@@ -4584,11 +4588,13 @@ EmitConditionalOperatorLValue(const AbstractConditionalOperator *expr) {
   EmitBlock(contBlock);
 
   if (lhs && rhs) {
-    llvm::PHINode *phi =
-        Builder.CreatePHI(lhs->getPointer(*this)->getType(), 2, "cond-lvalue");
-    phi->addIncoming(lhs->getPointer(*this), lhsBlock);
-    phi->addIncoming(rhs->getPointer(*this), rhsBlock);
-    Address result(phi, std::min(lhs->getAlignment(), rhs->getAlignment()));
+    Address lhsAddr = lhs->getAddress(*this);
+    Address rhsAddr = rhs->getAddress(*this);
+    llvm::PHINode *phi = Builder.CreatePHI(lhsAddr.getType(), 2, "cond-lvalue");
+    phi->addIncoming(lhsAddr.getPointer(), lhsBlock);
+    phi->addIncoming(rhsAddr.getPointer(), rhsBlock);
+    Address result(phi, lhsAddr.getElementType(),
+                   std::min(lhsAddr.getAlignment(), rhsAddr.getAlignment()));
     AlignmentSource alignSource =
       std::max(lhs->getBaseInfo().getAlignmentSource(),
                rhs->getBaseInfo().getAlignmentSource());
diff --git a/clang/lib/CodeGen/CGExprAgg.cpp b/clang/lib/CodeGen/CGExprAgg.cpp
index 5b56a587fa5f..3b996b89a1d7 100644
--- a/clang/lib/CodeGen/CGExprAgg.cpp
+++ b/clang/lib/CodeGen/CGExprAgg.cpp
@@ -301,7 +301,7 @@ void AggExprEmitter::withReturnValueSlot(
   if (!UseTemp)
     return;
 
-  assert(Dest.getPointer() != Src.getAggregatePointer());
+  assert(Dest.isIgnored() || Dest.getPointer() != Src.getAggregatePointer());
   EmitFinalDestCopy(E->getType(), Src);
 
   if (!RequiresDestruction && LifetimeStartInst) {
@@ -493,7 +493,7 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr, llvm::ArrayType *AType,
   CharUnits elementSize = CGF.getContext().getTypeSizeInChars(elementType);
   CharUnits elementAlign =
     DestPtr.getAlignment().alignmentOfArrayElement(elementSize);
-  llvm::Type *llvmElementType = begin->getType()->getPointerElementType();
+  llvm::Type *llvmElementType = CGF.ConvertTypeForMem(elementType);
 
   // Consider initializing the array by copying from a global. For this to be
   // more efficient than per-element initialization, the size of the elements
@@ -513,7 +513,8 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr, llvm::ArrayType *AType,
       Emitter.finalize(GV);
       CharUnits Align = CGM.getContext().getTypeAlignInChars(ArrayQTy);
       GV->setAlignment(Align.getAsAlign());
-      EmitFinalDestCopy(ArrayQTy, CGF.MakeAddrLValue(GV, ArrayQTy, Align));
+      Address GVAddr(GV, GV->getValueType(), Align);
+      EmitFinalDestCopy(ArrayQTy, CGF.MakeAddrLValue(GVAddr, ArrayQTy));
       return;
     }
   }
@@ -565,8 +566,8 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr, llvm::ArrayType *AType,
       if (endOfInit.isValid()) Builder.CreateStore(element, endOfInit);
     }
 
-    LValue elementLV =
-      CGF.MakeAddrLValue(Address(element, elementAlign), elementType);
+    LValue elementLV = CGF.MakeAddrLValue(
+        Address(element, llvmElementType, elementAlign), elementType);
     EmitInitializationToLValue(E->getInit(i), elementLV);
   }
 
diff --git a/clang/lib/CodeGen/CGExprCXX.cpp b/clang/lib/CodeGen/CGExprCXX.cpp
index cc838bf38c6c..0571c498c377 100644
--- a/clang/lib/CodeGen/CGExprCXX.cpp
+++ b/clang/lib/CodeGen/CGExprCXX.cpp
@@ -1052,13 +1052,8 @@ void CodeGenFunction::EmitNewArrayInitializer(
       InitListElements =
           cast<ConstantArrayType>(ILE->getType()->getAsArrayTypeUnsafe())
               ->getSize().getZExtValue();
-      CurPtr =
-          Address(Builder.CreateInBoundsGEP(CurPtr.getElementType(),
-                                            CurPtr.getPointer(),
-                                            Builder.getSize(InitListElements),
-                                            "string.init.end"),
-                  CurPtr.getAlignment().alignmentAtOffset(InitListElements *
-                                                          ElementSize));
+      CurPtr = Builder.CreateConstInBoundsGEP(
+          CurPtr, InitListElements, "string.init.end");
 
       // Zero out the rest, if any remain.
       llvm::ConstantInt *ConstNum = dyn_cast<llvm::ConstantInt>(NumElements);
@@ -1135,7 +1130,7 @@ void CodeGenFunction::EmitNewArrayInitializer(
     }
 
     // Switch back to initializing one base element at a time.
-    CurPtr = Builder.CreateBitCast(CurPtr, BeginPtr.getType());
+    CurPtr = Builder.CreateElementBitCast(CurPtr, BeginPtr.getElementType());
   }
 
   // If all elements have already been initialized, skip any further
@@ -1594,7 +1589,7 @@ llvm::Value *CodeGenFunction::EmitCXXNewExpr(const CXXNewExpr *E) {
     // In these cases, discard the computed alignment and use the
     // formal alignment of the allocated type.
     if (BaseInfo.getAlignmentSource() != AlignmentSource::Decl)
-      allocation = Address(allocation.getPointer(), allocAlign);
+      allocation = allocation.withAlignment(allocAlign);
 
     // Set up allocatorArgs for the call to operator delete if it's not
     // the reserved global operator.
@@ -1664,7 +1659,7 @@ llvm::Value *CodeGenFunction::EmitCXXNewExpr(const CXXNewExpr *E) {
           allocationAlign, getContext().toCharUnitsFromBits(AllocatorAlign));
     }
 
-    allocation = Address(RV.getScalarVal(), allocationAlign);
+    allocation = Address(RV.getScalarVal(), Int8Ty, allocationAlign);
   }
 
   // Emit a null check on the allocation result if the allocation
@@ -1725,8 +1720,7 @@ llvm::Value *CodeGenFunction::EmitCXXNewExpr(const CXXNewExpr *E) {
   // of optimization level.
   if (CGM.getCodeGenOpts().StrictVTablePointers &&
       allocator->isReservedGlobalPlacementOperator())
-    result = Address(Builder.CreateLaunderInvariantGroup(result.getPointer()),
-                     result.getAlignment());
+    result = Builder.CreateLaunderInvariantGroup(result);
 
   // Emit sanitizer checks for pointer value now, so that in the case of an
   // array it was checked only once and not at each constructor call. We may
diff --git a/clang/lib/CodeGen/CGExprConstant.cpp b/clang/lib/CodeGen/CGExprConstant.cpp
index ff900ed077e6..cf1f2e0eab92 100644
--- a/clang/lib/CodeGen/CGExprConstant.cpp
+++ b/clang/lib/CodeGen/CGExprConstant.cpp
@@ -899,7 +899,7 @@ static ConstantAddress tryEmitGlobalCompoundLiteral(CodeGenModule &CGM,
   CharUnits Align = CGM.getContext().getTypeAlignInChars(E->getType());
   if (llvm::GlobalVariable *Addr =
           CGM.getAddrOfConstantCompoundLiteralIfEmitted(E))
-    return ConstantAddress(Addr, Align);
+    return ConstantAddress(Addr, Addr->getValueType(), Align);
 
   LangAS addressSpace = E->getType().getAddressSpace();
 
@@ -921,7 +921,7 @@ static ConstantAddress tryEmitGlobalCompoundLiteral(CodeGenModule &CGM,
   emitter.finalize(GV);
   GV->setAlignment(Align.getAsAlign());
   CGM.setAddrOfConstantCompoundLiteral(E, GV);
-  return ConstantAddress(GV, Align);
+  return ConstantAddress(GV, GV->getValueType(), Align);
 }
 
 static llvm::Constant *
@@ -1988,6 +1988,9 @@ ConstantLValueEmitter::VisitAddrLabelExpr(const AddrLabelExpr *E) {
 ConstantLValue
 ConstantLValueEmitter::VisitCallExpr(const CallExpr *E) {
   unsigned builtin = E->getBuiltinCallee();
+  if (builtin == Builtin::BI__builtin_function_start)
+    return CGM.GetFunctionStart(
+        E->getArg(0)->getAsBuiltinConstantDeclRef(CGM.getContext()));
   if (builtin != Builtin::BI__builtin___CFStringMakeConstantString &&
       builtin != Builtin::BI__builtin___NSStringMakeConstantString)
     return nullptr;
diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
index ae9434f96529..e32462eb635c 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -1240,7 +1240,18 @@ Value *ScalarExprEmitter::EmitScalarCast(Value *Src, QualType SrcType,
 
   if (isa<llvm::IntegerType>(DstElementTy)) {
     assert(SrcElementTy->isFloatingPointTy() && "Unknown real conversion");
-    if (DstElementType->isSignedIntegerOrEnumerationType())
+    bool IsSigned = DstElementType->isSignedIntegerOrEnumerationType();
+
+    // If we can't recognize overflow as undefined behavior, assume that
+    // overflow saturates. This protects against normal optimizations if we are
+    // compiling with non-standard FP semantics.
+    if (!CGF.CGM.getCodeGenOpts().StrictFloatCastOverflow) {
+      llvm::Intrinsic::ID IID =
+          IsSigned ? llvm::Intrinsic::fptosi_sat : llvm::Intrinsic::fptoui_sat;
+      return Builder.CreateCall(CGF.CGM.getIntrinsic(IID, {DstTy, SrcTy}), Src);
+    }
+
+    if (IsSigned)
       return Builder.CreateFPToSI(Src, DstTy, "conv");
     return Builder.CreateFPToUI(Src, DstTy, "conv");
   }
@@ -2631,12 +2642,12 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
           = CGF.getContext().getAsVariableArrayType(type)) {
       llvm::Value *numElts = CGF.getVLASize(vla).NumElts;
       if (!isInc) numElts = Builder.CreateNSWNeg(numElts, "vla.negsize");
+      llvm::Type *elemTy = value->getType()->getPointerElementType();
       if (CGF.getLangOpts().isSignedOverflowDefined())
-        value = Builder.CreateGEP(value->getType()->getPointerElementType(),
-                                  value, numElts, "vla.inc");
+        value = Builder.CreateGEP(elemTy, value, numElts, "vla.inc");
       else
         value = CGF.EmitCheckedInBoundsGEP(
-            value, numElts, /*SignedIndices=*/false, isSubtraction,
+            elemTy, value, numElts, /*SignedIndices=*/false, isSubtraction,
             E->getExprLoc(), "vla.inc");
 
     // Arithmetic on function pointers (!) is just +-1.
@@ -2647,7 +2658,8 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
       if (CGF.getLangOpts().isSignedOverflowDefined())
         value = Builder.CreateGEP(CGF.Int8Ty, value, amt, "incdec.funcptr");
       else
-        value = CGF.EmitCheckedInBoundsGEP(value, amt, /*SignedIndices=*/false,
+        value = CGF.EmitCheckedInBoundsGEP(CGF.Int8Ty, value, amt,
+                                           /*SignedIndices=*/false,
                                            isSubtraction, E->getExprLoc(),
                                            "incdec.funcptr");
       value = Builder.CreateBitCast(value, input->getType());
@@ -2655,13 +2667,13 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
     // For everything else, we can just do a simple increment.
     } else {
       llvm::Value *amt = Builder.getInt32(amount);
+      llvm::Type *elemTy = CGF.ConvertTypeForMem(type);
       if (CGF.getLangOpts().isSignedOverflowDefined())
-        value = Builder.CreateGEP(value->getType()->getPointerElementType(),
-                                  value, amt, "incdec.ptr");
+        value = Builder.CreateGEP(elemTy, value, amt, "incdec.ptr");
       else
-        value = CGF.EmitCheckedInBoundsGEP(value, amt, /*SignedIndices=*/false,
-                                           isSubtraction, E->getExprLoc(),
-                                           "incdec.ptr");
+        value = CGF.EmitCheckedInBoundsGEP(
+            elemTy, value, amt, /*SignedIndices=*/false, isSubtraction,
+            E->getExprLoc(), "incdec.ptr");
     }
 
   // Vector increment/decrement.
@@ -2771,9 +2783,9 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
     if (CGF.getLangOpts().isSignedOverflowDefined())
       value = Builder.CreateGEP(CGF.Int8Ty, value, sizeValue, "incdec.objptr");
     else
-      value = CGF.EmitCheckedInBoundsGEP(value, sizeValue,
-                                         /*SignedIndices=*/false, isSubtraction,
-                                         E->getExprLoc(), "incdec.objptr");
+      value = CGF.EmitCheckedInBoundsGEP(
+          CGF.Int8Ty, value, sizeValue, /*SignedIndices=*/false, isSubtraction,
+          E->getExprLoc(), "incdec.objptr");
     value = Builder.CreateBitCast(value, input->getType());
   }
 
@@ -3508,16 +3520,15 @@ static Value *emitPointerArithmetic(CodeGenFunction &CGF,
     // GEP indexes are signed, and scaling an index isn't permitted to
     // signed-overflow, so we use the same semantics for our explicit
     // multiply.  We suppress this if overflow is not undefined behavior.
+    llvm::Type *elemTy = pointer->getType()->getPointerElementType();
     if (CGF.getLangOpts().isSignedOverflowDefined()) {
       index = CGF.Builder.CreateMul(index, numElements, "vla.index");
-      pointer = CGF.Builder.CreateGEP(
-          pointer->getType()->getPointerElementType(), pointer, index,
-          "add.ptr");
+      pointer = CGF.Builder.CreateGEP(elemTy, pointer, index, "add.ptr");
     } else {
       index = CGF.Builder.CreateNSWMul(index, numElements, "vla.index");
-      pointer =
-          CGF.EmitCheckedInBoundsGEP(pointer, index, isSigned, isSubtraction,
-                                     op.E->getExprLoc(), "add.ptr");
+      pointer = CGF.EmitCheckedInBoundsGEP(
+          elemTy, pointer, index, isSigned, isSubtraction, op.E->getExprLoc(),
+          "add.ptr");
     }
     return pointer;
   }
@@ -3531,12 +3542,13 @@ static Value *emitPointerArithmetic(CodeGenFunction &CGF,
     return CGF.Builder.CreateBitCast(result, pointer->getType());
   }
 
+  llvm::Type *elemTy = CGF.ConvertTypeForMem(elementType);
   if (CGF.getLangOpts().isSignedOverflowDefined())
-    return CGF.Builder.CreateGEP(
-        pointer->getType()->getPointerElementType(), pointer, index, "add.ptr");
+    return CGF.Builder.CreateGEP(elemTy, pointer, index, "add.ptr");
 
-  return CGF.EmitCheckedInBoundsGEP(pointer, index, isSigned, isSubtraction,
-                                    op.E->getExprLoc(), "add.ptr");
+  return CGF.EmitCheckedInBoundsGEP(
+      elemTy, pointer, index, isSigned, isSubtraction, op.E->getExprLoc(),
+      "add.ptr");
 }
 
 // Construct an fmuladd intrinsic to represent a fused mul-add of MulOp and
@@ -5057,12 +5069,12 @@ static GEPOffsetAndOverflow EmitGEPOffsetInBytes(Value *BasePtr, Value *GEPVal,
 }
 
 Value *
-CodeGenFunction::EmitCheckedInBoundsGEP(Value *Ptr, ArrayRef<Value *> IdxList,
+CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type *ElemTy, Value *Ptr,
+                                        ArrayRef<Value *> IdxList,
                                         bool SignedIndices, bool IsSubtraction,
                                         SourceLocation Loc, const Twine &Name) {
   llvm::Type *PtrTy = Ptr->getType();
-  Value *GEPVal = Builder.CreateInBoundsGEP(
-      PtrTy->getPointerElementType(), Ptr, IdxList, Name);
+  Value *GEPVal = Builder.CreateInBoundsGEP(ElemTy, Ptr, IdxList, Name);
 
   // If the pointer overflow sanitizer isn't enabled, do nothing.
   if (!SanOpts.has(SanitizerKind::PointerOverflow))
diff --git a/clang/lib/CodeGen/CGNonTrivialStruct.cpp b/clang/lib/CodeGen/CGNonTrivialStruct.cpp
index ad505fc5a0d4..e3b0e069b830 100644
--- a/clang/lib/CodeGen/CGNonTrivialStruct.cpp
+++ b/clang/lib/CodeGen/CGNonTrivialStruct.cpp
@@ -366,11 +366,11 @@ template <class Derived> struct GenFuncBase {
         llvm::ConstantInt::get(NumElts->getType(), BaseEltSize);
     llvm::Value *SizeInBytes =
         CGF.Builder.CreateNUWMul(BaseEltSizeVal, NumElts);
-    Address BC = CGF.Builder.CreateBitCast(DstAddr, CGF.CGM.Int8PtrTy);
+    Address BC = CGF.Builder.CreateElementBitCast(DstAddr, CGF.CGM.Int8Ty);
     llvm::Value *DstArrayEnd =
         CGF.Builder.CreateInBoundsGEP(CGF.Int8Ty, BC.getPointer(), SizeInBytes);
-    DstArrayEnd = CGF.Builder.CreateBitCast(DstArrayEnd, CGF.CGM.Int8PtrPtrTy,
-                                            "dstarray.end");
+    DstArrayEnd = CGF.Builder.CreateBitCast(
+        DstArrayEnd, CGF.CGM.Int8PtrPtrTy, "dstarray.end");
     llvm::BasicBlock *PreheaderBB = CGF.Builder.GetInsertBlock();
 
     // Create the header block and insert the phi instructions.
@@ -426,9 +426,9 @@ template <class Derived> struct GenFuncBase {
     assert(Addr.isValid() && "invalid address");
     if (Offset.getQuantity() == 0)
       return Addr;
-    Addr = CGF->Builder.CreateBitCast(Addr, CGF->CGM.Int8PtrTy);
+    Addr = CGF->Builder.CreateElementBitCast(Addr, CGF->CGM.Int8Ty);
     Addr = CGF->Builder.CreateConstInBoundsGEP(Addr, Offset.getQuantity());
-    return CGF->Builder.CreateBitCast(Addr, CGF->CGM.Int8PtrPtrTy);
+    return CGF->Builder.CreateElementBitCast(Addr, CGF->CGM.Int8PtrTy);
   }
 
   Address getAddrWithOffset(Address Addr, CharUnits StructFieldOffset,
@@ -491,9 +491,8 @@ template <class Derived> struct GenFuncBase {
 
     for (unsigned I = 0; I < N; ++I) {
       Alignments[I] = Addrs[I].getAlignment();
-      Ptrs[I] =
-          CallerCGF.Builder.CreateBitCast(Addrs[I], CallerCGF.CGM.Int8PtrPtrTy)
-              .getPointer();
+      Ptrs[I] = CallerCGF.Builder.CreateElementBitCast(
+          Addrs[I], CallerCGF.CGM.Int8PtrTy).getPointer();
     }
 
     if (llvm::Function *F =
@@ -554,19 +553,21 @@ struct GenBinaryFunc : CopyStructVisitor<Derived, IsMove>,
         return;
 
       QualType RT = QualType(FD->getParent()->getTypeForDecl(), 0);
-      llvm::PointerType *PtrTy = this->CGF->ConvertType(RT)->getPointerTo();
+      llvm::Type *Ty = this->CGF->ConvertType(RT);
       Address DstAddr = this->getAddrWithOffset(Addrs[DstIdx], Offset);
       LValue DstBase = this->CGF->MakeAddrLValue(
-          this->CGF->Builder.CreateBitCast(DstAddr, PtrTy), FT);
+          this->CGF->Builder.CreateElementBitCast(DstAddr, Ty), FT);
       DstLV = this->CGF->EmitLValueForField(DstBase, FD);
       Address SrcAddr = this->getAddrWithOffset(Addrs[SrcIdx], Offset);
       LValue SrcBase = this->CGF->MakeAddrLValue(
-          this->CGF->Builder.CreateBitCast(SrcAddr, PtrTy), FT);
+          this->CGF->Builder.CreateElementBitCast(SrcAddr, Ty), FT);
       SrcLV = this->CGF->EmitLValueForField(SrcBase, FD);
     } else {
-      llvm::PointerType *Ty = this->CGF->ConvertTypeForMem(FT)->getPointerTo();
-      Address DstAddr = this->CGF->Builder.CreateBitCast(Addrs[DstIdx], Ty);
-      Address SrcAddr = this->CGF->Builder.CreateBitCast(Addrs[SrcIdx], Ty);
+      llvm::Type *Ty = this->CGF->ConvertTypeForMem(FT);
+      Address DstAddr =
+          this->CGF->Builder.CreateElementBitCast(Addrs[DstIdx], Ty);
+      Address SrcAddr =
+          this->CGF->Builder.CreateElementBitCast(Addrs[SrcIdx], Ty);
       DstLV = this->CGF->MakeAddrLValue(DstAddr, FT);
       SrcLV = this->CGF->MakeAddrLValue(SrcAddr, FT);
     }
@@ -817,7 +818,7 @@ void CodeGenFunction::destroyNonTrivialCStruct(CodeGenFunction &CGF,
 void CodeGenFunction::defaultInitNonTrivialCStructVar(LValue Dst) {
   GenDefaultInitialize Gen(getContext());
   Address DstPtr =
-      Builder.CreateBitCast(Dst.getAddress(*this), CGM.Int8PtrPtrTy);
+      Builder.CreateElementBitCast(Dst.getAddress(*this), CGM.Int8PtrTy);
   Gen.setCGF(this);
   QualType QT = Dst.getType();
   QT = Dst.isVolatile() ? QT.withVolatile() : QT;
@@ -830,7 +831,7 @@ static void callSpecialFunction(G &&Gen, StringRef FuncName, QualType QT,
                                 std::array<Address, N> Addrs) {
   auto SetArtificialLoc = ApplyDebugLocation::CreateArtificial(CGF);
   for (unsigned I = 0; I < N; ++I)
-    Addrs[I] = CGF.Builder.CreateBitCast(Addrs[I], CGF.CGM.Int8PtrPtrTy);
+    Addrs[I] = CGF.Builder.CreateElementBitCast(Addrs[I], CGF.CGM.Int8PtrTy);
   QT = IsVolatile ? QT.withVolatile() : QT;
   Gen.callFunc(FuncName, QT, Addrs, CGF);
 }
diff --git a/clang/lib/CodeGen/CGObjC.cpp b/clang/lib/CodeGen/CGObjC.cpp
index ac26f0d4232c..b5bcf157036d 100644
--- a/clang/lib/CodeGen/CGObjC.cpp
+++ b/clang/lib/CodeGen/CGObjC.cpp
@@ -3915,8 +3915,8 @@ static llvm::Value *emitIsPlatformVersionAtLeast(CodeGenFunction &CGF,
     Args.push_back(
         llvm::ConstantInt::get(CGM.Int32Ty, getBaseMachOPlatformID(TT)));
     Args.push_back(llvm::ConstantInt::get(CGM.Int32Ty, Version.getMajor()));
-    Args.push_back(llvm::ConstantInt::get(CGM.Int32Ty, Min ? *Min : 0));
-    Args.push_back(llvm::ConstantInt::get(CGM.Int32Ty, SMin ? *SMin : 0));
+    Args.push_back(llvm::ConstantInt::get(CGM.Int32Ty, Min.getValueOr(0)));
+    Args.push_back(llvm::ConstantInt::get(CGM.Int32Ty, SMin.getValueOr(0)));
   };
 
   assert(!Version.empty() && "unexpected empty version");
@@ -3952,8 +3952,8 @@ CodeGenFunction::EmitBuiltinAvailable(const VersionTuple &Version) {
   Optional<unsigned> Min = Version.getMinor(), SMin = Version.getSubminor();
   llvm::Value *Args[] = {
       llvm::ConstantInt::get(CGM.Int32Ty, Version.getMajor()),
-      llvm::ConstantInt::get(CGM.Int32Ty, Min ? *Min : 0),
-      llvm::ConstantInt::get(CGM.Int32Ty, SMin ? *SMin : 0),
+      llvm::ConstantInt::get(CGM.Int32Ty, Min.getValueOr(0)),
+      llvm::ConstantInt::get(CGM.Int32Ty, SMin.getValueOr(0))
   };
 
   llvm::Value *CallRes =
diff --git a/clang/lib/CodeGen/CGObjCGNU.cpp b/clang/lib/CodeGen/CGObjCGNU.cpp
index e016644150b4..b2bf60d2c0fc 100644
--- a/clang/lib/CodeGen/CGObjCGNU.cpp
+++ b/clang/lib/CodeGen/CGObjCGNU.cpp
@@ -978,7 +978,9 @@ class CGObjCGNUstep2 : public CGObjCGNUstep {
     // Look for an existing one
     llvm::StringMap<llvm::Constant*>::iterator old = ObjCStrings.find(Str);
     if (old != ObjCStrings.end())
-      return ConstantAddress(old->getValue(), Align);
+      return ConstantAddress(
+          old->getValue(), old->getValue()->getType()->getPointerElementType(),
+          Align);
 
     bool isNonASCII = SL->containsNonAscii();
 
@@ -1000,7 +1002,7 @@ class CGObjCGNUstep2 : public CGObjCGNUstep {
       auto *ObjCStr = llvm::ConstantExpr::getIntToPtr(
           llvm::ConstantInt::get(Int64Ty, str), IdTy);
       ObjCStrings[Str] = ObjCStr;
-      return ConstantAddress(ObjCStr, Align);
+      return ConstantAddress(ObjCStr, IdTy->getPointerElementType(), Align);
     }
 
     StringRef StringClass = CGM.getLangOpts().ObjCConstantStringClass;
@@ -1114,7 +1116,7 @@ class CGObjCGNUstep2 : public CGObjCGNUstep {
     llvm::Constant *ObjCStr = llvm::ConstantExpr::getBitCast(ObjCStrGV, IdTy);
     ObjCStrings[Str] = ObjCStr;
     ConstantStrings.push_back(ObjCStr);
-    return ConstantAddress(ObjCStr, Align);
+    return ConstantAddress(ObjCStr, IdTy->getPointerElementType(), Align);
   }
 
   void PushProperty(ConstantArrayBuilder &PropertiesArray,
@@ -2476,7 +2478,7 @@ ConstantAddress CGObjCGNU::GenerateConstantString(const StringLiteral *SL) {
   // Look for an existing one
   llvm::StringMap<llvm::Constant*>::iterator old = ObjCStrings.find(Str);
   if (old != ObjCStrings.end())
-    return ConstantAddress(old->getValue(), Align);
+    return ConstantAddress(old->getValue(), Int8Ty, Align);
 
   StringRef StringClass = CGM.getLangOpts().ObjCConstantStringClass;
 
@@ -2503,7 +2505,7 @@ ConstantAddress CGObjCGNU::GenerateConstantString(const StringLiteral *SL) {
   ObjCStr = llvm::ConstantExpr::getBitCast(ObjCStr, PtrToInt8Ty);
   ObjCStrings[Str] = ObjCStr;
   ConstantStrings.push_back(ObjCStr);
-  return ConstantAddress(ObjCStr, Align);
+  return ConstantAddress(ObjCStr, Int8Ty, Align);
 }
 
 ///Generates a message send where the super is the receiver.  This is a message
diff --git a/clang/lib/CodeGen/CGObjCMac.cpp b/clang/lib/CodeGen/CGObjCMac.cpp
index 5b925359ac25..425d1a793439 100644
--- a/clang/lib/CodeGen/CGObjCMac.cpp
+++ b/clang/lib/CodeGen/CGObjCMac.cpp
@@ -1983,7 +1983,8 @@ CGObjCCommonMac::GenerateConstantNSString(const StringLiteral *Literal) {
     GetConstantStringEntry(NSConstantStringMap, Literal, StringLength);
 
   if (auto *C = Entry.second)
-    return ConstantAddress(C, CharUnits::fromQuantity(C->getAlignment()));
+    return ConstantAddress(
+        C, C->getValueType(), CharUnits::fromQuantity(C->getAlignment()));
 
   // If we don't already have it, get _NSConstantStringClassReference.
   llvm::Constant *Class = getNSConstantStringClassRef();
@@ -2036,7 +2037,7 @@ CGObjCCommonMac::GenerateConstantNSString(const StringLiteral *Literal) {
                      : NSStringSection);
   Entry.second = GV;
 
-  return ConstantAddress(GV, Alignment);
+  return ConstantAddress(GV, GV->getValueType(), Alignment);
 }
 
 enum {
diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.cpp b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
index 75709b3c7e78..e35c15421520 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -15,6 +15,7 @@
 #include "CGCleanup.h"
 #include "CGRecordLayout.h"
 #include "CodeGenFunction.h"
+#include "TargetInfo.h"
 #include "clang/AST/APValue.h"
 #include "clang/AST/Attr.h"
 #include "clang/AST/Decl.h"
@@ -687,8 +688,6 @@ static void EmitOMPAggregateInit(CodeGenFunction &CGF, Address DestAddr,
   // Drill down to the base element type on both arrays.
   const ArrayType *ArrayTy = Type->getAsArrayTypeUnsafe();
   llvm::Value *NumElements = CGF.emitArrayLength(ArrayTy, ElementTy, DestAddr);
-  DestAddr =
-      CGF.Builder.CreateElementBitCast(DestAddr, DestAddr.getElementType());
   if (DRD)
     SrcAddr =
         CGF.Builder.CreateElementBitCast(SrcAddr, DestAddr.getElementType());
@@ -775,7 +774,7 @@ LValue ReductionCodeGen::emitSharedLValueUB(CodeGenFunction &CGF,
 }
 
 void ReductionCodeGen::emitAggregateInitialization(
-    CodeGenFunction &CGF, unsigned N, Address PrivateAddr, LValue SharedLVal,
+    CodeGenFunction &CGF, unsigned N, Address PrivateAddr, Address SharedAddr,
     const OMPDeclareReductionDecl *DRD) {
   // Emit VarDecl with copy init for arrays.
   // Get the address of the original variable captured in current
@@ -788,7 +787,7 @@ void ReductionCodeGen::emitAggregateInitialization(
                        EmitDeclareReductionInit,
                        EmitDeclareReductionInit ? ClausesData[N].ReductionOp
                                                 : PrivateVD->getInit(),
-                       DRD, SharedLVal.getAddress(CGF));
+                       DRD, SharedAddr);
 }
 
 ReductionCodeGen::ReductionCodeGen(ArrayRef<const Expr *> Shareds,
@@ -882,7 +881,7 @@ void ReductionCodeGen::emitAggregateType(CodeGenFunction &CGF, unsigned N,
 }
 
 void ReductionCodeGen::emitInitialization(
-    CodeGenFunction &CGF, unsigned N, Address PrivateAddr, LValue SharedLVal,
+    CodeGenFunction &CGF, unsigned N, Address PrivateAddr, Address SharedAddr,
     llvm::function_ref<bool(CodeGenFunction &)> DefaultInit) {
   assert(SharedAddresses.size() > N && "No variable was generated");
   const auto *PrivateVD =
@@ -892,21 +891,15 @@ void ReductionCodeGen::emitInitialization(
   QualType PrivateType = PrivateVD->getType();
   PrivateAddr = CGF.Builder.CreateElementBitCast(
       PrivateAddr, CGF.ConvertTypeForMem(PrivateType));
-  QualType SharedType = SharedAddresses[N].first.getType();
-  SharedLVal = CGF.MakeAddrLValue(
-      CGF.Builder.CreateElementBitCast(SharedLVal.getAddress(CGF),
-                                       CGF.ConvertTypeForMem(SharedType)),
-      SharedType, SharedAddresses[N].first.getBaseInfo(),
-      CGF.CGM.getTBAAInfoForSubobject(SharedAddresses[N].first, SharedType));
   if (CGF.getContext().getAsArrayType(PrivateVD->getType())) {
     if (DRD && DRD->getInitializer())
       (void)DefaultInit(CGF);
-    emitAggregateInitialization(CGF, N, PrivateAddr, SharedLVal, DRD);
+    emitAggregateInitialization(CGF, N, PrivateAddr, SharedAddr, DRD);
   } else if (DRD && (DRD->getInitializer() || !PrivateVD->hasInit())) {
     (void)DefaultInit(CGF);
+    QualType SharedType = SharedAddresses[N].first.getType();
     emitInitWithReductionInitializer(CGF, DRD, ClausesData[N].ReductionOp,
-                                     PrivateAddr, SharedLVal.getAddress(CGF),
-                                     SharedLVal.getType());
+                                     PrivateAddr, SharedAddr, SharedType);
   } else if (!DefaultInit(CGF) && PrivateVD->hasInit() &&
              !CGF.isTrivialInitializer(PrivateVD->getInit())) {
     CGF.EmitAnyExprToMem(PrivateVD->getInit(), PrivateAddr,
@@ -2016,12 +2009,13 @@ Address CGOpenMPRuntime::getAddrOfArtificialThreadPrivate(CodeGenFunction &CGF,
                                                           StringRef Name) {
   std::string Suffix = getName({"artificial", ""});
   llvm::Type *VarLVType = CGF.ConvertTypeForMem(VarType);
-  llvm::Value *GAddr =
+  llvm::GlobalVariable *GAddr =
       getOrCreateInternalVariable(VarLVType, Twine(Name).concat(Suffix));
   if (CGM.getLangOpts().OpenMP && CGM.getLangOpts().OpenMPUseTLS &&
       CGM.getTarget().isTLSSupported()) {
-    cast<llvm::GlobalVariable>(GAddr)->setThreadLocal(/*Val=*/true);
-    return Address(GAddr, CGM.getContext().getTypeAlignInChars(VarType));
+    GAddr->setThreadLocal(/*Val=*/true);
+    return Address(GAddr, GAddr->getValueType(),
+                   CGM.getContext().getTypeAlignInChars(VarType));
   }
   std::string CacheSuffix = getName({"cache", ""});
   llvm::Value *Args[] = {
@@ -2084,7 +2078,8 @@ void CGOpenMPRuntime::emitIfClause(CodeGenFunction &CGF, const Expr *Cond,
 void CGOpenMPRuntime::emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
                                        llvm::Function *OutlinedFn,
                                        ArrayRef<llvm::Value *> CapturedVars,
-                                       const Expr *IfCond) {
+                                       const Expr *IfCond,
+                                       llvm::Value *NumThreads) {
   if (!CGF.HaveInsertPoint())
     return;
   llvm::Value *RTLoc = emitUpdateLocation(CGF, Loc);
@@ -2175,7 +2170,7 @@ Address CGOpenMPRuntime::emitThreadIDAddress(CodeGenFunction &CGF,
   return ThreadIDTemp;
 }
 
-llvm::Constant *CGOpenMPRuntime::getOrCreateInternalVariable(
+llvm::GlobalVariable *CGOpenMPRuntime::getOrCreateInternalVariable(
     llvm::Type *Ty, const llvm::Twine &Name, unsigned AddressSpace) {
   SmallString<256> Buffer;
   llvm::raw_svector_ostream Out(Buffer);
@@ -2183,7 +2178,7 @@ llvm::Constant *CGOpenMPRuntime::getOrCreateInternalVariable(
   StringRef RuntimeName = Out.str();
   auto &Elem = *InternalVars.try_emplace(RuntimeName, nullptr).first;
   if (Elem.second) {
-    assert(Elem.second->getType()->getPointerElementType() == Ty &&
+    assert(Elem.second->getType()->isOpaqueOrPointeeTypeMatches(Ty) &&
            "OMP internal variable has different type than requested");
     return &*Elem.second;
   }
@@ -4498,10 +4493,7 @@ CGOpenMPRuntime::emitTaskInit(CodeGenFunction &CGF, SourceLocation Loc,
         std::tie(Addr, Size) = getPointerAndSize(CGF, E);
         llvm::Value *Idx = CGF.EmitLoadOfScalar(PosLVal, E->getExprLoc());
         LValue Base = CGF.MakeAddrLValue(
-            Address(CGF.Builder.CreateGEP(AffinitiesArray.getElementType(),
-                                          AffinitiesArray.getPointer(), Idx),
-                    AffinitiesArray.getAlignment()),
-            KmpTaskAffinityInfoTy);
+            CGF.Builder.CreateGEP(AffinitiesArray, Idx), KmpTaskAffinityInfoTy);
         // affs[i].base_addr = &<Affinities[i].second>;
         LValue BaseAddrLVal = CGF.EmitLValueForField(
             Base, *std::next(KmpAffinityInfoRD->field_begin(), BaseAddr));
@@ -4665,12 +4657,10 @@ CGOpenMPRuntime::getDepobjElements(CodeGenFunction &CGF, LValue DepobjLVal,
           Base.getAddress(CGF), CGF.ConvertTypeForMem(KmpDependInfoPtrTy));
   Base = CGF.MakeAddrLValue(Addr, KmpDependInfoTy, Base.getBaseInfo(),
                             Base.getTBAAInfo());
-  llvm::Value *DepObjAddr = CGF.Builder.CreateGEP(
-      Addr.getElementType(), Addr.getPointer(),
-      llvm::ConstantInt::get(CGF.IntPtrTy, -1, /*isSigned=*/true));
+  Address DepObjAddr = CGF.Builder.CreateGEP(
+      Addr, llvm::ConstantInt::get(CGF.IntPtrTy, -1, /*isSigned=*/true));
   LValue NumDepsBase = CGF.MakeAddrLValue(
-      Address(DepObjAddr, Addr.getAlignment()), KmpDependInfoTy,
-      Base.getBaseInfo(), Base.getTBAAInfo());
+      DepObjAddr, KmpDependInfoTy, Base.getBaseInfo(), Base.getTBAAInfo());
   // NumDeps = deps[i].base_addr;
   LValue BaseAddrLVal = CGF.EmitLValueForField(
       NumDepsBase, *std::next(KmpDependInfoRD->field_begin(), BaseAddr));
@@ -4706,10 +4696,7 @@ static void emitDependData(CodeGenFunction &CGF, QualType &KmpDependInfoTy,
       LValue &PosLVal = *Pos.get<LValue *>();
       llvm::Value *Idx = CGF.EmitLoadOfScalar(PosLVal, E->getExprLoc());
       Base = CGF.MakeAddrLValue(
-          Address(CGF.Builder.CreateGEP(DependenciesArray.getElementType(),
-                                        DependenciesArray.getPointer(), Idx),
-                  DependenciesArray.getAlignment()),
-          KmpDependInfoTy);
+          CGF.Builder.CreateGEP(DependenciesArray, Idx), KmpDependInfoTy);
     }
     // deps[i].base_addr = &<Dependencies[i].second>;
     LValue BaseAddrLVal = CGF.EmitLValueForField(
@@ -4766,12 +4753,10 @@ emitDepobjElementsSizes(CodeGenFunction &CGF, QualType &KmpDependInfoTy,
           Base.getAddress(CGF), KmpDependInfoPtrT);
       Base = CGF.MakeAddrLValue(Addr, KmpDependInfoTy, Base.getBaseInfo(),
                                 Base.getTBAAInfo());
-      llvm::Value *DepObjAddr = CGF.Builder.CreateGEP(
-          Addr.getElementType(), Addr.getPointer(),
-          llvm::ConstantInt::get(CGF.IntPtrTy, -1, /*isSigned=*/true));
+      Address DepObjAddr = CGF.Builder.CreateGEP(
+          Addr, llvm::ConstantInt::get(CGF.IntPtrTy, -1, /*isSigned=*/true));
       LValue NumDepsBase = CGF.MakeAddrLValue(
-          Address(DepObjAddr, Addr.getAlignment()), KmpDependInfoTy,
-          Base.getBaseInfo(), Base.getTBAAInfo());
+          DepObjAddr, KmpDependInfoTy, Base.getBaseInfo(), Base.getTBAAInfo());
       // NumDeps = deps[i].base_addr;
       LValue BaseAddrLVal = CGF.EmitLValueForField(
           NumDepsBase, *std::next(KmpDependInfoRD->field_begin(), BaseAddr));
@@ -4827,12 +4812,10 @@ static void emitDepobjElements(CodeGenFunction &CGF, QualType &KmpDependInfoTy,
                                 Base.getTBAAInfo());
 
       // Get number of elements in a single depobj.
-      llvm::Value *DepObjAddr = CGF.Builder.CreateGEP(
-          Addr.getElementType(), Addr.getPointer(),
-          llvm::ConstantInt::get(CGF.IntPtrTy, -1, /*isSigned=*/true));
+      Address DepObjAddr = CGF.Builder.CreateGEP(
+          Addr, llvm::ConstantInt::get(CGF.IntPtrTy, -1, /*isSigned=*/true));
       LValue NumDepsBase = CGF.MakeAddrLValue(
-          Address(DepObjAddr, Addr.getAlignment()), KmpDependInfoTy,
-          Base.getBaseInfo(), Base.getTBAAInfo());
+          DepObjAddr, KmpDependInfoTy, Base.getBaseInfo(), Base.getTBAAInfo());
       // NumDeps = deps[i].base_addr;
       LValue BaseAddrLVal = CGF.EmitLValueForField(
           NumDepsBase, *std::next(KmpDependInfoRD->field_begin(), BaseAddr));
@@ -4844,10 +4827,7 @@ static void emitDepobjElements(CodeGenFunction &CGF, QualType &KmpDependInfoTy,
           ElSize,
           CGF.Builder.CreateIntCast(NumDeps, CGF.SizeTy, /*isSigned=*/false));
       llvm::Value *Pos = CGF.EmitLoadOfScalar(PosLVal, E->getExprLoc());
-      Address DepAddr =
-          Address(CGF.Builder.CreateGEP(DependenciesArray.getElementType(),
-                                        DependenciesArray.getPointer(), Pos),
-                  DependenciesArray.getAlignment());
+      Address DepAddr = CGF.Builder.CreateGEP(DependenciesArray, Pos);
       CGF.Builder.CreateMemCpy(DepAddr, Base.getAddress(CGF), Size);
 
       // Increase pos.
@@ -5929,25 +5909,20 @@ static llvm::Value *emitReduceInitFunction(CodeGenModule &CGM,
                                 CGM.getContext().getSizeType(), Loc);
   }
   RCG.emitAggregateType(CGF, N, Size);
-  LValue OrigLVal;
+  Address OrigAddr = Address::invalid();
   // If initializer uses initializer from declare reduction construct, emit a
   // pointer to the address of the original reduction item (reuired by reduction
   // initializer)
   if (RCG.usesReductionInitializer(N)) {
     Address SharedAddr = CGF.GetAddrOfLocalVar(&ParamOrig);
-    SharedAddr = CGF.EmitLoadOfPointer(
+    OrigAddr = CGF.EmitLoadOfPointer(
         SharedAddr,
         CGM.getContext().VoidPtrTy.castAs<PointerType>()->getTypePtr());
-    OrigLVal = CGF.MakeAddrLValue(SharedAddr, CGM.getContext().VoidPtrTy);
-  } else {
-    OrigLVal = CGF.MakeNaturalAlignAddrLValue(
-        llvm::ConstantPointerNull::get(CGM.VoidPtrTy),
-        CGM.getContext().VoidPtrTy);
   }
   // Emit the initializer:
   // %0 = bitcast void* %arg to <type>*
   // store <type> <init>, <type>* %0
-  RCG.emitInitialization(CGF, N, PrivateAddr, OrigLVal,
+  RCG.emitInitialization(CGF, N, PrivateAddr, OrigAddr,
                          [](CodeGenFunction &) { return false; });
   CGF.FinishFunction();
   return Fn;
@@ -6122,7 +6097,7 @@ llvm::Value *CGOpenMPRuntime::emitTaskReductionInit(
     llvm::Value *Idxs[] = {llvm::ConstantInt::get(CGM.SizeTy, /*V=*/0),
                            llvm::ConstantInt::get(CGM.SizeTy, Cnt)};
     llvm::Value *GEP = CGF.EmitCheckedInBoundsGEP(
-        TaskRedInput.getPointer(), Idxs,
+        TaskRedInput.getElementType(), TaskRedInput.getPointer(), Idxs,
         /*SignedIndices=*/false, /*IsSubtraction=*/false, Loc,
         ".rd_input.gep.");
     LValue ElemLVal = CGF.MakeNaturalAlignAddrLValue(GEP, RDType);
@@ -6620,6 +6595,8 @@ void CGOpenMPRuntime::emitTargetOutlinedFunctionHelper(
     OutlinedFn->addFnAttr("omp_target_thread_limit",
                           std::to_string(DefaultValThreads));
   }
+
+  CGM.getTargetCodeGenInfo().setTargetAttributes(nullptr, OutlinedFn, CGM);
 }
 
 /// Checks if the expression is constant or does not have non-trivial function
@@ -12680,12 +12657,11 @@ void CGOpenMPRuntime::emitLastprivateConditionalUpdate(CodeGenFunction &CGF,
 
   // Last value of the lastprivate conditional.
   // decltype(priv_a) last_a;
-  llvm::Constant *Last = getOrCreateInternalVariable(
+  llvm::GlobalVariable *Last = getOrCreateInternalVariable(
       CGF.ConvertTypeForMem(LVal.getType()), UniqueDeclName);
-  cast<llvm::GlobalVariable>(Last)->setAlignment(
-      LVal.getAlignment().getAsAlign());
-  LValue LastLVal =
-      CGF.MakeAddrLValue(Last, LVal.getType(), LVal.getAlignment());
+  Last->setAlignment(LVal.getAlignment().getAsAlign());
+  LValue LastLVal = CGF.MakeAddrLValue(
+      Address(Last, Last->getValueType(), LVal.getAlignment()), LVal.getType());
 
   // Global loop counter. Required to handle inner parallel-for regions.
   // iv
@@ -12812,7 +12788,7 @@ void CGOpenMPRuntime::checkAndEmitSharedLastprivateConditional(
   const CapturedStmt *CS = D.getCapturedStmt(CaptureRegions.back());
   for (const auto &Pair : It->DeclToUniqueName) {
     const auto *VD = cast<VarDecl>(Pair.first->getCanonicalDecl());
-    if (!CS->capturesVariable(VD) || IgnoredDecls.count(VD) > 0)
+    if (!CS->capturesVariable(VD) || IgnoredDecls.contains(VD))
       continue;
     auto I = LPCI->getSecond().find(Pair.first);
     assert(I != LPCI->getSecond().end() &&
@@ -12858,7 +12834,8 @@ void CGOpenMPRuntime::emitLastprivateConditionalFinalUpdate(
   if (!GV)
     return;
   LValue LPLVal = CGF.MakeAddrLValue(
-      GV, PrivLVal.getType().getNonReferenceType(), PrivLVal.getAlignment());
+      Address(GV, GV->getValueType(), PrivLVal.getAlignment()),
+      PrivLVal.getType().getNonReferenceType());
   llvm::Value *Res = CGF.EmitLoadOfScalar(LPLVal, Loc);
   CGF.EmitStoreOfScalar(Res, PrivLVal);
 }
@@ -12887,7 +12864,8 @@ void CGOpenMPSIMDRuntime::emitParallelCall(CodeGenFunction &CGF,
                                            SourceLocation Loc,
                                            llvm::Function *OutlinedFn,
                                            ArrayRef<llvm::Value *> CapturedVars,
-                                           const Expr *IfCond) {
+                                           const Expr *IfCond,
+                                           llvm::Value *NumThreads) {
   llvm_unreachable("Not supported in SIMD-only mode");
 }
 
diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.h b/clang/lib/CodeGen/CGOpenMPRuntime.h
index 527a23a8af6a..b83ec78696d1 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.h
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.h
@@ -162,10 +162,10 @@ private:
   /// Performs aggregate initialization.
   /// \param N Number of reduction item in the common list.
   /// \param PrivateAddr Address of the corresponding private item.
-  /// \param SharedLVal Address of the original shared variable.
+  /// \param SharedAddr Address of the original shared variable.
   /// \param DRD Declare reduction construct used for reduction item.
   void emitAggregateInitialization(CodeGenFunction &CGF, unsigned N,
-                                   Address PrivateAddr, LValue SharedLVal,
+                                   Address PrivateAddr, Address SharedAddr,
                                    const OMPDeclareReductionDecl *DRD);
 
 public:
@@ -187,10 +187,10 @@ public:
   /// \param PrivateAddr Address of the corresponding private item.
   /// \param DefaultInit Default initialization sequence that should be
   /// performed if no reduction specific initialization is found.
-  /// \param SharedLVal Address of the original shared variable.
+  /// \param SharedAddr Address of the original shared variable.
   void
   emitInitialization(CodeGenFunction &CGF, unsigned N, Address PrivateAddr,
-                     LValue SharedLVal,
+                     Address SharedAddr,
                      llvm::function_ref<bool(CodeGenFunction &)> DefaultInit);
   /// Returns true if the private copy requires cleanups.
   bool needCleanups(unsigned N);
@@ -471,8 +471,8 @@ private:
   /// <critical_section_name> + ".var" for "omp critical" directives; 2)
   /// <mangled_name_for_global_var> + ".cache." for cache for threadprivate
   /// variables.
-  llvm::StringMap<llvm::AssertingVH<llvm::Constant>, llvm::BumpPtrAllocator>
-      InternalVars;
+  llvm::StringMap<llvm::AssertingVH<llvm::GlobalVariable>,
+                  llvm::BumpPtrAllocator> InternalVars;
   /// Type typedef kmp_int32 (* kmp_routine_entry_t)(kmp_int32, void *);
   llvm::Type *KmpRoutineEntryPtrTy = nullptr;
   QualType KmpRoutineEntryPtrQTy;
@@ -829,9 +829,9 @@ private:
   /// \param Ty Type of the global variable. If it is exist already the type
   /// must be the same.
   /// \param Name Name of the variable.
-  llvm::Constant *getOrCreateInternalVariable(llvm::Type *Ty,
-                                              const llvm::Twine &Name,
-                                              unsigned AddressSpace = 0);
+  llvm::GlobalVariable *getOrCreateInternalVariable(llvm::Type *Ty,
+                                                    const llvm::Twine &Name,
+                                                    unsigned AddressSpace = 0);
 
   /// Set of threadprivate variables with the generated initializer.
   llvm::StringSet<> ThreadPrivateWithDefinition;
@@ -1015,11 +1015,13 @@ public:
   /// variables used in \a OutlinedFn function.
   /// \param IfCond Condition in the associated 'if' clause, if it was
   /// specified, nullptr otherwise.
+  /// \param NumThreads The value corresponding to the num_threads clause, if
+  /// any, or nullptr.
   ///
   virtual void emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
                                 llvm::Function *OutlinedFn,
                                 ArrayRef<llvm::Value *> CapturedVars,
-                                const Expr *IfCond);
+                                const Expr *IfCond, llvm::Value *NumThreads);
 
   /// Emits a critical region.
   /// \param CriticalName Name of the critical region.
@@ -1991,11 +1993,13 @@ public:
   /// variables used in \a OutlinedFn function.
   /// \param IfCond Condition in the associated 'if' clause, if it was
   /// specified, nullptr otherwise.
+  /// \param NumThreads The value corresponding to the num_threads clause, if
+  /// any, or nullptr.
   ///
   void emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
                         llvm::Function *OutlinedFn,
                         ArrayRef<llvm::Value *> CapturedVars,
-                        const Expr *IfCond) override;
+                        const Expr *IfCond, llvm::Value *NumThreads) override;
 
   /// Emits a critical region.
   /// \param CriticalName Name of the critical region.
diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
index dcb224f33156..866454ddeaed 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
@@ -1221,11 +1221,7 @@ void CGOpenMPRuntimeGPU::emitProcBindClause(CodeGenFunction &CGF,
 void CGOpenMPRuntimeGPU::emitNumThreadsClause(CodeGenFunction &CGF,
                                                 llvm::Value *NumThreads,
                                                 SourceLocation Loc) {
-  // Do nothing in case of SPMD mode and L0 parallel.
-  if (getExecutionMode() == CGOpenMPRuntimeGPU::EM_SPMD)
-    return;
-
-  CGOpenMPRuntime::emitNumThreadsClause(CGF, NumThreads, Loc);
+  // Nothing to do.
 }
 
 void CGOpenMPRuntimeGPU::emitNumTeamsClause(CodeGenFunction &CGF,
@@ -1510,13 +1506,16 @@ void CGOpenMPRuntimeGPU::emitParallelCall(CodeGenFunction &CGF,
                                           SourceLocation Loc,
                                           llvm::Function *OutlinedFn,
                                           ArrayRef<llvm::Value *> CapturedVars,
-                                          const Expr *IfCond) {
+                                          const Expr *IfCond,
+                                          llvm::Value *NumThreads) {
   if (!CGF.HaveInsertPoint())
     return;
 
-  auto &&ParallelGen = [this, Loc, OutlinedFn, CapturedVars,
-                        IfCond](CodeGenFunction &CGF, PrePostActionTy &Action) {
+  auto &&ParallelGen = [this, Loc, OutlinedFn, CapturedVars, IfCond,
+                        NumThreads](CodeGenFunction &CGF,
+                                    PrePostActionTy &Action) {
     CGBuilderTy &Bld = CGF.Builder;
+    llvm::Value *NumThreadsVal = NumThreads;
     llvm::Function *WFn = WrapperFunctionsMap[OutlinedFn];
     llvm::Value *ID = llvm::ConstantPointerNull::get(CGM.Int8PtrTy);
     if (WFn)
@@ -1556,13 +1555,18 @@ void CGOpenMPRuntimeGPU::emitParallelCall(CodeGenFunction &CGF,
     else
       IfCondVal = llvm::ConstantInt::get(CGF.Int32Ty, 1);
 
-    assert(IfCondVal && "Expected a value");
+    if (!NumThreadsVal)
+      NumThreadsVal = llvm::ConstantInt::get(CGF.Int32Ty, -1);
+    else
+      NumThreadsVal = Bld.CreateZExtOrTrunc(NumThreadsVal, CGF.Int32Ty),
+
+      assert(IfCondVal && "Expected a value");
     llvm::Value *RTLoc = emitUpdateLocation(CGF, Loc);
     llvm::Value *Args[] = {
         RTLoc,
         getThreadID(CGF, Loc),
         IfCondVal,
-        llvm::ConstantInt::get(CGF.Int32Ty, -1),
+        NumThreadsVal,
         llvm::ConstantInt::get(CGF.Int32Ty, -1),
         FnPtr,
         ID,
@@ -2186,11 +2190,8 @@ static llvm::Value *emitInterWarpCopyFunction(CodeGenModule &CGM,
       // elemptr = ((CopyType*)(elemptrptr)) + I
       Address ElemPtr = Address(ElemPtrPtr, Align);
       ElemPtr = Bld.CreateElementBitCast(ElemPtr, CopyType);
-      if (NumIters > 1) {
-        ElemPtr = Address(Bld.CreateGEP(ElemPtr.getElementType(),
-                                        ElemPtr.getPointer(), Cnt),
-                          ElemPtr.getAlignment());
-      }
+      if (NumIters > 1)
+        ElemPtr = Bld.CreateGEP(ElemPtr, Cnt);
 
       // Get pointer to location in transfer medium.
       // MediumPtr = &medium[warp_id]
@@ -2256,11 +2257,8 @@ static llvm::Value *emitInterWarpCopyFunction(CodeGenModule &CGM,
           TargetElemPtrPtr, /*Volatile=*/false, C.VoidPtrTy, Loc);
       Address TargetElemPtr = Address(TargetElemPtrVal, Align);
       TargetElemPtr = Bld.CreateElementBitCast(TargetElemPtr, CopyType);
-      if (NumIters > 1) {
-        TargetElemPtr = Address(Bld.CreateGEP(TargetElemPtr.getElementType(),
-                                              TargetElemPtr.getPointer(), Cnt),
-                                TargetElemPtr.getAlignment());
-      }
+      if (NumIters > 1)
+        TargetElemPtr = Bld.CreateGEP(TargetElemPtr, Cnt);
 
       // *TargetElemPtr = SrcMediumVal;
       llvm::Value *SrcMediumValue =
@@ -3899,6 +3897,7 @@ void CGOpenMPRuntimeGPU::processRequiresDirective(
       case CudaArch::GFX1033:
       case CudaArch::GFX1034:
       case CudaArch::GFX1035:
+      case CudaArch::Generic:
       case CudaArch::UNUSED:
       case CudaArch::UNKNOWN:
         break;
diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.h b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.h
index ac51264d7685..1d30c5061743 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.h
+++ b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.h
@@ -257,10 +257,13 @@ public:
   /// variables used in \a OutlinedFn function.
   /// \param IfCond Condition in the associated 'if' clause, if it was
   /// specified, nullptr otherwise.
+  /// \param NumThreads The value corresponding to the num_threads clause, if
+  /// any,
+  ///                   or nullptr.
   void emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
                         llvm::Function *OutlinedFn,
                         ArrayRef<llvm::Value *> CapturedVars,
-                        const Expr *IfCond) override;
+                        const Expr *IfCond, llvm::Value *NumThreads) override;
 
   /// Emit an implicit/explicit barrier for OpenMP threads.
   /// \param Kind Directive for which this implicit barrier call must be
diff --git a/clang/lib/CodeGen/CGStmt.cpp b/clang/lib/CodeGen/CGStmt.cpp
index d399ff919cc3..ef0068cd3b0c 100644
--- a/clang/lib/CodeGen/CGStmt.cpp
+++ b/clang/lib/CodeGen/CGStmt.cpp
@@ -2454,7 +2454,7 @@ void CodeGenFunction::EmitAsmStmt(const AsmStmt &S) {
     const ABIArgInfo &RetAI = CurFnInfo->getReturnInfo();
     if (RetAI.isDirect() || RetAI.isExtend()) {
       // Make a fake lvalue for the return value slot.
-      LValue ReturnSlot = MakeAddrLValue(ReturnValue, FnRetTy);
+      LValue ReturnSlot = MakeAddrLValueWithoutTBAA(ReturnValue, FnRetTy);
       CGM.getTargetCodeGenInfo().addReturnRegisterOutputs(
           *this, ReturnSlot, Constraints, ResultRegTypes, ResultTruncRegTypes,
           ResultRegDests, AsmString, S.getNumOutputs());
diff --git a/clang/lib/CodeGen/CGStmtOpenMP.cpp b/clang/lib/CodeGen/CGStmtOpenMP.cpp
index f6853a22cd36..4c11f7d67534 100644
--- a/clang/lib/CodeGen/CGStmtOpenMP.cpp
+++ b/clang/lib/CodeGen/CGStmtOpenMP.cpp
@@ -24,10 +24,13 @@
 #include "clang/AST/StmtVisitor.h"
 #include "clang/Basic/OpenMPKinds.h"
 #include "clang/Basic/PrettyStackTrace.h"
+#include "llvm/BinaryFormat/Dwarf.h"
 #include "llvm/Frontend/OpenMP/OMPConstants.h"
 #include "llvm/Frontend/OpenMP/OMPIRBuilder.h"
 #include "llvm/IR/Constants.h"
+#include "llvm/IR/DebugInfoMetadata.h"
 #include "llvm/IR/Instructions.h"
+#include "llvm/IR/Metadata.h"
 #include "llvm/Support/AtomicOrdering.h"
 using namespace clang;
 using namespace CodeGen;
@@ -375,8 +378,7 @@ static Address castValueFromUintptr(CodeGenFunction &CGF, SourceLocation Loc,
       AddrLV.getAddress(CGF).getPointer(), Ctx.getUIntPtrType(),
       Ctx.getPointerType(DstType), Loc);
   Address TmpAddr =
-      CGF.MakeNaturalAlignAddrLValue(CastedPtr, Ctx.getPointerType(DstType))
-          .getAddress(CGF);
+      CGF.MakeNaturalAlignAddrLValue(CastedPtr, DstType).getAddress(CGF);
   return TmpAddr;
 }
 
@@ -1245,7 +1247,7 @@ void CodeGenFunction::EmitOMPReductionClauseInit(
     RedCG.emitAggregateType(*this, Count);
     AutoVarEmission Emission = EmitAutoVarAlloca(*PrivateVD);
     RedCG.emitInitialization(*this, Count, Emission.getAllocatedAddress(),
-                             RedCG.getSharedLValue(Count),
+                             RedCG.getSharedLValue(Count).getAddress(*this),
                              [&Emission](CodeGenFunction &CGF) {
                                CGF.EmitAutoVarInit(Emission);
                                return true;
@@ -1557,14 +1559,14 @@ static void emitCommonOMPParallelDirective(
     OpenMPDirectiveKind InnermostKind, const RegionCodeGenTy &CodeGen,
     const CodeGenBoundParametersTy &CodeGenBoundParameters) {
   const CapturedStmt *CS = S.getCapturedStmt(OMPD_parallel);
+  llvm::Value *NumThreads = nullptr;
   llvm::Function *OutlinedFn =
       CGF.CGM.getOpenMPRuntime().emitParallelOutlinedFunction(
           S, *CS->getCapturedDecl()->param_begin(), InnermostKind, CodeGen);
   if (const auto *NumThreadsClause = S.getSingleClause<OMPNumThreadsClause>()) {
     CodeGenFunction::RunCleanupsScope NumThreadsScope(CGF);
-    llvm::Value *NumThreads =
-        CGF.EmitScalarExpr(NumThreadsClause->getNumThreads(),
-                           /*IgnoreResultAssign=*/true);
+    NumThreads = CGF.EmitScalarExpr(NumThreadsClause->getNumThreads(),
+                                    /*IgnoreResultAssign=*/true);
     CGF.CGM.getOpenMPRuntime().emitNumThreadsClause(
         CGF, NumThreads, NumThreadsClause->getBeginLoc());
   }
@@ -1591,7 +1593,7 @@ static void emitCommonOMPParallelDirective(
   CodeGenBoundParameters(CGF, S, CapturedVars);
   CGF.GenerateOpenMPCapturedVars(*CS, CapturedVars);
   CGF.CGM.getOpenMPRuntime().emitParallelCall(CGF, S.getBeginLoc(), OutlinedFn,
-                                              CapturedVars, IfCond);
+                                              CapturedVars, IfCond, NumThreads);
 }
 
 static bool isAllocatableDecl(const VarDecl *VD) {
@@ -1972,7 +1974,7 @@ CodeGenFunction::EmitOMPCollapsedCanonicalLoopNest(const Stmt *S, int Depth) {
 
   // Pop the \p Depth loops requested by the call from that stack and restore
   // the previous context.
-  OMPLoopNestStack.set_size(OMPLoopNestStack.size() - Depth);
+  OMPLoopNestStack.pop_back_n(Depth);
   ExpectedOMPLoopDepth = ParentExpectedOMPLoopDepth;
 
   return Result;
@@ -4299,10 +4301,10 @@ public:
           PrivateDecls.push_back(VD);
     }
   }
-  void VisitOMPExecutableDirective(const OMPExecutableDirective *) { return; }
-  void VisitCapturedStmt(const CapturedStmt *) { return; }
-  void VisitLambdaExpr(const LambdaExpr *) { return; }
-  void VisitBlockExpr(const BlockExpr *) { return; }
+  void VisitOMPExecutableDirective(const OMPExecutableDirective *) {}
+  void VisitCapturedStmt(const CapturedStmt *) {}
+  void VisitLambdaExpr(const LambdaExpr *) {}
+  void VisitBlockExpr(const BlockExpr *) {}
   void VisitStmt(const Stmt *S) {
     if (!S)
       return;
@@ -4431,6 +4433,53 @@ void CodeGenFunction::EmitOMPTaskBasedDirective(
         UntiedLocalVars;
     // Set proper addresses for generated private copies.
     OMPPrivateScope Scope(CGF);
+    // Generate debug info for variables present in shared clause.
+    if (auto *DI = CGF.getDebugInfo()) {
+      llvm::SmallDenseMap<const VarDecl *, FieldDecl *> CaptureFields =
+          CGF.CapturedStmtInfo->getCaptureFields();
+      llvm::Value *ContextValue = CGF.CapturedStmtInfo->getContextValue();
+      if (CaptureFields.size() && ContextValue) {
+        unsigned CharWidth = CGF.getContext().getCharWidth();
+        // The shared variables are packed together as members of structure.
+        // So the address of each shared variable can be computed by adding
+        // offset of it (within record) to the base address of record. For each
+        // shared variable, debug intrinsic llvm.dbg.declare is generated with
+        // appropriate expressions (DIExpression).
+        // Ex:
+        //  %12 = load %struct.anon*, %struct.anon** %__context.addr.i
+        //  call void @llvm.dbg.declare(metadata %struct.anon* %12,
+        //            metadata !svar1,
+        //            metadata !DIExpression(DW_OP_deref))
+        //  call void @llvm.dbg.declare(metadata %struct.anon* %12,
+        //            metadata !svar2,
+        //            metadata !DIExpression(DW_OP_plus_uconst, 8, DW_OP_deref))
+        for (auto It = CaptureFields.begin(); It != CaptureFields.end(); ++It) {
+          const VarDecl *SharedVar = It->first;
+          RecordDecl *CaptureRecord = It->second->getParent();
+          const ASTRecordLayout &Layout =
+              CGF.getContext().getASTRecordLayout(CaptureRecord);
+          unsigned Offset =
+              Layout.getFieldOffset(It->second->getFieldIndex()) / CharWidth;
+          (void)DI->EmitDeclareOfAutoVariable(SharedVar, ContextValue,
+                                              CGF.Builder, false);
+          llvm::Instruction &Last = CGF.Builder.GetInsertBlock()->back();
+          // Get the call dbg.declare instruction we just created and update
+          // its DIExpression to add offset to base address.
+          if (auto DDI = dyn_cast<llvm::DbgVariableIntrinsic>(&Last)) {
+            SmallVector<uint64_t, 8> Ops;
+            // Add offset to the base address if non zero.
+            if (Offset) {
+              Ops.push_back(llvm::dwarf::DW_OP_plus_uconst);
+              Ops.push_back(Offset);
+            }
+            Ops.push_back(llvm::dwarf::DW_OP_deref);
+            auto &Ctx = DDI->getContext();
+            llvm::DIExpression *DIExpr = llvm::DIExpression::get(Ctx, Ops);
+            Last.setOperand(2, llvm::MetadataAsValue::get(Ctx, DIExpr));
+          }
+        }
+      }
+    }
     llvm::SmallVector<std::pair<const VarDecl *, Address>, 16> FirstprivatePtrs;
     if (!Data.PrivateVars.empty() || !Data.FirstprivateVars.empty() ||
         !Data.LastprivateVars.empty() || !Data.PrivateLocals.empty()) {
@@ -5918,6 +5967,9 @@ static void emitOMPAtomicExpr(CodeGenFunction &CGF, OpenMPClauseKind Kind,
     emitOMPAtomicCaptureExpr(CGF, AO, IsPostfixUpdate, V, X, E, UE,
                              IsXLHSInRHSPart, Loc);
     break;
+  case OMPC_compare:
+    // Do nothing here as we already emit an error.
+    break;
   case OMPC_if:
   case OMPC_final:
   case OMPC_num_threads:
diff --git a/clang/lib/CodeGen/CGValue.h b/clang/lib/CodeGen/CGValue.h
index 4b39a0520833..f01eece042f8 100644
--- a/clang/lib/CodeGen/CGValue.h
+++ b/clang/lib/CodeGen/CGValue.h
@@ -47,6 +47,8 @@ class RValue {
   llvm::PointerIntPair<llvm::Value *, 2, Flavor> V1;
   // Stores second value and volatility.
   llvm::PointerIntPair<llvm::Value *, 1, bool> V2;
+  // Stores element type for aggregate values.
+  llvm::Type *ElementType;
 
 public:
   bool isScalar() const { return V1.getInt() == Scalar; }
@@ -71,7 +73,8 @@ public:
   Address getAggregateAddress() const {
     assert(isAggregate() && "Not an aggregate!");
     auto align = reinterpret_cast<uintptr_t>(V2.getPointer()) >> AggAlignShift;
-    return Address(V1.getPointer(), CharUnits::fromQuantity(align));
+    return Address(
+        V1.getPointer(), ElementType, CharUnits::fromQuantity(align));
   }
   llvm::Value *getAggregatePointer() const {
     assert(isAggregate() && "Not an aggregate!");
@@ -108,6 +111,7 @@ public:
     RValue ER;
     ER.V1.setPointer(addr.getPointer());
     ER.V1.setInt(Aggregate);
+    ER.ElementType = addr.getElementType();
 
     auto align = static_cast<uintptr_t>(addr.getAlignment().getQuantity());
     ER.V2.setPointer(reinterpret_cast<llvm::Value*>(align << AggAlignShift));
@@ -175,6 +179,7 @@ class LValue {
   } LVType;
 
   llvm::Value *V;
+  llvm::Type *ElementType;
 
   union {
     // Index into a vector subscript: V[i]
@@ -230,6 +235,13 @@ private:
                   LValueBaseInfo BaseInfo, TBAAAccessInfo TBAAInfo) {
     assert((!Alignment.isZero() || Type->isIncompleteType()) &&
            "initializing l-value with zero alignment!");
+    if (isGlobalReg())
+      assert(ElementType == nullptr && "Global reg does not store elem type");
+    else
+      assert(llvm::cast<llvm::PointerType>(V->getType())
+                 ->isOpaqueOrPointeeTypeMatches(ElementType) &&
+             "Pointer element type mismatch");
+
     this->Type = Type;
     this->Quals = Quals;
     const unsigned MaxAlign = 1U << 31;
@@ -327,17 +339,18 @@ public:
     return V;
   }
   Address getAddress(CodeGenFunction &CGF) const {
-    return Address(getPointer(CGF), getAlignment());
+    return Address(getPointer(CGF), ElementType, getAlignment());
   }
   void setAddress(Address address) {
     assert(isSimple());
     V = address.getPointer();
+    ElementType = address.getElementType();
     Alignment = address.getAlignment().getQuantity();
   }
 
   // vector elt lvalue
   Address getVectorAddress() const {
-    return Address(getVectorPointer(), getAlignment());
+    return Address(getVectorPointer(), ElementType, getAlignment());
   }
   llvm::Value *getVectorPointer() const {
     assert(isVectorElt());
@@ -349,7 +362,7 @@ public:
   }
 
   Address getMatrixAddress() const {
-    return Address(getMatrixPointer(), getAlignment());
+    return Address(getMatrixPointer(), ElementType, getAlignment());
   }
   llvm::Value *getMatrixPointer() const {
     assert(isMatrixElt());
@@ -362,7 +375,7 @@ public:
 
   // extended vector elements.
   Address getExtVectorAddress() const {
-    return Address(getExtVectorPointer(), getAlignment());
+    return Address(getExtVectorPointer(), ElementType, getAlignment());
   }
   llvm::Value *getExtVectorPointer() const {
     assert(isExtVectorElt());
@@ -375,7 +388,7 @@ public:
 
   // bitfield lvalue
   Address getBitFieldAddress() const {
-    return Address(getBitFieldPointer(), getAlignment());
+    return Address(getBitFieldPointer(), ElementType, getAlignment());
   }
   llvm::Value *getBitFieldPointer() const { assert(isBitField()); return V; }
   const CGBitFieldInfo &getBitFieldInfo() const {
@@ -395,6 +408,7 @@ public:
     R.LVType = Simple;
     assert(address.getPointer()->getType()->isPointerTy());
     R.V = address.getPointer();
+    R.ElementType = address.getElementType();
     R.Initialize(type, qs, address.getAlignment(), BaseInfo, TBAAInfo);
     return R;
   }
@@ -405,6 +419,7 @@ public:
     LValue R;
     R.LVType = VectorElt;
     R.V = vecAddress.getPointer();
+    R.ElementType = vecAddress.getElementType();
     R.VectorIdx = Idx;
     R.Initialize(type, type.getQualifiers(), vecAddress.getAlignment(),
                  BaseInfo, TBAAInfo);
@@ -417,6 +432,7 @@ public:
     LValue R;
     R.LVType = ExtVectorElt;
     R.V = vecAddress.getPointer();
+    R.ElementType = vecAddress.getElementType();
     R.VectorElts = Elts;
     R.Initialize(type, type.getQualifiers(), vecAddress.getAlignment(),
                  BaseInfo, TBAAInfo);
@@ -435,17 +451,20 @@ public:
     LValue R;
     R.LVType = BitField;
     R.V = Addr.getPointer();
+    R.ElementType = Addr.getElementType();
     R.BitFieldInfo = &Info;
     R.Initialize(type, type.getQualifiers(), Addr.getAlignment(), BaseInfo,
                  TBAAInfo);
     return R;
   }
 
-  static LValue MakeGlobalReg(Address Reg, QualType type) {
+  static LValue MakeGlobalReg(llvm::Value *V, CharUnits alignment,
+                              QualType type) {
     LValue R;
     R.LVType = GlobalReg;
-    R.V = Reg.getPointer();
-    R.Initialize(type, type.getQualifiers(), Reg.getAlignment(),
+    R.V = V;
+    R.ElementType = nullptr;
+    R.Initialize(type, type.getQualifiers(), alignment,
                  LValueBaseInfo(AlignmentSource::Decl), TBAAAccessInfo());
     return R;
   }
@@ -456,6 +475,7 @@ public:
     LValue R;
     R.LVType = MatrixElt;
     R.V = matAddress.getPointer();
+    R.ElementType = matAddress.getElementType();
     R.VectorIdx = Idx;
     R.Initialize(type, type.getQualifiers(), matAddress.getAlignment(),
                  BaseInfo, TBAAInfo);
@@ -470,13 +490,11 @@ public:
 /// An aggregate value slot.
 class AggValueSlot {
   /// The address.
-  llvm::Value *Addr;
+  Address Addr;
 
   // Qualifiers
   Qualifiers Quals;
 
-  unsigned Alignment;
-
   /// DestructedFlag - This is set to true if some external code is
   /// responsible for setting up a destructor for the slot.  Otherwise
   /// the code which constructs it should push the appropriate cleanup.
@@ -520,6 +538,14 @@ class AggValueSlot {
   /// them.
   bool SanitizerCheckedFlag : 1;
 
+  AggValueSlot(Address Addr, Qualifiers Quals, bool DestructedFlag,
+               bool ObjCGCFlag, bool ZeroedFlag, bool AliasedFlag,
+               bool OverlapFlag, bool SanitizerCheckedFlag)
+      : Addr(Addr), Quals(Quals), DestructedFlag(DestructedFlag),
+        ObjCGCFlag(ObjCGCFlag), ZeroedFlag(ZeroedFlag),
+        AliasedFlag(AliasedFlag), OverlapFlag(OverlapFlag),
+        SanitizerCheckedFlag(SanitizerCheckedFlag) {}
+
 public:
   enum IsAliased_t { IsNotAliased, IsAliased };
   enum IsDestructed_t { IsNotDestructed, IsDestructed };
@@ -553,22 +579,8 @@ public:
                               Overlap_t mayOverlap,
                               IsZeroed_t isZeroed = IsNotZeroed,
                        IsSanitizerChecked_t isChecked = IsNotSanitizerChecked) {
-    AggValueSlot AV;
-    if (addr.isValid()) {
-      AV.Addr = addr.getPointer();
-      AV.Alignment = addr.getAlignment().getQuantity();
-    } else {
-      AV.Addr = nullptr;
-      AV.Alignment = 0;
-    }
-    AV.Quals = quals;
-    AV.DestructedFlag = isDestructed;
-    AV.ObjCGCFlag = needsGC;
-    AV.ZeroedFlag = isZeroed;
-    AV.AliasedFlag = isAliased;
-    AV.OverlapFlag = mayOverlap;
-    AV.SanitizerCheckedFlag = isChecked;
-    return AV;
+    return AggValueSlot(addr, quals, isDestructed, needsGC, isZeroed, isAliased,
+                        mayOverlap, isChecked);
   }
 
   static AggValueSlot
@@ -609,19 +621,19 @@ public:
   }
 
   llvm::Value *getPointer() const {
-    return Addr;
+    return Addr.getPointer();
   }
 
   Address getAddress() const {
-    return Address(Addr, getAlignment());
+    return Addr;
   }
 
   bool isIgnored() const {
-    return Addr == nullptr;
+    return !Addr.isValid();
   }
 
   CharUnits getAlignment() const {
-    return CharUnits::fromQuantity(Alignment);
+    return Addr.getAlignment();
   }
 
   IsAliased_t isPotentiallyAliased() const {
diff --git a/clang/lib/CodeGen/CodeGenAction.cpp b/clang/lib/CodeGen/CodeGenAction.cpp
index 52c54d3c7a72..b72b16cf2b5f 100644
--- a/clang/lib/CodeGen/CodeGenAction.cpp
+++ b/clang/lib/CodeGen/CodeGenAction.cpp
@@ -571,7 +571,6 @@ void BackendConsumer::SrcMgrDiagHandler(const llvm::DiagnosticInfoSrcMgr &DI) {
   // If Loc is invalid, we still need to report the issue, it just gets no
   // location info.
   Diags.Report(Loc, DiagID).AddString(Message);
-  return;
 }
 
 bool
diff --git a/clang/lib/CodeGen/CodeGenFunction.cpp b/clang/lib/CodeGen/CodeGenFunction.cpp
index d87cf2d49720..e6adec6948af 100644
--- a/clang/lib/CodeGen/CodeGenFunction.cpp
+++ b/clang/lib/CodeGen/CodeGenFunction.cpp
@@ -188,8 +188,8 @@ LValue CodeGenFunction::MakeNaturalAlignAddrLValue(llvm::Value *V, QualType T) {
   LValueBaseInfo BaseInfo;
   TBAAAccessInfo TBAAInfo;
   CharUnits Alignment = CGM.getNaturalTypeAlignment(T, &BaseInfo, &TBAAInfo);
-  return LValue::MakeAddr(Address(V, Alignment), T, getContext(), BaseInfo,
-                          TBAAInfo);
+  Address Addr(V, ConvertTypeForMem(T), Alignment);
+  return LValue::MakeAddr(Addr, T, getContext(), BaseInfo, TBAAInfo);
 }
 
 /// Given a value of type T* that may not be to a complete object,
@@ -200,7 +200,8 @@ CodeGenFunction::MakeNaturalAlignPointeeAddrLValue(llvm::Value *V, QualType T) {
   TBAAAccessInfo TBAAInfo;
   CharUnits Align = CGM.getNaturalTypeAlignment(T, &BaseInfo, &TBAAInfo,
                                                 /* forPointeeType= */ true);
-  return MakeAddrLValue(Address(V, Align), T, BaseInfo, TBAAInfo);
+  Address Addr(V, ConvertTypeForMem(T), Align);
+  return MakeAddrLValue(Addr, T, BaseInfo, TBAAInfo);
 }
 
 
@@ -243,7 +244,7 @@ TypeEvaluationKind CodeGenFunction::getEvaluationKind(QualType type) {
     case Type::Enum:
     case Type::ObjCObjectPointer:
     case Type::Pipe:
-    case Type::ExtInt:
+    case Type::BitInt:
       return TEK_Scalar;
 
     // Complexes.
@@ -1070,7 +1071,8 @@ void CodeGenFunction::StartFunction(GlobalDecl GD, QualType RetTy,
     auto AI = CurFn->arg_begin();
     if (CurFnInfo->getReturnInfo().isSRetAfterThis())
       ++AI;
-    ReturnValue = Address(&*AI, CurFnInfo->getReturnInfo().getIndirectAlign());
+    ReturnValue = Address(&*AI, ConvertType(RetTy),
+                          CurFnInfo->getReturnInfo().getIndirectAlign());
     if (!CurFnInfo->getReturnInfo().getIndirectByVal()) {
       ReturnValuePointer =
           CreateDefaultAlignTempAlloca(Int8PtrTy, "result.ptr");
@@ -1298,47 +1300,44 @@ QualType CodeGenFunction::BuildFunctionArgList(GlobalDecl GD,
 
 void CodeGenFunction::GenerateCode(GlobalDecl GD, llvm::Function *Fn,
                                    const CGFunctionInfo &FnInfo) {
+  assert(Fn && "generating code for null Function");
   const FunctionDecl *FD = cast<FunctionDecl>(GD.getDecl());
   CurGD = GD;
 
   FunctionArgList Args;
   QualType ResTy = BuildFunctionArgList(GD, Args);
 
-  // When generating code for a builtin with an inline declaration, use a
-  // mangled name to hold the actual body, while keeping an external definition
-  // in case the function pointer is referenced somewhere.
-  if (Fn) {
-    if (FD->isInlineBuiltinDeclaration()) {
-      std::string FDInlineName = (Fn->getName() + ".inline").str();
-      llvm::Module *M = Fn->getParent();
-      llvm::Function *Clone = M->getFunction(FDInlineName);
-      if (!Clone) {
-        Clone = llvm::Function::Create(Fn->getFunctionType(),
-                                       llvm::GlobalValue::InternalLinkage,
-                                       Fn->getAddressSpace(), FDInlineName, M);
-        Clone->addFnAttr(llvm::Attribute::AlwaysInline);
-      }
-      Fn->setLinkage(llvm::GlobalValue::ExternalLinkage);
-      Fn = Clone;
+  if (FD->isInlineBuiltinDeclaration()) {
+    // When generating code for a builtin with an inline declaration, use a
+    // mangled name to hold the actual body, while keeping an external
+    // definition in case the function pointer is referenced somewhere.
+    std::string FDInlineName = (Fn->getName() + ".inline").str();
+    llvm::Module *M = Fn->getParent();
+    llvm::Function *Clone = M->getFunction(FDInlineName);
+    if (!Clone) {
+      Clone = llvm::Function::Create(Fn->getFunctionType(),
+                                     llvm::GlobalValue::InternalLinkage,
+                                     Fn->getAddressSpace(), FDInlineName, M);
+      Clone->addFnAttr(llvm::Attribute::AlwaysInline);
     }
-
+    Fn->setLinkage(llvm::GlobalValue::ExternalLinkage);
+    Fn = Clone;
+  } else {
     // Detect the unusual situation where an inline version is shadowed by a
     // non-inline version. In that case we should pick the external one
     // everywhere. That's GCC behavior too. Unfortunately, I cannot find a way
     // to detect that situation before we reach codegen, so do some late
     // replacement.
-    else {
-      for (const FunctionDecl *PD = FD->getPreviousDecl(); PD;
-           PD = PD->getPreviousDecl()) {
-        if (LLVM_UNLIKELY(PD->isInlineBuiltinDeclaration())) {
-          std::string FDInlineName = (Fn->getName() + ".inline").str();
-          llvm::Module *M = Fn->getParent();
-          if (llvm::Function *Clone = M->getFunction(FDInlineName)) {
-            Clone->replaceAllUsesWith(Fn);
-            Clone->eraseFromParent();
-          }
-          break;
+    for (const FunctionDecl *PD = FD->getPreviousDecl(); PD;
+         PD = PD->getPreviousDecl()) {
+      if (LLVM_UNLIKELY(PD->isInlineBuiltinDeclaration())) {
+        std::string FDInlineName = (Fn->getName() + ".inline").str();
+        llvm::Module *M = Fn->getParent();
+        if (llvm::Function *Clone = M->getFunction(FDInlineName)) {
+          Clone->replaceAllUsesWith(Fn);
+          Clone->eraseFromParent();
         }
+        break;
       }
     }
   }
@@ -1347,8 +1346,7 @@ void CodeGenFunction::GenerateCode(GlobalDecl GD, llvm::Function *Fn,
   if (FD->hasAttr<NoDebugAttr>()) {
     // Clear non-distinct debug info that was possibly attached to the function
     // due to an earlier declaration without the nodebug attribute
-    if (Fn)
-      Fn->setSubprogram(nullptr);
+    Fn->setSubprogram(nullptr);
     // Disable debug info indefinitely for this function
     DebugInfo = nullptr;
   }
@@ -2202,12 +2200,13 @@ void CodeGenFunction::EmitVariablyModifiedType(QualType type) {
     case Type::Record:
     case Type::Enum:
     case Type::Elaborated:
+    case Type::Using:
     case Type::TemplateSpecialization:
     case Type::ObjCTypeParam:
     case Type::ObjCObject:
     case Type::ObjCInterface:
     case Type::ObjCObjectPointer:
-    case Type::ExtInt:
+    case Type::BitInt:
       llvm_unreachable("type class is never variably-modified!");
 
     case Type::Adjusted:
diff --git a/clang/lib/CodeGen/CodeGenFunction.h b/clang/lib/CodeGen/CodeGenFunction.h
index ff5b6634da1c..f76ce8a6400d 100644
--- a/clang/lib/CodeGen/CodeGenFunction.h
+++ b/clang/lib/CodeGen/CodeGenFunction.h
@@ -459,6 +459,11 @@ public:
     /// Get the name of the capture helper.
     virtual StringRef getHelperName() const { return "__captured_stmt"; }
 
+    /// Get the CaptureFields
+    llvm::SmallDenseMap<const VarDecl *, FieldDecl *> getCaptureFields() {
+      return CaptureFields;
+    }
+
   private:
     /// The kind of captured statement being generated.
     CapturedRegionKind Kind;
@@ -2494,14 +2499,16 @@ public:
 
   LValue MakeAddrLValue(llvm::Value *V, QualType T, CharUnits Alignment,
                         AlignmentSource Source = AlignmentSource::Type) {
-    return LValue::MakeAddr(Address(V, Alignment), T, getContext(),
-                            LValueBaseInfo(Source), CGM.getTBAAAccessInfo(T));
+    Address Addr(V, ConvertTypeForMem(T), Alignment);
+    return LValue::MakeAddr(Addr, T, getContext(), LValueBaseInfo(Source),
+                            CGM.getTBAAAccessInfo(T));
   }
 
-  LValue MakeAddrLValue(llvm::Value *V, QualType T, CharUnits Alignment,
-                        LValueBaseInfo BaseInfo, TBAAAccessInfo TBAAInfo) {
-    return LValue::MakeAddr(Address(V, Alignment), T, getContext(),
-                            BaseInfo, TBAAInfo);
+  LValue
+  MakeAddrLValueWithoutTBAA(Address Addr, QualType T,
+                            AlignmentSource Source = AlignmentSource::Type) {
+    return LValue::MakeAddr(Addr, T, getContext(), LValueBaseInfo(Source),
+                            TBAAAccessInfo());
   }
 
   LValue MakeNaturalAlignPointeeAddrLValue(llvm::Value *V, QualType T);
@@ -3128,15 +3135,18 @@ public:
 
   class ParamValue {
     llvm::Value *Value;
+    llvm::Type *ElementType;
     unsigned Alignment;
-    ParamValue(llvm::Value *V, unsigned A) : Value(V), Alignment(A) {}
+    ParamValue(llvm::Value *V, llvm::Type *T, unsigned A)
+        : Value(V), ElementType(T), Alignment(A) {}
   public:
     static ParamValue forDirect(llvm::Value *value) {
-      return ParamValue(value, 0);
+      return ParamValue(value, nullptr, 0);
     }
     static ParamValue forIndirect(Address addr) {
       assert(!addr.getAlignment().isZero());
-      return ParamValue(addr.getPointer(), addr.getAlignment().getQuantity());
+      return ParamValue(addr.getPointer(), addr.getElementType(),
+                        addr.getAlignment().getQuantity());
     }
 
     bool isIndirect() const { return Alignment != 0; }
@@ -3149,7 +3159,7 @@ public:
 
     Address getIndirectAddress() const {
       assert(isIndirect());
-      return Address(Value, CharUnits::fromQuantity(Alignment));
+      return Address(Value, ElementType, CharUnits::fromQuantity(Alignment));
     }
   };
 
@@ -4405,7 +4415,7 @@ public:
 
   /// EmitCXXGlobalVarDeclInit - Create the initializer for a C++
   /// variable with global storage.
-  void EmitCXXGlobalVarDeclInit(const VarDecl &D, llvm::Constant *DeclPtr,
+  void EmitCXXGlobalVarDeclInit(const VarDecl &D, llvm::GlobalVariable *GV,
                                 bool PerformInit);
 
   llvm::Function *createAtExitStub(const VarDecl &VD, llvm::FunctionCallee Dtor,
@@ -4556,7 +4566,7 @@ public:
   /// \p SignedIndices indicates whether any of the GEP indices are signed.
   /// \p IsSubtraction indicates whether the expression used to form the GEP
   /// is a subtraction.
-  llvm::Value *EmitCheckedInBoundsGEP(llvm::Value *Ptr,
+  llvm::Value *EmitCheckedInBoundsGEP(llvm::Type *ElemTy, llvm::Value *Ptr,
                                       ArrayRef<llvm::Value *> IdxList,
                                       bool SignedIndices,
                                       bool IsSubtraction,
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
index 9ba1a5c25e81..36b7ce87336c 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -2832,7 +2832,7 @@ ConstantAddress CodeGenModule::GetAddrOfMSGuidDecl(const MSGuidDecl *GD) {
 
   // Look for an existing global.
   if (llvm::GlobalVariable *GV = getModule().getNamedGlobal(Name))
-    return ConstantAddress(GV, Alignment);
+    return ConstantAddress(GV, GV->getValueType(), Alignment);
 
   ConstantEmitter Emitter(*this);
   llvm::Constant *Init;
@@ -2866,15 +2866,15 @@ ConstantAddress CodeGenModule::GetAddrOfMSGuidDecl(const MSGuidDecl *GD) {
     GV->setComdat(TheModule.getOrInsertComdat(GV->getName()));
   setDSOLocal(GV);
 
-  llvm::Constant *Addr = GV;
   if (!V.isAbsent()) {
     Emitter.finalize(GV);
-  } else {
-    llvm::Type *Ty = getTypes().ConvertTypeForMem(GD->getType());
-    Addr = llvm::ConstantExpr::getBitCast(
-        GV, Ty->getPointerTo(GV->getAddressSpace()));
+    return ConstantAddress(GV, GV->getValueType(), Alignment);
   }
-  return ConstantAddress(Addr, Alignment);
+
+  llvm::Type *Ty = getTypes().ConvertTypeForMem(GD->getType());
+  llvm::Constant *Addr = llvm::ConstantExpr::getBitCast(
+      GV, Ty->getPointerTo(GV->getAddressSpace()));
+  return ConstantAddress(Addr, Ty, Alignment);
 }
 
 ConstantAddress CodeGenModule::GetAddrOfTemplateParamObject(
@@ -2883,7 +2883,7 @@ ConstantAddress CodeGenModule::GetAddrOfTemplateParamObject(
   CharUnits Alignment = getNaturalTypeAlignment(TPO->getType());
 
   if (llvm::GlobalVariable *GV = getModule().getNamedGlobal(Name))
-    return ConstantAddress(GV, Alignment);
+    return ConstantAddress(GV, GV->getValueType(), Alignment);
 
   ConstantEmitter Emitter(*this);
   llvm::Constant *Init = Emitter.emitForInitializer(
@@ -2901,7 +2901,7 @@ ConstantAddress CodeGenModule::GetAddrOfTemplateParamObject(
     GV->setComdat(TheModule.getOrInsertComdat(GV->getName()));
   Emitter.finalize(GV);
 
-  return ConstantAddress(GV, Alignment);
+  return ConstantAddress(GV, GV->getValueType(), Alignment);
 }
 
 ConstantAddress CodeGenModule::GetWeakRefReference(const ValueDecl *VD) {
@@ -2916,7 +2916,7 @@ ConstantAddress CodeGenModule::GetWeakRefReference(const ValueDecl *VD) {
   if (Entry) {
     unsigned AS = getContext().getTargetAddressSpace(VD->getType());
     auto Ptr = llvm::ConstantExpr::getBitCast(Entry, DeclTy->getPointerTo(AS));
-    return ConstantAddress(Ptr, Alignment);
+    return ConstantAddress(Ptr, DeclTy, Alignment);
   }
 
   llvm::Constant *Aliasee;
@@ -2932,7 +2932,7 @@ ConstantAddress CodeGenModule::GetWeakRefReference(const ValueDecl *VD) {
   F->setLinkage(llvm::Function::ExternalWeakLinkage);
   WeakRefReferences.insert(F);
 
-  return ConstantAddress(Aliasee, Alignment);
+  return ConstantAddress(Aliasee, DeclTy, Alignment);
 }
 
 void CodeGenModule::EmitGlobal(GlobalDecl GD) {
@@ -3886,6 +3886,14 @@ llvm::Constant *CodeGenModule::GetAddrOfFunction(GlobalDecl GD,
   return F;
 }
 
+llvm::Constant *CodeGenModule::GetFunctionStart(const ValueDecl *Decl) {
+  llvm::GlobalValue *F =
+      cast<llvm::GlobalValue>(GetAddrOfFunction(Decl)->stripPointerCasts());
+
+  return llvm::ConstantExpr::getBitCast(llvm::NoCFIValue::get(F),
+                                        llvm::Type::getInt8PtrTy(VMContext));
+}
+
 static const FunctionDecl *
 GetRuntimeFunctionDecl(ASTContext &C, StringRef Name) {
   TranslationUnitDecl *TUDecl = C.getTranslationUnitDecl();
@@ -5228,7 +5236,8 @@ CodeGenModule::GetAddrOfConstantCFString(const StringLiteral *Literal) {
                                StringLength);
 
   if (auto *C = Entry.second)
-    return ConstantAddress(C, CharUnits::fromQuantity(C->getAlignment()));
+    return ConstantAddress(
+        C, C->getValueType(), CharUnits::fromQuantity(C->getAlignment()));
 
   llvm::Constant *Zero = llvm::Constant::getNullValue(Int32Ty);
   llvm::Constant *Zeros[] = { Zero, Zero };
@@ -5409,7 +5418,7 @@ CodeGenModule::GetAddrOfConstantCFString(const StringLiteral *Literal) {
   }
   Entry.second = GV;
 
-  return ConstantAddress(GV, Alignment);
+  return ConstantAddress(GV, GV->getValueType(), Alignment);
 }
 
 bool CodeGenModule::getExpressionLocationsEnabled() const {
@@ -5527,7 +5536,7 @@ CodeGenModule::GetAddrOfConstantStringFromLiteral(const StringLiteral *S,
       if (uint64_t(Alignment.getQuantity()) > GV->getAlignment())
         GV->setAlignment(Alignment.getAsAlign());
       return ConstantAddress(castStringLiteralToDefaultAddressSpace(*this, GV),
-                             Alignment);
+                             GV->getValueType(), Alignment);
     }
   }
 
@@ -5557,7 +5566,7 @@ CodeGenModule::GetAddrOfConstantStringFromLiteral(const StringLiteral *S,
                                   QualType());
 
   return ConstantAddress(castStringLiteralToDefaultAddressSpace(*this, GV),
-                         Alignment);
+                         GV->getValueType(), Alignment);
 }
 
 /// GetAddrOfConstantStringFromObjCEncode - Return a pointer to a constant
@@ -5590,7 +5599,7 @@ ConstantAddress CodeGenModule::GetAddrOfConstantCString(
       if (uint64_t(Alignment.getQuantity()) > GV->getAlignment())
         GV->setAlignment(Alignment.getAsAlign());
       return ConstantAddress(castStringLiteralToDefaultAddressSpace(*this, GV),
-                             Alignment);
+                             GV->getValueType(), Alignment);
     }
   }
 
@@ -5604,7 +5613,7 @@ ConstantAddress CodeGenModule::GetAddrOfConstantCString(
     *Entry = GV;
 
   return ConstantAddress(castStringLiteralToDefaultAddressSpace(*this, GV),
-                         Alignment);
+                         GV->getValueType(), Alignment);
 }
 
 ConstantAddress CodeGenModule::GetAddrOfGlobalTemporary(
@@ -5634,7 +5643,9 @@ ConstantAddress CodeGenModule::GetAddrOfGlobalTemporary(
           getModule(), Type, false, llvm::GlobalVariable::InternalLinkage,
           nullptr);
     }
-    return ConstantAddress(InsertResult.first->second, Align);
+    return ConstantAddress(
+        InsertResult.first->second,
+        InsertResult.first->second->getType()->getPointerElementType(), Align);
   }
 
   // FIXME: If an externally-visible declaration extends multiple temporaries,
@@ -5725,7 +5736,7 @@ ConstantAddress CodeGenModule::GetAddrOfGlobalTemporary(
   }
   Entry = CV;
 
-  return ConstantAddress(CV, Align);
+  return ConstantAddress(CV, Type, Align);
 }
 
 /// EmitObjCPropertyImplementations - Emit information for synthesized
@@ -6398,6 +6409,11 @@ void CodeGenModule::EmitOMPThreadPrivateDecl(const OMPThreadPrivateDecl *D) {
 llvm::Metadata *
 CodeGenModule::CreateMetadataIdentifierImpl(QualType T, MetadataTypeMap &Map,
                                             StringRef Suffix) {
+  if (auto *FnType = T->getAs<FunctionProtoType>())
+    T = getContext().getFunctionType(
+        FnType->getReturnType(), FnType->getParamTypes(),
+        FnType->getExtProtoInfo().withExceptionSpec(EST_None));
+
   llvm::Metadata *&InternalId = Map[T.getCanonicalType()];
   if (InternalId)
     return InternalId;
diff --git a/clang/lib/CodeGen/CodeGenModule.h b/clang/lib/CodeGen/CodeGenModule.h
index e1c7f486d334..f1565511f98a 100644
--- a/clang/lib/CodeGen/CodeGenModule.h
+++ b/clang/lib/CodeGen/CodeGenModule.h
@@ -881,6 +881,9 @@ public:
                                     ForDefinition_t IsForDefinition
                                       = NotForDefinition);
 
+  // Return the function body address of the given function.
+  llvm::Constant *GetFunctionStart(const ValueDecl *Decl);
+
   /// Get the address of the RTTI descriptor for the given type.
   llvm::Constant *GetAddrOfRTTIDescriptor(QualType Ty, bool ForEH = false);
 
diff --git a/clang/lib/CodeGen/CodeGenTBAA.cpp b/clang/lib/CodeGen/CodeGenTBAA.cpp
index f4ebe6885675..95763d8e18b7 100644
--- a/clang/lib/CodeGen/CodeGenTBAA.cpp
+++ b/clang/lib/CodeGen/CodeGenTBAA.cpp
@@ -209,12 +209,12 @@ llvm::MDNode *CodeGenTBAA::getTypeInfoHelper(const Type *Ty) {
     return createScalarTypeNode(OutName, getChar(), Size);
   }
 
-  if (const auto *EIT = dyn_cast<ExtIntType>(Ty)) {
+  if (const auto *EIT = dyn_cast<BitIntType>(Ty)) {
     SmallString<256> OutName;
     llvm::raw_svector_ostream Out(OutName);
     // Don't specify signed/unsigned since integer types can alias despite sign
     // differences.
-    Out << "_ExtInt(" << EIT->getNumBits() << ')';
+    Out << "_BitInt(" << EIT->getNumBits() << ')';
     return createScalarTypeNode(OutName, getChar(), Size);
   }
 
diff --git a/clang/lib/CodeGen/CodeGenTypes.cpp b/clang/lib/CodeGen/CodeGenTypes.cpp
index fb05475a4e8c..77721510dfd0 100644
--- a/clang/lib/CodeGen/CodeGenTypes.cpp
+++ b/clang/lib/CodeGen/CodeGenTypes.cpp
@@ -97,10 +97,10 @@ llvm::Type *CodeGenTypes::ConvertTypeForMem(QualType T, bool ForBitField) {
 
   llvm::Type *R = ConvertType(T);
 
-  // If this is a bool type, or an ExtIntType in a bitfield representation,
-  // map this integer to the target-specified size.
-  if ((ForBitField && T->isExtIntType()) ||
-      (!T->isExtIntType() && R->isIntegerTy(1)))
+  // If this is a bool type, or a bit-precise integer type in a bitfield
+  // representation, map this integer to the target-specified size.
+  if ((ForBitField && T->isBitIntType()) ||
+      (!T->isBitIntType() && R->isIntegerTy(1)))
     return llvm::IntegerType::get(getLLVMContext(),
                                   (unsigned)Context.getTypeSize(T));
 
@@ -786,8 +786,8 @@ llvm::Type *CodeGenTypes::ConvertType(QualType T) {
     ResultType = CGM.getOpenCLRuntime().getPipeType(cast<PipeType>(Ty));
     break;
   }
-  case Type::ExtInt: {
-    const auto &EIT = cast<ExtIntType>(Ty);
+  case Type::BitInt: {
+    const auto &EIT = cast<BitIntType>(Ty);
     ResultType = llvm::Type::getIntNTy(getLLVMContext(), EIT->getNumBits());
     break;
   }
diff --git a/clang/lib/CodeGen/ItaniumCXXABI.cpp b/clang/lib/CodeGen/ItaniumCXXABI.cpp
index 04163aeaddc5..1a15b09c7b2b 100644
--- a/clang/lib/CodeGen/ItaniumCXXABI.cpp
+++ b/clang/lib/CodeGen/ItaniumCXXABI.cpp
@@ -1345,7 +1345,8 @@ void ItaniumCXXABI::emitThrow(CodeGenFunction &CGF, const CXXThrowExpr *E) {
       AllocExceptionFn, llvm::ConstantInt::get(SizeTy, TypeSize), "exception");
 
   CharUnits ExnAlign = CGF.getContext().getExnObjectAlignment();
-  CGF.EmitAnyExprToExn(E->getSubExpr(), Address(ExceptionPtr, ExnAlign));
+  CGF.EmitAnyExprToExn(
+      E->getSubExpr(), Address(ExceptionPtr, CGM.Int8Ty, ExnAlign));
 
   // Now throw the exception.
   llvm::Constant *TypeInfo = CGM.GetAddrOfRTTIDescriptor(ThrowType,
@@ -2465,7 +2466,7 @@ void ItaniumCXXABI::EmitGuardedInit(CodeGenFunction &CGF,
     CGM.setStaticLocalDeclGuardAddress(&D, guard);
   }
 
-  Address guardAddr = Address(guard, guardAlignment);
+  Address guardAddr = Address(guard, guard->getValueType(), guardAlignment);
 
   // Test whether the variable has completed initialization.
   //
@@ -2880,7 +2881,7 @@ void ItaniumCXXABI::EmitThreadLocalInitFuncs(
     Guard->setAlignment(GuardAlign.getAsAlign());
 
     CodeGenFunction(CGM).GenerateCXXGlobalInitFunc(
-        InitFunc, OrderedInits, ConstantAddress(Guard, GuardAlign));
+        InitFunc, OrderedInits, ConstantAddress(Guard, CGM.Int8Ty, GuardAlign));
     // On Darwin platforms, use CXX_FAST_TLS calling convention.
     if (CGM.getTarget().getTriple().isOSDarwin()) {
       InitFunc->setCallingConv(llvm::CallingConv::CXX_FAST_TLS);
@@ -3529,7 +3530,7 @@ void ItaniumRTTIBuilder::BuildVTablePointer(const Type *Ty) {
     llvm_unreachable("Pipe types shouldn't get here");
 
   case Type::Builtin:
-  case Type::ExtInt:
+  case Type::BitInt:
   // GCC treats vector and complex types as fundamental types.
   case Type::Vector:
   case Type::ExtVector:
@@ -3802,7 +3803,7 @@ llvm::Constant *ItaniumRTTIBuilder::BuildTypeInfo(
   case Type::Pipe:
     break;
 
-  case Type::ExtInt:
+  case Type::BitInt:
     break;
 
   case Type::ConstantArray:
diff --git a/clang/lib/CodeGen/MicrosoftCXXABI.cpp b/clang/lib/CodeGen/MicrosoftCXXABI.cpp
index 0fd5a0ffe06c..5971a7709304 100644
--- a/clang/lib/CodeGen/MicrosoftCXXABI.cpp
+++ b/clang/lib/CodeGen/MicrosoftCXXABI.cpp
@@ -917,7 +917,7 @@ void MicrosoftCXXABI::emitBeginCatch(CodeGenFunction &CGF,
 std::tuple<Address, llvm::Value *, const CXXRecordDecl *>
 MicrosoftCXXABI::performBaseAdjustment(CodeGenFunction &CGF, Address Value,
                                        QualType SrcRecordTy) {
-  Value = CGF.Builder.CreateBitCast(Value, CGF.Int8PtrTy);
+  Value = CGF.Builder.CreateElementBitCast(Value, CGF.Int8Ty);
   const CXXRecordDecl *SrcDecl = SrcRecordTy->getAsCXXRecordDecl();
   const ASTContext &Context = getContext();
 
@@ -2408,14 +2408,14 @@ static ConstantAddress getInitThreadEpochPtr(CodeGenModule &CGM) {
   StringRef VarName("_Init_thread_epoch");
   CharUnits Align = CGM.getIntAlign();
   if (auto *GV = CGM.getModule().getNamedGlobal(VarName))
-    return ConstantAddress(GV, Align);
+    return ConstantAddress(GV, GV->getValueType(), Align);
   auto *GV = new llvm::GlobalVariable(
       CGM.getModule(), CGM.IntTy,
       /*isConstant=*/false, llvm::GlobalVariable::ExternalLinkage,
       /*Initializer=*/nullptr, VarName,
       /*InsertBefore=*/nullptr, llvm::GlobalVariable::GeneralDynamicTLSModel);
   GV->setAlignment(Align.getAsAlign());
-  return ConstantAddress(GV, Align);
+  return ConstantAddress(GV, GV->getValueType(), Align);
 }
 
 static llvm::FunctionCallee getInitThreadHeaderFn(CodeGenModule &CGM) {
@@ -2567,7 +2567,7 @@ void MicrosoftCXXABI::EmitGuardedInit(CodeGenFunction &CGF, const VarDecl &D,
       GI->Guard = GuardVar;
   }
 
-  ConstantAddress GuardAddr(GuardVar, GuardAlign);
+  ConstantAddress GuardAddr(GuardVar, GuardTy, GuardAlign);
 
   assert(GuardVar->getLinkage() == GV->getLinkage() &&
          "static local from the same function had different linkage");
diff --git a/clang/lib/CodeGen/TargetInfo.cpp b/clang/lib/CodeGen/TargetInfo.cpp
index 36e0319c8ab9..85089cdb2200 100644
--- a/clang/lib/CodeGen/TargetInfo.cpp
+++ b/clang/lib/CodeGen/TargetInfo.cpp
@@ -104,7 +104,7 @@ bool ABIInfo::isPromotableIntegerTypeForABI(QualType Ty) const {
   if (Ty->isPromotableIntegerType())
     return true;
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() < getContext().getTypeSize(getContext().IntTy))
       return true;
 
@@ -431,7 +431,7 @@ static Address emitMergePHI(CodeGenFunction &CGF,
   PHI->addIncoming(Addr1.getPointer(), Block1);
   PHI->addIncoming(Addr2.getPointer(), Block2);
   CharUnits Align = std::min(Addr1.getAlignment(), Addr2.getAlignment());
-  return Address(PHI, Align);
+  return Address(PHI, Addr1.getElementType(), Align);
 }
 
 TargetCodeGenInfo::~TargetCodeGenInfo() = default;
@@ -762,7 +762,7 @@ ABIArgInfo DefaultABIInfo::classifyArgumentType(QualType Ty) const {
     Ty = EnumTy->getDecl()->getIntegerType();
 
   ASTContext &Context = getContext();
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() >
         Context.getTypeSize(Context.getTargetInfo().hasInt128Type()
                                 ? Context.Int128Ty
@@ -784,7 +784,7 @@ ABIArgInfo DefaultABIInfo::classifyReturnType(QualType RetTy) const {
   if (const EnumType *EnumTy = RetTy->getAs<EnumType>())
     RetTy = EnumTy->getDecl()->getIntegerType();
 
-  if (const auto *EIT = RetTy->getAs<ExtIntType>())
+  if (const auto *EIT = RetTy->getAs<BitIntType>())
     if (EIT->getNumBits() >
         getContext().getTypeSize(getContext().getTargetInfo().hasInt128Type()
                                      ? getContext().Int128Ty
@@ -1008,8 +1008,9 @@ ABIArgInfo PNaClABIInfo::classifyArgumentType(QualType Ty) const {
   } else if (Ty->isFloatingType()) {
     // Floating-point types don't go inreg.
     return ABIArgInfo::getDirect();
-  } else if (const auto *EIT = Ty->getAs<ExtIntType>()) {
-    // Treat extended integers as integers if <=64, otherwise pass indirectly.
+  } else if (const auto *EIT = Ty->getAs<BitIntType>()) {
+    // Treat bit-precise integers as integers if <= 64, otherwise pass
+    // indirectly.
     if (EIT->getNumBits() > 64)
       return getNaturalAlignIndirect(Ty);
     return ABIArgInfo::getDirect();
@@ -1027,8 +1028,8 @@ ABIArgInfo PNaClABIInfo::classifyReturnType(QualType RetTy) const {
   if (isAggregateTypeForABI(RetTy))
     return getNaturalAlignIndirect(RetTy);
 
-  // Treat extended integers as integers if <=64, otherwise pass indirectly.
-  if (const auto *EIT = RetTy->getAs<ExtIntType>()) {
+  // Treat bit-precise integers as integers if <= 64, otherwise pass indirectly.
+  if (const auto *EIT = RetTy->getAs<BitIntType>()) {
     if (EIT->getNumBits() > 64)
       return getNaturalAlignIndirect(RetTy);
     return ABIArgInfo::getDirect();
@@ -1590,7 +1591,7 @@ ABIArgInfo X86_32ABIInfo::classifyReturnType(QualType RetTy,
   if (const EnumType *EnumTy = RetTy->getAs<EnumType>())
     RetTy = EnumTy->getDecl()->getIntegerType();
 
-  if (const auto *EIT = RetTy->getAs<ExtIntType>())
+  if (const auto *EIT = RetTy->getAs<BitIntType>())
     if (EIT->getNumBits() > 64)
       return getIndirectReturnResult(RetTy, State);
 
@@ -1926,7 +1927,7 @@ ABIArgInfo X86_32ABIInfo::classifyArgumentType(QualType Ty,
     return ABIArgInfo::getExtend(Ty);
   }
 
-  if (const auto * EIT = Ty->getAs<ExtIntType>()) {
+  if (const auto *EIT = Ty->getAs<BitIntType>()) {
     if (EIT->getNumBits() <= 64) {
       if (InReg)
         return ABIArgInfo::getDirectInReg();
@@ -3009,7 +3010,7 @@ void X86_64ABIInfo::classify(QualType Ty, uint64_t OffsetBase,
     return;
   }
 
-  if (const auto *EITy = Ty->getAs<ExtIntType>()) {
+  if (const auto *EITy = Ty->getAs<BitIntType>()) {
     if (EITy->getNumBits() <= 64)
       Current = Integer;
     else if (EITy->getNumBits() <= 128)
@@ -3200,7 +3201,7 @@ ABIArgInfo X86_64ABIInfo::getIndirectReturnResult(QualType Ty) const {
     if (const EnumType *EnumTy = Ty->getAs<EnumType>())
       Ty = EnumTy->getDecl()->getIntegerType();
 
-    if (Ty->isExtIntType())
+    if (Ty->isBitIntType())
       return getNaturalAlignIndirect(Ty);
 
     return (isPromotableIntegerTypeForABI(Ty) ? ABIArgInfo::getExtend(Ty)
@@ -3237,7 +3238,7 @@ ABIArgInfo X86_64ABIInfo::getIndirectResult(QualType Ty,
   // but this code would be much safer if we could mark the argument with
   // 'onstack'. See PR12193.
   if (!isAggregateTypeForABI(Ty) && !IsIllegalVectorType(Ty) &&
-      !Ty->isExtIntType()) {
+      !Ty->isBitIntType()) {
     // Treat an enum type as its underlying type.
     if (const EnumType *EnumTy = Ty->getAs<EnumType>())
       Ty = EnumTy->getDecl()->getIntegerType();
@@ -4033,7 +4034,7 @@ static Address EmitX86_64VAArgFromMemory(CodeGenFunction &CGF,
   CGF.Builder.CreateStore(overflow_arg_area, overflow_arg_area_p);
 
   // AMD64-ABI 3.5.7p5: Step 11. Return the fetched type.
-  return Address(Res, Align);
+  return Address(Res, LTy, Align);
 }
 
 Address X86_64ABIInfo::EmitVAArg(CodeGenFunction &CGF, Address VAListAddr,
@@ -4146,7 +4147,7 @@ Address X86_64ABIInfo::EmitVAArg(CodeGenFunction &CGF, Address VAListAddr,
     RegAddr = CGF.Builder.CreateElementBitCast(Tmp, LTy);
   } else if (neededInt) {
     RegAddr = Address(CGF.Builder.CreateGEP(CGF.Int8Ty, RegSaveArea, gp_offset),
-                      CharUnits::fromQuantity(8));
+                      CGF.Int8Ty, CharUnits::fromQuantity(8));
     RegAddr = CGF.Builder.CreateElementBitCast(RegAddr, LTy);
 
     // Copy to a temporary if necessary to ensure the appropriate alignment.
@@ -4164,7 +4165,7 @@ Address X86_64ABIInfo::EmitVAArg(CodeGenFunction &CGF, Address VAListAddr,
 
   } else if (neededSSE == 1) {
     RegAddr = Address(CGF.Builder.CreateGEP(CGF.Int8Ty, RegSaveArea, fp_offset),
-                      CharUnits::fromQuantity(16));
+                      CGF.Int8Ty, CharUnits::fromQuantity(16));
     RegAddr = CGF.Builder.CreateElementBitCast(RegAddr, LTy);
   } else {
     assert(neededSSE == 2 && "Invalid number of needed registers!");
@@ -4176,7 +4177,7 @@ Address X86_64ABIInfo::EmitVAArg(CodeGenFunction &CGF, Address VAListAddr,
     // all the SSE registers to the RSA.
     Address RegAddrLo = Address(CGF.Builder.CreateGEP(CGF.Int8Ty, RegSaveArea,
                                                       fp_offset),
-                                CharUnits::fromQuantity(16));
+                                CGF.Int8Ty, CharUnits::fromQuantity(16));
     Address RegAddrHi =
       CGF.Builder.CreateConstInBoundsByteGEP(RegAddrLo,
                                              CharUnits::fromQuantity(16));
@@ -4357,12 +4358,12 @@ ABIArgInfo WinX86_64ABIInfo::classify(QualType Ty, unsigned &FreeSSERegs,
     }
   }
 
-  if (Ty->isExtIntType()) {
+  if (Ty->isBitIntType()) {
     // MS x64 ABI requirement: "Any argument that doesn't fit in 8 bytes, or is
     // not 1, 2, 4, or 8 bytes, must be passed by reference."
-    // However, non-power-of-two _ExtInts will be passed as 1,2,4 or 8 bytes
-    // anyway as long is it fits in them, so we don't have to check the power of
-    // 2.
+    // However, non-power-of-two bit-precise integers will be passed as 1, 2, 4,
+    // or 8 bytes anyway as long is it fits in them, so we don't have to check
+    // the power of 2.
     if (Width <= 64)
       return ABIArgInfo::getDirect();
     return ABIArgInfo::getIndirect(Align, /*ByVal=*/false);
@@ -5069,7 +5070,7 @@ PPC64_SVR4_ABIInfo::isPromotableTypeForABI(QualType Ty) const {
       break;
     }
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() < 64)
       return true;
 
@@ -5083,13 +5084,16 @@ CharUnits PPC64_SVR4_ABIInfo::getParamTypeAlignment(QualType Ty) const {
   if (const ComplexType *CTy = Ty->getAs<ComplexType>())
     Ty = CTy->getElementType();
 
+  auto FloatUsesVector = [this](QualType Ty){
+    return Ty->isRealFloatingType() && &getContext().getFloatTypeSemantics(
+                                           Ty) == &llvm::APFloat::IEEEquad();
+  };
+
   // Only vector types of size 16 bytes need alignment (larger types are
   // passed via reference, smaller types are not aligned).
   if (Ty->isVectorType()) {
     return CharUnits::fromQuantity(getContext().getTypeSize(Ty) == 128 ? 16 : 8);
-  } else if (Ty->isRealFloatingType() &&
-             &getContext().getFloatTypeSemantics(Ty) ==
-                 &llvm::APFloat::IEEEquad()) {
+  } else if (FloatUsesVector(Ty)) {
     // According to ABI document section 'Optional Save Areas': If extended
     // precision floating-point values in IEEE BINARY 128 QUADRUPLE PRECISION
     // format are supported, map them to a single quadword, quadword aligned.
@@ -5116,7 +5120,9 @@ CharUnits PPC64_SVR4_ABIInfo::getParamTypeAlignment(QualType Ty) const {
 
   // With special case aggregates, only vector base types need alignment.
   if (AlignAsType) {
-    return CharUnits::fromQuantity(AlignAsType->isVectorType() ? 16 : 8);
+    bool UsesVector = AlignAsType->isVectorType() ||
+                      FloatUsesVector(QualType(AlignAsType, 0));
+    return CharUnits::fromQuantity(UsesVector ? 16 : 8);
   }
 
   // Otherwise, we only need alignment for any aggregate type that
@@ -5289,7 +5295,7 @@ PPC64_SVR4_ABIInfo::classifyArgumentType(QualType Ty) const {
     }
   }
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() > 128)
       return getNaturalAlignIndirect(Ty, /*ByVal=*/true);
 
@@ -5365,7 +5371,7 @@ PPC64_SVR4_ABIInfo::classifyReturnType(QualType RetTy) const {
     }
   }
 
-  if (const auto *EIT = RetTy->getAs<ExtIntType>())
+  if (const auto *EIT = RetTy->getAs<BitIntType>())
     if (EIT->getNumBits() > 128)
       return getNaturalAlignIndirect(RetTy, /*ByVal=*/false);
 
@@ -5717,7 +5723,7 @@ AArch64ABIInfo::classifyArgumentType(QualType Ty, bool IsVariadic,
     if (const EnumType *EnumTy = Ty->getAs<EnumType>())
       Ty = EnumTy->getDecl()->getIntegerType();
 
-    if (const auto *EIT = Ty->getAs<ExtIntType>())
+    if (const auto *EIT = Ty->getAs<BitIntType>())
       if (EIT->getNumBits() > 128)
         return getNaturalAlignIndirect(Ty);
 
@@ -5819,7 +5825,7 @@ ABIArgInfo AArch64ABIInfo::classifyReturnType(QualType RetTy,
     if (const EnumType *EnumTy = RetTy->getAs<EnumType>())
       RetTy = EnumTy->getDecl()->getIntegerType();
 
-    if (const auto *EIT = RetTy->getAs<ExtIntType>())
+    if (const auto *EIT = RetTy->getAs<BitIntType>())
       if (EIT->getNumBits() > 128)
         return getNaturalAlignIndirect(RetTy);
 
@@ -6561,7 +6567,7 @@ ABIArgInfo ARMABIInfo::classifyArgumentType(QualType Ty, bool isVariadic,
       Ty = EnumTy->getDecl()->getIntegerType();
     }
 
-    if (const auto *EIT = Ty->getAs<ExtIntType>())
+    if (const auto *EIT = Ty->getAs<BitIntType>())
       if (EIT->getNumBits() > 64)
         return getNaturalAlignIndirect(Ty, /*ByVal=*/true);
 
@@ -6763,7 +6769,7 @@ ABIArgInfo ARMABIInfo::classifyReturnType(QualType RetTy, bool isVariadic,
     if (const EnumType *EnumTy = RetTy->getAs<EnumType>())
       RetTy = EnumTy->getDecl()->getIntegerType();
 
-    if (const auto *EIT = RetTy->getAs<ExtIntType>())
+    if (const auto *EIT = RetTy->getAs<BitIntType>())
       if (EIT->getNumBits() > 64)
         return getNaturalAlignIndirect(RetTy, /*ByVal=*/false);
 
@@ -7100,7 +7106,7 @@ bool NVPTXABIInfo::isUnsupportedType(QualType T) const {
       (T->isFloat128Type() ||
        (T->isRealFloatingType() && Context.getTypeSize(T) == 128)))
     return true;
-  if (const auto *EIT = T->getAs<ExtIntType>())
+  if (const auto *EIT = T->getAs<BitIntType>())
     return EIT->getNumBits() >
            (Context.getTargetInfo().hasInt128Type() ? 128U : 64U);
   if (!Context.getTargetInfo().hasInt128Type() && T->isIntegerType() &&
@@ -7177,7 +7183,7 @@ ABIArgInfo NVPTXABIInfo::classifyArgumentType(QualType Ty) const {
     return getNaturalAlignIndirect(Ty, /* byval */ true);
   }
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>()) {
+  if (const auto *EIT = Ty->getAs<BitIntType>()) {
     if ((EIT->getNumBits() > 128) ||
         (!getContext().getTargetInfo().hasInt128Type() &&
          EIT->getNumBits() > 64))
@@ -7391,7 +7397,7 @@ bool SystemZABIInfo::isPromotableIntegerTypeForABI(QualType Ty) const {
   if (ABIInfo::isPromotableIntegerTypeForABI(Ty))
     return true;
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() < 64)
       return true;
 
@@ -7994,7 +8000,7 @@ MipsABIInfo::classifyArgumentType(QualType Ty, uint64_t &Offset) const {
     Ty = EnumTy->getDecl()->getIntegerType();
 
   // Make sure we pass indirectly things that are too large.
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() > 128 ||
         (EIT->getNumBits() > 64 &&
          !getContext().getTargetInfo().hasInt128Type()))
@@ -8085,7 +8091,7 @@ ABIArgInfo MipsABIInfo::classifyReturnType(QualType RetTy) const {
     RetTy = EnumTy->getDecl()->getIntegerType();
 
   // Make sure we pass indirectly things that are too large.
-  if (const auto *EIT = RetTy->getAs<ExtIntType>())
+  if (const auto *EIT = RetTy->getAs<BitIntType>())
     if (EIT->getNumBits() > 128 ||
         (EIT->getNumBits() > 64 &&
          !getContext().getTargetInfo().hasInt128Type()))
@@ -8460,7 +8466,7 @@ ABIArgInfo HexagonABIInfo::classifyArgumentType(QualType Ty,
     if (Size <= 64)
       HexagonAdjustRegsLeft(Size, RegsLeft);
 
-    if (Size > 64 && Ty->isExtIntType())
+    if (Size > 64 && Ty->isBitIntType())
       return getNaturalAlignIndirect(Ty, /*ByVal=*/true);
 
     return isPromotableIntegerTypeForABI(Ty) ? ABIArgInfo::getExtend(Ty)
@@ -8516,7 +8522,7 @@ ABIArgInfo HexagonABIInfo::classifyReturnType(QualType RetTy) const {
     if (const EnumType *EnumTy = RetTy->getAs<EnumType>())
       RetTy = EnumTy->getDecl()->getIntegerType();
 
-    if (Size > 64 && RetTy->isExtIntType())
+    if (Size > 64 && RetTy->isBitIntType())
       return getNaturalAlignIndirect(RetTy, /*ByVal=*/false);
 
     return isPromotableIntegerTypeForABI(RetTy) ? ABIArgInfo::getExtend(RetTy)
@@ -8887,7 +8893,7 @@ ABIArgInfo LanaiABIInfo::classifyArgumentType(QualType Ty,
   bool InReg = shouldUseInReg(Ty, State);
 
   // Don't pass >64 bit integers in registers.
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() > 64)
       return getIndirectResult(Ty, /*ByVal=*/true, State);
 
@@ -9161,6 +9167,10 @@ class AMDGPUTargetCodeGenInfo : public TargetCodeGenInfo {
 public:
   AMDGPUTargetCodeGenInfo(CodeGenTypes &CGT)
       : TargetCodeGenInfo(std::make_unique<AMDGPUABIInfo>(CGT)) {}
+
+  void setFunctionDeclAttributes(const FunctionDecl *FD, llvm::Function *F,
+                                 CodeGenModule &CGM) const;
+
   void setTargetAttributes(const Decl *D, llvm::GlobalValue *GV,
                            CodeGen::CodeGenModule &M) const override;
   unsigned getOpenCLKernelCallingConv() const override;
@@ -9200,36 +9210,13 @@ static bool requiresAMDGPUProtectedVisibility(const Decl *D,
            cast<VarDecl>(D)->getType()->isCUDADeviceBuiltinTextureType()));
 }
 
-void AMDGPUTargetCodeGenInfo::setTargetAttributes(
-    const Decl *D, llvm::GlobalValue *GV, CodeGen::CodeGenModule &M) const {
-  if (requiresAMDGPUProtectedVisibility(D, GV)) {
-    GV->setVisibility(llvm::GlobalValue::ProtectedVisibility);
-    GV->setDSOLocal(true);
-  }
-
-  if (GV->isDeclaration())
-    return;
-  const FunctionDecl *FD = dyn_cast_or_null<FunctionDecl>(D);
-  if (!FD)
-    return;
-
-  llvm::Function *F = cast<llvm::Function>(GV);
-
-  const auto *ReqdWGS = M.getLangOpts().OpenCL ?
-    FD->getAttr<ReqdWorkGroupSizeAttr>() : nullptr;
-
-
-  const bool IsOpenCLKernel = M.getLangOpts().OpenCL &&
-                              FD->hasAttr<OpenCLKernelAttr>();
-  const bool IsHIPKernel = M.getLangOpts().HIP &&
-                           FD->hasAttr<CUDAGlobalAttr>();
-  if ((IsOpenCLKernel || IsHIPKernel) &&
-      (M.getTriple().getOS() == llvm::Triple::AMDHSA))
-    F->addFnAttr("amdgpu-implicitarg-num-bytes", "56");
-
-  if (IsHIPKernel)
-    F->addFnAttr("uniform-work-group-size", "true");
-
+void AMDGPUTargetCodeGenInfo::setFunctionDeclAttributes(
+    const FunctionDecl *FD, llvm::Function *F, CodeGenModule &M) const {
+  const auto *ReqdWGS =
+      M.getLangOpts().OpenCL ? FD->getAttr<ReqdWorkGroupSizeAttr>() : nullptr;
+  const bool IsOpenCLKernel =
+      M.getLangOpts().OpenCL && FD->hasAttr<OpenCLKernelAttr>();
+  const bool IsHIPKernel = M.getLangOpts().HIP && FD->hasAttr<CUDAGlobalAttr>();
 
   const auto *FlatWGS = FD->getAttr<AMDGPUFlatWorkGroupSizeAttr>();
   if (ReqdWGS || FlatWGS) {
@@ -9297,6 +9284,38 @@ void AMDGPUTargetCodeGenInfo::setTargetAttributes(
     if (NumVGPR != 0)
       F->addFnAttr("amdgpu-num-vgpr", llvm::utostr(NumVGPR));
   }
+}
+
+void AMDGPUTargetCodeGenInfo::setTargetAttributes(
+    const Decl *D, llvm::GlobalValue *GV, CodeGen::CodeGenModule &M) const {
+  if (requiresAMDGPUProtectedVisibility(D, GV)) {
+    GV->setVisibility(llvm::GlobalValue::ProtectedVisibility);
+    GV->setDSOLocal(true);
+  }
+
+  if (GV->isDeclaration())
+    return;
+
+  llvm::Function *F = dyn_cast<llvm::Function>(GV);
+  if (!F)
+    return;
+
+  const FunctionDecl *FD = dyn_cast_or_null<FunctionDecl>(D);
+  if (FD)
+    setFunctionDeclAttributes(FD, F, M);
+
+  const bool IsOpenCLKernel =
+      M.getLangOpts().OpenCL && FD && FD->hasAttr<OpenCLKernelAttr>();
+  const bool IsHIPKernel =
+      M.getLangOpts().HIP && FD && FD->hasAttr<CUDAGlobalAttr>();
+
+  const bool IsOpenMP = M.getLangOpts().OpenMP && !FD;
+  if ((IsOpenCLKernel || IsHIPKernel || IsOpenMP) &&
+      (M.getTriple().getOS() == llvm::Triple::AMDHSA))
+    F->addFnAttr("amdgpu-implicitarg-num-bytes", "56");
+
+  if (IsHIPKernel)
+    F->addFnAttr("uniform-work-group-size", "true");
 
   if (M.getContext().getTargetInfo().allowAMDGPUUnsafeFPAtomics())
     F->addFnAttr("amdgpu-unsafe-fp-atomics", "true");
@@ -9343,7 +9362,9 @@ AMDGPUTargetCodeGenInfo::getGlobalVarAddressSpace(CodeGenModule &CGM,
   if (AddrSpace != LangAS::Default)
     return AddrSpace;
 
-  if (CGM.isTypeConstant(D->getType(), false)) {
+  // Only promote to address space 4 if VarDecl has constant initialization.
+  if (CGM.isTypeConstant(D->getType(), false) &&
+      D->hasConstantInitialization()) {
     if (auto ConstAS = CGM.getTarget().getConstantAddressSpace())
       return ConstAS.getValue();
   }
@@ -9606,7 +9627,7 @@ SparcV9ABIInfo::classifyType(QualType Ty, unsigned SizeLimit) const {
   if (Size < 64 && Ty->isIntegerType())
     return ABIArgInfo::getExtend(Ty);
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() < 64)
       return ABIArgInfo::getExtend(Ty);
 
@@ -9860,7 +9881,7 @@ ABIArgInfo ARCABIInfo::classifyArgumentType(QualType Ty,
         ABIArgInfo::getDirect(Result, 0, nullptr, false);
   }
 
-  if (const auto *EIT = Ty->getAs<ExtIntType>())
+  if (const auto *EIT = Ty->getAs<BitIntType>())
     if (EIT->getNumBits() > 64)
       return getIndirectByValue(Ty);
 
@@ -10209,12 +10230,23 @@ public:
 private:
   void setCCs();
 };
+
+class SPIRVABIInfo : public CommonSPIRABIInfo {
+public:
+  SPIRVABIInfo(CodeGenTypes &CGT) : CommonSPIRABIInfo(CGT) {}
+  void computeInfo(CGFunctionInfo &FI) const override;
+
+private:
+  ABIArgInfo classifyKernelArgumentType(QualType Ty) const;
+};
 } // end anonymous namespace
 namespace {
 class CommonSPIRTargetCodeGenInfo : public TargetCodeGenInfo {
 public:
   CommonSPIRTargetCodeGenInfo(CodeGen::CodeGenTypes &CGT)
       : TargetCodeGenInfo(std::make_unique<CommonSPIRABIInfo>(CGT)) {}
+  CommonSPIRTargetCodeGenInfo(std::unique_ptr<ABIInfo> ABIInfo)
+      : TargetCodeGenInfo(std::move(ABIInfo)) {}
 
   LangAS getASTAllocaAddressSpace() const override {
     return getLangASFromTargetAS(
@@ -10223,18 +10255,60 @@ public:
 
   unsigned getOpenCLKernelCallingConv() const override;
 };
-
+class SPIRVTargetCodeGenInfo : public CommonSPIRTargetCodeGenInfo {
+public:
+  SPIRVTargetCodeGenInfo(CodeGen::CodeGenTypes &CGT)
+      : CommonSPIRTargetCodeGenInfo(std::make_unique<SPIRVABIInfo>(CGT)) {}
+  void setCUDAKernelCallingConvention(const FunctionType *&FT) const override;
+};
 } // End anonymous namespace.
+
 void CommonSPIRABIInfo::setCCs() {
   assert(getRuntimeCC() == llvm::CallingConv::C);
   RuntimeCC = llvm::CallingConv::SPIR_FUNC;
 }
 
+ABIArgInfo SPIRVABIInfo::classifyKernelArgumentType(QualType Ty) const {
+  if (getContext().getLangOpts().HIP) {
+    // Coerce pointer arguments with default address space to CrossWorkGroup
+    // pointers for HIPSPV. When the language mode is HIP, the SPIRTargetInfo
+    // maps cuda_device to SPIR-V's CrossWorkGroup address space.
+    llvm::Type *LTy = CGT.ConvertType(Ty);
+    auto DefaultAS = getContext().getTargetAddressSpace(LangAS::Default);
+    auto GlobalAS = getContext().getTargetAddressSpace(LangAS::cuda_device);
+    if (LTy->isPointerTy() && LTy->getPointerAddressSpace() == DefaultAS) {
+      LTy = llvm::PointerType::get(
+          cast<llvm::PointerType>(LTy)->getElementType(), GlobalAS);
+      return ABIArgInfo::getDirect(LTy, 0, nullptr, false);
+    }
+  }
+  return classifyArgumentType(Ty);
+}
+
+void SPIRVABIInfo::computeInfo(CGFunctionInfo &FI) const {
+  // The logic is same as in DefaultABIInfo with an exception on the kernel
+  // arguments handling.
+  llvm::CallingConv::ID CC = FI.getCallingConvention();
+
+  if (!getCXXABI().classifyReturnType(FI))
+    FI.getReturnInfo() = classifyReturnType(FI.getReturnType());
+
+  for (auto &I : FI.arguments()) {
+    if (CC == llvm::CallingConv::SPIR_KERNEL) {
+      I.info = classifyKernelArgumentType(I.type);
+    } else {
+      I.info = classifyArgumentType(I.type);
+    }
+  }
+}
+
 namespace clang {
 namespace CodeGen {
 void computeSPIRKernelABIInfo(CodeGenModule &CGM, CGFunctionInfo &FI) {
-  DefaultABIInfo SPIRABI(CGM.getTypes());
-  SPIRABI.computeInfo(FI);
+  if (CGM.getTarget().getTriple().isSPIRV())
+    SPIRVABIInfo(CGM.getTypes()).computeInfo(FI);
+  else
+    CommonSPIRABIInfo(CGM.getTypes()).computeInfo(FI);
 }
 }
 }
@@ -10243,6 +10317,16 @@ unsigned CommonSPIRTargetCodeGenInfo::getOpenCLKernelCallingConv() const {
   return llvm::CallingConv::SPIR_KERNEL;
 }
 
+void SPIRVTargetCodeGenInfo::setCUDAKernelCallingConvention(
+    const FunctionType *&FT) const {
+  // Convert HIP kernels to SPIR-V kernels.
+  if (getABIInfo().getContext().getLangOpts().HIP) {
+    FT = getABIInfo().getContext().adjustFunctionType(
+        FT, FT->getExtInfo().withCallingConv(CC_OpenCLKernel));
+    return;
+  }
+}
+
 static bool appendType(SmallStringEnc &Enc, QualType QType,
                        const CodeGen::CodeGenModule &CGM,
                        TypeStringCache &TSC);
@@ -10943,7 +11027,7 @@ ABIArgInfo RISCVABIInfo::classifyArgumentType(QualType Ty, bool IsFixed,
       return extendType(Ty);
     }
 
-    if (const auto *EIT = Ty->getAs<ExtIntType>()) {
+    if (const auto *EIT = Ty->getAs<BitIntType>()) {
       if (EIT->getNumBits() < XLen && !MustUseStack)
         return extendType(Ty);
       if (EIT->getNumBits() > 128 ||
@@ -11308,9 +11392,10 @@ const TargetCodeGenInfo &CodeGenModule::getTargetCodeGenInfo() {
     return SetCGInfo(new ARCTargetCodeGenInfo(Types));
   case llvm::Triple::spir:
   case llvm::Triple::spir64:
+    return SetCGInfo(new CommonSPIRTargetCodeGenInfo(Types));
   case llvm::Triple::spirv32:
   case llvm::Triple::spirv64:
-    return SetCGInfo(new CommonSPIRTargetCodeGenInfo(Types));
+    return SetCGInfo(new SPIRVTargetCodeGenInfo(Types));
   case llvm::Triple::ve:
     return SetCGInfo(new VETargetCodeGenInfo(Types));
   }
diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
index d501bd026219..3b551ea94cc2 100644
--- a/clang/lib/Driver/Driver.cpp
+++ b/clang/lib/Driver/Driver.cpp
@@ -23,7 +23,8 @@
 #include "ToolChains/FreeBSD.h"
 #include "ToolChains/Fuchsia.h"
 #include "ToolChains/Gnu.h"
-#include "ToolChains/HIP.h"
+#include "ToolChains/HIPAMD.h"
+#include "ToolChains/HIPSPV.h"
 #include "ToolChains/Haiku.h"
 #include "ToolChains/Hexagon.h"
 #include "ToolChains/Hurd.h"
@@ -42,6 +43,7 @@
 #include "ToolChains/PPCLinux.h"
 #include "ToolChains/PS4CPU.h"
 #include "ToolChains/RISCVToolchain.h"
+#include "ToolChains/SPIRV.h"
 #include "ToolChains/Solaris.h"
 #include "ToolChains/TCE.h"
 #include "ToolChains/VEToolchain.h"
@@ -99,8 +101,39 @@ using namespace clang::driver;
 using namespace clang;
 using namespace llvm::opt;
 
-static llvm::Triple getHIPOffloadTargetTriple() {
-  static const llvm::Triple T("amdgcn-amd-amdhsa");
+static llvm::Optional<llvm::Triple>
+getHIPOffloadTargetTriple(const Driver &D, const ArgList &Args) {
+  if (Args.hasArg(options::OPT_offload_EQ)) {
+    auto HIPOffloadTargets = Args.getAllArgValues(options::OPT_offload_EQ);
+
+    // HIP compilation flow does not support multiple targets for now. We need
+    // the HIPActionBuilder (and possibly the CudaActionBuilder{,Base}too) to
+    // support multiple tool chains first.
+    switch (HIPOffloadTargets.size()) {
+    default:
+      D.Diag(diag::err_drv_only_one_offload_target_supported_in) << "HIP";
+      return llvm::None;
+    case 0:
+      D.Diag(diag::err_drv_invalid_or_unsupported_offload_target) << "";
+      return llvm::None;
+    case 1:
+      break;
+    }
+    llvm::Triple TT(HIPOffloadTargets[0]);
+    if (TT.getArch() == llvm::Triple::amdgcn &&
+        TT.getVendor() == llvm::Triple::AMD &&
+        TT.getOS() == llvm::Triple::AMDHSA)
+      return TT;
+    if (TT.getArch() == llvm::Triple::spirv64 &&
+        TT.getVendor() == llvm::Triple::UnknownVendor &&
+        TT.getOS() == llvm::Triple::UnknownOS)
+      return TT;
+    D.Diag(diag::err_drv_invalid_or_unsupported_offload_target)
+        << HIPOffloadTargets[0];
+    return llvm::None;
+  }
+
+  static const llvm::Triple T("amdgcn-amd-amdhsa"); // Default HIP triple.
   return T;
 }
 
@@ -694,17 +727,14 @@ void Driver::CreateOffloadingDeviceToolChains(Compilation &C,
       return;
     }
     const ToolChain *HostTC = C.getSingleOffloadToolChain<Action::OFK_Host>();
-    const llvm::Triple &HostTriple = HostTC->getTriple();
     auto OFK = Action::OFK_HIP;
-    llvm::Triple HIPTriple = getHIPOffloadTargetTriple();
-    // Use the HIP and host triples as the key into the ToolChains map,
-    // because the device toolchain we create depends on both.
-    auto &HIPTC = ToolChains[HIPTriple.str() + "/" + HostTriple.str()];
-    if (!HIPTC) {
-      HIPTC = std::make_unique<toolchains::HIPToolChain>(
-          *this, HIPTriple, *HostTC, C.getInputArgs());
-    }
-    C.addOffloadDeviceToolChain(HIPTC.get(), OFK);
+    auto HIPTriple = getHIPOffloadTargetTriple(*this, C.getInputArgs());
+    if (!HIPTriple)
+      return;
+    auto *HIPTC = &getOffloadingDeviceToolChain(C.getInputArgs(), *HIPTriple,
+                                                *HostTC, OFK);
+    assert(HIPTC && "Could not create offloading device tool chain.");
+    C.addOffloadDeviceToolChain(HIPTC, OFK);
   }
 
   //
@@ -1545,7 +1575,7 @@ int Driver::ExecuteCompilation(
   if (Diags.hasErrorOccurred())
     return 1;
 
-  // Set up response file names for each command, if necessary
+  // Set up response file names for each command, if necessary.
   for (auto &Job : C.getJobs())
     setUpResponseFiles(C, Job);
 
@@ -2040,11 +2070,7 @@ static bool ContainsCompileOrAssembleAction(const Action *A) {
       isa<AssembleJobAction>(A))
     return true;
 
-  for (const Action *Input : A->inputs())
-    if (ContainsCompileOrAssembleAction(Input))
-      return true;
-
-  return false;
+  return llvm::any_of(A->inputs(), ContainsCompileOrAssembleAction);
 }
 
 void Driver::BuildUniversalActions(Compilation &C, const ToolChain &TC,
@@ -2733,6 +2759,14 @@ class OffloadingActionBuilder final {
         }
       }
 
+      // --offload and --offload-arch options are mutually exclusive.
+      if (Args.hasArgNoClaim(options::OPT_offload_EQ) &&
+          Args.hasArgNoClaim(options::OPT_offload_arch_EQ,
+                             options::OPT_no_offload_arch_EQ)) {
+        C.getDriver().Diag(diag::err_opt_not_valid_with_opt) << "--offload-arch"
+                                                             << "--offload";
+      }
+
       // Collect all cuda_gpu_arch parameters, removing duplicates.
       std::set<StringRef> GpuArchs;
       bool Error = false;
@@ -2775,8 +2809,12 @@ class OffloadingActionBuilder final {
       // Default to sm_20 which is the lowest common denominator for
       // supported GPUs.  sm_20 code should work correctly, if
       // suboptimally, on all newer GPUs.
-      if (GpuArchList.empty())
-        GpuArchList.push_back(DefaultCudaArch);
+      if (GpuArchList.empty()) {
+        if (ToolChains.front()->getTriple().isSPIRV())
+          GpuArchList.push_back(CudaArch::Generic);
+        else
+          GpuArchList.push_back(DefaultCudaArch);
+      }
 
       return Error;
     }
@@ -2937,8 +2975,11 @@ class OffloadingActionBuilder final {
 
     StringRef getCanonicalOffloadArch(StringRef IdStr) override {
       llvm::StringMap<bool> Features;
-      auto ArchStr =
-          parseTargetID(getHIPOffloadTargetTriple(), IdStr, &Features);
+      // getHIPOffloadTargetTriple() is known to return valid value as it has
+      // been called successfully in the CreateOffloadingDeviceToolChains().
+      auto ArchStr = parseTargetID(
+          *getHIPOffloadTargetTriple(C.getDriver(), C.getInputArgs()), IdStr,
+          &Features);
       if (!ArchStr) {
         C.getDriver().Diag(clang::diag::err_drv_bad_target_id) << IdStr;
         C.setContainsError();
@@ -2992,9 +3033,19 @@ class OffloadingActionBuilder final {
             // When LTO is not enabled, we follow the conventional
             // compiler phases, including backend and assemble phases.
             ActionList AL;
-            auto BackendAction = C.getDriver().ConstructPhaseAction(
-                C, Args, phases::Backend, CudaDeviceActions[I],
-                AssociatedOffloadKind);
+            Action *BackendAction = nullptr;
+            if (ToolChains.front()->getTriple().isSPIRV()) {
+              // Emit LLVM bitcode for SPIR-V targets. SPIR-V device tool chain
+              // (HIPSPVToolChain) runs post-link LLVM IR passes.
+              types::ID Output = Args.hasArg(options::OPT_S)
+                                     ? types::TY_LLVM_IR
+                                     : types::TY_LLVM_BC;
+              BackendAction =
+                  C.MakeAction<BackendJobAction>(CudaDeviceActions[I], Output);
+            } else
+              BackendAction = C.getDriver().ConstructPhaseAction(
+                  C, Args, phases::Backend, CudaDeviceActions[I],
+                  AssociatedOffloadKind);
             auto AssembleAction = C.getDriver().ConstructPhaseAction(
                 C, Args, phases::Assemble, BackendAction,
                 AssociatedOffloadKind);
@@ -3724,6 +3775,14 @@ void Driver::BuildActions(Compilation &C, DerivedArgList &Args,
     }
   }
 
+  // FIXME: Linking separate translation units for SPIR-V is not supported yet.
+  // It can be done either by LLVM IR linking before conversion of the final
+  // linked module to SPIR-V or external SPIR-V linkers can be used e.g.
+  // spirv-link.
+  if (C.getDefaultToolChain().getTriple().isSPIRV() && Inputs.size() > 1) {
+    Diag(clang::diag::warn_drv_spirv_linking_multiple_inputs_unsupported);
+  }
+
   handleArguments(C, Args, Inputs, Actions);
 
   // Builder to be used to build offloading actions.
@@ -3763,8 +3822,15 @@ void Driver::BuildActions(Compilation &C, DerivedArgList &Args,
       // Queue linker inputs.
       if (Phase == phases::Link) {
         assert(Phase == PL.back() && "linking must be final compilation step.");
-        LinkerInputs.push_back(Current);
-        Current = nullptr;
+        // Compilation phases are setup per language, however for SPIR-V the
+        // final linking phase is meaningless since the compilation phase
+        // produces the final binary.
+        // FIXME: OpenCL - we could strip linking phase out from OpenCL
+        // compilation phases if we could verify it is not needed by any target.
+        if (!C.getDefaultToolChain().getTriple().isSPIRV()) {
+          LinkerInputs.push_back(Current);
+          Current = nullptr;
+        }
         break;
       }
 
@@ -4337,6 +4403,12 @@ class ToolSelector final {
     if (!T)
       return nullptr;
 
+    // Can't collapse if we don't have codegen support unless we are
+    // emitting LLVM IR.
+    bool OutputIsLLVM = types::isLLVMIR(ActionInfo[0].JA->getType());
+    if (!T->hasIntegratedBackend() && !(OutputIsLLVM && T->canEmitIR()))
+      return nullptr;
+
     // When using -fembed-bitcode, it is required to have the same tool (clang)
     // for both CompilerJA and BackendJA. Otherwise, combine two stages.
     if (EmbedBitcode) {
@@ -4406,6 +4478,12 @@ class ToolSelector final {
     if (!T)
       return nullptr;
 
+    // Can't collapse if we don't have codegen support unless we are
+    // emitting LLVM IR.
+    bool OutputIsLLVM = types::isLLVMIR(ActionInfo[0].JA->getType());
+    if (!T->hasIntegratedBackend() && !(OutputIsLLVM && T->canEmitIR()))
+      return nullptr;
+
     if (T->canEmitIR() && ((SaveTemps && !InputIsBitcode) || EmbedBitcode))
       return nullptr;
 
@@ -5429,6 +5507,10 @@ const ToolChain &Driver::getToolChain(const ArgList &Args,
       case llvm::Triple::ve:
         TC = std::make_unique<toolchains::VEToolChain>(*this, Target, Args);
         break;
+      case llvm::Triple::spirv32:
+      case llvm::Triple::spirv64:
+        TC = std::make_unique<toolchains::SPIRVToolChain>(*this, Target, Args);
+        break;
       default:
         if (Target.getVendor() == llvm::Triple::Myriad)
           TC = std::make_unique<toolchains::MyriadToolChain>(*this, Target,
@@ -5453,6 +5535,38 @@ const ToolChain &Driver::getToolChain(const ArgList &Args,
   return *TC;
 }
 
+const ToolChain &Driver::getOffloadingDeviceToolChain(
+    const ArgList &Args, const llvm::Triple &Target, const ToolChain &HostTC,
+    const Action::OffloadKind &TargetDeviceOffloadKind) const {
+  // Use device / host triples as the key into the ToolChains map because the
+  // device ToolChain we create depends on both.
+  auto &TC = ToolChains[Target.str() + "/" + HostTC.getTriple().str()];
+  if (!TC) {
+    // Categorized by offload kind > arch rather than OS > arch like
+    // the normal getToolChain call, as it seems a reasonable way to categorize
+    // things.
+    switch (TargetDeviceOffloadKind) {
+    case Action::OFK_HIP: {
+      if (Target.getArch() == llvm::Triple::amdgcn &&
+          Target.getVendor() == llvm::Triple::AMD &&
+          Target.getOS() == llvm::Triple::AMDHSA)
+        TC = std::make_unique<toolchains::HIPAMDToolChain>(*this, Target,
+                                                           HostTC, Args);
+      else if (Target.getArch() == llvm::Triple::spirv64 &&
+               Target.getVendor() == llvm::Triple::UnknownVendor &&
+               Target.getOS() == llvm::Triple::UnknownOS)
+        TC = std::make_unique<toolchains::HIPSPVToolChain>(*this, Target,
+                                                           HostTC, Args);
+      break;
+    }
+    default:
+      break;
+    }
+  }
+
+  return *TC;
+}
+
 bool Driver::ShouldUseClangCompiler(const JobAction &JA) const {
   // Say "no" if there is not exactly one input of a type clang understands.
   if (JA.size() != 1 ||
diff --git a/clang/lib/Driver/Job.cpp b/clang/lib/Driver/Job.cpp
index 5b87106b6565..f63763effaff 100644
--- a/clang/lib/Driver/Job.cpp
+++ b/clang/lib/Driver/Job.cpp
@@ -388,6 +388,8 @@ int CC1Command::Execute(ArrayRef<llvm::Optional<StringRef>> Redirects,
   Argv.push_back(getExecutable());
   Argv.append(getArguments().begin(), getArguments().end());
   Argv.push_back(nullptr);
+  Argv.pop_back(); // The terminating null element shall not be part of the
+                   // slice (main() behavior).
 
   // This flag simply indicates that the program couldn't start, which isn't
   // applicable here.
diff --git a/clang/lib/Driver/ToolChain.cpp b/clang/lib/Driver/ToolChain.cpp
index 6588cdf9fecd..50c89aaadc18 100644
--- a/clang/lib/Driver/ToolChain.cpp
+++ b/clang/lib/Driver/ToolChain.cpp
@@ -260,7 +260,7 @@ bool ToolChain::IsUnwindTablesDefault(const ArgList &Args) const {
 
 Tool *ToolChain::getClang() const {
   if (!Clang)
-    Clang.reset(new tools::Clang(*this));
+    Clang.reset(new tools::Clang(*this, useIntegratedBackend()));
   return Clang.get();
 }
 
@@ -541,12 +541,9 @@ std::string ToolChain::GetProgramPath(const char *Name) const {
   return D.GetProgramPath(Name, *this);
 }
 
-std::string ToolChain::GetLinkerPath(bool *LinkerIsLLD,
-                                     bool *LinkerIsLLDDarwinNew) const {
+std::string ToolChain::GetLinkerPath(bool *LinkerIsLLD) const {
   if (LinkerIsLLD)
     *LinkerIsLLD = false;
-  if (LinkerIsLLDDarwinNew)
-    *LinkerIsLLDDarwinNew = false;
 
   // Get -fuse-ld= first to prevent -Wunused-command-line-argument. -fuse-ld= is
   // considered as the linker flavor, e.g. "bfd", "gold", or "lld".
@@ -599,11 +596,8 @@ std::string ToolChain::GetLinkerPath(bool *LinkerIsLLD,
 
     std::string LinkerPath(GetProgramPath(LinkerName.c_str()));
     if (llvm::sys::fs::can_execute(LinkerPath)) {
-      // FIXME: Remove LinkerIsLLDDarwinNew once there's only one MachO lld.
       if (LinkerIsLLD)
-        *LinkerIsLLD = UseLinker == "lld" || UseLinker == "lld.darwinold";
-      if (LinkerIsLLDDarwinNew)
-        *LinkerIsLLDDarwinNew = UseLinker == "lld";
+        *LinkerIsLLD = UseLinker == "lld";
       return LinkerPath;
     }
   }
diff --git a/clang/lib/Driver/ToolChains/AMDGPU.cpp b/clang/lib/Driver/ToolChains/AMDGPU.cpp
index b5eaf1adca6b..43ce33750eba 100644
--- a/clang/lib/Driver/ToolChains/AMDGPU.cpp
+++ b/clang/lib/Driver/ToolChains/AMDGPU.cpp
@@ -478,7 +478,8 @@ void RocmInstallationDetector::print(raw_ostream &OS) const {
 
 void RocmInstallationDetector::AddHIPIncludeArgs(const ArgList &DriverArgs,
                                                  ArgStringList &CC1Args) const {
-  bool UsesRuntimeWrapper = VersionMajorMinor > llvm::VersionTuple(3, 5);
+  bool UsesRuntimeWrapper = VersionMajorMinor > llvm::VersionTuple(3, 5) &&
+                            !DriverArgs.hasArg(options::OPT_nohipwrapperinc);
 
   if (!DriverArgs.hasArg(options::OPT_nobuiltininc)) {
     // HIP header includes standard library wrapper headers under clang
diff --git a/clang/lib/Driver/ToolChains/AMDGPUOpenMP.cpp b/clang/lib/Driver/ToolChains/AMDGPUOpenMP.cpp
index 863e2c597d53..f282f04b7931 100644
--- a/clang/lib/Driver/ToolChains/AMDGPUOpenMP.cpp
+++ b/clang/lib/Driver/ToolChains/AMDGPUOpenMP.cpp
@@ -267,7 +267,7 @@ void AMDGPUOpenMPToolChain::addClangTargetOptions(
 
   std::string BitcodeSuffix;
   if (DriverArgs.hasFlag(options::OPT_fopenmp_target_new_runtime,
-                         options::OPT_fno_openmp_target_new_runtime, false))
+                         options::OPT_fno_openmp_target_new_runtime, true))
     BitcodeSuffix = "new-amdgpu-" + GPUArch;
   else
     BitcodeSuffix = "amdgcn-" + GPUArch;
diff --git a/clang/lib/Driver/ToolChains/Arch/AArch64.cpp b/clang/lib/Driver/ToolChains/Arch/AArch64.cpp
index abc32f22d2a1..be13d6d583ce 100644
--- a/clang/lib/Driver/ToolChains/Arch/AArch64.cpp
+++ b/clang/lib/Driver/ToolChains/Arch/AArch64.cpp
@@ -568,4 +568,15 @@ fp16_fml_fallthrough:
 
   if (Args.hasArg(options::OPT_mno_neg_immediates))
     Features.push_back("+no-neg-immediates");
+
+  if (Arg *A = Args.getLastArg(options::OPT_mfix_cortex_a53_835769,
+                               options::OPT_mno_fix_cortex_a53_835769)) {
+    if (A->getOption().matches(options::OPT_mfix_cortex_a53_835769))
+      Features.push_back("+fix-cortex-a53-835769");
+    else
+      Features.push_back("-fix-cortex-a53-835769");
+  } else if (Triple.isAndroid()) {
+    // Enabled A53 errata (835769) workaround by default on android
+    Features.push_back("+fix-cortex-a53-835769");
+  }
 }
diff --git a/clang/lib/Driver/ToolChains/Arch/ARM.cpp b/clang/lib/Driver/ToolChains/Arch/ARM.cpp
index 21c091e1a0ba..4013cf230026 100644
--- a/clang/lib/Driver/ToolChains/Arch/ARM.cpp
+++ b/clang/lib/Driver/ToolChains/Arch/ARM.cpp
@@ -147,19 +147,27 @@ bool arm::useAAPCSForMachO(const llvm::Triple &T) {
          T.getOS() == llvm::Triple::UnknownOS || isARMMProfile(T);
 }
 
+// We follow GCC and support when the backend has support for the MRC/MCR
+// instructions that are used to set the hard thread pointer ("CP15 C13
+// Thread id").
+bool arm::isHardTPSupported(const llvm::Triple &Triple) {
+  int Ver = getARMSubArchVersionNumber(Triple);
+  llvm::ARM::ArchKind AK = llvm::ARM::parseArch(Triple.getArchName());
+  return Triple.isARM() || AK == llvm::ARM::ArchKind::ARMV6T2 ||
+         (Ver >= 7 && AK != llvm::ARM::ArchKind::ARMV8MBaseline);
+}
+
 // Select mode for reading thread pointer (-mtp=soft/cp15).
 arm::ReadTPMode arm::getReadTPMode(const Driver &D, const ArgList &Args,
-                                   const llvm::Triple &Triple) {
+                                   const llvm::Triple &Triple, bool ForAS) {
   if (Arg *A = Args.getLastArg(options::OPT_mtp_mode_EQ)) {
     arm::ReadTPMode ThreadPointer =
         llvm::StringSwitch<arm::ReadTPMode>(A->getValue())
             .Case("cp15", ReadTPMode::Cp15)
             .Case("soft", ReadTPMode::Soft)
             .Default(ReadTPMode::Invalid);
-    if (ThreadPointer == ReadTPMode::Cp15 &&
-        getARMSubArchVersionNumber(Triple) < 7 &&
-        llvm::ARM::parseArch(Triple.getArchName()) !=
-            llvm::ARM::ArchKind::ARMV6T2) {
+    if (ThreadPointer == ReadTPMode::Cp15 && !isHardTPSupported(Triple) &&
+        !ForAS) {
       D.Diag(diag::err_target_unsupported_tp_hard) << Triple.getArchName();
       return ReadTPMode::Invalid;
     }
@@ -430,7 +438,6 @@ void arm::getARMTargetFeatures(const Driver &D, const llvm::Triple &Triple,
   bool KernelOrKext =
       Args.hasArg(options::OPT_mkernel, options::OPT_fapple_kext);
   arm::FloatABI ABI = arm::getARMFloatABI(D, Triple, Args);
-  arm::ReadTPMode ThreadPointer = arm::getReadTPMode(D, Args, Triple);
   llvm::Optional<std::pair<const Arg *, StringRef>> WaCPU, WaFPU, WaHDiv,
       WaArch;
 
@@ -482,7 +489,7 @@ void arm::getARMTargetFeatures(const Driver &D, const llvm::Triple &Triple,
     }
   }
 
-  if (ThreadPointer == arm::ReadTPMode::Cp15)
+  if (getReadTPMode(D, Args, Triple, ForAS) == ReadTPMode::Cp15)
     Features.push_back("+read-tp-hard");
 
   const Arg *ArchArg = Args.getLastArg(options::OPT_march_EQ);
@@ -869,6 +876,8 @@ fp16_fml_fallthrough:
     }
   }
 
+  if (Args.getLastArg(options::OPT_mno_bti_at_return_twice))
+    Features.push_back("+no-bti-at-return-twice");
 }
 
 std::string arm::getARMArch(StringRef Arch, const llvm::Triple &Triple) {
diff --git a/clang/lib/Driver/ToolChains/Arch/ARM.h b/clang/lib/Driver/ToolChains/Arch/ARM.h
index b6fd68fbb9c6..881b63bd36b9 100644
--- a/clang/lib/Driver/ToolChains/Arch/ARM.h
+++ b/clang/lib/Driver/ToolChains/Arch/ARM.h
@@ -53,8 +53,9 @@ FloatABI getARMFloatABI(const Driver &D, const llvm::Triple &Triple,
                         const llvm::opt::ArgList &Args);
 void setFloatABIInTriple(const Driver &D, const llvm::opt::ArgList &Args,
                          llvm::Triple &triple);
+bool isHardTPSupported(const llvm::Triple &Triple);
 ReadTPMode getReadTPMode(const Driver &D, const llvm::opt::ArgList &Args,
-                         const llvm::Triple &Triple);
+                         const llvm::Triple &Triple, bool ForAS);
 void setArchNameInTriple(const Driver &D, const llvm::opt::ArgList &Args,
                          types::ID InputType, llvm::Triple &Triple);
 
diff --git a/clang/lib/Driver/ToolChains/Arch/RISCV.cpp b/clang/lib/Driver/ToolChains/Arch/RISCV.cpp
index 323f588c8269..7ad8ca69bed5 100644
--- a/clang/lib/Driver/ToolChains/Arch/RISCV.cpp
+++ b/clang/lib/Driver/ToolChains/Arch/RISCV.cpp
@@ -194,27 +194,11 @@ StringRef riscv::getRISCVABI(const ArgList &Args, const llvm::Triple &Triple) {
 
   auto ParseResult = llvm::RISCVISAInfo::parseArchString(
       Arch, /* EnableExperimentalExtension */ true);
-  if (!ParseResult) {
+  if (!ParseResult)
     // Ignore parsing error, just go 3rd step.
     consumeError(ParseResult.takeError());
-  } else {
-    auto &ISAInfo = *ParseResult;
-    bool HasD = ISAInfo->hasExtension("d");
-    unsigned XLen = ISAInfo->getXLen();
-    if (XLen == 32) {
-      bool HasE = ISAInfo->hasExtension("e");
-      if (HasD)
-        return "ilp32d";
-      if (HasE)
-        return "ilp32e";
-      return "ilp32";
-    } else if (XLen == 64) {
-      if (HasD)
-        return "lp64d";
-      return "lp64";
-    }
-    llvm_unreachable("unhandled XLen");
-  }
+  else
+    return llvm::RISCV::computeDefaultABIFromArch(**ParseResult);
 
   // 3. Choose a default based on the triple
   //
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index c5aaa067c4f5..65347a38490e 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -625,8 +625,9 @@ getFramePointerKind(const ArgList &Args, const llvm::Triple &Triple) {
 }
 
 /// Add a CC1 option to specify the debug compilation directory.
-static void addDebugCompDirArg(const ArgList &Args, ArgStringList &CmdArgs,
-                               const llvm::vfs::FileSystem &VFS) {
+static const char *addDebugCompDirArg(const ArgList &Args,
+                                      ArgStringList &CmdArgs,
+                                      const llvm::vfs::FileSystem &VFS) {
   if (Arg *A = Args.getLastArg(options::OPT_ffile_compilation_dir_EQ,
                                options::OPT_fdebug_compilation_dir_EQ)) {
     if (A->getOption().matches(options::OPT_ffile_compilation_dir_EQ))
@@ -638,6 +639,31 @@ static void addDebugCompDirArg(const ArgList &Args, ArgStringList &CmdArgs,
                  VFS.getCurrentWorkingDirectory()) {
     CmdArgs.push_back(Args.MakeArgString("-fdebug-compilation-dir=" + *CWD));
   }
+  StringRef Path(CmdArgs.back());
+  return Path.substr(Path.find('=') + 1).data();
+}
+
+static void addDebugObjectName(const ArgList &Args, ArgStringList &CmdArgs,
+                               const char *DebugCompilationDir,
+                               const char *OutputFileName) {
+  // No need to generate a value for -object-file-name if it was provided.
+  for (auto *Arg : Args.filtered(options::OPT_Xclang))
+    if (StringRef(Arg->getValue()).startswith("-object-file-name"))
+      return;
+
+  if (Args.hasArg(options::OPT_object_file_name_EQ))
+    return;
+
+  SmallString<128> ObjFileNameForDebug(OutputFileName);
+  if (ObjFileNameForDebug != "-" &&
+      !llvm::sys::path::is_absolute(ObjFileNameForDebug) &&
+      (!DebugCompilationDir ||
+       llvm::sys::path::is_absolute(DebugCompilationDir))) {
+    // Make the path absolute in the debug infos like MSVC does.
+    llvm::sys::fs::make_absolute(ObjFileNameForDebug);
+  }
+  CmdArgs.push_back(
+      Args.MakeArgString(Twine("-object-file-name=") + ObjFileNameForDebug));
 }
 
 /// Add a CC1 and CC1AS option to specify the debug file path prefix map.
@@ -976,11 +1002,7 @@ static bool ContainsCompileAction(const Action *A) {
   if (isa<CompileJobAction>(A) || isa<BackendJobAction>(A))
     return true;
 
-  for (const auto &AI : A->inputs())
-    if (ContainsCompileAction(AI))
-      return true;
-
-  return false;
+  return llvm::any_of(A->inputs(), ContainsCompileAction);
 }
 
 /// Check if -relax-all should be passed to the internal assembler.
@@ -1806,19 +1828,6 @@ void Clang::AddAArch64TargetArgs(const ArgList &Args,
 
   RenderAArch64ABI(Triple, Args, CmdArgs);
 
-  if (Arg *A = Args.getLastArg(options::OPT_mfix_cortex_a53_835769,
-                               options::OPT_mno_fix_cortex_a53_835769)) {
-    CmdArgs.push_back("-mllvm");
-    if (A->getOption().matches(options::OPT_mfix_cortex_a53_835769))
-      CmdArgs.push_back("-aarch64-fix-cortex-a53-835769=1");
-    else
-      CmdArgs.push_back("-aarch64-fix-cortex-a53-835769=0");
-  } else if (Triple.isAndroid()) {
-    // Enabled A53 errata (835769) workaround by default on android
-    CmdArgs.push_back("-mllvm");
-    CmdArgs.push_back("-aarch64-fix-cortex-a53-835769=1");
-  }
-
   // Forward the -mglobal-merge option for explicit control over the pass.
   if (Arg *A = Args.getLastArg(options::OPT_mglobal_merge,
                                options::OPT_mno_global_merge)) {
@@ -5666,7 +5675,8 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
     CmdArgs.push_back("-fno-autolink");
 
   // Add in -fdebug-compilation-dir if necessary.
-  addDebugCompDirArg(Args, CmdArgs, D.getVFS());
+  const char *DebugCompilationDir =
+      addDebugCompDirArg(Args, CmdArgs, D.getVFS());
 
   addDebugPrefixMapArg(D, Args, CmdArgs);
 
@@ -5904,7 +5914,7 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
       // runtime.
       if (Args.hasFlag(options::OPT_fopenmp_target_new_runtime,
                        options::OPT_fno_openmp_target_new_runtime,
-                       /*Default=*/!getToolChain().getTriple().isAMDGCN()))
+                       /*Default=*/true))
         CmdArgs.push_back("-fopenmp-target-new-runtime");
 
       // When in OpenMP offloading mode, enable debugging on the device.
@@ -6997,18 +7007,18 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
 
   if (Arg *A = Args.getLastArg(options::OPT_moutline_atomics,
                                options::OPT_mno_outline_atomics)) {
-    if (A->getOption().matches(options::OPT_moutline_atomics)) {
-      // Option -moutline-atomics supported for AArch64 target only.
-      if (!Triple.isAArch64()) {
-        D.Diag(diag::warn_drv_moutline_atomics_unsupported_opt)
-            << Triple.getArchName();
-      } else {
+    // Option -moutline-atomics supported for AArch64 target only.
+    if (!Triple.isAArch64()) {
+      D.Diag(diag::warn_drv_moutline_atomics_unsupported_opt)
+          << Triple.getArchName() << A->getOption().getName();
+    } else {
+      if (A->getOption().matches(options::OPT_moutline_atomics)) {
         CmdArgs.push_back("-target-feature");
         CmdArgs.push_back("+outline-atomics");
+      } else {
+        CmdArgs.push_back("-target-feature");
+        CmdArgs.push_back("-outline-atomics");
       }
-    } else {
-      CmdArgs.push_back("-target-feature");
-      CmdArgs.push_back("-outline-atomics");
     }
   } else if (Triple.isAArch64() &&
              getToolChain().IsAArch64OutlineAtomicsDefault(Args)) {
@@ -7038,6 +7048,11 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
     CmdArgs.push_back(Args.MakeArgString(Str));
   }
 
+  // Add the output path to the object file for CodeView debug infos.
+  if (EmitCodeView && Output.isFilename())
+    addDebugObjectName(Args, CmdArgs, DebugCompilationDir,
+                       Output.getFilename());
+
   // Add the "-o out -x type src.c" flags last. This is done primarily to make
   // the -cc1 command easier to edit when reproducing compiler crashes.
   if (Output.getType() == types::TY_Dependencies) {
@@ -7111,11 +7126,11 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
   Args.ClaimAllArgs(options::OPT_emit_llvm);
 }
 
-Clang::Clang(const ToolChain &TC)
+Clang::Clang(const ToolChain &TC, bool HasIntegratedBackend)
     // CAUTION! The first constructor argument ("clang") is not arbitrary,
     // as it is for other tools. Some operations on a Tool actually test
     // whether that tool is Clang based on the Tool's Name as a string.
-    : Tool("clang", "clang frontend", TC) {}
+    : Tool("clang", "clang frontend", TC), HasBackend(HasIntegratedBackend) {}
 
 Clang::~Clang() {}
 
@@ -7655,11 +7670,14 @@ void ClangAs::ConstructJob(Compilation &C, const JobAction &JA,
   Args.AddAllArgs(CmdArgs, options::OPT_I_Group);
 
   // Determine the original source input.
-  const Action *SourceAction = &JA;
-  while (SourceAction->getKind() != Action::InputClass) {
-    assert(!SourceAction->getInputs().empty() && "unexpected root action!");
-    SourceAction = SourceAction->getInputs()[0];
-  }
+  auto FindSource = [](const Action *S) -> const Action * {
+    while (S->getKind() != Action::InputClass) {
+      assert(!S->getInputs().empty() && "unexpected root action!");
+      S = S->getInputs()[0];
+    }
+    return S;
+  };
+  const Action *SourceAction = FindSource(&JA);
 
   // Forward -g and handle debug info related flags, assuming we are dealing
   // with an actual assembly file.
@@ -7678,6 +7696,10 @@ void ClangAs::ConstructJob(Compilation &C, const JobAction &JA,
 
   codegenoptions::DebugInfoKind DebugInfoKind = codegenoptions::NoDebugInfo;
 
+  // Add the -fdebug-compilation-dir flag if needed.
+  const char *DebugCompilationDir =
+      addDebugCompDirArg(Args, CmdArgs, C.getDriver().getVFS());
+
   if (SourceAction->getType() == types::TY_Asm ||
       SourceAction->getType() == types::TY_PP_Asm) {
     // You might think that it would be ok to set DebugInfoKind outside of
@@ -7686,8 +7708,6 @@ void ClangAs::ConstructJob(Compilation &C, const JobAction &JA,
     // and it's not clear whether that test is just overly restrictive.
     DebugInfoKind = (WantDebug ? codegenoptions::DebugInfoConstructor
                                : codegenoptions::NoDebugInfo);
-    // Add the -fdebug-compilation-dir flag if needed.
-    addDebugCompDirArg(Args, CmdArgs, C.getDriver().getVFS());
 
     addDebugPrefixMapArg(getToolChain().getDriver(), Args, CmdArgs);
 
@@ -7798,6 +7818,29 @@ void ClangAs::ConstructJob(Compilation &C, const JobAction &JA,
 
   Args.AddAllArgs(CmdArgs, options::OPT_mllvm);
 
+  if (DebugInfoKind > codegenoptions::NoDebugInfo && Output.isFilename())
+    addDebugObjectName(Args, CmdArgs, DebugCompilationDir,
+                       Output.getFilename());
+
+  // Fixup any previous commands that use -object-file-name because when we
+  // generated them, the final .obj name wasn't yet known.
+  for (Command &J : C.getJobs()) {
+    if (SourceAction != FindSource(&J.getSource()))
+      continue;
+    auto &JArgs = J.getArguments();
+    for (unsigned I = 0; I < JArgs.size(); ++I) {
+      if (StringRef(JArgs[I]).startswith("-object-file-name=") &&
+          Output.isFilename()) {
+        ArgStringList NewArgs(JArgs.begin(), JArgs.begin() + I);
+        addDebugObjectName(Args, NewArgs, DebugCompilationDir,
+                           Output.getFilename());
+        NewArgs.append(JArgs.begin() + I + 1, JArgs.end());
+        J.replaceArguments(NewArgs);
+        break;
+      }
+    }
+  }
+
   assert(Output.isFilename() && "Unexpected lipo output.");
   CmdArgs.push_back("-o");
   CmdArgs.push_back(Output.getFilename());
@@ -7878,7 +7921,7 @@ void OffloadBundler::ConstructJob(Compilation &C, const JobAction &JA,
     Triples += '-';
     Triples += CurTC->getTriple().normalize();
     if ((CurKind == Action::OFK_HIP || CurKind == Action::OFK_Cuda) &&
-        CurDep->getOffloadingArch()) {
+        !StringRef(CurDep->getOffloadingArch()).empty()) {
       Triples += '-';
       Triples += CurDep->getOffloadingArch();
     }
diff --git a/clang/lib/Driver/ToolChains/Clang.h b/clang/lib/Driver/ToolChains/Clang.h
index d4b4988b4a8c..00e0490e069b 100644
--- a/clang/lib/Driver/ToolChains/Clang.h
+++ b/clang/lib/Driver/ToolChains/Clang.h
@@ -26,6 +26,10 @@ namespace tools {
 
 /// Clang compiler tool.
 class LLVM_LIBRARY_VISIBILITY Clang : public Tool {
+  // Indicates whether this instance has integrated backend using
+  // internal LLVM infrastructure.
+  bool HasBackend;
+
 public:
   static const char *getBaseInputName(const llvm::opt::ArgList &Args,
                                       const InputInfo &Input);
@@ -99,11 +103,12 @@ private:
       const InputInfo &Input, const llvm::opt::ArgList &Args) const;
 
 public:
-  Clang(const ToolChain &TC);
+  Clang(const ToolChain &TC, bool HasIntegratedBackend = true);
   ~Clang() override;
 
   bool hasGoodDiagnostics() const override { return true; }
   bool hasIntegratedAssembler() const override { return true; }
+  bool hasIntegratedBackend() const override { return HasBackend; }
   bool hasIntegratedCPP() const override { return true; }
   bool canEmitIR() const override { return true; }
 
diff --git a/clang/lib/Driver/ToolChains/CommonArgs.cpp b/clang/lib/Driver/ToolChains/CommonArgs.cpp
index 630baf9d6ae6..407f81a2ae09 100644
--- a/clang/lib/Driver/ToolChains/CommonArgs.cpp
+++ b/clang/lib/Driver/ToolChains/CommonArgs.cpp
@@ -15,7 +15,7 @@
 #include "Arch/SystemZ.h"
 #include "Arch/VE.h"
 #include "Arch/X86.h"
-#include "HIP.h"
+#include "HIPAMD.h"
 #include "Hexagon.h"
 #include "clang/Basic/CharInfo.h"
 #include "clang/Basic/LangOptions.h"
diff --git a/clang/lib/Driver/ToolChains/Darwin.cpp b/clang/lib/Driver/ToolChains/Darwin.cpp
index 06d3edc70e45..f7da3f187814 100644
--- a/clang/lib/Driver/ToolChains/Darwin.cpp
+++ b/clang/lib/Driver/ToolChains/Darwin.cpp
@@ -209,8 +209,7 @@ static bool shouldLinkerNotDedup(bool IsLinkerOnlyAction, const ArgList &Args) {
 void darwin::Linker::AddLinkArgs(Compilation &C, const ArgList &Args,
                                  ArgStringList &CmdArgs,
                                  const InputInfoList &Inputs,
-                                 unsigned Version[5], bool LinkerIsLLD,
-                                 bool LinkerIsLLDDarwinNew) const {
+                                 unsigned Version[5], bool LinkerIsLLD) const {
   const Driver &D = getToolChain().getDriver();
   const toolchains::MachO &MachOTC = getMachOToolChain();
 
@@ -343,7 +342,7 @@ void darwin::Linker::AddLinkArgs(Compilation &C, const ArgList &Args,
   Args.AddAllArgs(CmdArgs, options::OPT_init);
 
   // Add the deployment target.
-  if (Version[0] >= 520 || LinkerIsLLDDarwinNew)
+  if (Version[0] >= 520 || LinkerIsLLD)
     MachOTC.addPlatformVersionArgs(Args, CmdArgs);
   else
     MachOTC.addMinVersionArgs(Args, CmdArgs);
@@ -556,14 +555,13 @@ void darwin::Linker::ConstructJob(Compilation &C, const JobAction &JA,
           << A->getAsString(Args);
   }
 
-  bool LinkerIsLLD, LinkerIsLLDDarwinNew;
-  const char *Exec = Args.MakeArgString(
-      getToolChain().GetLinkerPath(&LinkerIsLLD, &LinkerIsLLDDarwinNew));
+  bool LinkerIsLLD;
+  const char *Exec =
+      Args.MakeArgString(getToolChain().GetLinkerPath(&LinkerIsLLD));
 
   // I'm not sure why this particular decomposition exists in gcc, but
   // we follow suite for ease of comparison.
-  AddLinkArgs(C, Args, CmdArgs, Inputs, Version, LinkerIsLLD,
-              LinkerIsLLDDarwinNew);
+  AddLinkArgs(C, Args, CmdArgs, Inputs, Version, LinkerIsLLD);
 
   if (willEmitRemarks(Args) &&
       checkRemarksOptions(getToolChain().getDriver(), Args,
@@ -715,7 +713,7 @@ void darwin::Linker::ConstructJob(Compilation &C, const JobAction &JA,
   }
 
   ResponseFileSupport ResponseSupport;
-  if (Version[0] >= 705 || LinkerIsLLDDarwinNew) {
+  if (Version[0] >= 705 || LinkerIsLLD) {
     ResponseSupport = ResponseFileSupport::AtFileUTF8();
   } else {
     // For older versions of the linker, use the legacy filelist method instead.
@@ -1412,8 +1410,8 @@ static std::string getSystemOrSDKMacOSVersion(StringRef MacOSSDKVersion) {
   llvm::Triple SystemTriple(llvm::sys::getProcessTriple());
   if (!SystemTriple.isMacOSX())
     return std::string(MacOSSDKVersion);
-  SystemTriple.getMacOSXVersion(Major, Minor, Micro);
-  VersionTuple SystemVersion(Major, Minor, Micro);
+  VersionTuple SystemVersion;
+  SystemTriple.getMacOSXVersion(SystemVersion);
   bool HadExtra;
   if (!Driver::GetReleaseVersion(MacOSSDKVersion, Major, Minor, Micro,
                                  HadExtra))
@@ -1554,12 +1552,10 @@ struct DarwinPlatform {
                    const Optional<DarwinSDKInfo> &SDKInfo) {
     DarwinPlatform Result(TargetArg, getPlatformFromOS(TT.getOS()), OSVersion,
                           A);
-    unsigned Major, Minor, Micro;
-    TT.getOSVersion(Major, Minor, Micro);
-    if (Major == 0)
+    VersionTuple OsVersion = TT.getOSVersion();
+    if (OsVersion.getMajor() == 0)
       Result.HasOSVersion = false;
-    Result.setEnvironment(TT.getEnvironment(),
-                          VersionTuple(Major, Minor, Micro), SDKInfo);
+    Result.setEnvironment(TT.getEnvironment(), OsVersion, SDKInfo);
     return Result;
   }
   static DarwinPlatform
@@ -1805,7 +1801,7 @@ inferDeploymentTargetFromSDK(DerivedArgList &Args,
 
 std::string getOSVersion(llvm::Triple::OSType OS, const llvm::Triple &Triple,
                          const Driver &TheDriver) {
-  unsigned Major, Minor, Micro;
+  VersionTuple OsVersion;
   llvm::Triple SystemTriple(llvm::sys::getProcessTriple());
   switch (OS) {
   case llvm::Triple::Darwin:
@@ -1814,24 +1810,22 @@ std::string getOSVersion(llvm::Triple::OSType OS, const llvm::Triple &Triple,
     // macos, use the host triple to infer OS version.
     if (Triple.isMacOSX() && SystemTriple.isMacOSX() &&
         !Triple.getOSMajorVersion())
-      SystemTriple.getMacOSXVersion(Major, Minor, Micro);
-    else if (!Triple.getMacOSXVersion(Major, Minor, Micro))
+      SystemTriple.getMacOSXVersion(OsVersion);
+    else if (!Triple.getMacOSXVersion(OsVersion))
       TheDriver.Diag(diag::err_drv_invalid_darwin_version)
           << Triple.getOSName();
     break;
   case llvm::Triple::IOS:
     if (Triple.isMacCatalystEnvironment() && !Triple.getOSMajorVersion()) {
-      Major = 13;
-      Minor = 1;
-      Micro = 0;
+      OsVersion = VersionTuple(13, 1);
     } else
-      Triple.getiOSVersion(Major, Minor, Micro);
+      OsVersion = Triple.getiOSVersion();
     break;
   case llvm::Triple::TvOS:
-    Triple.getOSVersion(Major, Minor, Micro);
+    OsVersion = Triple.getOSVersion();
     break;
   case llvm::Triple::WatchOS:
-    Triple.getWatchOSVersion(Major, Minor, Micro);
+    OsVersion = Triple.getWatchOSVersion();
     break;
   default:
     llvm_unreachable("Unexpected OS type");
@@ -1839,7 +1833,9 @@ std::string getOSVersion(llvm::Triple::OSType OS, const llvm::Triple &Triple,
   }
 
   std::string OSVersion;
-  llvm::raw_string_ostream(OSVersion) << Major << '.' << Minor << '.' << Micro;
+  llvm::raw_string_ostream(OSVersion)
+      << OsVersion.getMajor() << '.' << OsVersion.getMinor().getValueOr(0)
+      << '.' << OsVersion.getSubminor().getValueOr(0);
   return OSVersion;
 }
 
@@ -1909,15 +1905,13 @@ getDeploymentTargetFromMTargetOSArg(DerivedArgList &Args,
     return None;
   }
 
-  unsigned Major, Minor, Micro;
-  TT.getOSVersion(Major, Minor, Micro);
-  if (!Major) {
+  VersionTuple Version = TT.getOSVersion();
+  if (!Version.getMajor()) {
     TheDriver.Diag(diag::err_drv_invalid_version_number)
         << A->getAsString(Args);
     return None;
   }
-  return DarwinPlatform::createFromMTargetOS(TT.getOS(),
-                                             VersionTuple(Major, Minor, Micro),
+  return DarwinPlatform::createFromMTargetOS(TT.getOS(), Version,
                                              TT.getEnvironment(), A, SDKInfo);
 }
 
diff --git a/clang/lib/Driver/ToolChains/Darwin.h b/clang/lib/Driver/ToolChains/Darwin.h
index a307cd317ac3..5e23047a5512 100644
--- a/clang/lib/Driver/ToolChains/Darwin.h
+++ b/clang/lib/Driver/ToolChains/Darwin.h
@@ -64,7 +64,7 @@ class LLVM_LIBRARY_VISIBILITY Linker : public MachOTool {
   void AddLinkArgs(Compilation &C, const llvm::opt::ArgList &Args,
                    llvm::opt::ArgStringList &CmdArgs,
                    const InputInfoList &Inputs, unsigned Version[5],
-                   bool LinkerIsLLD, bool LinkerIsLLDDarwinNew) const;
+                   bool LinkerIsLLD) const;
 
 public:
   Linker(const ToolChain &TC) : MachOTool("darwin::Linker", "linker", TC) {}
diff --git a/clang/lib/Driver/ToolChains/FreeBSD.cpp b/clang/lib/Driver/ToolChains/FreeBSD.cpp
index d08ea282f6df..de635f5816cf 100644
--- a/clang/lib/Driver/ToolChains/FreeBSD.cpp
+++ b/clang/lib/Driver/ToolChains/FreeBSD.cpp
@@ -293,8 +293,8 @@ void freebsd::Linker::ConstructJob(Compilation &C, const JobAction &JA,
   addLinkerCompressDebugSectionsOption(ToolChain, Args, CmdArgs);
   AddLinkerInputs(ToolChain, Inputs, Args, CmdArgs, JA);
 
-  bool Profiling = Args.hasArg(options::OPT_pg) &&
-                   ToolChain.getTriple().getOSMajorVersion() < 14;
+  unsigned Major = ToolChain.getTriple().getOSMajorVersion();
+  bool Profiling = Args.hasArg(options::OPT_pg) && Major != 0 && Major < 14;
   if (!Args.hasArg(options::OPT_nostdlib, options::OPT_nodefaultlibs)) {
     // Use the static OpenMP runtime with -static-openmp
     bool StaticOpenMP = Args.hasArg(options::OPT_static_openmp) &&
@@ -419,8 +419,8 @@ void FreeBSD::addLibStdCxxIncludePaths(
 void FreeBSD::AddCXXStdlibLibArgs(const ArgList &Args,
                                   ArgStringList &CmdArgs) const {
   CXXStdlibType Type = GetCXXStdlibType(Args);
-  bool Profiling =
-      Args.hasArg(options::OPT_pg) && getTriple().getOSMajorVersion() < 14;
+  unsigned Major = getTriple().getOSMajorVersion();
+  bool Profiling = Args.hasArg(options::OPT_pg) && Major != 0 && Major < 14;
 
   switch (Type) {
   case ToolChain::CST_Libcxx:
diff --git a/clang/lib/Driver/ToolChains/HIP.cpp b/clang/lib/Driver/ToolChains/HIPAMD.cpp
index 07af1a0457c7..6d553791b394 100644
--- a/clang/lib/Driver/ToolChains/HIP.cpp
+++ b/clang/lib/Driver/ToolChains/HIPAMD.cpp
@@ -1,4 +1,4 @@
-//===--- HIP.cpp - HIP Tool and ToolChain Implementations -------*- C++ -*-===//
+//===--- HIPAMD.cpp - HIP Tool and ToolChain Implementations ----*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
@@ -6,9 +6,10 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include "HIP.h"
+#include "HIPAMD.h"
 #include "AMDGPU.h"
 #include "CommonArgs.h"
+#include "HIPUtility.h"
 #include "clang/Basic/Cuda.h"
 #include "clang/Basic/TargetID.h"
 #include "clang/Driver/Compilation.h"
@@ -34,10 +35,6 @@ using namespace llvm::opt;
 #define NULL_FILE "/dev/null"
 #endif
 
-namespace {
-const unsigned HIPCodeObjectAlign = 4096;
-} // namespace
-
 static bool shouldSkipSanitizeOption(const ToolChain &TC,
                                      const llvm::opt::ArgList &DriverArgs,
                                      StringRef TargetID,
@@ -76,9 +73,9 @@ static bool shouldSkipSanitizeOption(const ToolChain &TC,
 }
 
 void AMDGCN::Linker::constructLldCommand(Compilation &C, const JobAction &JA,
-                                          const InputInfoList &Inputs,
-                                          const InputInfo &Output,
-                                          const llvm::opt::ArgList &Args) const {
+                                         const InputInfoList &Inputs,
+                                         const InputInfo &Output,
+                                         const llvm::opt::ArgList &Args) const {
   // Construct lld command.
   // The output from ld.lld is an HSA code object file.
   ArgStringList LldArgs{"-flavor", "gnu", "--no-undefined", "-shared",
@@ -129,142 +126,28 @@ void AMDGCN::Linker::constructLldCommand(Compilation &C, const JobAction &JA,
                                          Lld, LldArgs, Inputs, Output));
 }
 
-// Construct a clang-offload-bundler command to bundle code objects for
-// different GPU's into a HIP fat binary.
-void AMDGCN::constructHIPFatbinCommand(Compilation &C, const JobAction &JA,
-                  StringRef OutputFileName, const InputInfoList &Inputs,
-                  const llvm::opt::ArgList &Args, const Tool& T) {
-  // Construct clang-offload-bundler command to bundle object files for
-  // for different GPU archs.
-  ArgStringList BundlerArgs;
-  BundlerArgs.push_back(Args.MakeArgString("-type=o"));
-  BundlerArgs.push_back(
-      Args.MakeArgString("-bundle-align=" + Twine(HIPCodeObjectAlign)));
-
-  // ToDo: Remove the dummy host binary entry which is required by
-  // clang-offload-bundler.
-  std::string BundlerTargetArg = "-targets=host-x86_64-unknown-linux";
-  std::string BundlerInputArg = "-inputs=" NULL_FILE;
-
-  // For code object version 2 and 3, the offload kind in bundle ID is 'hip'
-  // for backward compatibility. For code object version 4 and greater, the
-  // offload kind in bundle ID is 'hipv4'.
-  std::string OffloadKind = "hip";
-  if (getAMDGPUCodeObjectVersion(C.getDriver(), Args) >= 4)
-    OffloadKind = OffloadKind + "v4";
-  for (const auto &II : Inputs) {
-    const auto* A = II.getAction();
-    BundlerTargetArg = BundlerTargetArg + "," + OffloadKind +
-                       "-amdgcn-amd-amdhsa--" +
-                       StringRef(A->getOffloadingArch()).str();
-    BundlerInputArg = BundlerInputArg + "," + II.getFilename();
-  }
-  BundlerArgs.push_back(Args.MakeArgString(BundlerTargetArg));
-  BundlerArgs.push_back(Args.MakeArgString(BundlerInputArg));
-
-  std::string Output = std::string(OutputFileName);
-  auto BundlerOutputArg =
-      Args.MakeArgString(std::string("-outputs=").append(Output));
-  BundlerArgs.push_back(BundlerOutputArg);
-
-  const char *Bundler = Args.MakeArgString(
-      T.getToolChain().GetProgramPath("clang-offload-bundler"));
-  C.addCommand(std::make_unique<Command>(
-      JA, T, ResponseFileSupport::None(), Bundler, BundlerArgs, Inputs,
-      InputInfo(&JA, Args.MakeArgString(Output))));
-}
-
-/// Add Generated HIP Object File which has device images embedded into the
-/// host to the argument list for linking. Using MC directives, embed the
-/// device code and also define symbols required by the code generation so that
-/// the image can be retrieved at runtime.
-void AMDGCN::Linker::constructGenerateObjFileFromHIPFatBinary(
-    Compilation &C, const InputInfo &Output,
-    const InputInfoList &Inputs, const ArgList &Args,
-    const JobAction &JA) const {
-  const ToolChain &TC = getToolChain();
-  std::string Name =
-      std::string(llvm::sys::path::stem(Output.getFilename()));
-
-  // Create Temp Object File Generator,
-  // Offload Bundled file and Bundled Object file.
-  // Keep them if save-temps is enabled.
-  const char *McinFile;
-  const char *BundleFile;
-  if (C.getDriver().isSaveTempsEnabled()) {
-    McinFile = C.getArgs().MakeArgString(Name + ".mcin");
-    BundleFile = C.getArgs().MakeArgString(Name + ".hipfb");
-  } else {
-    auto TmpNameMcin = C.getDriver().GetTemporaryPath(Name, "mcin");
-    McinFile = C.addTempFile(C.getArgs().MakeArgString(TmpNameMcin));
-    auto TmpNameFb = C.getDriver().GetTemporaryPath(Name, "hipfb");
-    BundleFile = C.addTempFile(C.getArgs().MakeArgString(TmpNameFb));
-  }
-  constructHIPFatbinCommand(C, JA, BundleFile, Inputs, Args, *this);
-
-  // Create a buffer to write the contents of the temp obj generator.
-  std::string ObjBuffer;
-  llvm::raw_string_ostream ObjStream(ObjBuffer);
-
-  // Add MC directives to embed target binaries. We ensure that each
-  // section and image is 16-byte aligned. This is not mandatory, but
-  // increases the likelihood of data to be aligned with a cache block
-  // in several main host machines.
-  ObjStream << "#       HIP Object Generator\n";
-  ObjStream << "# *** Automatically generated by Clang ***\n";
-  ObjStream << "  .protected __hip_fatbin\n";
-  ObjStream << "  .type __hip_fatbin,@object\n";
-  ObjStream << "  .section .hip_fatbin,\"a\",@progbits\n";
-  ObjStream << "  .globl __hip_fatbin\n";
-  ObjStream << "  .p2align " << llvm::Log2(llvm::Align(HIPCodeObjectAlign))
-            << "\n";
-  ObjStream << "__hip_fatbin:\n";
-  ObjStream << "  .incbin \"" << BundleFile << "\"\n";
-  ObjStream.flush();
-
-  // Dump the contents of the temp object file gen if the user requested that.
-  // We support this option to enable testing of behavior with -###.
-  if (C.getArgs().hasArg(options::OPT_fhip_dump_offload_linker_script))
-    llvm::errs() << ObjBuffer;
-
-  // Open script file and write the contents.
-  std::error_code EC;
-  llvm::raw_fd_ostream Objf(McinFile, EC, llvm::sys::fs::OF_None);
-
-  if (EC) {
-    C.getDriver().Diag(clang::diag::err_unable_to_make_temp) << EC.message();
-    return;
-  }
-
-  Objf << ObjBuffer;
-
-  ArgStringList McArgs{"-o",      Output.getFilename(),
-                       McinFile,  "--filetype=obj"};
-  const char *Mc = Args.MakeArgString(TC.GetProgramPath("llvm-mc"));
-  C.addCommand(std::make_unique<Command>(JA, *this, ResponseFileSupport::None(),
-                                         Mc, McArgs, Inputs, Output));
-}
-
 // For amdgcn the inputs of the linker job are device bitcode and output is
 // object file. It calls llvm-link, opt, llc, then lld steps.
 void AMDGCN::Linker::ConstructJob(Compilation &C, const JobAction &JA,
-                                   const InputInfo &Output,
-                                   const InputInfoList &Inputs,
-                                   const ArgList &Args,
-                                   const char *LinkingOutput) const {
+                                  const InputInfo &Output,
+                                  const InputInfoList &Inputs,
+                                  const ArgList &Args,
+                                  const char *LinkingOutput) const {
   if (Inputs.size() > 0 &&
       Inputs[0].getType() == types::TY_Image &&
       JA.getType() == types::TY_Object)
-    return constructGenerateObjFileFromHIPFatBinary(C, Output, Inputs, Args, JA);
+    return HIP::constructGenerateObjFileFromHIPFatBinary(C, Output, Inputs,
+                                                         Args, JA, *this);
 
   if (JA.getType() == types::TY_HIP_FATBIN)
-    return constructHIPFatbinCommand(C, JA, Output.getFilename(), Inputs, Args, *this);
+    return HIP::constructHIPFatbinCommand(C, JA, Output.getFilename(), Inputs,
+                                          Args, *this);
 
   return constructLldCommand(C, JA, Inputs, Output, Args);
 }
 
-HIPToolChain::HIPToolChain(const Driver &D, const llvm::Triple &Triple,
-                             const ToolChain &HostTC, const ArgList &Args)
+HIPAMDToolChain::HIPAMDToolChain(const Driver &D, const llvm::Triple &Triple,
+                                 const ToolChain &HostTC, const ArgList &Args)
     : ROCMToolChain(D, Triple, Args), HostTC(HostTC) {
   // Lookup binaries into the driver directory, this is used to
   // discover the clang-offload-bundler executable.
@@ -279,9 +162,8 @@ HIPToolChain::HIPToolChain(const Driver &D, const llvm::Triple &Triple,
   }
 }
 
-void HIPToolChain::addClangTargetOptions(
-    const llvm::opt::ArgList &DriverArgs,
-    llvm::opt::ArgStringList &CC1Args,
+void HIPAMDToolChain::addClangTargetOptions(
+    const llvm::opt::ArgList &DriverArgs, llvm::opt::ArgStringList &CC1Args,
     Action::OffloadKind DeviceOffloadingKind) const {
   HostTC.addClangTargetOptions(DriverArgs, CC1Args, DeviceOffloadingKind);
 
@@ -324,9 +206,9 @@ void HIPToolChain::addClangTargetOptions(
 }
 
 llvm::opt::DerivedArgList *
-HIPToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
-                             StringRef BoundArch,
-                             Action::OffloadKind DeviceOffloadKind) const {
+HIPAMDToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
+                               StringRef BoundArch,
+                               Action::OffloadKind DeviceOffloadKind) const {
   DerivedArgList *DAL =
       HostTC.TranslateArgs(Args, BoundArch, DeviceOffloadKind);
   if (!DAL)
@@ -349,44 +231,44 @@ HIPToolChain::TranslateArgs(const llvm::opt::DerivedArgList &Args,
   return DAL;
 }
 
-Tool *HIPToolChain::buildLinker() const {
+Tool *HIPAMDToolChain::buildLinker() const {
   assert(getTriple().getArch() == llvm::Triple::amdgcn);
   return new tools::AMDGCN::Linker(*this);
 }
 
-void HIPToolChain::addClangWarningOptions(ArgStringList &CC1Args) const {
+void HIPAMDToolChain::addClangWarningOptions(ArgStringList &CC1Args) const {
   HostTC.addClangWarningOptions(CC1Args);
 }
 
 ToolChain::CXXStdlibType
-HIPToolChain::GetCXXStdlibType(const ArgList &Args) const {
+HIPAMDToolChain::GetCXXStdlibType(const ArgList &Args) const {
   return HostTC.GetCXXStdlibType(Args);
 }
 
-void HIPToolChain::AddClangSystemIncludeArgs(const ArgList &DriverArgs,
-                                              ArgStringList &CC1Args) const {
+void HIPAMDToolChain::AddClangSystemIncludeArgs(const ArgList &DriverArgs,
+                                                ArgStringList &CC1Args) const {
   HostTC.AddClangSystemIncludeArgs(DriverArgs, CC1Args);
 }
 
-void HIPToolChain::AddClangCXXStdlibIncludeArgs(const ArgList &Args,
-                                                 ArgStringList &CC1Args) const {
+void HIPAMDToolChain::AddClangCXXStdlibIncludeArgs(
+    const ArgList &Args, ArgStringList &CC1Args) const {
   HostTC.AddClangCXXStdlibIncludeArgs(Args, CC1Args);
 }
 
-void HIPToolChain::AddIAMCUIncludeArgs(const ArgList &Args,
-                                        ArgStringList &CC1Args) const {
+void HIPAMDToolChain::AddIAMCUIncludeArgs(const ArgList &Args,
+                                          ArgStringList &CC1Args) const {
   HostTC.AddIAMCUIncludeArgs(Args, CC1Args);
 }
 
-void HIPToolChain::AddHIPIncludeArgs(const ArgList &DriverArgs,
-                                     ArgStringList &CC1Args) const {
+void HIPAMDToolChain::AddHIPIncludeArgs(const ArgList &DriverArgs,
+                                        ArgStringList &CC1Args) const {
   RocmInstallation.AddHIPIncludeArgs(DriverArgs, CC1Args);
 }
 
-SanitizerMask HIPToolChain::getSupportedSanitizers() const {
-  // The HIPToolChain only supports sanitizers in the sense that it allows
+SanitizerMask HIPAMDToolChain::getSupportedSanitizers() const {
+  // The HIPAMDToolChain only supports sanitizers in the sense that it allows
   // sanitizer arguments on the command line if they are supported by the host
-  // toolchain. The HIPToolChain will actually ignore any command line
+  // toolchain. The HIPAMDToolChain will actually ignore any command line
   // arguments for any of these "supported" sanitizers. That means that no
   // sanitization of device code is actually supported at this time.
   //
@@ -396,13 +278,13 @@ SanitizerMask HIPToolChain::getSupportedSanitizers() const {
   return HostTC.getSupportedSanitizers();
 }
 
-VersionTuple HIPToolChain::computeMSVCVersion(const Driver *D,
-                                               const ArgList &Args) const {
+VersionTuple HIPAMDToolChain::computeMSVCVersion(const Driver *D,
+                                                 const ArgList &Args) const {
   return HostTC.computeMSVCVersion(D, Args);
 }
 
 llvm::SmallVector<ToolChain::BitCodeLibraryInfo, 12>
-HIPToolChain::getHIPDeviceLibs(const llvm::opt::ArgList &DriverArgs) const {
+HIPAMDToolChain::getHIPDeviceLibs(const llvm::opt::ArgList &DriverArgs) const {
   llvm::SmallVector<BitCodeLibraryInfo, 12> BCLibs;
   if (DriverArgs.hasArg(options::OPT_nogpulib))
     return {};
@@ -476,11 +358,11 @@ HIPToolChain::getHIPDeviceLibs(const llvm::opt::ArgList &DriverArgs) const {
   return BCLibs;
 }
 
-void HIPToolChain::checkTargetID(const llvm::opt::ArgList &DriverArgs) const {
+void HIPAMDToolChain::checkTargetID(
+    const llvm::opt::ArgList &DriverArgs) const {
   auto PTID = getParsedTargetID(DriverArgs);
   if (PTID.OptionalTargetID && !PTID.OptionalGPUArch) {
     getDriver().Diag(clang::diag::err_drv_bad_target_id)
         << PTID.OptionalTargetID.getValue();
   }
-  return;
 }
diff --git a/clang/lib/Driver/ToolChains/HIP.h b/clang/lib/Driver/ToolChains/HIPAMD.h
index 60b3d69b3f52..cc472a595db9 100644
--- a/clang/lib/Driver/ToolChains/HIP.h
+++ b/clang/lib/Driver/ToolChains/HIPAMD.h
@@ -1,4 +1,4 @@
-//===--- HIP.h - HIP ToolChain Implementations ------------------*- C++ -*-===//
+//===--- HIPAMD.h - HIP ToolChain Implementations ---------------*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
@@ -6,12 +6,12 @@
 //
 //===----------------------------------------------------------------------===//
 
-#ifndef LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIP_H
-#define LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIP_H
+#ifndef LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPAMD_H
+#define LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPAMD_H
 
-#include "clang/Driver/ToolChain.h"
-#include "clang/Driver/Tool.h"
 #include "AMDGPU.h"
+#include "clang/Driver/Tool.h"
+#include "clang/Driver/ToolChain.h"
 
 namespace clang {
 namespace driver {
@@ -19,11 +19,6 @@ namespace driver {
 namespace tools {
 
 namespace AMDGCN {
-  // Construct command for creating HIP fatbin.
-  void constructHIPFatbinCommand(Compilation &C, const JobAction &JA,
-                  StringRef OutputFileName, const InputInfoList &Inputs,
-                  const llvm::opt::ArgList &TCArgs, const Tool& T);
-
 // Runs llvm-link/opt/llc/lld, which links multiple LLVM bitcode, together with
 // device library, then compiles it to ISA in a shared object.
 class LLVM_LIBRARY_VISIBILITY Linker : public Tool {
@@ -38,17 +33,9 @@ public:
                     const char *LinkingOutput) const override;
 
 private:
-
   void constructLldCommand(Compilation &C, const JobAction &JA,
                            const InputInfoList &Inputs, const InputInfo &Output,
                            const llvm::opt::ArgList &Args) const;
-
-  // Construct command for creating Object from HIP fatbin.
-  void constructGenerateObjFileFromHIPFatBinary(Compilation &C,
-                                                const InputInfo &Output,
-                                                const InputInfoList &Inputs,
-                                                const llvm::opt::ArgList &Args,
-                                                const JobAction &JA) const;
 };
 
 } // end namespace AMDGCN
@@ -56,10 +43,10 @@ private:
 
 namespace toolchains {
 
-class LLVM_LIBRARY_VISIBILITY HIPToolChain final : public ROCMToolChain {
+class LLVM_LIBRARY_VISIBILITY HIPAMDToolChain final : public ROCMToolChain {
 public:
-  HIPToolChain(const Driver &D, const llvm::Triple &Triple,
-                const ToolChain &HostTC, const llvm::opt::ArgList &Args);
+  HIPAMDToolChain(const Driver &D, const llvm::Triple &Triple,
+                  const ToolChain &HostTC, const llvm::opt::ArgList &Args);
 
   const llvm::Triple *getAuxTriple() const override {
     return &HostTC.getTriple();
@@ -68,9 +55,10 @@ public:
   llvm::opt::DerivedArgList *
   TranslateArgs(const llvm::opt::DerivedArgList &Args, StringRef BoundArch,
                 Action::OffloadKind DeviceOffloadKind) const override;
-  void addClangTargetOptions(const llvm::opt::ArgList &DriverArgs,
-                             llvm::opt::ArgStringList &CC1Args,
-                             Action::OffloadKind DeviceOffloadKind) const override;
+  void
+  addClangTargetOptions(const llvm::opt::ArgList &DriverArgs,
+                        llvm::opt::ArgStringList &CC1Args,
+                        Action::OffloadKind DeviceOffloadKind) const override;
   void addClangWarningOptions(llvm::opt::ArgStringList &CC1Args) const override;
   CXXStdlibType GetCXXStdlibType(const llvm::opt::ArgList &Args) const override;
   void
@@ -105,4 +93,4 @@ protected:
 } // end namespace driver
 } // end namespace clang
 
-#endif // LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIP_H
+#endif // LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPAMD_H
diff --git a/clang/lib/Driver/ToolChains/HIPSPV.cpp b/clang/lib/Driver/ToolChains/HIPSPV.cpp
new file mode 100644
index 000000000000..d68c87e9b3e7
--- /dev/null
+++ b/clang/lib/Driver/ToolChains/HIPSPV.cpp
@@ -0,0 +1,292 @@
+//===--- HIPSPV.cpp - HIPSPV ToolChain Implementation -----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "HIPSPV.h"
+#include "CommonArgs.h"
+#include "HIPUtility.h"
+#include "clang/Driver/Compilation.h"
+#include "clang/Driver/Driver.h"
+#include "clang/Driver/DriverDiagnostic.h"
+#include "clang/Driver/InputInfo.h"
+#include "clang/Driver/Options.h"
+#include "llvm/Support/FileSystem.h"
+#include "llvm/Support/Path.h"
+
+using namespace clang::driver;
+using namespace clang::driver::toolchains;
+using namespace clang::driver::tools;
+using namespace clang;
+using namespace llvm::opt;
+
+// Convenience function for creating temporary file for both modes of
+// isSaveTempsEnabled().
+static const char *getTempFile(Compilation &C, StringRef Prefix,
+                               StringRef Extension) {
+  if (C.getDriver().isSaveTempsEnabled()) {
+    return C.getArgs().MakeArgString(Prefix + "." + Extension);
+  }
+  auto TmpFile = C.getDriver().GetTemporaryPath(Prefix, Extension);
+  return C.addTempFile(C.getArgs().MakeArgString(TmpFile));
+}
+
+// Locates HIP pass plugin.
+static std::string findPassPlugin(const Driver &D,
+                                  const llvm::opt::ArgList &Args) {
+  StringRef Path = Args.getLastArgValue(options::OPT_hipspv_pass_plugin_EQ);
+  if (!Path.empty()) {
+    if (llvm::sys::fs::exists(Path))
+      return Path.str();
+    D.Diag(diag::err_drv_no_such_file) << Path;
+  }
+
+  StringRef hipPath = Args.getLastArgValue(options::OPT_hip_path_EQ);
+  if (!hipPath.empty()) {
+    SmallString<128> PluginPath(hipPath);
+    llvm::sys::path::append(PluginPath, "lib", "libLLVMHipSpvPasses.so");
+    if (llvm::sys::fs::exists(PluginPath))
+      return PluginPath.str().str();
+    PluginPath.assign(hipPath);
+    llvm::sys::path::append(PluginPath, "lib", "llvm",
+                            "libLLVMHipSpvPasses.so");
+    if (llvm::sys::fs::exists(PluginPath))
+      return PluginPath.str().str();
+  }
+
+  return std::string();
+}
+
+void HIPSPV::Linker::constructLinkAndEmitSpirvCommand(
+    Compilation &C, const JobAction &JA, const InputInfoList &Inputs,
+    const InputInfo &Output, const llvm::opt::ArgList &Args) const {
+
+  assert(!Inputs.empty() && "Must have at least one input.");
+  std::string Name = std::string(llvm::sys::path::stem(Output.getFilename()));
+  const char *TempFile = getTempFile(C, Name + "-link", "bc");
+
+  // Link LLVM bitcode.
+  ArgStringList LinkArgs{};
+  for (auto Input : Inputs)
+    LinkArgs.push_back(Input.getFilename());
+  LinkArgs.append({"-o", TempFile});
+  const char *LlvmLink =
+      Args.MakeArgString(getToolChain().GetProgramPath("llvm-link"));
+  C.addCommand(std::make_unique<Command>(JA, *this, ResponseFileSupport::None(),
+                                         LlvmLink, LinkArgs, Inputs, Output));
+
+  // Post-link HIP lowering.
+
+  // Run LLVM IR passes to lower/expand/emulate HIP code that does not translate
+  // to SPIR-V (E.g. dynamic shared memory).
+  auto PassPluginPath = findPassPlugin(C.getDriver(), Args);
+  if (!PassPluginPath.empty()) {
+    const char *PassPathCStr = C.getArgs().MakeArgString(PassPluginPath);
+    const char *OptOutput = getTempFile(C, Name + "-lower", "bc");
+    ArgStringList OptArgs{TempFile,     "-load-pass-plugin",
+                          PassPathCStr, "-passes=hip-post-link-passes",
+                          "-o",         OptOutput};
+    const char *Opt = Args.MakeArgString(getToolChain().GetProgramPath("opt"));
+    C.addCommand(std::make_unique<Command>(
+        JA, *this, ResponseFileSupport::None(), Opt, OptArgs, Inputs, Output));
+    TempFile = OptOutput;
+  }
+
+  // Emit SPIR-V binary.
+
+  llvm::opt::ArgStringList TrArgs{"--spirv-max-version=1.1",
+                                  "--spirv-ext=+all"};
+  InputInfo TrInput = InputInfo(types::TY_LLVM_BC, TempFile, "");
+  SPIRV::constructTranslateCommand(C, *this, JA, Output, TrInput, TrArgs);
+}
+
+void HIPSPV::Linker::ConstructJob(Compilation &C, const JobAction &JA,
+                                  const InputInfo &Output,
+                                  const InputInfoList &Inputs,
+                                  const ArgList &Args,
+                                  const char *LinkingOutput) const {
+  if (Inputs.size() > 0 && Inputs[0].getType() == types::TY_Image &&
+      JA.getType() == types::TY_Object)
+    return HIP::constructGenerateObjFileFromHIPFatBinary(C, Output, Inputs,
+                                                         Args, JA, *this);
+
+  if (JA.getType() == types::TY_HIP_FATBIN)
+    return HIP::constructHIPFatbinCommand(C, JA, Output.getFilename(), Inputs,
+                                          Args, *this);
+
+  constructLinkAndEmitSpirvCommand(C, JA, Inputs, Output, Args);
+}
+
+HIPSPVToolChain::HIPSPVToolChain(const Driver &D, const llvm::Triple &Triple,
+                                 const ToolChain &HostTC, const ArgList &Args)
+    : ToolChain(D, Triple, Args), HostTC(HostTC) {
+  // Lookup binaries into the driver directory, this is used to
+  // discover the clang-offload-bundler executable.
+  getProgramPaths().push_back(getDriver().Dir);
+}
+
+void HIPSPVToolChain::addClangTargetOptions(
+    const llvm::opt::ArgList &DriverArgs, llvm::opt::ArgStringList &CC1Args,
+    Action::OffloadKind DeviceOffloadingKind) const {
+  HostTC.addClangTargetOptions(DriverArgs, CC1Args, DeviceOffloadingKind);
+
+  assert(DeviceOffloadingKind == Action::OFK_HIP &&
+         "Only HIP offloading kinds are supported for GPUs.");
+
+  CC1Args.append(
+      {"-fcuda-is-device", "-fcuda-allow-variadic-functions",
+       // A crude workaround for llvm-spirv which does not handle the
+       // autovectorized code well (vector reductions, non-i{8,16,32,64} types).
+       // TODO: Allow autovectorization when SPIR-V backend arrives.
+       "-mllvm", "-vectorize-loops=false", "-mllvm", "-vectorize-slp=false"});
+
+  if (DriverArgs.hasFlag(options::OPT_fcuda_approx_transcendentals,
+                         options::OPT_fno_cuda_approx_transcendentals, false))
+    CC1Args.push_back("-fcuda-approx-transcendentals");
+
+  // Default to "hidden" visibility, as object level linking will not be
+  // supported for the foreseeable future.
+  if (!DriverArgs.hasArg(options::OPT_fvisibility_EQ,
+                         options::OPT_fvisibility_ms_compat))
+    CC1Args.append(
+        {"-fvisibility", "hidden", "-fapply-global-visibility-to-externs"});
+
+  llvm::for_each(getHIPDeviceLibs(DriverArgs),
+                 [&](const BitCodeLibraryInfo &BCFile) {
+                   CC1Args.append({"-mlink-builtin-bitcode",
+                                   DriverArgs.MakeArgString(BCFile.Path)});
+                 });
+}
+
+Tool *HIPSPVToolChain::buildLinker() const {
+  assert(getTriple().getArch() == llvm::Triple::spirv64);
+  return new tools::HIPSPV::Linker(*this);
+}
+
+void HIPSPVToolChain::addClangWarningOptions(ArgStringList &CC1Args) const {
+  HostTC.addClangWarningOptions(CC1Args);
+}
+
+ToolChain::CXXStdlibType
+HIPSPVToolChain::GetCXXStdlibType(const ArgList &Args) const {
+  return HostTC.GetCXXStdlibType(Args);
+}
+
+void HIPSPVToolChain::AddClangSystemIncludeArgs(const ArgList &DriverArgs,
+                                                ArgStringList &CC1Args) const {
+  HostTC.AddClangSystemIncludeArgs(DriverArgs, CC1Args);
+}
+
+void HIPSPVToolChain::AddClangCXXStdlibIncludeArgs(
+    const ArgList &Args, ArgStringList &CC1Args) const {
+  HostTC.AddClangCXXStdlibIncludeArgs(Args, CC1Args);
+}
+
+void HIPSPVToolChain::AddIAMCUIncludeArgs(const ArgList &Args,
+                                          ArgStringList &CC1Args) const {
+  HostTC.AddIAMCUIncludeArgs(Args, CC1Args);
+}
+
+void HIPSPVToolChain::AddHIPIncludeArgs(const ArgList &DriverArgs,
+                                        ArgStringList &CC1Args) const {
+  if (DriverArgs.hasArg(options::OPT_nogpuinc))
+    return;
+
+  StringRef hipPath = DriverArgs.getLastArgValue(options::OPT_hip_path_EQ);
+  if (hipPath.empty()) {
+    getDriver().Diag(diag::err_drv_hipspv_no_hip_path) << 1 << "'-nogpuinc'";
+    return;
+  }
+  SmallString<128> P(hipPath);
+  llvm::sys::path::append(P, "include");
+  CC1Args.append({"-isystem", DriverArgs.MakeArgString(P)});
+}
+
+llvm::SmallVector<ToolChain::BitCodeLibraryInfo, 12>
+HIPSPVToolChain::getHIPDeviceLibs(const llvm::opt::ArgList &DriverArgs) const {
+  llvm::SmallVector<ToolChain::BitCodeLibraryInfo, 12> BCLibs;
+  if (DriverArgs.hasArg(options::OPT_nogpulib))
+    return {};
+
+  ArgStringList LibraryPaths;
+  // Find device libraries in --hip-device-lib-path and HIP_DEVICE_LIB_PATH.
+  auto HipDeviceLibPathArgs = DriverArgs.getAllArgValues(
+      // --hip-device-lib-path is alias to this option.
+      clang::driver::options::OPT_rocm_device_lib_path_EQ);
+  for (auto Path : HipDeviceLibPathArgs)
+    LibraryPaths.push_back(DriverArgs.MakeArgString(Path));
+
+  StringRef HipPath = DriverArgs.getLastArgValue(options::OPT_hip_path_EQ);
+  if (!HipPath.empty()) {
+    SmallString<128> Path(HipPath);
+    llvm::sys::path::append(Path, "lib", "hip-device-lib");
+    LibraryPaths.push_back(DriverArgs.MakeArgString(Path));
+  }
+
+  addDirectoryList(DriverArgs, LibraryPaths, "", "HIP_DEVICE_LIB_PATH");
+
+  // Maintain compatability with --hip-device-lib.
+  auto BCLibArgs = DriverArgs.getAllArgValues(options::OPT_hip_device_lib_EQ);
+  if (!BCLibArgs.empty()) {
+    llvm::for_each(BCLibArgs, [&](StringRef BCName) {
+      StringRef FullName;
+      for (std::string LibraryPath : LibraryPaths) {
+        SmallString<128> Path(LibraryPath);
+        llvm::sys::path::append(Path, BCName);
+        FullName = Path;
+        if (llvm::sys::fs::exists(FullName)) {
+          BCLibs.emplace_back(FullName.str());
+          return;
+        }
+      }
+      getDriver().Diag(diag::err_drv_no_such_file) << BCName;
+    });
+  } else {
+    // Search device library named as 'hipspv-<triple>.bc'.
+    auto TT = getTriple().normalize();
+    std::string BCName = "hipspv-" + TT + ".bc";
+    for (auto *LibPath : LibraryPaths) {
+      SmallString<128> Path(LibPath);
+      llvm::sys::path::append(Path, BCName);
+      if (llvm::sys::fs::exists(Path)) {
+        BCLibs.emplace_back(Path.str().str());
+        return BCLibs;
+      }
+    }
+    getDriver().Diag(diag::err_drv_no_hipspv_device_lib)
+        << 1 << ("'" + TT + "' target");
+    return {};
+  }
+
+  return BCLibs;
+}
+
+SanitizerMask HIPSPVToolChain::getSupportedSanitizers() const {
+  // The HIPSPVToolChain only supports sanitizers in the sense that it allows
+  // sanitizer arguments on the command line if they are supported by the host
+  // toolchain. The HIPSPVToolChain will actually ignore any command line
+  // arguments for any of these "supported" sanitizers. That means that no
+  // sanitization of device code is actually supported at this time.
+  //
+  // This behavior is necessary because the host and device toolchains
+  // invocations often share the command line, so the device toolchain must
+  // tolerate flags meant only for the host toolchain.
+  return HostTC.getSupportedSanitizers();
+}
+
+VersionTuple HIPSPVToolChain::computeMSVCVersion(const Driver *D,
+                                                 const ArgList &Args) const {
+  return HostTC.computeMSVCVersion(D, Args);
+}
+
+void HIPSPVToolChain::adjustDebugInfoKind(
+    codegenoptions::DebugInfoKind &DebugInfoKind,
+    const llvm::opt::ArgList &Args) const {
+  // Debug info generation is disabled for SPIRV-LLVM-Translator
+  // which currently aborts on the presence of DW_OP_LLVM_convert.
+  // TODO: Enable debug info when the SPIR-V backend arrives.
+  DebugInfoKind = codegenoptions::NoDebugInfo;
+}
diff --git a/clang/lib/Driver/ToolChains/HIPSPV.h b/clang/lib/Driver/ToolChains/HIPSPV.h
new file mode 100644
index 000000000000..79520f77c742
--- /dev/null
+++ b/clang/lib/Driver/ToolChains/HIPSPV.h
@@ -0,0 +1,103 @@
+//===--- HIPSPV.h - HIP ToolChain Implementations ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPSPV_H
+#define LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPSPV_H
+
+#include "SPIRV.h"
+#include "clang/Driver/Tool.h"
+#include "clang/Driver/ToolChain.h"
+
+namespace clang {
+namespace driver {
+namespace tools {
+namespace HIPSPV {
+
+// Runs llvm-link/opt/llc/lld, which links multiple LLVM bitcode, together with
+// device library, then compiles it to SPIR-V in a shared object.
+class LLVM_LIBRARY_VISIBILITY Linker : public Tool {
+public:
+  Linker(const ToolChain &TC) : Tool("HIPSPV::Linker", "hipspv-link", TC) {}
+
+  bool hasIntegratedCPP() const override { return false; }
+
+  void ConstructJob(Compilation &C, const JobAction &JA,
+                    const InputInfo &Output, const InputInfoList &Inputs,
+                    const llvm::opt::ArgList &TCArgs,
+                    const char *LinkingOutput) const override;
+
+private:
+  void constructLinkAndEmitSpirvCommand(Compilation &C, const JobAction &JA,
+                                        const InputInfoList &Inputs,
+                                        const InputInfo &Output,
+                                        const llvm::opt::ArgList &Args) const;
+};
+
+} // namespace HIPSPV
+} // namespace tools
+
+namespace toolchains {
+
+class LLVM_LIBRARY_VISIBILITY HIPSPVToolChain final : public ToolChain {
+public:
+  HIPSPVToolChain(const Driver &D, const llvm::Triple &Triple,
+                  const ToolChain &HostTC, const llvm::opt::ArgList &Args);
+
+  const llvm::Triple *getAuxTriple() const override {
+    return &HostTC.getTriple();
+  }
+
+  void
+  addClangTargetOptions(const llvm::opt::ArgList &DriverArgs,
+                        llvm::opt::ArgStringList &CC1Args,
+                        Action::OffloadKind DeviceOffloadKind) const override;
+  void addClangWarningOptions(llvm::opt::ArgStringList &CC1Args) const override;
+  CXXStdlibType GetCXXStdlibType(const llvm::opt::ArgList &Args) const override;
+  void
+  AddClangSystemIncludeArgs(const llvm::opt::ArgList &DriverArgs,
+                            llvm::opt::ArgStringList &CC1Args) const override;
+  void AddClangCXXStdlibIncludeArgs(
+      const llvm::opt::ArgList &Args,
+      llvm::opt::ArgStringList &CC1Args) const override;
+  void AddIAMCUIncludeArgs(const llvm::opt::ArgList &DriverArgs,
+                           llvm::opt::ArgStringList &CC1Args) const override;
+  void AddHIPIncludeArgs(const llvm::opt::ArgList &DriverArgs,
+                         llvm::opt::ArgStringList &CC1Args) const override;
+  llvm::SmallVector<BitCodeLibraryInfo, 12>
+  getHIPDeviceLibs(const llvm::opt::ArgList &Args) const override;
+
+  SanitizerMask getSupportedSanitizers() const override;
+
+  VersionTuple
+  computeMSVCVersion(const Driver *D,
+                     const llvm::opt::ArgList &Args) const override;
+
+  void adjustDebugInfoKind(codegenoptions::DebugInfoKind &DebugInfoKind,
+                           const llvm::opt::ArgList &Args) const override;
+  bool IsIntegratedAssemblerDefault() const override { return true; }
+  bool IsMathErrnoDefault() const override { return false; }
+  bool useIntegratedAs() const override { return true; }
+  bool isCrossCompiling() const override { return true; }
+  bool isPICDefault() const override { return false; }
+  bool isPIEDefault(const llvm::opt::ArgList &Args) const override {
+    return false;
+  }
+  bool isPICDefaultForced() const override { return false; }
+  bool SupportsProfiling() const override { return false; }
+
+  const ToolChain &HostTC;
+
+protected:
+  Tool *buildLinker() const override;
+};
+
+} // end namespace toolchains
+} // end namespace driver
+} // end namespace clang
+
+#endif // LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPSPV_H
diff --git a/clang/lib/Driver/ToolChains/HIPUtility.cpp b/clang/lib/Driver/ToolChains/HIPUtility.cpp
new file mode 100644
index 000000000000..1b04a20bacbf
--- /dev/null
+++ b/clang/lib/Driver/ToolChains/HIPUtility.cpp
@@ -0,0 +1,167 @@
+//===--- HIPUtility.cpp - Common HIP Tool Chain Utilities -------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "HIPUtility.h"
+#include "CommonArgs.h"
+#include "clang/Driver/Compilation.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Triple.h"
+#include "llvm/Support/Path.h"
+
+using namespace clang::driver;
+using namespace clang::driver::tools;
+using namespace llvm::opt;
+
+#if defined(_WIN32) || defined(_WIN64)
+#define NULL_FILE "nul"
+#else
+#define NULL_FILE "/dev/null"
+#endif
+
+namespace {
+const unsigned HIPCodeObjectAlign = 4096;
+} // namespace
+
+// Constructs a triple string for clang offload bundler.
+static std::string normalizeForBundler(const llvm::Triple &T,
+                                       bool HasTargetID) {
+  return HasTargetID ? (T.getArchName() + "-" + T.getVendorName() + "-" +
+                        T.getOSName() + "-" + T.getEnvironmentName())
+                           .str()
+                     : T.normalize();
+}
+
+// Construct a clang-offload-bundler command to bundle code objects for
+// different devices into a HIP fat binary.
+void HIP::constructHIPFatbinCommand(Compilation &C, const JobAction &JA,
+                                    llvm::StringRef OutputFileName,
+                                    const InputInfoList &Inputs,
+                                    const llvm::opt::ArgList &Args,
+                                    const Tool &T) {
+  // Construct clang-offload-bundler command to bundle object files for
+  // for different GPU archs.
+  ArgStringList BundlerArgs;
+  BundlerArgs.push_back(Args.MakeArgString("-type=o"));
+  BundlerArgs.push_back(
+      Args.MakeArgString("-bundle-align=" + Twine(HIPCodeObjectAlign)));
+
+  // ToDo: Remove the dummy host binary entry which is required by
+  // clang-offload-bundler.
+  std::string BundlerTargetArg = "-targets=host-x86_64-unknown-linux";
+  std::string BundlerInputArg = "-inputs=" NULL_FILE;
+
+  // AMDGCN:
+  // For code object version 2 and 3, the offload kind in bundle ID is 'hip'
+  // for backward compatibility. For code object version 4 and greater, the
+  // offload kind in bundle ID is 'hipv4'.
+  std::string OffloadKind = "hip";
+  auto &TT = T.getToolChain().getTriple();
+  if (TT.isAMDGCN() && getAMDGPUCodeObjectVersion(C.getDriver(), Args) >= 4)
+    OffloadKind = OffloadKind + "v4";
+  for (const auto &II : Inputs) {
+    const auto *A = II.getAction();
+    auto ArchStr = llvm::StringRef(A->getOffloadingArch());
+    BundlerTargetArg +=
+        "," + OffloadKind + "-" + normalizeForBundler(TT, !ArchStr.empty());
+    if (!ArchStr.empty())
+      BundlerTargetArg += "-" + ArchStr.str();
+    BundlerInputArg = BundlerInputArg + "," + II.getFilename();
+  }
+  BundlerArgs.push_back(Args.MakeArgString(BundlerTargetArg));
+  BundlerArgs.push_back(Args.MakeArgString(BundlerInputArg));
+
+  std::string Output = std::string(OutputFileName);
+  auto *BundlerOutputArg =
+      Args.MakeArgString(std::string("-outputs=").append(Output));
+  BundlerArgs.push_back(BundlerOutputArg);
+
+  const char *Bundler = Args.MakeArgString(
+      T.getToolChain().GetProgramPath("clang-offload-bundler"));
+  C.addCommand(std::make_unique<Command>(
+      JA, T, ResponseFileSupport::None(), Bundler, BundlerArgs, Inputs,
+      InputInfo(&JA, Args.MakeArgString(Output))));
+}
+
+/// Add Generated HIP Object File which has device images embedded into the
+/// host to the argument list for linking. Using MC directives, embed the
+/// device code and also define symbols required by the code generation so that
+/// the image can be retrieved at runtime.
+void HIP::constructGenerateObjFileFromHIPFatBinary(
+    Compilation &C, const InputInfo &Output, const InputInfoList &Inputs,
+    const ArgList &Args, const JobAction &JA, const Tool &T) {
+  const ToolChain &TC = T.getToolChain();
+  std::string Name = std::string(llvm::sys::path::stem(Output.getFilename()));
+
+  // Create Temp Object File Generator,
+  // Offload Bundled file and Bundled Object file.
+  // Keep them if save-temps is enabled.
+  const char *McinFile;
+  const char *BundleFile;
+  if (C.getDriver().isSaveTempsEnabled()) {
+    McinFile = C.getArgs().MakeArgString(Name + ".mcin");
+    BundleFile = C.getArgs().MakeArgString(Name + ".hipfb");
+  } else {
+    auto TmpNameMcin = C.getDriver().GetTemporaryPath(Name, "mcin");
+    McinFile = C.addTempFile(C.getArgs().MakeArgString(TmpNameMcin));
+    auto TmpNameFb = C.getDriver().GetTemporaryPath(Name, "hipfb");
+    BundleFile = C.addTempFile(C.getArgs().MakeArgString(TmpNameFb));
+  }
+  HIP::constructHIPFatbinCommand(C, JA, BundleFile, Inputs, Args, T);
+
+  // Create a buffer to write the contents of the temp obj generator.
+  std::string ObjBuffer;
+  llvm::raw_string_ostream ObjStream(ObjBuffer);
+
+  auto HostTriple =
+      C.getSingleOffloadToolChain<Action::OFK_Host>()->getTriple();
+
+  // Add MC directives to embed target binaries. We ensure that each
+  // section and image is 16-byte aligned. This is not mandatory, but
+  // increases the likelihood of data to be aligned with a cache block
+  // in several main host machines.
+  ObjStream << "#       HIP Object Generator\n";
+  ObjStream << "# *** Automatically generated by Clang ***\n";
+  if (HostTriple.isWindowsMSVCEnvironment()) {
+    ObjStream << "  .section .hip_fatbin, \"dw\"\n";
+  } else {
+    ObjStream << "  .protected __hip_fatbin\n";
+    ObjStream << "  .type __hip_fatbin,@object\n";
+    ObjStream << "  .section .hip_fatbin,\"a\",@progbits\n";
+  }
+  ObjStream << "  .globl __hip_fatbin\n";
+  ObjStream << "  .p2align " << llvm::Log2(llvm::Align(HIPCodeObjectAlign))
+            << "\n";
+  ObjStream << "__hip_fatbin:\n";
+  ObjStream << "  .incbin ";
+  llvm::sys::printArg(ObjStream, BundleFile, /*Quote=*/true);
+  ObjStream << "\n";
+  ObjStream.flush();
+
+  // Dump the contents of the temp object file gen if the user requested that.
+  // We support this option to enable testing of behavior with -###.
+  if (C.getArgs().hasArg(options::OPT_fhip_dump_offload_linker_script))
+    llvm::errs() << ObjBuffer;
+
+  // Open script file and write the contents.
+  std::error_code EC;
+  llvm::raw_fd_ostream Objf(McinFile, EC, llvm::sys::fs::OF_None);
+
+  if (EC) {
+    C.getDriver().Diag(clang::diag::err_unable_to_make_temp) << EC.message();
+    return;
+  }
+
+  Objf << ObjBuffer;
+
+  ArgStringList McArgs{"-triple", Args.MakeArgString(HostTriple.normalize()),
+                       "-o",      Output.getFilename(),
+                       McinFile,  "--filetype=obj"};
+  const char *Mc = Args.MakeArgString(TC.GetProgramPath("llvm-mc"));
+  C.addCommand(std::make_unique<Command>(JA, T, ResponseFileSupport::None(), Mc,
+                                         McArgs, Inputs, Output));
+}
diff --git a/clang/lib/Driver/ToolChains/HIPUtility.h b/clang/lib/Driver/ToolChains/HIPUtility.h
new file mode 100644
index 000000000000..29e5a922024a
--- /dev/null
+++ b/clang/lib/Driver/ToolChains/HIPUtility.h
@@ -0,0 +1,35 @@
+//===--- HIPUtility.h - Common HIP Tool Chain Utilities ---------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPUTILITY_H
+#define LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPUTILITY_H
+
+#include "clang/Driver/Tool.h"
+
+namespace clang {
+namespace driver {
+namespace tools {
+namespace HIP {
+
+// Construct command for creating HIP fatbin.
+void constructHIPFatbinCommand(Compilation &C, const JobAction &JA,
+                               StringRef OutputFileName,
+                               const InputInfoList &Inputs,
+                               const llvm::opt::ArgList &TCArgs, const Tool &T);
+
+// Construct command for creating Object from HIP fatbin.
+void constructGenerateObjFileFromHIPFatBinary(
+    Compilation &C, const InputInfo &Output, const InputInfoList &Inputs,
+    const llvm::opt::ArgList &Args, const JobAction &JA, const Tool &T);
+
+} // namespace HIP
+} // namespace tools
+} // namespace driver
+} // namespace clang
+
+#endif // LLVM_CLANG_LIB_DRIVER_TOOLCHAINS_HIPUTILITY_H
diff --git a/clang/lib/Driver/ToolChains/Hexagon.cpp b/clang/lib/Driver/ToolChains/Hexagon.cpp
index 18270818d158..ba3040636604 100644
--- a/clang/lib/Driver/ToolChains/Hexagon.cpp
+++ b/clang/lib/Driver/ToolChains/Hexagon.cpp
@@ -26,8 +26,8 @@ using namespace clang;
 using namespace llvm::opt;
 
 // Default hvx-length for various versions.
-static StringRef getDefaultHvxLength(StringRef Cpu) {
-  return llvm::StringSwitch<StringRef>(Cpu)
+static StringRef getDefaultHvxLength(StringRef HvxVer) {
+  return llvm::StringSwitch<StringRef>(HvxVer)
       .Case("v60", "64b")
       .Case("v62", "64b")
       .Case("v65", "64b")
@@ -51,42 +51,107 @@ static void handleHVXTargetFeatures(const Driver &D, const ArgList &Args,
   // Handle HVX warnings.
   handleHVXWarnings(D, Args);
 
-  // Add the +hvx* features based on commandline flags.
-  StringRef HVXFeature, HVXLength;
-
-  // Handle -mhvx, -mhvx=, -mno-hvx.
-  if (Arg *A = Args.getLastArg(options::OPT_mno_hexagon_hvx,
-                               options::OPT_mhexagon_hvx,
-                               options::OPT_mhexagon_hvx_EQ)) {
-    if (A->getOption().matches(options::OPT_mno_hexagon_hvx))
-      return;
-    if (A->getOption().matches(options::OPT_mhexagon_hvx_EQ)) {
-      HasHVX = true;
-      HVXFeature = Cpu = A->getValue();
-      HVXFeature = Args.MakeArgString(llvm::Twine("+hvx") + HVXFeature.lower());
-    } else if (A->getOption().matches(options::OPT_mhexagon_hvx)) {
-      HasHVX = true;
-      HVXFeature = Args.MakeArgString(llvm::Twine("+hvx") + Cpu);
+  auto makeFeature = [&Args](Twine T, bool Enable) -> StringRef {
+    const std::string &S = T.str();
+    StringRef Opt(S);
+    if (Opt.endswith("="))
+      Opt = Opt.drop_back(1);
+    if (Opt.startswith("mno-"))
+      Opt = Opt.drop_front(4);
+    else if (Opt.startswith("m"))
+      Opt = Opt.drop_front(1);
+    return Args.MakeArgString(Twine(Enable ? "+" : "-") + Twine(Opt));
+  };
+
+  auto withMinus = [](StringRef S) -> std::string {
+    return "-" + S.str();
+  };
+
+  // Drop tiny core suffix for HVX version.
+  std::string HvxVer =
+      (Cpu.back() == 'T' || Cpu.back() == 't' ? Cpu.drop_back(1) : Cpu).str();
+  HasHVX = false;
+
+  // Handle -mhvx, -mhvx=, -mno-hvx. If both present, -mhvx= wins over -mhvx.
+  auto argOrNull = [&Args](auto FlagOn, auto FlagOff) -> Arg* {
+    if (Arg *A = Args.getLastArg(FlagOn, FlagOff)) {
+      if (A->getOption().matches(FlagOn))
+        return A;
     }
-    Features.push_back(HVXFeature);
+    return nullptr;
+  };
+
+  Arg *HvxBareA =
+      argOrNull(options::OPT_mhexagon_hvx, options::OPT_mno_hexagon_hvx);
+  Arg *HvxVerA =
+      argOrNull(options::OPT_mhexagon_hvx_EQ, options::OPT_mno_hexagon_hvx);
+
+  if (Arg *A = HvxVerA ? HvxVerA : HvxBareA) {
+    if (A->getOption().matches(options::OPT_mhexagon_hvx_EQ))
+      HvxVer = StringRef(A->getValue()).lower(); // lower produces std:string
+    HasHVX = true;
+    Features.push_back(makeFeature(Twine("hvx") + HvxVer, true));
+  } else if (Arg *A = Args.getLastArg(options::OPT_mno_hexagon_hvx)) {
+    // If there was an explicit -mno-hvx, add -hvx to target features.
+    Features.push_back(makeFeature(A->getOption().getName(), false));
   }
 
+  StringRef HvxLen = getDefaultHvxLength(HvxVer);
+
   // Handle -mhvx-length=.
   if (Arg *A = Args.getLastArg(options::OPT_mhexagon_hvx_length_EQ)) {
     // These flags are valid only if HVX in enabled.
     if (!HasHVX)
-      D.Diag(diag::err_drv_invalid_hvx_length);
+      D.Diag(diag::err_drv_needs_hvx) << withMinus(A->getOption().getName());
     else if (A->getOption().matches(options::OPT_mhexagon_hvx_length_EQ))
-      HVXLength = A->getValue();
+      HvxLen = A->getValue();
+  }
+
+  if (HasHVX) {
+    StringRef L = makeFeature(Twine("hvx-length") + HvxLen.lower(), true);
+    Features.push_back(L);
   }
-  // Default hvx-length based on Cpu.
-  else if (HasHVX)
-    HVXLength = getDefaultHvxLength(Cpu);
-
-  if (!HVXLength.empty()) {
-    HVXFeature =
-        Args.MakeArgString(llvm::Twine("+hvx-length") + HVXLength.lower());
-    Features.push_back(HVXFeature);
+
+  unsigned HvxVerNum;
+  // getAsInteger returns 'true' on error.
+  if (StringRef(HvxVer).drop_front(1).getAsInteger(10, HvxVerNum))
+    HvxVerNum = 0;
+
+  // Handle HVX floating point flags.
+  auto checkFlagHvxVersion = [&](auto FlagOn, auto FlagOff,
+                                 unsigned MinVerNum) -> Optional<StringRef> {
+    // Return an Optional<StringRef>:
+    // - None indicates a verification failure, or that the flag was not
+    //   present in Args.
+    // - Otherwise the returned value is that name of the feature to add
+    //   to Features.
+    Arg *A = Args.getLastArg(FlagOn, FlagOff);
+    if (!A)
+      return None;
+
+    StringRef OptName = A->getOption().getName();
+    if (A->getOption().matches(FlagOff))
+      return makeFeature(OptName, false);
+
+    if (!HasHVX) {
+      D.Diag(diag::err_drv_needs_hvx) << withMinus(OptName);
+      return None;
+    }
+    if (HvxVerNum < MinVerNum) {
+      D.Diag(diag::err_drv_needs_hvx_version)
+          << withMinus(OptName) << ("v" + std::to_string(HvxVerNum));
+      return None;
+    }
+    return makeFeature(OptName, true);
+  };
+
+  if (auto F = checkFlagHvxVersion(options::OPT_mhexagon_hvx_qfloat,
+                                   options::OPT_mno_hexagon_hvx_qfloat, 68)) {
+    Features.push_back(*F);
+  }
+  if (auto F = checkFlagHvxVersion(options::OPT_mhexagon_hvx_ieee_fp,
+                                   options::OPT_mno_hexagon_hvx_ieee_fp, 68)) {
+    Features.push_back(*F);
   }
 }
 
@@ -117,7 +182,7 @@ void hexagon::getHexagonTargetFeatures(const Driver &D, const ArgList &Args,
   handleHVXTargetFeatures(D, Args, Features, Cpu, HasHVX);
 
   if (HexagonToolChain::isAutoHVXEnabled(Args) && !HasHVX)
-    D.Diag(diag::warn_drv_vectorize_needs_hvx);
+    D.Diag(diag::warn_drv_needs_hvx) << "auto-vectorization";
 }
 
 // Hexagon tools start.
@@ -156,6 +221,12 @@ void hexagon::Assembler::ConstructJob(Compilation &C, const JobAction &JA,
     CmdArgs.push_back("-fsyntax-only");
   }
 
+  if (Arg *A = Args.getLastArg(options::OPT_mhexagon_hvx_ieee_fp,
+                               options::OPT_mno_hexagon_hvx_ieee_fp)) {
+    if (A->getOption().matches(options::OPT_mhexagon_hvx_ieee_fp))
+      CmdArgs.push_back("-mhvx-ieee-fp");
+  }
+
   if (auto G = toolchains::HexagonToolChain::getSmallDataThreshold(Args)) {
     CmdArgs.push_back(Args.MakeArgString("-gpsize=" + Twine(G.getValue())));
   }
@@ -226,6 +297,7 @@ constructHexagonLinkArgs(Compilation &C, const JobAction &JA,
   StringRef CpuVer = toolchains::HexagonToolChain::GetTargetCPUVersion(Args);
 
   bool NeedsSanitizerDeps = addSanitizerRuntimes(HTC, Args, CmdArgs);
+  bool NeedsXRayDeps = addXRayRuntime(HTC, Args, CmdArgs);
 
   //----------------------------------------------------------------------------
   // Silence warnings for various options
@@ -297,6 +369,8 @@ constructHexagonLinkArgs(Compilation &C, const JobAction &JA,
 
         CmdArgs.push_back("-lunwind");
       }
+      if (NeedsXRayDeps)
+        linkXRayRuntimeDeps(HTC, CmdArgs);
 
       CmdArgs.push_back("-lclang_rt.builtins-hexagon");
       CmdArgs.push_back("-lc");
diff --git a/clang/lib/Driver/ToolChains/Linux.cpp b/clang/lib/Driver/ToolChains/Linux.cpp
index 198774506e5e..e413640abad3 100644
--- a/clang/lib/Driver/ToolChains/Linux.cpp
+++ b/clang/lib/Driver/ToolChains/Linux.cpp
@@ -277,14 +277,11 @@ Linux::Linux(const Driver &D, const llvm::Triple &Triple, const ArgList &Args)
     // Android sysroots contain a library directory for each supported OS
     // version as well as some unversioned libraries in the usual multiarch
     // directory.
-    unsigned Major;
-    unsigned Minor;
-    unsigned Micro;
-    Triple.getEnvironmentVersion(Major, Minor, Micro);
-    addPathIfExists(D,
-                    SysRoot + "/usr/lib/" + MultiarchTriple + "/" +
-                        llvm::to_string(Major),
-                    Paths);
+    addPathIfExists(
+        D,
+        SysRoot + "/usr/lib/" + MultiarchTriple + "/" +
+            llvm::to_string(Triple.getEnvironmentVersion().getMajor()),
+        Paths);
   }
 
   addPathIfExists(D, SysRoot + "/usr/lib/" + MultiarchTriple, Paths);
@@ -666,8 +663,8 @@ void Linux::AddIAMCUIncludeArgs(const ArgList &DriverArgs,
 }
 
 bool Linux::isPIEDefault(const llvm::opt::ArgList &Args) const {
-  return getTriple().isAndroid() || getTriple().isMusl() ||
-         getSanitizerArgs(Args).requiresPIE();
+  return CLANG_DEFAULT_PIE_ON_LINUX || getTriple().isAndroid() ||
+         getTriple().isMusl() || getSanitizerArgs(Args).requiresPIE();
 }
 
 bool Linux::IsAArch64OutlineAtomicsDefault(const ArgList &Args) const {
diff --git a/clang/lib/Driver/ToolChains/MSVC.cpp b/clang/lib/Driver/ToolChains/MSVC.cpp
index 792b0a51fea0..66e9d8ab525a 100644
--- a/clang/lib/Driver/ToolChains/MSVC.cpp
+++ b/clang/lib/Driver/ToolChains/MSVC.cpp
@@ -1194,14 +1194,6 @@ bool MSVCToolChain::getUniversalCRTLibraryPath(const ArgList &Args,
   return true;
 }
 
-static VersionTuple getMSVCVersionFromTriple(const llvm::Triple &Triple) {
-  unsigned Major, Minor, Micro;
-  Triple.getEnvironmentVersion(Major, Minor, Micro);
-  if (Major || Minor || Micro)
-    return VersionTuple(Major, Minor, Micro);
-  return VersionTuple();
-}
-
 static VersionTuple getMSVCVersionFromExe(const std::string &BinDir) {
   VersionTuple Version;
 #ifdef _WIN32
@@ -1374,7 +1366,7 @@ VersionTuple MSVCToolChain::computeMSVCVersion(const Driver *D,
   bool IsWindowsMSVC = getTriple().isWindowsMSVCEnvironment();
   VersionTuple MSVT = ToolChain::computeMSVCVersion(D, Args);
   if (MSVT.empty())
-    MSVT = getMSVCVersionFromTriple(getTriple());
+    MSVT = getTriple().getEnvironmentVersion();
   if (MSVT.empty() && IsWindowsMSVC)
     MSVT = getMSVCVersionFromExe(getSubDirectoryPath(SubDirectoryType::Bin));
   if (MSVT.empty() &&
diff --git a/clang/lib/Driver/ToolChains/NetBSD.cpp b/clang/lib/Driver/ToolChains/NetBSD.cpp
index 7571398b7cc6..37b1fc5215ff 100644
--- a/clang/lib/Driver/ToolChains/NetBSD.cpp
+++ b/clang/lib/Driver/ToolChains/NetBSD.cpp
@@ -270,10 +270,9 @@ void netbsd::Linker::ConstructJob(Compilation &C, const JobAction &JA,
     CmdArgs.push_back(Args.MakeArgString(ToolChain.getCompilerRTPath()));
   }
 
-  unsigned Major, Minor, Micro;
-  Triple.getOSVersion(Major, Minor, Micro);
+  VersionTuple OsVersion = Triple.getOSVersion();
   bool useLibgcc = true;
-  if (Major >= 7 || Major == 0) {
+  if (OsVersion >= VersionTuple(7) || OsVersion.getMajor() == 0) {
     switch (ToolChain.getArch()) {
     case llvm::Triple::aarch64:
     case llvm::Triple::aarch64_be:
@@ -409,9 +408,8 @@ Tool *NetBSD::buildAssembler() const {
 Tool *NetBSD::buildLinker() const { return new tools::netbsd::Linker(*this); }
 
 ToolChain::CXXStdlibType NetBSD::GetDefaultCXXStdlibType() const {
-  unsigned Major, Minor, Micro;
-  getTriple().getOSVersion(Major, Minor, Micro);
-  if (Major >= 7 || Major == 0) {
+  VersionTuple OsVersion = getTriple().getOSVersion();
+  if (OsVersion >= VersionTuple(7) || OsVersion.getMajor() == 0) {
     switch (getArch()) {
     case llvm::Triple::aarch64:
     case llvm::Triple::aarch64_be:
@@ -505,14 +503,13 @@ void NetBSD::addClangTargetOptions(const ArgList &DriverArgs,
   if (SanArgs.hasAnySanitizer())
     CC1Args.push_back("-D_REENTRANT");
 
-  unsigned Major, Minor, Micro;
-  getTriple().getOSVersion(Major, Minor, Micro);
+  VersionTuple OsVersion = getTriple().getOSVersion();
   bool UseInitArrayDefault =
-    Major >= 9 || Major == 0 ||
-    getTriple().getArch() == llvm::Triple::aarch64 ||
-    getTriple().getArch() == llvm::Triple::aarch64_be ||
-    getTriple().getArch() == llvm::Triple::arm ||
-    getTriple().getArch() == llvm::Triple::armeb;
+      OsVersion >= VersionTuple(9) || OsVersion.getMajor() == 0 ||
+      getTriple().getArch() == llvm::Triple::aarch64 ||
+      getTriple().getArch() == llvm::Triple::aarch64_be ||
+      getTriple().getArch() == llvm::Triple::arm ||
+      getTriple().getArch() == llvm::Triple::armeb;
 
   if (!DriverArgs.hasFlag(options::OPT_fuse_init_array,
                           options::OPT_fno_use_init_array, UseInitArrayDefault))
diff --git a/clang/lib/Driver/ToolChains/SPIRV.cpp b/clang/lib/Driver/ToolChains/SPIRV.cpp
index 16e72d3c733f..50d03e79bbb0 100644
--- a/clang/lib/Driver/ToolChains/SPIRV.cpp
+++ b/clang/lib/Driver/ToolChains/SPIRV.cpp
@@ -13,6 +13,7 @@
 #include "clang/Driver/Options.h"
 
 using namespace clang::driver;
+using namespace clang::driver::toolchains;
 using namespace clang::driver::tools;
 using namespace llvm::opt;
 
@@ -27,7 +28,7 @@ void SPIRV::constructTranslateCommand(Compilation &C, const Tool &T,
   if (Input.getType() == types::TY_PP_Asm)
     CmdArgs.push_back("-to-binary");
   if (Output.getType() == types::TY_PP_Asm)
-    CmdArgs.push_back("-spirv-text");
+    CmdArgs.push_back("--spirv-tools-dis");
 
   CmdArgs.append({"-o", Output.getFilename()});
 
@@ -47,3 +48,25 @@ void SPIRV::Translator::ConstructJob(Compilation &C, const JobAction &JA,
     llvm_unreachable("Invalid number of input files.");
   constructTranslateCommand(C, *this, JA, Output, Inputs[0], {});
 }
+
+clang::driver::Tool *SPIRVToolChain::getTranslator() const {
+  if (!Translator)
+    Translator = std::make_unique<SPIRV::Translator>(*this);
+  return Translator.get();
+}
+
+clang::driver::Tool *SPIRVToolChain::SelectTool(const JobAction &JA) const {
+  Action::ActionClass AC = JA.getKind();
+  return SPIRVToolChain::getTool(AC);
+}
+
+clang::driver::Tool *SPIRVToolChain::getTool(Action::ActionClass AC) const {
+  switch (AC) {
+  default:
+    break;
+  case Action::BackendJobClass:
+  case Action::AssembleJobClass:
+    return SPIRVToolChain::getTranslator();
+  }
+  return ToolChain::getTool(AC);
+}
diff --git a/clang/lib/Driver/ToolChains/SPIRV.h b/clang/lib/Driver/ToolChains/SPIRV.h
index 35d0446bd8b8..229f7018e3b5 100644
--- a/clang/lib/Driver/ToolChains/SPIRV.h
+++ b/clang/lib/Driver/ToolChains/SPIRV.h
@@ -41,6 +41,39 @@ public:
 
 } // namespace SPIRV
 } // namespace tools
+
+namespace toolchains {
+
+class LLVM_LIBRARY_VISIBILITY SPIRVToolChain final : public ToolChain {
+  mutable std::unique_ptr<Tool> Translator;
+
+public:
+  SPIRVToolChain(const Driver &D, const llvm::Triple &Triple,
+                 const llvm::opt::ArgList &Args)
+      : ToolChain(D, Triple, Args) {}
+
+  bool useIntegratedAs() const override { return true; }
+  bool useIntegratedBackend() const override { return false; }
+
+  bool IsMathErrnoDefault() const override { return false; }
+  bool isCrossCompiling() const override { return true; }
+  bool isPICDefault() const override { return false; }
+  bool isPIEDefault(const llvm::opt::ArgList &Args) const override {
+    return false;
+  }
+  bool isPICDefaultForced() const override { return false; }
+  bool SupportsProfiling() const override { return false; }
+
+  clang::driver::Tool *SelectTool(const JobAction &JA) const override;
+
+protected:
+  clang::driver::Tool *getTool(Action::ActionClass AC) const override;
+
+private:
+  clang::driver::Tool *getTranslator() const;
+};
+
+} // namespace toolchains
 } // namespace driver
 } // namespace clang
 #endif
diff --git a/clang/lib/Driver/ToolChains/VEToolchain.cpp b/clang/lib/Driver/ToolChains/VEToolchain.cpp
index 1fcc52684baa..4cdeec7f9d8a 100644
--- a/clang/lib/Driver/ToolChains/VEToolchain.cpp
+++ b/clang/lib/Driver/ToolChains/VEToolchain.cpp
@@ -28,17 +28,27 @@ VEToolChain::VEToolChain(const Driver &D, const llvm::Triple &Triple,
   getProgramPaths().push_back("/opt/nec/ve/bin");
   // ProgramPaths are found via 'PATH' environment variable.
 
-  // default file paths are:
-  //   ${RESOURCEDIR}/lib/linux/ve (== getArchSpecificLibPath)
-  //   /lib/../lib64
-  //   /usr/lib/../lib64
-  //   ${BINPATH}/../lib
-  //   /lib
-  //   /usr/lib
-  //
-  // These are OK for host, but no go for VE.  So, defines them all
-  // from scratch here.
+  // Default library paths are following:
+  //   ${RESOURCEDIR}/lib/ve-unknown-linux-gnu,
+  // These are OK.
+
+  // Default file paths are following:
+  //   ${RESOURCEDIR}/lib/linux/ve, (== getArchSpecificLibPath)
+  //   /lib/../lib64,
+  //   /usr/lib/../lib64,
+  //   ${BINPATH}/../lib,
+  //   /lib,
+  //   /usr/lib,
+  // These are OK for host, but no go for VE.
+
+  // Define file paths from scratch here.
   getFilePaths().clear();
+
+  // Add library directories:
+  //   ${BINPATH}/../lib/ve-unknown-linux-gnu, (== getStdlibPath)
+  //   ${RESOURCEDIR}/lib/linux/ve, (== getArchSpecificLibPath)
+  //   ${SYSROOT}/opt/nec/ve/lib,
+  getFilePaths().push_back(getStdlibPath());
   getFilePaths().push_back(getArchSpecificLibPath());
   getFilePaths().push_back(computeSysRoot() + "/opt/nec/ve/lib");
 }
@@ -115,9 +125,10 @@ void VEToolChain::AddClangCXXStdlibIncludeArgs(const ArgList &DriverArgs,
     ArrayRef<StringRef> DirVec(Dirs);
     addSystemIncludes(DriverArgs, CC1Args, DirVec);
   } else {
-    SmallString<128> P(getDriver().ResourceDir);
-    llvm::sys::path::append(P, "include/c++/v1");
-    addSystemInclude(DriverArgs, CC1Args, P);
+    // Add following paths for multiple target installation.
+    //   ${INSTALLDIR}/include/ve-unknown-linux-gnu/c++/v1,
+    //   ${INSTALLDIR}/include/c++/v1,
+    addLibCxxIncludePaths(DriverArgs, CC1Args);
   }
 }
 
diff --git a/clang/lib/Driver/XRayArgs.cpp b/clang/lib/Driver/XRayArgs.cpp
index b44509ad3b88..63b575178bd1 100644
--- a/clang/lib/Driver/XRayArgs.cpp
+++ b/clang/lib/Driver/XRayArgs.cpp
@@ -40,6 +40,7 @@ XRayArgs::XRayArgs(const ToolChain &TC, const ArgList &Args) {
     case llvm::Triple::x86_64:
     case llvm::Triple::arm:
     case llvm::Triple::aarch64:
+    case llvm::Triple::hexagon:
     case llvm::Triple::ppc64le:
     case llvm::Triple::mips:
     case llvm::Triple::mipsel:
diff --git a/clang/lib/Format/BreakableToken.cpp b/clang/lib/Format/BreakableToken.cpp
index 968b35bfda23..5d03c9811e1b 100644
--- a/clang/lib/Format/BreakableToken.cpp
+++ b/clang/lib/Format/BreakableToken.cpp
@@ -91,7 +91,7 @@ getCommentSplit(StringRef Text, unsigned ContentStartColumn,
   // In JavaScript, some @tags can be followed by {, and machinery that parses
   // these comments will fail to understand the comment if followed by a line
   // break. So avoid ever breaking before a {.
-  if (Style.Language == FormatStyle::LK_JavaScript) {
+  if (Style.isJavaScript()) {
     StringRef::size_type SpaceOffset =
         Text.find_first_of(Blanks, MaxSplitBytes);
     if (SpaceOffset != StringRef::npos && SpaceOffset + 1 < Text.size() &&
@@ -127,8 +127,7 @@ getCommentSplit(StringRef Text, unsigned ContentStartColumn,
     }
 
     // Avoid ever breaking before a @tag or a { in JavaScript.
-    if (Style.Language == FormatStyle::LK_JavaScript &&
-        SpaceOffset + 1 < Text.size() &&
+    if (Style.isJavaScript() && SpaceOffset + 1 < Text.size() &&
         (Text[SpaceOffset + 1] == '{' || Text[SpaceOffset + 1] == '@')) {
       SpaceOffset = Text.find_last_of(Blanks, SpaceOffset);
       continue;
@@ -460,8 +459,7 @@ BreakableBlockComment::BreakableBlockComment(
   IndentAtLineBreak = std::max<unsigned>(IndentAtLineBreak, Decoration.size());
 
   // Detect a multiline jsdoc comment and set DelimitersOnNewline in that case.
-  if (Style.Language == FormatStyle::LK_JavaScript ||
-      Style.Language == FormatStyle::LK_Java) {
+  if (Style.isJavaScript() || Style.Language == FormatStyle::LK_Java) {
     if ((Lines[0] == "*" || Lines[0].startswith("* ")) && Lines.size() > 1) {
       // This is a multiline jsdoc comment.
       DelimitersOnNewline = true;
@@ -580,8 +578,7 @@ const llvm::StringSet<>
 };
 
 unsigned BreakableBlockComment::getContentIndent(unsigned LineIndex) const {
-  if (Style.Language != FormatStyle::LK_Java &&
-      Style.Language != FormatStyle::LK_JavaScript)
+  if (Style.Language != FormatStyle::LK_Java && !Style.isJavaScript())
     return 0;
   // The content at LineIndex 0 of a comment like:
   // /** line 0 */
diff --git a/clang/lib/Format/ContinuationIndenter.cpp b/clang/lib/Format/ContinuationIndenter.cpp
index 5073f5105d05..4225d6b67b0e 100644
--- a/clang/lib/Format/ContinuationIndenter.cpp
+++ b/clang/lib/Format/ContinuationIndenter.cpp
@@ -422,8 +422,7 @@ bool ContinuationIndenter::mustBreak(const LineState &State) {
       //     ...
       //   }.bind(...));
       // FIXME: We should find a more generic solution to this problem.
-      !(State.Column <= NewLineColumn &&
-        Style.Language == FormatStyle::LK_JavaScript) &&
+      !(State.Column <= NewLineColumn && Style.isJavaScript()) &&
       !(Previous.closesScopeAfterBlock() && State.Column <= NewLineColumn))
     return true;
 
@@ -493,14 +492,14 @@ bool ContinuationIndenter::mustBreak(const LineState &State) {
       return true;
   }
 
-  // Break after the closing parenthesis of TypeScript decorators before
-  // functions, getters and setters.
-  static const llvm::StringSet<> BreakBeforeDecoratedTokens = {"get", "set",
-                                                               "function"};
-  if (Style.Language == FormatStyle::LK_JavaScript &&
-      BreakBeforeDecoratedTokens.contains(Current.TokenText) &&
-      Previous.is(tok::r_paren) && Previous.is(TT_JavaAnnotation)) {
-    return true;
+  if (Style.isJavaScript() && Previous.is(tok::r_paren) &&
+      Previous.is(TT_JavaAnnotation)) {
+    // Break after the closing parenthesis of TypeScript decorators before
+    // functions, getters and setters.
+    static const llvm::StringSet<> BreakBeforeDecoratedTokens = {"get", "set",
+                                                                 "function"};
+    if (BreakBeforeDecoratedTokens.contains(Current.TokenText))
+      return true;
   }
 
   // If the return type spans multiple lines, wrap before the function name.
@@ -510,7 +509,7 @@ bool ContinuationIndenter::mustBreak(const LineState &State) {
          Style.AlwaysBreakAfterReturnType != FormatStyle::RTBS_None) &&
         // Don't always break between a JavaScript `function` and the function
         // name.
-        Style.Language != FormatStyle::LK_JavaScript) ||
+        !Style.isJavaScript()) ||
        (Current.is(tok::kw_operator) && !Previous.is(tok::coloncolon))) &&
       !Previous.is(tok::kw_template) && State.Stack.back().BreakBeforeParameter)
     return true;
@@ -827,9 +826,8 @@ unsigned ContinuationIndenter::addTokenOnNewLine(LineState &State,
   // is common and should be formatted like a free-standing function. The same
   // goes for wrapping before the lambda return type arrow.
   if (!Current.is(TT_LambdaArrow) &&
-      (Style.Language != FormatStyle::LK_JavaScript ||
-       Current.NestingLevel != 0 || !PreviousNonComment ||
-       !PreviousNonComment->is(tok::equal) ||
+      (!Style.isJavaScript() || Current.NestingLevel != 0 ||
+       !PreviousNonComment || !PreviousNonComment->is(tok::equal) ||
        !Current.isOneOf(Keywords.kw_async, Keywords.kw_function)))
     State.Stack.back().NestedBlockIndent = State.Column;
 
@@ -1337,6 +1335,9 @@ unsigned ContinuationIndenter::moveStateToNextToken(LineState &State,
 void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
                                                     bool Newline) {
   const FormatToken &Current = *State.NextToken;
+  if (Current.FakeLParens.empty())
+    return;
+
   const FormatToken *Previous = Current.getPreviousNonComment();
 
   // Don't add extra indentation for the first fake parenthesis after
@@ -1348,10 +1349,7 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
                     (Previous->getPrecedence() == prec::Assignment &&
                      Style.AlignOperands != FormatStyle::OAS_DontAlign) ||
                     Previous->is(TT_ObjCMethodExpr)));
-  for (SmallVectorImpl<prec::Level>::const_reverse_iterator
-           I = Current.FakeLParens.rbegin(),
-           E = Current.FakeLParens.rend();
-       I != E; ++I) {
+  for (const auto &PrecedenceLevel : llvm::reverse(Current.FakeLParens)) {
     ParenState NewParenState = State.Stack.back();
     NewParenState.Tok = nullptr;
     NewParenState.ContainsLineBreak = false;
@@ -1363,7 +1361,7 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
         NewParenState.NoLineBreak || State.Stack.back().NoLineBreakInOperand;
 
     // Don't propagate AvoidBinPacking into subexpressions of arg/param lists.
-    if (*I > prec::Comma)
+    if (PrecedenceLevel > prec::Comma)
       NewParenState.AvoidBinPacking = false;
 
     // Indent from 'LastSpace' unless these are fake parentheses encapsulating
@@ -1371,11 +1369,11 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
     // brackets is disabled.
     if (!Current.isTrailingComment() &&
         (Style.AlignOperands != FormatStyle::OAS_DontAlign ||
-         *I < prec::Assignment) &&
+         PrecedenceLevel < prec::Assignment) &&
         (!Previous || Previous->isNot(tok::kw_return) ||
-         (Style.Language != FormatStyle::LK_Java && *I > 0)) &&
+         (Style.Language != FormatStyle::LK_Java && PrecedenceLevel > 0)) &&
         (Style.AlignAfterOpenBracket != FormatStyle::BAS_DontAlign ||
-         *I != prec::Comma || Current.NestingLevel == 0)) {
+         PrecedenceLevel != prec::Comma || Current.NestingLevel == 0)) {
       NewParenState.Indent =
           std::max(std::max(State.Column, NewParenState.Indent),
                    State.Stack.back().LastSpace);
@@ -1384,7 +1382,7 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
     if (Previous &&
         (Previous->getPrecedence() == prec::Assignment ||
          Previous->is(tok::kw_return) ||
-         (*I == prec::Conditional && Previous->is(tok::question) &&
+         (PrecedenceLevel == prec::Conditional && Previous->is(tok::question) &&
           Previous->is(TT_ConditionalExpr))) &&
         !Newline) {
       // If BreakBeforeBinaryOperators is set, un-indent a bit to account for
@@ -1402,9 +1400,9 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
     //       ParameterToInnerFunction));
     //   OuterFunction(SomeObject.InnerFunctionCall( // break
     //       ParameterToInnerFunction));
-    if (*I > prec::Unknown)
+    if (PrecedenceLevel > prec::Unknown)
       NewParenState.LastSpace = std::max(NewParenState.LastSpace, State.Column);
-    if (*I != prec::Conditional && !Current.is(TT_UnaryOperator) &&
+    if (PrecedenceLevel != prec::Conditional && !Current.is(TT_UnaryOperator) &&
         Style.AlignAfterOpenBracket != FormatStyle::BAS_DontAlign)
       NewParenState.StartOfFunctionCall = State.Column;
 
@@ -1413,17 +1411,18 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State,
     // an assignment (i.e. *I <= prec::Assignment) as those have different
     // indentation rules. Indent other expression, unless the indentation needs
     // to be skipped.
-    if (*I == prec::Conditional && Previous && Previous->is(tok::colon) &&
-        Previous->is(TT_ConditionalExpr) && I == Current.FakeLParens.rbegin() &&
+    if (PrecedenceLevel == prec::Conditional && Previous &&
+        Previous->is(tok::colon) && Previous->is(TT_ConditionalExpr) &&
+        &PrecedenceLevel == &Current.FakeLParens.back() &&
         !State.Stack.back().IsWrappedConditional) {
       NewParenState.IsChainedConditional = true;
       NewParenState.UnindentOperator = State.Stack.back().UnindentOperator;
-    } else if (*I == prec::Conditional ||
-               (!SkipFirstExtraIndent && *I > prec::Assignment &&
+    } else if (PrecedenceLevel == prec::Conditional ||
+               (!SkipFirstExtraIndent && PrecedenceLevel > prec::Assignment &&
                 !Current.isTrailingComment())) {
       NewParenState.Indent += Style.ContinuationIndentWidth;
     }
-    if ((Previous && !Previous->opensScope()) || *I != prec::Comma)
+    if ((Previous && !Previous->opensScope()) || PrecedenceLevel != prec::Comma)
       NewParenState.BreakBeforeParameter = false;
     State.Stack.push_back(NewParenState);
     SkipFirstExtraIndent = false;
@@ -1518,7 +1517,7 @@ void ContinuationIndenter::moveStatePastScopeOpener(LineState &State,
 
     AvoidBinPacking =
         (State.Stack.back().IsCSharpGenericTypeConstraint) ||
-        (Style.Language == FormatStyle::LK_JavaScript && EndsInComma) ||
+        (Style.isJavaScript() && EndsInComma) ||
         (State.Line->MustBeDeclaration && !BinPackDeclaration) ||
         (!State.Line->MustBeDeclaration && !Style.BinPackArguments) ||
         (Style.ExperimentalAutoDetectBinPacking &&
@@ -1547,7 +1546,7 @@ void ContinuationIndenter::moveStatePastScopeOpener(LineState &State,
       }
     }
 
-    if (Style.Language == FormatStyle::LK_JavaScript && EndsInComma)
+    if (Style.isJavaScript() && EndsInComma)
       BreakBeforeParameter = true;
   }
   // Generally inherit NoLineBreak from the current scope to nested scope.
@@ -1924,9 +1923,9 @@ ContinuationIndenter::createBreakableToken(const FormatToken &Current,
     // FIXME: String literal breaking is currently disabled for C#, Java, Json
     // and JavaScript, as it requires strings to be merged using "+" which we
     // don't support.
-    if (Style.Language == FormatStyle::LK_Java ||
-        Style.Language == FormatStyle::LK_JavaScript || Style.isCSharp() ||
-        Style.isJson() || !Style.BreakStringLiterals || !AllowBreak)
+    if (Style.Language == FormatStyle::LK_Java || Style.isJavaScript() ||
+        Style.isCSharp() || Style.isJson() || !Style.BreakStringLiterals ||
+        !AllowBreak)
       return nullptr;
 
     // Don't break string literals inside preprocessor directives (except for
diff --git a/clang/lib/Format/Format.cpp b/clang/lib/Format/Format.cpp
index 17de1075aeaa..be01daa38929 100644
--- a/clang/lib/Format/Format.cpp
+++ b/clang/lib/Format/Format.cpp
@@ -2586,12 +2586,31 @@ tooling::Replacements sortCppIncludes(const FormatStyle &Style, StringRef Code,
   bool MainIncludeFound = false;
   bool FormattingOff = false;
 
+  llvm::Regex RawStringRegex(
+      "R\"(([\\[A-Za-z0-9_{}#<>%:;.?*+/^&\\$|~!=,'\\-]|])*)\\(");
+  SmallVector<StringRef, 2> RawStringMatches;
+  std::string RawStringTermination = ")\"";
+
   for (;;) {
     auto Pos = Code.find('\n', SearchFrom);
     StringRef Line =
         Code.substr(Prev, (Pos != StringRef::npos ? Pos : Code.size()) - Prev);
 
     StringRef Trimmed = Line.trim();
+
+    // #includes inside raw string literals need to be ignored.
+    // or we will sort the contents of the string.
+    // Skip past until we think we are at the rawstring literal close.
+    if (RawStringRegex.match(Trimmed, &RawStringMatches)) {
+      std::string CharSequence = RawStringMatches[1].str();
+      RawStringTermination = ")" + CharSequence + "\"";
+      FormattingOff = true;
+    }
+
+    if (Trimmed.contains(RawStringTermination)) {
+      FormattingOff = false;
+    }
+
     if (Trimmed == "// clang-format off" || Trimmed == "/* clang-format off */")
       FormattingOff = true;
     else if (Trimmed == "// clang-format on" ||
@@ -3031,8 +3050,7 @@ reformat(const FormatStyle &Style, StringRef Code,
       });
   }
 
-  if (Style.Language == FormatStyle::LK_JavaScript &&
-      Style.JavaScriptQuotes != FormatStyle::JSQS_Leave)
+  if (Style.isJavaScript() && Style.JavaScriptQuotes != FormatStyle::JSQS_Leave)
     Passes.emplace_back([&](const Environment &Env) {
       return JavaScriptRequoter(Env, Expanded).process();
     });
@@ -3041,7 +3059,7 @@ reformat(const FormatStyle &Style, StringRef Code,
     return Formatter(Env, Expanded, Status).process();
   });
 
-  if (Style.Language == FormatStyle::LK_JavaScript &&
+  if (Style.isJavaScript() &&
       Style.InsertTrailingCommas == FormatStyle::TCS_Wrapped)
     Passes.emplace_back([&](const Environment &Env) {
       return TrailingCommaInserter(Env, Expanded).process();
diff --git a/clang/lib/Format/FormatToken.cpp b/clang/lib/Format/FormatToken.cpp
index 6768f041135c..57f8a5a45cbb 100644
--- a/clang/lib/Format/FormatToken.cpp
+++ b/clang/lib/Format/FormatToken.cpp
@@ -70,6 +70,10 @@ bool FormatToken::isSimpleTypeSpecifier() const {
   }
 }
 
+bool FormatToken::isTypeOrIdentifier() const {
+  return isSimpleTypeSpecifier() || Tok.isOneOf(tok::kw_auto, tok::identifier);
+}
+
 TokenRole::~TokenRole() {}
 
 void TokenRole::precomputeFormattingInfos(const FormatToken *Token) {}
diff --git a/clang/lib/Format/FormatToken.h b/clang/lib/Format/FormatToken.h
index 1a2858018fde..d410ede32240 100644
--- a/clang/lib/Format/FormatToken.h
+++ b/clang/lib/Format/FormatToken.h
@@ -521,7 +521,9 @@ public:
   }
 
   /// Determine whether the token is a simple-type-specifier.
-  bool isSimpleTypeSpecifier() const;
+  LLVM_NODISCARD bool isSimpleTypeSpecifier() const;
+
+  LLVM_NODISCARD bool isTypeOrIdentifier() const;
 
   bool isObjCAccessSpecifier() const {
     return is(tok::at) && Next &&
diff --git a/clang/lib/Format/FormatTokenLexer.cpp b/clang/lib/Format/FormatTokenLexer.cpp
index 64fbd2d5d45b..7736a7042f86 100644
--- a/clang/lib/Format/FormatTokenLexer.cpp
+++ b/clang/lib/Format/FormatTokenLexer.cpp
@@ -78,7 +78,7 @@ ArrayRef<FormatToken *> FormatTokenLexer::lex() {
   assert(FirstInLineIndex == 0);
   do {
     Tokens.push_back(getNextToken());
-    if (Style.Language == FormatStyle::LK_JavaScript) {
+    if (Style.isJavaScript()) {
       tryParseJSRegexLiteral();
       handleTemplateStrings();
     }
@@ -107,7 +107,7 @@ void FormatTokenLexer::tryMergePreviousTokens() {
   if (Style.isCpp() && tryTransformTryUsageForC())
     return;
 
-  if (Style.Language == FormatStyle::LK_JavaScript || Style.isCSharp()) {
+  if (Style.isJavaScript() || Style.isCSharp()) {
     static const tok::TokenKind NullishCoalescingOperator[] = {tok::question,
                                                                tok::question};
     static const tok::TokenKind NullPropagatingOperator[] = {tok::question,
@@ -152,7 +152,7 @@ void FormatTokenLexer::tryMergePreviousTokens() {
   if (tryMergeNSStringLiteral())
     return;
 
-  if (Style.Language == FormatStyle::LK_JavaScript) {
+  if (Style.isJavaScript()) {
     static const tok::TokenKind JSIdentity[] = {tok::equalequal, tok::equal};
     static const tok::TokenKind JSNotIdentity[] = {tok::exclaimequal,
                                                    tok::equal};
@@ -920,8 +920,7 @@ FormatToken *FormatTokenLexer::getNextToken() {
   // finds comments that contain a backslash followed by a line break, truncates
   // the comment token at the backslash, and resets the lexer to restart behind
   // the backslash.
-  if ((Style.Language == FormatStyle::LK_JavaScript ||
-       Style.Language == FormatStyle::LK_Java) &&
+  if ((Style.isJavaScript() || Style.Language == FormatStyle::LK_Java) &&
       FormatTok->is(tok::comment) && FormatTok->TokenText.startswith("//")) {
     size_t BackslashPos = FormatTok->TokenText.find('\\');
     while (BackslashPos != StringRef::npos) {
@@ -982,7 +981,7 @@ FormatToken *FormatTokenLexer::getNextToken() {
                            tok::kw_operator)) {
       FormatTok->Tok.setKind(tok::identifier);
       FormatTok->Tok.setIdentifierInfo(nullptr);
-    } else if (Style.Language == FormatStyle::LK_JavaScript &&
+    } else if (Style.isJavaScript() &&
                FormatTok->isOneOf(tok::kw_struct, tok::kw_union,
                                   tok::kw_operator)) {
       FormatTok->Tok.setKind(tok::identifier);
@@ -1060,14 +1059,12 @@ void FormatTokenLexer::readRawToken(FormatToken &Tok) {
     if (!Tok.TokenText.empty() && Tok.TokenText[0] == '"') {
       Tok.Tok.setKind(tok::string_literal);
       Tok.IsUnterminatedLiteral = true;
-    } else if (Style.Language == FormatStyle::LK_JavaScript &&
-               Tok.TokenText == "''") {
+    } else if (Style.isJavaScript() && Tok.TokenText == "''") {
       Tok.Tok.setKind(tok::string_literal);
     }
   }
 
-  if ((Style.Language == FormatStyle::LK_JavaScript ||
-       Style.Language == FormatStyle::LK_Proto ||
+  if ((Style.isJavaScript() || Style.Language == FormatStyle::LK_Proto ||
        Style.Language == FormatStyle::LK_TextProto) &&
       Tok.is(tok::char_constant)) {
     Tok.Tok.setKind(tok::string_literal);
diff --git a/clang/lib/Format/NamespaceEndCommentsFixer.cpp b/clang/lib/Format/NamespaceEndCommentsFixer.cpp
index def551f863cd..38ab5b9df76d 100644
--- a/clang/lib/Format/NamespaceEndCommentsFixer.cpp
+++ b/clang/lib/Format/NamespaceEndCommentsFixer.cpp
@@ -180,9 +180,13 @@ getNamespaceToken(const AnnotatedLine *Line,
   if (NamespaceTok->is(tok::l_brace)) {
     // "namespace" keyword can be on the line preceding '{', e.g. in styles
     // where BraceWrapping.AfterNamespace is true.
-    if (StartLineIndex > 0)
+    if (StartLineIndex > 0) {
       NamespaceTok = AnnotatedLines[StartLineIndex - 1]->First;
+      if (AnnotatedLines[StartLineIndex - 1]->endsWith(tok::semi))
+        return nullptr;
+    }
   }
+
   return NamespaceTok->getNamespaceToken();
 }
 
diff --git a/clang/lib/Format/TokenAnnotator.cpp b/clang/lib/Format/TokenAnnotator.cpp
index a94d8cdc3b04..505a7250572b 100644
--- a/clang/lib/Format/TokenAnnotator.cpp
+++ b/clang/lib/Format/TokenAnnotator.cpp
@@ -242,7 +242,7 @@ private:
       bool OperatorCalledAsMemberFunction =
           Prev->Previous && Prev->Previous->isOneOf(tok::period, tok::arrow);
       Contexts.back().IsExpression = OperatorCalledAsMemberFunction;
-    } else if (Style.Language == FormatStyle::LK_JavaScript &&
+    } else if (Style.isJavaScript() &&
                (Line.startsWith(Keywords.kw_type, tok::identifier) ||
                 Line.startsWith(tok::kw_export, Keywords.kw_type,
                                 tok::identifier))) {
@@ -256,13 +256,13 @@ private:
                 Left->Previous->is(TT_BinaryOperator))) {
       // static_assert, if and while usually contain expressions.
       Contexts.back().IsExpression = true;
-    } else if (Style.Language == FormatStyle::LK_JavaScript && Left->Previous &&
+    } else if (Style.isJavaScript() && Left->Previous &&
                (Left->Previous->is(Keywords.kw_function) ||
                 (Left->Previous->endsSequence(tok::identifier,
                                               Keywords.kw_function)))) {
       // function(...) or function f(...)
       Contexts.back().IsExpression = false;
-    } else if (Style.Language == FormatStyle::LK_JavaScript && Left->Previous &&
+    } else if (Style.isJavaScript() && Left->Previous &&
                Left->Previous->is(TT_JsTypeColon)) {
       // let x: (SomeType);
       Contexts.back().IsExpression = false;
@@ -582,7 +582,7 @@ private:
         Left->setType(TT_InlineASMSymbolicNameLSquare);
       } else if (IsCpp11AttributeSpecifier) {
         Left->setType(TT_AttributeSquare);
-      } else if (Style.Language == FormatStyle::LK_JavaScript && Parent &&
+      } else if (Style.isJavaScript() && Parent &&
                  Contexts.back().ContextKind == tok::l_brace &&
                  Parent->isOneOf(tok::l_brace, tok::comma)) {
         Left->setType(TT_JsComputedPropertyName);
@@ -646,8 +646,7 @@ private:
 
     ScopedContextCreator ContextCreator(*this, tok::l_square, BindingIncrease);
     Contexts.back().IsExpression = true;
-    if (Style.Language == FormatStyle::LK_JavaScript && Parent &&
-        Parent->is(TT_JsTypeColon))
+    if (Style.isJavaScript() && Parent && Parent->is(TT_JsTypeColon))
       Contexts.back().IsExpression = false;
 
     Contexts.back().ColonIsObjCMethodExpr = StartsObjCMethodExpr;
@@ -774,7 +773,7 @@ private:
       Contexts.back().ColonIsDictLiteral = true;
       if (Left->is(BK_BracedInit))
         Contexts.back().IsExpression = true;
-      if (Style.Language == FormatStyle::LK_JavaScript && Left->Previous &&
+      if (Style.isJavaScript() && Left->Previous &&
           Left->Previous->is(TT_JsTypeColon))
         Contexts.back().IsExpression = false;
 
@@ -808,12 +807,11 @@ private:
                 Previous->is(tok::string_literal))
               Previous->setType(TT_SelectorName);
           }
-          if (CurrentToken->is(tok::colon) ||
-              Style.Language == FormatStyle::LK_JavaScript)
+          if (CurrentToken->is(tok::colon) || Style.isJavaScript())
             Left->setType(TT_DictLiteral);
         }
         if (CurrentToken->is(tok::comma)) {
-          if (Style.Language == FormatStyle::LK_JavaScript)
+          if (Style.isJavaScript())
             Left->setType(TT_DictLiteral);
           ++CommaCount;
         }
@@ -879,7 +877,7 @@ private:
       if (!Tok->Previous)
         return false;
       // Colons from ?: are handled in parseConditional().
-      if (Style.Language == FormatStyle::LK_JavaScript) {
+      if (Style.isJavaScript()) {
         if (Contexts.back().ColonIsForRangeExpr || // colon in for loop
             (Contexts.size() == 1 &&               // switch/case labels
              !Line.First->isOneOf(tok::kw_enum, tok::kw_case)) ||
@@ -979,8 +977,7 @@ private:
     case tok::amp:
       // | and & in declarations/type expressions represent union and
       // intersection types, respectively.
-      if (Style.Language == FormatStyle::LK_JavaScript &&
-          !Contexts.back().IsExpression)
+      if (Style.isJavaScript() && !Contexts.back().IsExpression)
         Tok->setType(TT_JsTypeOperator);
       break;
     case tok::kw_if:
@@ -995,7 +992,7 @@ private:
       }
       break;
     case tok::kw_for:
-      if (Style.Language == FormatStyle::LK_JavaScript) {
+      if (Style.isJavaScript()) {
         // x.for and {for: ...}
         if ((Tok->Previous && Tok->Previous->is(tok::period)) ||
             (Tok->Next && Tok->Next->is(tok::colon)))
@@ -1106,7 +1103,7 @@ private:
         CurrentToken->Previous->setType(TT_OverloadedOperator);
       break;
     case tok::question:
-      if (Style.Language == FormatStyle::LK_JavaScript && Tok->Next &&
+      if (Style.isJavaScript() && Tok->Next &&
           Tok->Next->isOneOf(tok::semi, tok::comma, tok::colon, tok::r_paren,
                              tok::r_brace)) {
         // Question marks before semicolons, colons, etc. indicate optional
@@ -1119,7 +1116,7 @@ private:
       // Declarations cannot be conditional expressions, this can only be part
       // of a type declaration.
       if (Line.MustBeDeclaration && !Contexts.back().IsExpression &&
-          Style.Language == FormatStyle::LK_JavaScript)
+          Style.isJavaScript())
         break;
       if (Style.isCSharp()) {
         // `Type?)`, `Type?>`, `Type? name;` and `Type? name =` can only be
@@ -1252,7 +1249,7 @@ private:
     if (!CurrentToken)
       return Type;
 
-    if (Style.Language == FormatStyle::LK_JavaScript && IsFirstToken) {
+    if (Style.isJavaScript() && IsFirstToken) {
       // JavaScript files can contain shebang lines of the form:
       // #!/usr/bin/env node
       // Treat these like C++ #include directives.
@@ -1354,14 +1351,13 @@ public:
     bool ImportStatement = false;
 
     // import {...} from '...';
-    if (Style.Language == FormatStyle::LK_JavaScript &&
-        CurrentToken->is(Keywords.kw_import))
+    if (Style.isJavaScript() && CurrentToken->is(Keywords.kw_import))
       ImportStatement = true;
 
     while (CurrentToken) {
       if (CurrentToken->is(tok::kw_virtual))
         KeywordVirtualFound = true;
-      if (Style.Language == FormatStyle::LK_JavaScript) {
+      if (Style.isJavaScript()) {
         // export {...} from '...';
         // An export followed by "from 'some string';" is a re-export from
         // another module identified by a URI and is treated as a
@@ -1504,7 +1500,7 @@ private:
         !Line.First->isOneOf(tok::kw_template, tok::kw_using, tok::kw_return) &&
         // Type aliases use `type X = ...;` in TypeScript and can be exported
         // using `export type ...`.
-        !(Style.Language == FormatStyle::LK_JavaScript &&
+        !(Style.isJavaScript() &&
           (Line.startsWith(Keywords.kw_type, tok::identifier) ||
            Line.startsWith(tok::kw_export, Keywords.kw_type,
                            tok::identifier))) &&
@@ -1633,11 +1629,11 @@ private:
       // The token type is already known.
       return;
 
-    if ((Style.Language == FormatStyle::LK_JavaScript || Style.isCSharp()) &&
+    if ((Style.isJavaScript() || Style.isCSharp()) &&
         Current.is(tok::exclaim)) {
       if (Current.Previous) {
         bool IsIdentifier =
-            Style.Language == FormatStyle::LK_JavaScript
+            Style.isJavaScript()
                 ? Keywords.IsJavaScriptIdentifier(
                       *Current.Previous, /* AcceptIdentifierName= */ true)
                 : Current.Previous->is(tok::identifier);
@@ -1679,7 +1675,7 @@ private:
       Current.setType(TT_LambdaArrow);
     } else if (Current.is(tok::arrow) && AutoFound && Line.MustBeDeclaration &&
                Current.NestingLevel == 0 &&
-               !Current.Previous->is(tok::kw_operator)) {
+               !Current.Previous->isOneOf(tok::kw_operator, tok::identifier)) {
       // not auto operator->() -> xxx;
       Current.setType(TT_TrailingReturnArrow);
     } else if (Current.is(tok::arrow) && Current.Previous &&
@@ -1704,8 +1700,8 @@ private:
     } else if (Current.isOneOf(tok::exclaim, tok::tilde)) {
       Current.setType(TT_UnaryOperator);
     } else if (Current.is(tok::question)) {
-      if (Style.Language == FormatStyle::LK_JavaScript &&
-          Line.MustBeDeclaration && !Contexts.back().IsExpression) {
+      if (Style.isJavaScript() && Line.MustBeDeclaration &&
+          !Contexts.back().IsExpression) {
         // In JavaScript, `interface X { foo?(): bar; }` is an optional method
         // on the interface, not a ternary expression.
         Current.setType(TT_JsTypeOptionalQuestion);
@@ -1748,8 +1744,7 @@ private:
                 Current.setType(TT_FunctionAnnotationRParen);
           }
         }
-    } else if (Current.is(tok::at) && Current.Next &&
-               Style.Language != FormatStyle::LK_JavaScript &&
+    } else if (Current.is(tok::at) && Current.Next && !Style.isJavaScript() &&
                Style.Language != FormatStyle::LK_Java) {
       // In Java & JavaScript, "@..." is a decorator or annotation. In ObjC, it
       // marks declarations and properties that need special formatting.
@@ -1796,7 +1791,7 @@ private:
       // function declaration have been found.
       Current.setType(TT_TrailingAnnotation);
     } else if ((Style.Language == FormatStyle::LK_Java ||
-                Style.Language == FormatStyle::LK_JavaScript) &&
+                Style.isJavaScript()) &&
                Current.Previous) {
       if (Current.Previous->is(tok::at) &&
           Current.isNot(Keywords.kw_interface)) {
@@ -1826,14 +1821,16 @@ private:
     if (Tok.Previous->isOneOf(TT_LeadingJavaAnnotation, Keywords.kw_instanceof,
                               Keywords.kw_as))
       return false;
-    if (Style.Language == FormatStyle::LK_JavaScript &&
-        Tok.Previous->is(Keywords.kw_in))
+    if (Style.isJavaScript() && Tok.Previous->is(Keywords.kw_in))
       return false;
 
     // Skip "const" as it does not have an influence on whether this is a name.
     FormatToken *PreviousNotConst = Tok.getPreviousNonComment();
-    while (PreviousNotConst && PreviousNotConst->is(tok::kw_const))
-      PreviousNotConst = PreviousNotConst->getPreviousNonComment();
+
+    // For javascript const can be like "let" or "var"
+    if (!Style.isJavaScript())
+      while (PreviousNotConst && PreviousNotConst->is(tok::kw_const))
+        PreviousNotConst = PreviousNotConst->getPreviousNonComment();
 
     if (!PreviousNotConst)
       return false;
@@ -1852,10 +1849,24 @@ private:
         PreviousNotConst->is(TT_TypeDeclarationParen))
       return true;
 
-    return (!IsPPKeyword &&
-            PreviousNotConst->isOneOf(tok::identifier, tok::kw_auto)) ||
-           PreviousNotConst->is(TT_PointerOrReference) ||
-           PreviousNotConst->isSimpleTypeSpecifier();
+    // If is a preprocess keyword like #define.
+    if (IsPPKeyword)
+      return false;
+
+    // int a or auto a.
+    if (PreviousNotConst->isOneOf(tok::identifier, tok::kw_auto))
+      return true;
+
+    // *a or &a or &&a.
+    if (PreviousNotConst->is(TT_PointerOrReference))
+      return true;
+
+    // MyClass a;
+    if (PreviousNotConst->isSimpleTypeSpecifier())
+      return true;
+
+    // const a = in JavaScript.
+    return (Style.isJavaScript() && PreviousNotConst->is(tok::kw_const));
   }
 
   /// Determine whether ')' is ending a cast.
@@ -2006,7 +2017,7 @@ private:
   /// Return the type of the given token assuming it is * or &.
   TokenType determineStarAmpUsage(const FormatToken &Tok, bool IsExpression,
                                   bool InTemplateArgument) {
-    if (Style.Language == FormatStyle::LK_JavaScript)
+    if (Style.isJavaScript())
       return TT_BinaryOperator;
 
     // && in C# must be a binary operator.
@@ -2034,9 +2045,8 @@ private:
                            tok::comma, tok::semi, tok::kw_return, tok::colon,
                            tok::kw_co_return, tok::kw_co_await,
                            tok::kw_co_yield, tok::equal, tok::kw_delete,
-                           tok::kw_sizeof, tok::kw_throw) ||
-        PrevToken->isOneOf(TT_BinaryOperator, TT_ConditionalExpr,
-                           TT_UnaryOperator, TT_CastRParen))
+                           tok::kw_sizeof, tok::kw_throw, TT_BinaryOperator,
+                           TT_ConditionalExpr, TT_UnaryOperator, TT_CastRParen))
       return TT_UnaryOperator;
 
     if (NextToken->is(tok::l_square) && NextToken->isNot(TT_LambdaLSquare))
@@ -2174,8 +2184,8 @@ public:
 
       int CurrentPrecedence = getCurrentPrecedence();
 
-      if (Current && Current->is(TT_SelectorName) &&
-          Precedence == CurrentPrecedence) {
+      if (Precedence == CurrentPrecedence && Current &&
+          Current->is(TT_SelectorName)) {
         if (LatestOperator)
           addFakeParenthesis(Start, prec::Level(Precedence));
         Start = Current;
@@ -2251,19 +2261,17 @@ private:
         return 0;
       if (Current->is(TT_RangeBasedForLoopColon))
         return prec::Comma;
-      if ((Style.Language == FormatStyle::LK_Java ||
-           Style.Language == FormatStyle::LK_JavaScript) &&
+      if ((Style.Language == FormatStyle::LK_Java || Style.isJavaScript()) &&
           Current->is(Keywords.kw_instanceof))
         return prec::Relational;
-      if (Style.Language == FormatStyle::LK_JavaScript &&
+      if (Style.isJavaScript() &&
           Current->isOneOf(Keywords.kw_in, Keywords.kw_as))
         return prec::Relational;
       if (Current->is(TT_BinaryOperator) || Current->is(tok::comma))
         return Current->getPrecedence();
       if (Current->isOneOf(tok::period, tok::arrow))
         return PrecedenceArrowAndPeriod;
-      if ((Style.Language == FormatStyle::LK_Java ||
-           Style.Language == FormatStyle::LK_JavaScript) &&
+      if ((Style.Language == FormatStyle::LK_Java || Style.isJavaScript()) &&
           Current->isOneOf(Keywords.kw_extends, Keywords.kw_implements,
                            Keywords.kw_throws))
         return 0;
@@ -2374,11 +2382,9 @@ static unsigned maxNestingDepth(const AnnotatedLine &Line) {
 }
 
 void TokenAnnotator::annotate(AnnotatedLine &Line) {
-  for (SmallVectorImpl<AnnotatedLine *>::iterator I = Line.Children.begin(),
-                                                  E = Line.Children.end();
-       I != E; ++I) {
-    annotate(**I);
-  }
+  for (auto &Child : Line.Children)
+    annotate(*Child);
+
   AnnotatingParser Parser(Style, Line, Keywords);
   Line.Type = Parser.parseLine();
 
@@ -2734,7 +2740,7 @@ unsigned TokenAnnotator::splitPenalty(const AnnotatedLine &Line,
       return 2;
     if (Left.is(tok::comma) && Left.NestingLevel == 0)
       return 3;
-  } else if (Style.Language == FormatStyle::LK_JavaScript) {
+  } else if (Style.isJavaScript()) {
     if (Right.is(Keywords.kw_function) && Left.isNot(tok::comma))
       return 100;
     if (Left.is(TT_JsTypeColon))
@@ -3024,8 +3030,14 @@ bool TokenAnnotator::spaceRequiredBetween(const AnnotatedLine &Line,
          Style.SpaceAroundPointerQualifiers == FormatStyle::SAPQ_Both) &&
         (Left.is(TT_AttributeParen) || Left.canBePointerOrReferenceQualifier()))
       return true;
+    if (Left.Tok.isLiteral())
+      return true;
+    // for (auto a = 0, b = 0; const auto & c : {1, 2, 3})
+    if (Left.isTypeOrIdentifier() && Right.Next && Right.Next->Next &&
+        Right.Next->Next->is(TT_RangeBasedForLoopColon))
+      return getTokenPointerOrReferenceAlignment(Right) !=
+             FormatStyle::PAS_Left;
     return (
-        Left.Tok.isLiteral() ||
         (!Left.isOneOf(TT_PointerOrReference, tok::l_paren) &&
          (getTokenPointerOrReferenceAlignment(Right) != FormatStyle::PAS_Left ||
           (Line.IsMultiVariableDeclStmt &&
@@ -3044,18 +3056,32 @@ bool TokenAnnotator::spaceRequiredBetween(const AnnotatedLine &Line,
          Style.SpaceAroundPointerQualifiers == FormatStyle::SAPQ_Both) &&
         Right.canBePointerOrReferenceQualifier())
       return true;
-    return Right.Tok.isLiteral() || Right.is(TT_BlockComment) ||
-           (Right.isOneOf(Keywords.kw_override, Keywords.kw_final) &&
-            !Right.is(TT_StartOfName)) ||
-           (Right.is(tok::l_brace) && Right.is(BK_Block)) ||
-           (!Right.isOneOf(TT_PointerOrReference, TT_ArraySubscriptLSquare,
-                           tok::l_paren) &&
-            (getTokenPointerOrReferenceAlignment(Left) !=
-                 FormatStyle::PAS_Right &&
-             !Line.IsMultiVariableDeclStmt) &&
-            Left.Previous &&
-            !Left.Previous->isOneOf(tok::l_paren, tok::coloncolon,
-                                    tok::l_square));
+    // & 1
+    if (Right.Tok.isLiteral())
+      return true;
+    // & /* comment
+    if (Right.is(TT_BlockComment))
+      return true;
+    // foo() -> const Bar * override/final
+    if (Right.isOneOf(Keywords.kw_override, Keywords.kw_final) &&
+        !Right.is(TT_StartOfName))
+      return true;
+    // & {
+    if (Right.is(tok::l_brace) && Right.is(BK_Block))
+      return true;
+    // for (auto a = 0, b = 0; const auto& c : {1, 2, 3})
+    if (Left.Previous && Left.Previous->isTypeOrIdentifier() && Right.Next &&
+        Right.Next->is(TT_RangeBasedForLoopColon))
+      return getTokenPointerOrReferenceAlignment(Left) !=
+             FormatStyle::PAS_Right;
+    return !Right.isOneOf(TT_PointerOrReference, TT_ArraySubscriptLSquare,
+                          tok::l_paren) &&
+           (getTokenPointerOrReferenceAlignment(Left) !=
+                FormatStyle::PAS_Right &&
+            !Line.IsMultiVariableDeclStmt) &&
+           Left.Previous &&
+           !Left.Previous->isOneOf(tok::l_paren, tok::coloncolon,
+                                   tok::l_square);
   }
   // Ensure right pointer alignment with ellipsis e.g. int *...P
   if (Left.is(tok::ellipsis) && Left.Previous &&
@@ -3146,8 +3172,7 @@ bool TokenAnnotator::spaceRequiredBetween(const AnnotatedLine &Line,
     return Style.Cpp11BracedListStyle ? Style.SpacesInParentheses : true;
   if (Left.is(TT_BlockComment))
     // No whitespace in x(/*foo=*/1), except for JavaScript.
-    return Style.Language == FormatStyle::LK_JavaScript ||
-           !Left.TokenText.endswith("=*/");
+    return Style.isJavaScript() || !Left.TokenText.endswith("=*/");
 
   // Space between template and attribute.
   // e.g. template <typename T> [[nodiscard]] ...
@@ -3225,7 +3250,13 @@ bool TokenAnnotator::spaceRequiredBetween(const AnnotatedLine &Line,
     return false;
   if (Left.is(tok::period) || Right.is(tok::period))
     return false;
-  if (Right.is(tok::hash) && Left.is(tok::identifier) && Left.TokenText == "L")
+  // u#str, U#str, L#str, u8#str
+  // uR#str, UR#str, LR#str, u8R#str
+  if (Right.is(tok::hash) && Left.is(tok::identifier) &&
+      (Left.TokenText == "L" || Left.TokenText == "u" ||
+       Left.TokenText == "U" || Left.TokenText == "u8" ||
+       Left.TokenText == "LR" || Left.TokenText == "uR" ||
+       Left.TokenText == "UR" || Left.TokenText == "u8R"))
     return false;
   if (Left.is(TT_TemplateCloser) && Left.MatchingParen &&
       Left.MatchingParen->Previous &&
@@ -3396,7 +3427,7 @@ bool TokenAnnotator::spaceRequiredBefore(const AnnotatedLine &Line,
                      Keywords.kw_async, Keywords.kw_unsafe) &&
         Right.is(tok::l_paren))
       return true;
-  } else if (Style.Language == FormatStyle::LK_JavaScript) {
+  } else if (Style.isJavaScript()) {
     if (Left.is(TT_FatArrow))
       return true;
     // for await ( ...
@@ -3694,11 +3725,18 @@ bool TokenAnnotator::mustBreakBefore(const AnnotatedLine &Line,
     return true;
 
   if (Style.isCSharp()) {
+    if (Left.is(TT_FatArrow) && Right.is(tok::l_brace) &&
+        Style.BraceWrapping.AfterFunction)
+      return true;
     if (Right.is(TT_CSharpNamedArgumentColon) ||
         Left.is(TT_CSharpNamedArgumentColon))
       return false;
     if (Right.is(TT_CSharpGenericTypeConstraint))
       return true;
+    if (Right.Next && Right.Next->is(TT_FatArrow) &&
+        (Right.is(tok::numeric_constant) ||
+         (Right.is(tok::identifier) && Right.TokenText == "_")))
+      return true;
 
     // Break after C# [...] and before public/protected/private/internal.
     if (Left.is(TT_AttributeSquare) && Left.is(tok::r_square) &&
@@ -3710,7 +3748,7 @@ bool TokenAnnotator::mustBreakBefore(const AnnotatedLine &Line,
         Left.is(tok::r_square) && Right.is(tok::l_square))
       return true;
 
-  } else if (Style.Language == FormatStyle::LK_JavaScript) {
+  } else if (Style.isJavaScript()) {
     // FIXME: This might apply to other languages and token kinds.
     if (Right.is(tok::string_literal) && Left.is(tok::plus) && Left.Previous &&
         Left.Previous->is(tok::string_literal))
@@ -3800,15 +3838,13 @@ bool TokenAnnotator::mustBreakBefore(const AnnotatedLine &Line,
   if (Style.JavaScriptWrapImports || Line.Type != LT_ImportStatement) {
     const FormatToken *BeforeClosingBrace = nullptr;
     if ((Left.isOneOf(tok::l_brace, TT_ArrayInitializerLSquare) ||
-         (Style.Language == FormatStyle::LK_JavaScript &&
-          Left.is(tok::l_paren))) &&
+         (Style.isJavaScript() && Left.is(tok::l_paren))) &&
         Left.isNot(BK_Block) && Left.MatchingParen)
       BeforeClosingBrace = Left.MatchingParen->Previous;
     else if (Right.MatchingParen &&
              (Right.MatchingParen->isOneOf(tok::l_brace,
                                            TT_ArrayInitializerLSquare) ||
-              (Style.Language == FormatStyle::LK_JavaScript &&
-               Right.MatchingParen->is(tok::l_paren))))
+              (Style.isJavaScript() && Right.MatchingParen->is(tok::l_paren))))
       BeforeClosingBrace = &Left;
     if (BeforeClosingBrace && (BeforeClosingBrace->is(tok::comma) ||
                                BeforeClosingBrace->isTrailingComment()))
@@ -3927,8 +3963,7 @@ bool TokenAnnotator::mustBreakBefore(const AnnotatedLine &Line,
   }
 
   // Put multiple Java annotation on a new line.
-  if ((Style.Language == FormatStyle::LK_Java ||
-       Style.Language == FormatStyle::LK_JavaScript) &&
+  if ((Style.Language == FormatStyle::LK_Java || Style.isJavaScript()) &&
       Left.is(TT_LeadingJavaAnnotation) &&
       Right.isNot(TT_LeadingJavaAnnotation) && Right.isNot(tok::l_paren) &&
       (Line.Last->is(tok::l_brace) || Style.BreakAfterJavaFieldAnnotations))
@@ -4071,7 +4106,7 @@ bool TokenAnnotator::canBreakBefore(const AnnotatedLine &Line,
     if (Right.isOneOf(Keywords.kw_throws, Keywords.kw_extends,
                       Keywords.kw_implements))
       return true;
-  } else if (Style.Language == FormatStyle::LK_JavaScript) {
+  } else if (Style.isJavaScript()) {
     const FormatToken *NonComment = Right.getPreviousNonComment();
     if (NonComment &&
         NonComment->isOneOf(
diff --git a/clang/lib/Format/UnwrappedLineFormatter.cpp b/clang/lib/Format/UnwrappedLineFormatter.cpp
index d099cfee9dea..f652a4e7088f 100644
--- a/clang/lib/Format/UnwrappedLineFormatter.cpp
+++ b/clang/lib/Format/UnwrappedLineFormatter.cpp
@@ -62,7 +62,7 @@ public:
       Indent = Line.Level * IndentWidth + AdditionalIndent;
     } else {
       IndentForLevel.resize(Line.Level + 1);
-      Indent = getIndent(IndentForLevel, Line.Level);
+      Indent = getIndent(Line.Level);
     }
     if (static_cast<int>(Indent) + Offset >= 0)
       Indent += Offset;
@@ -97,8 +97,8 @@ private:
   /// For example, 'public:' labels in classes are offset by 1 or 2
   /// characters to the left from their level.
   int getIndentOffset(const FormatToken &RootToken) {
-    if (Style.Language == FormatStyle::LK_Java ||
-        Style.Language == FormatStyle::LK_JavaScript || Style.isCSharp())
+    if (Style.Language == FormatStyle::LK_Java || Style.isJavaScript() ||
+        Style.isCSharp())
       return 0;
     if (RootToken.isAccessSpecifier(false) ||
         RootToken.isObjCAccessSpecifier() ||
@@ -118,12 +118,12 @@ private:
   /// \p IndentForLevel must contain the indent for the level \c l
   /// at \p IndentForLevel[l], or a value < 0 if the indent for
   /// that level is unknown.
-  unsigned getIndent(ArrayRef<int> IndentForLevel, unsigned Level) {
+  unsigned getIndent(unsigned Level) const {
     if (IndentForLevel[Level] != -1)
       return IndentForLevel[Level];
     if (Level == 0)
       return 0;
-    return getIndent(IndentForLevel, Level - 1) + Style.IndentWidth;
+    return getIndent(Level - 1) + Style.IndentWidth;
   }
 
   const FormatStyle &Style;
@@ -393,11 +393,24 @@ private:
 
     // Try to merge a block with left brace wrapped that wasn't yet covered
     if (TheLine->Last->is(tok::l_brace)) {
-      return !Style.BraceWrapping.AfterFunction ||
-                     (I[1]->First->is(tok::r_brace) &&
-                      !Style.BraceWrapping.SplitEmptyRecord)
-                 ? tryMergeSimpleBlock(I, E, Limit)
-                 : 0;
+      const FormatToken *Tok = TheLine->First;
+      bool ShouldMerge = false;
+      if (Tok->is(tok::kw_typedef)) {
+        Tok = Tok->getNextNonComment();
+        assert(Tok);
+      }
+      if (Tok->isOneOf(tok::kw_class, tok::kw_struct)) {
+        ShouldMerge = !Style.BraceWrapping.AfterClass ||
+                      (I[1]->First->is(tok::r_brace) &&
+                       !Style.BraceWrapping.SplitEmptyRecord);
+      } else if (Tok->is(tok::kw_enum)) {
+        ShouldMerge = Style.AllowShortEnumsOnASingleLine;
+      } else {
+        ShouldMerge = !Style.BraceWrapping.AfterFunction ||
+                      (I[1]->First->is(tok::r_brace) &&
+                       !Style.BraceWrapping.SplitEmptyFunction);
+      }
+      return ShouldMerge ? tryMergeSimpleBlock(I, E, Limit) : 0;
     }
     // Try to merge a function block with left brace wrapped
     if (I[1]->First->is(TT_FunctionLBrace) &&
@@ -584,6 +597,9 @@ private:
                             Keywords.kw___except)) {
       if (Style.AllowShortBlocksOnASingleLine == FormatStyle::SBS_Never)
         return 0;
+      if (Style.AllowShortBlocksOnASingleLine == FormatStyle::SBS_Empty &&
+          !I[1]->First->is(tok::r_brace))
+        return 0;
       // Don't merge when we can't except the case when
       // the control statement block is empty
       if (!Style.AllowShortIfStatementsOnASingleLine &&
@@ -1015,9 +1031,9 @@ private:
     QueueType Queue;
 
     // Insert start element into queue.
-    StateNode *Node =
+    StateNode *RootNode =
         new (Allocator.Allocate()) StateNode(InitialState, false, nullptr);
-    Queue.push(QueueItem(OrderedPenalty(0, Count), Node));
+    Queue.push(QueueItem(OrderedPenalty(0, Count), RootNode));
     ++Count;
 
     unsigned Penalty = 0;
@@ -1044,9 +1060,9 @@ private:
 
       FormatDecision LastFormat = Node->State.NextToken->getDecision();
       if (LastFormat == FD_Unformatted || LastFormat == FD_Continue)
-        addNextStateToQueue(Penalty, Node, /*NewLine=*/false, &Count, &Queue);
+        addNextStateToQueue(Penalty, Node, /*NewLine=*/false, Count, Queue);
       if (LastFormat == FD_Unformatted || LastFormat == FD_Break)
-        addNextStateToQueue(Penalty, Node, /*NewLine=*/true, &Count, &Queue);
+        addNextStateToQueue(Penalty, Node, /*NewLine=*/true, Count, Queue);
     }
 
     if (Queue.empty()) {
@@ -1072,7 +1088,7 @@ private:
   /// Assume the current state is \p PreviousNode and has been reached with a
   /// penalty of \p Penalty. Insert a line break if \p NewLine is \c true.
   void addNextStateToQueue(unsigned Penalty, StateNode *PreviousNode,
-                           bool NewLine, unsigned *Count, QueueType *Queue) {
+                           bool NewLine, unsigned &Count, QueueType &Queue) {
     if (NewLine && !Indenter->canBreak(PreviousNode->State))
       return;
     if (!NewLine && Indenter->mustBreak(PreviousNode->State))
@@ -1085,8 +1101,8 @@ private:
 
     Penalty += Indenter->addTokenToState(Node->State, NewLine, true);
 
-    Queue->push(QueueItem(OrderedPenalty(Penalty, *Count), Node));
-    ++(*Count);
+    Queue.push(QueueItem(OrderedPenalty(Penalty, Count), Node));
+    ++Count;
   }
 
   /// Applies the best formatting by reconstructing the path in the
@@ -1184,8 +1200,7 @@ unsigned UnwrappedLineFormatter::format(
       bool FitsIntoOneLine =
           TheLine.Last->TotalLength + Indent <= ColumnLimit ||
           (TheLine.Type == LT_ImportStatement &&
-           (Style.Language != FormatStyle::LK_JavaScript ||
-            !Style.JavaScriptWrapImports)) ||
+           (!Style.isJavaScript() || !Style.JavaScriptWrapImports)) ||
           (Style.isCSharp() &&
            TheLine.InPPDirective); // don't split #regions in C#
       if (Style.ColumnLimit == 0)
diff --git a/clang/lib/Format/UnwrappedLineParser.cpp b/clang/lib/Format/UnwrappedLineParser.cpp
index 5b9fe267aae6..b6e55aab708f 100644
--- a/clang/lib/Format/UnwrappedLineParser.cpp
+++ b/clang/lib/Format/UnwrappedLineParser.cpp
@@ -366,8 +366,7 @@ void UnwrappedLineParser::parse() {
 void UnwrappedLineParser::parseFile() {
   // The top-level context in a file always has declarations, except for pre-
   // processor directives and JavaScript files.
-  bool MustBeDeclaration =
-      !Line->InPPDirective && Style.Language != FormatStyle::LK_JavaScript;
+  bool MustBeDeclaration = !Line->InPPDirective && !Style.isJavaScript();
   ScopedDeclarationState DeclarationState(*Line, DeclarationScopeStack,
                                           MustBeDeclaration);
   if (Style.Language == FormatStyle::LK_TextProto)
@@ -478,8 +477,7 @@ void UnwrappedLineParser::parseLevel(bool HasOpeningBrace) {
       LLVM_FALLTHROUGH;
     }
     case tok::kw_case:
-      if (Style.Language == FormatStyle::LK_JavaScript &&
-          Line->MustBeDeclaration) {
+      if (Style.isJavaScript() && Line->MustBeDeclaration) {
         // A 'case: string' style field declaration.
         parseStructuralElement();
         break;
@@ -528,7 +526,7 @@ void UnwrappedLineParser::calculateBraceTypes(bool ExpectClassBody) {
 
     switch (Tok->Tok.getKind()) {
     case tok::l_brace:
-      if (Style.Language == FormatStyle::LK_JavaScript && PrevTok) {
+      if (Style.isJavaScript() && PrevTok) {
         if (PrevTok->isOneOf(tok::colon, tok::less))
           // A ':' indicates this code is in a type, or a braced list
           // following a label in an object literal ({a: {b: 1}}).
@@ -581,7 +579,7 @@ void UnwrappedLineParser::calculateBraceTypes(bool ExpectClassBody) {
           // FIXME: Some of these do not apply to JS, e.g. "} {" can never be a
           // braced list in JS.
           ProbablyBracedList =
-              (Style.Language == FormatStyle::LK_JavaScript &&
+              (Style.isJavaScript() &&
                NextTok->isOneOf(Keywords.kw_of, Keywords.kw_in,
                                 Keywords.kw_as)) ||
               (Style.isCpp() && NextTok->is(tok::l_paren)) ||
@@ -791,7 +789,7 @@ void UnwrappedLineParser::parseChildBlock() {
   FormatTok->setBlockKind(BK_Block);
   nextToken();
   {
-    bool SkipIndent = (Style.Language == FormatStyle::LK_JavaScript &&
+    bool SkipIndent = (Style.isJavaScript() &&
                        (isGoogScope(*Line) || isIIFE(*Line, Keywords)));
     ScopedLineState LineState(*this);
     ScopedDeclarationState DeclarationState(*Line, DeclarationScopeStack,
@@ -1140,7 +1138,6 @@ void UnwrappedLineParser::parseModuleImport() {
   }
 
   addUnwrappedLine();
-  return;
 }
 
 // readTokenWithJavaScriptASI reads the next token and terminates the current
@@ -1222,39 +1219,39 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
   case tok::kw_public:
   case tok::kw_protected:
   case tok::kw_private:
-    if (Style.Language == FormatStyle::LK_Java ||
-        Style.Language == FormatStyle::LK_JavaScript || Style.isCSharp())
+    if (Style.Language == FormatStyle::LK_Java || Style.isJavaScript() ||
+        Style.isCSharp())
       nextToken();
     else
       parseAccessSpecifier();
     return;
   case tok::kw_if:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // field/method declaration.
       break;
     parseIfThenElse();
     return;
   case tok::kw_for:
   case tok::kw_while:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // field/method declaration.
       break;
     parseForOrWhileLoop();
     return;
   case tok::kw_do:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // field/method declaration.
       break;
     parseDoWhile();
     return;
   case tok::kw_switch:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // 'switch: string' field declaration.
       break;
     parseSwitch();
     return;
   case tok::kw_default:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // 'default: string' field declaration.
       break;
     nextToken();
@@ -1265,14 +1262,14 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
     // e.g. "default void f() {}" in a Java interface.
     break;
   case tok::kw_case:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // 'case: string' field declaration.
       break;
     parseCaseLabel();
     return;
   case tok::kw_try:
   case tok::kw___try:
-    if (Style.Language == FormatStyle::LK_JavaScript && Line->MustBeDeclaration)
+    if (Style.isJavaScript() && Line->MustBeDeclaration)
       // field/method declaration.
       break;
     parseTryCatch();
@@ -1282,24 +1279,25 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
     if (FormatTok->Tok.is(tok::string_literal)) {
       nextToken();
       if (FormatTok->Tok.is(tok::l_brace)) {
-        if (!Style.IndentExternBlock) {
-          if (Style.BraceWrapping.AfterExternBlock) {
-            addUnwrappedLine();
-          }
-          unsigned AddLevels = Style.BraceWrapping.AfterExternBlock ? 1u : 0u;
-          parseBlock(/*MustBeDeclaration=*/true, AddLevels);
-        } else {
-          unsigned AddLevels =
-              Style.IndentExternBlock == FormatStyle::IEBS_Indent ? 1u : 0u;
-          parseBlock(/*MustBeDeclaration=*/true, AddLevels);
-        }
+        if (Style.BraceWrapping.AfterExternBlock)
+          addUnwrappedLine();
+        // Either we indent or for backwards compatibility we follow the
+        // AfterExternBlock style.
+        unsigned AddLevels =
+            (Style.IndentExternBlock == FormatStyle::IEBS_Indent) ||
+                    (Style.BraceWrapping.AfterExternBlock &&
+                     Style.IndentExternBlock ==
+                         FormatStyle::IEBS_AfterExternBlock)
+                ? 1u
+                : 0u;
+        parseBlock(/*MustBeDeclaration=*/true, AddLevels);
         addUnwrappedLine();
         return;
       }
     }
     break;
   case tok::kw_export:
-    if (Style.Language == FormatStyle::LK_JavaScript) {
+    if (Style.isJavaScript()) {
       parseJavaScriptEs6ImportExport();
       return;
     }
@@ -1325,7 +1323,7 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
       return;
     }
     if (FormatTok->is(Keywords.kw_import)) {
-      if (Style.Language == FormatStyle::LK_JavaScript) {
+      if (Style.isJavaScript()) {
         parseJavaScriptEs6ImportExport();
         return;
       }
@@ -1437,10 +1435,10 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
       break;
     case tok::kw_concept:
       parseConcept();
-      break;
+      return;
     case tok::kw_requires:
       parseRequires();
-      break;
+      return;
     case tok::kw_enum:
       // Ignore if this is part of "template <enum ...".
       if (Previous && Previous->is(tok::less)) {
@@ -1479,7 +1477,7 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
       if (Style.Language == FormatStyle::LK_Java && FormatTok &&
           FormatTok->is(tok::kw_class))
         nextToken();
-      if (Style.Language == FormatStyle::LK_JavaScript && FormatTok &&
+      if (Style.isJavaScript() && FormatTok &&
           FormatTok->Tok.getIdentifierInfo())
         // JavaScript only has pseudo keywords, all keywords are allowed to
         // appear in "IdentifierName" positions. See http://es5.github.io/#x7.6
@@ -1536,8 +1534,7 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
       // element continues.
       break;
     case tok::kw_try:
-      if (Style.Language == FormatStyle::LK_JavaScript &&
-          Line->MustBeDeclaration) {
+      if (Style.isJavaScript() && Line->MustBeDeclaration) {
         // field/method declaration.
         nextToken();
         break;
@@ -1564,17 +1561,15 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
       // expressions (functions that are not on their own line) must not create
       // a new unwrapped line, so they are special cased below.
       size_t TokenCount = Line->Tokens.size();
-      if (Style.Language == FormatStyle::LK_JavaScript &&
-          FormatTok->is(Keywords.kw_function) &&
+      if (Style.isJavaScript() && FormatTok->is(Keywords.kw_function) &&
           (TokenCount > 1 || (TokenCount == 1 && !Line->Tokens.front().Tok->is(
                                                      Keywords.kw_async)))) {
         tryToParseJSFunction();
         break;
       }
-      if ((Style.Language == FormatStyle::LK_JavaScript ||
-           Style.Language == FormatStyle::LK_Java) &&
+      if ((Style.isJavaScript() || Style.Language == FormatStyle::LK_Java) &&
           FormatTok->is(Keywords.kw_interface)) {
-        if (Style.Language == FormatStyle::LK_JavaScript) {
+        if (Style.isJavaScript()) {
           // In JavaScript/TypeScript, "interface" can be used as a standalone
           // identifier, e.g. in `var interface = 1;`. If "interface" is
           // followed by another identifier, it is very like to be an actual
@@ -1610,7 +1605,7 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
 
       // JS doesn't have macros, and within classes colons indicate fields, not
       // labels.
-      if (Style.Language == FormatStyle::LK_JavaScript)
+      if (Style.isJavaScript())
         break;
 
       TokenCount = Line->Tokens.size();
@@ -1641,19 +1636,9 @@ void UnwrappedLineParser::parseStructuralElement(bool IsTopLevel) {
       break;
     }
     case tok::equal:
-      // Fat arrows (=>) have tok::TokenKind tok::equal but TokenType
-      // TT_FatArrow. They always start an expression or a child block if
-      // followed by a curly brace.
-      if (FormatTok->is(TT_FatArrow)) {
-        nextToken();
-        if (FormatTok->is(tok::l_brace)) {
-          // C# may break after => if the next character is a newline.
-          if (Style.isCSharp() && Style.BraceWrapping.AfterFunction == true) {
-            // calling `addUnwrappedLine()` here causes odd parsing errors.
-            FormatTok->MustBreakBefore = true;
-          }
-          parseChildBlock();
-        }
+      if ((Style.isJavaScript() || Style.isCSharp()) &&
+          FormatTok->is(TT_FatArrow)) {
+        tryToParseChildBlock();
         break;
       }
 
@@ -1729,7 +1714,7 @@ bool UnwrappedLineParser::tryToParsePropertyAccessor() {
   // Try to parse the property accessor:
   // https://docs.microsoft.com/en-us/dotnet/csharp/programming-guide/classes-and-structs/properties
   Tokens->setPosition(StoredPosition);
-  if (!IsTrivialPropertyAccessor && Style.BraceWrapping.AfterFunction == true)
+  if (!IsTrivialPropertyAccessor && Style.BraceWrapping.AfterFunction)
     addUnwrappedLine();
   nextToken();
   do {
@@ -1944,6 +1929,19 @@ bool UnwrappedLineParser::tryToParseBracedList() {
   return true;
 }
 
+bool UnwrappedLineParser::tryToParseChildBlock() {
+  assert(Style.isJavaScript() || Style.isCSharp());
+  assert(FormatTok->is(TT_FatArrow));
+  // Fat arrows (=>) have tok::TokenKind tok::equal but TokenType TT_FatArrow.
+  // They always start an expression or a child block if followed by a curly
+  // brace.
+  nextToken();
+  if (FormatTok->isNot(tok::l_brace))
+    return false;
+  parseChildBlock();
+  return true;
+}
+
 bool UnwrappedLineParser::parseBracedList(bool ContinueOnSemicolons,
                                           bool IsEnum,
                                           tok::TokenKind ClosingBraceKind) {
@@ -1952,38 +1950,15 @@ bool UnwrappedLineParser::parseBracedList(bool ContinueOnSemicolons,
   // FIXME: Once we have an expression parser in the UnwrappedLineParser,
   // replace this by using parseAssignmentExpression() inside.
   do {
-    if (Style.isCSharp()) {
-      // Fat arrows (=>) have tok::TokenKind tok::equal but TokenType
-      // TT_FatArrow. They always start an expression or a child block if
-      // followed by a curly brace.
-      if (FormatTok->is(TT_FatArrow)) {
-        nextToken();
-        if (FormatTok->is(tok::l_brace)) {
-          // C# may break after => if the next character is a newline.
-          if (Style.isCSharp() && Style.BraceWrapping.AfterFunction == true) {
-            // calling `addUnwrappedLine()` here causes odd parsing errors.
-            FormatTok->MustBreakBefore = true;
-          }
-          parseChildBlock();
-          continue;
-        }
-      }
-    }
-    if (Style.Language == FormatStyle::LK_JavaScript) {
+    if (Style.isCSharp() && FormatTok->is(TT_FatArrow) &&
+        tryToParseChildBlock())
+      continue;
+    if (Style.isJavaScript()) {
       if (FormatTok->is(Keywords.kw_function) ||
           FormatTok->startsSequence(Keywords.kw_async, Keywords.kw_function)) {
         tryToParseJSFunction();
         continue;
       }
-      if (FormatTok->is(TT_FatArrow)) {
-        nextToken();
-        // Fat arrows can be followed by simple expressions or by child blocks
-        // in curly braces.
-        if (FormatTok->is(tok::l_brace)) {
-          parseChildBlock();
-          continue;
-        }
-      }
       if (FormatTok->is(tok::l_brace)) {
         // Could be a method inside of a braced list `{a() { return 1; }}`.
         if (tryToParseBracedList())
@@ -1998,12 +1973,6 @@ bool UnwrappedLineParser::parseBracedList(bool ContinueOnSemicolons,
       return !HasError;
     }
     switch (FormatTok->Tok.getKind()) {
-    case tok::caret:
-      nextToken();
-      if (FormatTok->is(tok::l_brace)) {
-        parseChildBlock();
-      }
-      break;
     case tok::l_square:
       if (Style.isCSharp())
         parseSquare();
@@ -2014,7 +1983,7 @@ bool UnwrappedLineParser::parseBracedList(bool ContinueOnSemicolons,
       parseParens();
       // JavaScript can just have free standing methods and getters/setters in
       // object literals. Detect them by a "{" following ")".
-      if (Style.Language == FormatStyle::LK_JavaScript) {
+      if (Style.isJavaScript()) {
         if (FormatTok->is(tok::l_brace))
           parseChildBlock();
         break;
@@ -2041,7 +2010,7 @@ bool UnwrappedLineParser::parseBracedList(bool ContinueOnSemicolons,
       // lists (in so-called TypeMemberLists). Thus, the semicolon cannot be
       // used for error recovery if we have otherwise determined that this is
       // a braced list.
-      if (Style.Language == FormatStyle::LK_JavaScript) {
+      if (Style.isJavaScript()) {
         nextToken();
         break;
       }
@@ -2095,18 +2064,18 @@ void UnwrappedLineParser::parseParens() {
       break;
     case tok::equal:
       if (Style.isCSharp() && FormatTok->is(TT_FatArrow))
-        parseStructuralElement();
+        tryToParseChildBlock();
       else
         nextToken();
       break;
     case tok::kw_class:
-      if (Style.Language == FormatStyle::LK_JavaScript)
+      if (Style.isJavaScript())
         parseRecord(/*ParseAsExpr=*/true);
       else
         nextToken();
       break;
     case tok::identifier:
-      if (Style.Language == FormatStyle::LK_JavaScript &&
+      if (Style.isJavaScript() &&
           (FormatTok->is(Keywords.kw_function) ||
            FormatTok->startsSequence(Keywords.kw_async, Keywords.kw_function)))
         tryToParseJSFunction();
@@ -2160,15 +2129,22 @@ void UnwrappedLineParser::parseSquare(bool LambdaIntroducer) {
 }
 
 void UnwrappedLineParser::parseIfThenElse() {
+  auto HandleAttributes = [this]() {
+    // Handle AttributeMacro, e.g. `if (x) UNLIKELY`.
+    if (FormatTok->is(TT_AttributeMacro))
+      nextToken();
+    // Handle [[likely]] / [[unlikely]] attributes.
+    if (FormatTok->is(tok::l_square) && tryToParseSimpleAttribute())
+      parseSquare();
+  };
+
   assert(FormatTok->Tok.is(tok::kw_if) && "'if' expected");
   nextToken();
   if (FormatTok->Tok.isOneOf(tok::kw_constexpr, tok::identifier))
     nextToken();
   if (FormatTok->Tok.is(tok::l_paren))
     parseParens();
-  // handle [[likely]] / [[unlikely]]
-  if (FormatTok->is(tok::l_square) && tryToParseSimpleAttribute())
-    parseSquare();
+  HandleAttributes();
   bool NeedsUnwrappedLine = false;
   if (FormatTok->Tok.is(tok::l_brace)) {
     CompoundStatementIndenter Indenter(this, Style, Line->Level);
@@ -2185,9 +2161,7 @@ void UnwrappedLineParser::parseIfThenElse() {
   }
   if (FormatTok->Tok.is(tok::kw_else)) {
     nextToken();
-    // handle [[likely]] / [[unlikely]]
-    if (FormatTok->Tok.is(tok::l_square) && tryToParseSimpleAttribute())
-      parseSquare();
+    HandleAttributes();
     if (FormatTok->Tok.is(tok::l_brace)) {
       CompoundStatementIndenter Indenter(this, Style, Line->Level);
       parseBlock();
@@ -2272,8 +2246,7 @@ void UnwrappedLineParser::parseTryCatch() {
       nextToken();
     if (!(FormatTok->isOneOf(tok::kw_catch, Keywords.kw___except,
                              tok::kw___finally) ||
-          ((Style.Language == FormatStyle::LK_Java ||
-            Style.Language == FormatStyle::LK_JavaScript) &&
+          ((Style.Language == FormatStyle::LK_Java || Style.isJavaScript()) &&
            FormatTok->is(Keywords.kw_finally)) ||
           (FormatTok->Tok.isObjCAtKeyword(tok::objc_catch) ||
            FormatTok->Tok.isObjCAtKeyword(tok::objc_finally))))
@@ -2396,8 +2369,7 @@ void UnwrappedLineParser::parseForOrWhileLoop() {
          "'for', 'while' or foreach macro expected");
   nextToken();
   // JS' for await ( ...
-  if (Style.Language == FormatStyle::LK_JavaScript &&
-      FormatTok->is(Keywords.kw_await))
+  if (Style.isJavaScript() && FormatTok->is(Keywords.kw_await))
     nextToken();
   if (Style.isCpp() && FormatTok->is(tok::kw_co_await))
     nextToken();
@@ -2643,8 +2615,7 @@ bool UnwrappedLineParser::parseEnum() {
   // In TypeScript, "enum" can also be used as property name, e.g. in interface
   // declarations. An "enum" keyword followed by a colon would be a syntax
   // error and thus assume it is just an identifier.
-  if (Style.Language == FormatStyle::LK_JavaScript &&
-      FormatTok->isOneOf(tok::colon, tok::question))
+  if (Style.isJavaScript() && FormatTok->isOneOf(tok::colon, tok::question))
     return false;
 
   // In protobuf, "enum" can be used as a field name.
@@ -2716,8 +2687,8 @@ bool UnwrappedLineParser::parseStructLike() {
   // record declaration or definition can start a structural element.
   parseRecord();
   // This does not apply to Java, JavaScript and C#.
-  if (Style.Language == FormatStyle::LK_Java ||
-      Style.Language == FormatStyle::LK_JavaScript || Style.isCSharp()) {
+  if (Style.Language == FormatStyle::LK_Java || Style.isJavaScript() ||
+      Style.isCSharp()) {
     if (FormatTok->is(tok::semi))
       nextToken();
     addUnwrappedLine();
@@ -2846,10 +2817,9 @@ void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
   while (FormatTok->isOneOf(tok::identifier, tok::coloncolon, tok::hashhash,
                             tok::kw___attribute, tok::kw___declspec,
                             tok::kw_alignas, tok::l_square, tok::r_square) ||
-         ((Style.Language == FormatStyle::LK_Java ||
-           Style.Language == FormatStyle::LK_JavaScript) &&
+         ((Style.Language == FormatStyle::LK_Java || Style.isJavaScript()) &&
           FormatTok->isOneOf(tok::period, tok::comma))) {
-    if (Style.Language == FormatStyle::LK_JavaScript &&
+    if (Style.isJavaScript() &&
         FormatTok->isOneOf(Keywords.kw_extends, Keywords.kw_implements)) {
       // JavaScript/TypeScript supports inline object types in
       // extends/implements positions:
@@ -3323,7 +3293,7 @@ void UnwrappedLineParser::nextToken(int LevelDifference) {
   flushComments(isOnNewLine(*FormatTok));
   pushToken(FormatTok);
   FormatToken *Previous = FormatTok;
-  if (Style.Language != FormatStyle::LK_JavaScript)
+  if (!Style.isJavaScript())
     readToken(LevelDifference);
   else
     readTokenWithJavaScriptASI();
diff --git a/clang/lib/Format/UnwrappedLineParser.h b/clang/lib/Format/UnwrappedLineParser.h
index b4c082654597..0c79723d50fc 100644
--- a/clang/lib/Format/UnwrappedLineParser.h
+++ b/clang/lib/Format/UnwrappedLineParser.h
@@ -138,6 +138,7 @@ private:
   // https://docs.microsoft.com/en-us/dotnet/csharp/language-reference/keywords/where-generic-type-constraint
   void parseCSharpGenericTypeConstraint();
   bool tryToParseLambda();
+  bool tryToParseChildBlock();
   bool tryToParseLambdaIntroducer();
   bool tryToParsePropertyAccessor();
   void tryToParseJSFunction();
diff --git a/clang/lib/Format/WhitespaceManager.cpp b/clang/lib/Format/WhitespaceManager.cpp
index fae8a1c3fdc6..96a66da0f82b 100644
--- a/clang/lib/Format/WhitespaceManager.cpp
+++ b/clang/lib/Format/WhitespaceManager.cpp
@@ -1282,9 +1282,12 @@ void WhitespaceManager::generateChanges() {
                                  C.EscapedNewlineColumn);
       else
         appendNewlineText(ReplacementText, C.NewlinesBefore);
+      // FIXME: This assert should hold if we computed the column correctly.
+      // assert((int)C.StartOfTokenColumn >= C.Spaces);
       appendIndentText(
           ReplacementText, C.Tok->IndentLevel, std::max(0, C.Spaces),
-          C.StartOfTokenColumn - std::max(0, C.Spaces), C.IsAligned);
+          std::max((int)C.StartOfTokenColumn, C.Spaces) - std::max(0, C.Spaces),
+          C.IsAligned);
       ReplacementText.append(C.CurrentLinePrefix);
       storeReplacement(C.OriginalWhitespaceRange, ReplacementText);
     }
diff --git a/clang/lib/Frontend/CompilerInstance.cpp b/clang/lib/Frontend/CompilerInstance.cpp
index 1432607204bd..31e7ea3d243d 100644
--- a/clang/lib/Frontend/CompilerInstance.cpp
+++ b/clang/lib/Frontend/CompilerInstance.cpp
@@ -1154,12 +1154,12 @@ compileModuleImpl(CompilerInstance &ImportingInstance, SourceLocation ImportLoc,
   // Remove any macro definitions that are explicitly ignored by the module.
   // They aren't supposed to affect how the module is built anyway.
   HeaderSearchOptions &HSOpts = Invocation->getHeaderSearchOpts();
-  llvm::erase_if(
-      PPOpts.Macros, [&HSOpts](const std::pair<std::string, bool> &def) {
-        StringRef MacroDef = def.first;
-        return HSOpts.ModulesIgnoreMacros.count(
-                   llvm::CachedHashString(MacroDef.split('=').first)) > 0;
-      });
+  llvm::erase_if(PPOpts.Macros,
+                 [&HSOpts](const std::pair<std::string, bool> &def) {
+                   StringRef MacroDef = def.first;
+                   return HSOpts.ModulesIgnoreMacros.contains(
+                       llvm::CachedHashString(MacroDef.split('=').first));
+                 });
 
   // If the original compiler invocation had -fmodule-name, pass it through.
   Invocation->getLangOpts()->ModuleName =
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index c104a6f40e20..b71addd84bfd 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -770,9 +770,7 @@ static void parseAnalyzerConfigs(AnalyzerOptions &AnOpts,
 static void getAllNoBuiltinFuncValues(ArgList &Args,
                                       std::vector<std::string> &Funcs) {
   std::vector<std::string> Values = Args.getAllArgValues(OPT_fno_builtin_);
-  auto BuiltinEnd = llvm::partition(Values, [](const std::string FuncName) {
-    return Builtin::Context::isBuiltinFunc(FuncName);
-  });
+  auto BuiltinEnd = llvm::partition(Values, Builtin::Context::isBuiltinFunc);
   Funcs.insert(Funcs.end(), Values.begin(), BuiltinEnd);
 }
 
@@ -1285,7 +1283,7 @@ static std::string serializeXRayInstrumentationBundle(const XRayInstrSet &S) {
   std::string Buffer;
   llvm::raw_string_ostream OS(Buffer);
   llvm::interleave(BundleParts, OS, [&OS](StringRef Part) { OS << Part; }, ",");
-  return OS.str();
+  return Buffer;
 }
 
 // Set the profile kind using fprofile-instrument-use-path.
@@ -4123,6 +4121,13 @@ bool CompilerInvocation::ParseLangArgs(LangOptions &Opts, ArgList &Args,
         {std::string(Split.first), std::string(Split.second)});
   }
 
+  // Error if -mvscale-min is unbounded.
+  if (Arg *A = Args.getLastArg(options::OPT_mvscale_min_EQ)) {
+    unsigned VScaleMin;
+    if (StringRef(A->getValue()).getAsInteger(10, VScaleMin) || VScaleMin == 0)
+      Diags.Report(diag::err_cc1_unbounded_vscale_min);
+  }
+
   return Diags.getNumErrors() == NumErrorsBefore;
 }
 
@@ -4513,7 +4518,7 @@ bool CompilerInvocation::CreateFromArgsImpl(
 
   // Store the command-line for using in the CodeView backend.
   Res.getCodeGenOpts().Argv0 = Argv0;
-  Res.getCodeGenOpts().CommandLineArgs = CommandLineArgs;
+  append_range(Res.getCodeGenOpts().CommandLineArgs, CommandLineArgs);
 
   FixupInvocation(Res, Diags, Args, DashX);
 
diff --git a/clang/lib/Frontend/InitPreprocessor.cpp b/clang/lib/Frontend/InitPreprocessor.cpp
index 0c153446142e..629f99110661 100644
--- a/clang/lib/Frontend/InitPreprocessor.cpp
+++ b/clang/lib/Frontend/InitPreprocessor.cpp
@@ -500,8 +500,12 @@ static void InitializeStandardPredefinedMacros(const TargetInfo &TI,
   // Not "standard" per se, but available even with the -undef flag.
   if (LangOpts.AsmPreprocessor)
     Builder.defineMacro("__ASSEMBLER__");
-  if (LangOpts.CUDA && !LangOpts.HIP)
-    Builder.defineMacro("__CUDA__");
+  if (LangOpts.CUDA) {
+    if (LangOpts.GPURelocatableDeviceCode)
+      Builder.defineMacro("__CLANG_RDC__");
+    if (!LangOpts.HIP)
+      Builder.defineMacro("__CUDA__");
+  }
   if (LangOpts.HIP) {
     Builder.defineMacro("__HIP__");
     Builder.defineMacro("__HIPCC__");
diff --git a/clang/lib/Frontend/TestModuleFileExtension.cpp b/clang/lib/Frontend/TestModuleFileExtension.cpp
index ea737e6891bf..2d5145d0c54c 100644
--- a/clang/lib/Frontend/TestModuleFileExtension.cpp
+++ b/clang/lib/Frontend/TestModuleFileExtension.cpp
@@ -133,5 +133,5 @@ std::string TestModuleFileExtension::str() const {
   llvm::raw_string_ostream OS(Buffer);
   OS << BlockName << ":" << MajorVersion << ":" << MinorVersion << ":" << Hashed
      << ":" << UserInfo;
-  return OS.str();
+  return Buffer;
 }
diff --git a/clang/lib/Headers/arm_neon_sve_bridge.h b/clang/lib/Headers/arm_neon_sve_bridge.h
new file mode 100644
index 000000000000..17699d8d11dd
--- /dev/null
+++ b/clang/lib/Headers/arm_neon_sve_bridge.h
@@ -0,0 +1,184 @@
+/*===---- arm_neon_sve_bridge.h - ARM NEON SVE Bridge intrinsics -----------===
+ *
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __ARM_NEON_SVE_BRIDGE_H
+#define __ARM_NEON_SVE_BRIDGE_H
+
+#include <arm_neon.h>
+#include <arm_sve.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Function attributes */
+#define __ai static __inline__ __attribute__((__always_inline__, __nodebug__))
+#define __aio                                                                  \
+  static __inline__                                                            \
+      __attribute__((__always_inline__, __nodebug__, __overloadable__))
+
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s8)))
+svint8_t svset_neonq(svint8_t, int8x16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s16)))
+svint16_t svset_neonq(svint16_t, int16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s32)))
+svint32_t svset_neonq(svint32_t, int32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s64)))
+svint64_t svset_neonq(svint64_t, int64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u8)))
+svuint8_t svset_neonq(svuint8_t, uint8x16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u16)))
+svuint16_t svset_neonq(svuint16_t, uint16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u32)))
+svuint32_t svset_neonq(svuint32_t, uint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u64)))
+svuint64_t svset_neonq(svuint64_t, uint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f16)))
+svfloat16_t svset_neonq(svfloat16_t, float16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f32)))
+svfloat32_t svset_neonq(svfloat32_t, float32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f64)))
+svfloat64_t svset_neonq(svfloat64_t, float64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s8)))
+svint8_t svset_neonq_s8(svint8_t, int8x16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s16)))
+svint16_t svset_neonq_s16(svint16_t, int16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s32)))
+svint32_t svset_neonq_s32(svint32_t, int32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_s64)))
+svint64_t svset_neonq_s64(svint64_t, int64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u8)))
+svuint8_t svset_neonq_u8(svuint8_t, uint8x16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u16)))
+svuint16_t svset_neonq_u16(svuint16_t, uint16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u32)))
+svuint32_t svset_neonq_u32(svuint32_t, uint32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_u64)))
+svuint64_t svset_neonq_u64(svuint64_t, uint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f16)))
+svfloat16_t svset_neonq_f16(svfloat16_t, float16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f32)))
+svfloat32_t svset_neonq_f32(svfloat32_t, float32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_f64)))
+svfloat64_t svset_neonq_f64(svfloat64_t, float64x2_t);
+
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s8)))
+int8x16_t svget_neonq(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s16)))
+int16x8_t svget_neonq(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s32)))
+int32x4_t svget_neonq(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s64)))
+int64x2_t svget_neonq(svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u8)))
+uint8x16_t svget_neonq(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u16)))
+uint16x8_t svget_neonq(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u32)))
+uint32x4_t svget_neonq(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u64)))
+uint64x2_t svget_neonq(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f16)))
+float16x8_t svget_neonq(svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f32)))
+float32x4_t svget_neonq(svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f64)))
+float64x2_t svget_neonq(svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s8)))
+int8x16_t svget_neonq_s8(svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s16)))
+int16x8_t svget_neonq_s16(svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s32)))
+int32x4_t svget_neonq_s32(svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_s64)))
+int64x2_t svget_neonq_s64(svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u8)))
+uint8x16_t svget_neonq_u8(svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u16)))
+uint16x8_t svget_neonq_u16(svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u32)))
+uint32x4_t svget_neonq_u32(svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_u64)))
+uint64x2_t svget_neonq_u64(svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f16)))
+float16x8_t svget_neonq_f16(svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f32)))
+float32x4_t svget_neonq_f32(svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_f64)))
+float64x2_t svget_neonq_f64(svfloat64_t);
+
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s8)))
+svint8_t svdup_neonq(int8x16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s16)))
+svint16_t svdup_neonq(int16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s32)))
+svint32_t svdup_neonq(int32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s64)))
+svint64_t svdup_neonq(int64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u8)))
+svuint8_t svdup_neonq(uint8x16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u16)))
+svuint16_t svdup_neonq(uint16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u32)))
+svuint32_t svdup_neonq(uint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u64)))
+svuint64_t svdup_neonq(uint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f16)))
+svfloat16_t svdup_neonq(float16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f32)))
+svfloat32_t svdup_neonq(float32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f64)))
+svfloat64_t svdup_neonq(float64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s8)))
+svint8_t svdup_neonq_s8(int8x16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s16)))
+svint16_t svdup_neonq_s16(int16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s32)))
+svint32_t svdup_neonq_s32(int32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_s64)))
+svint64_t svdup_neonq_s64(int64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u8)))
+svuint8_t svdup_neonq_u8(uint8x16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u16)))
+svuint16_t svdup_neonq_u16(uint16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u32)))
+svuint32_t svdup_neonq_u32(uint32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_u64)))
+svuint64_t svdup_neonq_u64(uint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f16)))
+svfloat16_t svdup_neonq_f16(float16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f32)))
+svfloat32_t svdup_neonq_f32(float32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_f64)))
+svfloat64_t svdup_neonq_f64(float64x2_t);
+
+#if defined(__ARM_FEATURE_SVE_BF16)
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_bf16)))
+svbfloat16_t svset_neonq(svbfloat16_t, bfloat16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset_neonq_bf16)))
+svbfloat16_t svset_neonq_bf16(svbfloat16_t, bfloat16x8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_bf16)))
+bfloat16x8_t svget_neonq(svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget_neonq_bf16)))
+bfloat16x8_t svget_neonq_bf16(svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_bf16)))
+svbfloat16_t svdup_neonq(bfloat16x8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_neonq_bf16)))
+svbfloat16_t svdup_neonq_bf16(bfloat16x8_t);
+#endif // defined(__ARM_FEATURE_SVE_BF16)
+
+#undef __ai
+#undef __aio
+
+#ifdef __cplusplus
+} // extern "C"
+#endif
+
+#endif //__ARM_NEON_SVE_BRIDGE_H
diff --git a/clang/lib/Headers/hexagon_protos.h b/clang/lib/Headers/hexagon_protos.h
index cdffd93bb859..2642f3c8428d 100644
--- a/clang/lib/Headers/hexagon_protos.h
+++ b/clang/lib/Headers/hexagon_protos.h
@@ -8003,17 +8003,6 @@
 #define Q6_P_vtrunohb_PP __builtin_HEXAGON_S6_vtrunohb_ppp
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
-#if __HEXAGON_ARCH__ >= 62
-/* ==========================================================================
-   Assembly Syntax:       Vd32=vmem(Rt32):nt
-   C Intrinsic Prototype: HVX_Vector Q6_V_vmem_R_nt(Word32 Rt)
-   Instruction Type:      MAPPING
-   Execution Slots:       SLOT0123
-   ========================================================================== */
-
-#define Q6_V_vmem_R_nt __builtin_HEXAGON_V6_ldntnt0
-#endif /* __HEXAGON_ARCH___ >= 62 */
-
 #if __HEXAGON_ARCH__ >= 65
 /* ==========================================================================
    Assembly Syntax:       Pd4=!any8(vcmpb.eq(Rss32,Rtt32))
diff --git a/clang/lib/Headers/hexagon_types.h b/clang/lib/Headers/hexagon_types.h
index 6958809418d8..029727cc4817 100644
--- a/clang/lib/Headers/hexagon_types.h
+++ b/clang/lib/Headers/hexagon_types.h
@@ -1177,37 +1177,6 @@ private:
 
 #endif /* __cplusplus */
 
-// V65 Silver types
-#if __Q6S_ARCH__ >= 65
-  // Silver vector types are 128 bytes, and pairs are 256. The vector predicate
-  // types are 16 bytes and 32 bytes for pairs.
-  typedef long HEXAGON_VecPred128 __attribute__((__vector_size__(16)))
-    __attribute__((aligned(128)));
-
-  typedef long HEXAGON_VecPred256 __attribute__((__vector_size__(32)))
-    __attribute__((aligned(128)));
-
-  typedef long HEXAGON_Vect1024 __attribute__((__vector_size__(128)))
-    __attribute__((aligned(128)));
-
-  typedef long HEXAGON_Vect2048 __attribute__((__vector_size__(256)))
-    __attribute__((aligned(256)));
-
-  typedef long HEXAGON_UVect1024 __attribute__((__vector_size__(128)))
-    __attribute__((aligned(4)));
-
-  typedef long HEXAGON_UVect2048 __attribute__((__vector_size__(256)))
-    __attribute__((aligned(4)));
-
-  #define Q6S_VectorPredPair HEXAGON_VecPred256
-  #define Q6S_VectorPred     HEXAGON_VecPred128
-  #define Q6S_Vector         HEXAGON_Vect1024
-  #define Q6S_VectorPair     HEXAGON_Vect2048
-  #define Q6S_UVector        HEXAGON_UVect1024
-  #define Q6S_UVectorPair    HEXAGON_UVect2048
-
-#else /* __Q6S_ARCH__ >= 65 */
-
 // V65 Vector types
 #if __HVX_ARCH__ >= 65
 #if defined __HVX__ && (__HVX_LENGTH__ == 128)
@@ -1256,7 +1225,6 @@ private:
 #endif /* defined __HVX__ &&  (__HVX_LENGTH__ == 64) */
 #endif /* defined __HVX__ && (__HVX_LENGTH__ == 128) */
 #endif /* __HVX_ARCH__ >= 65 */
-#endif /* __Q6S_ARCH__ >= 65 */
 
 /* Predicates */
 
diff --git a/clang/lib/Headers/hvx_hexagon_protos.h b/clang/lib/Headers/hvx_hexagon_protos.h
index 41ce7a6b93e9..7e3679a38b2c 100644
--- a/clang/lib/Headers/hvx_hexagon_protos.h
+++ b/clang/lib/Headers/hvx_hexagon_protos.h
@@ -9,7 +9,6 @@
 //===----------------------------------------------------------------------===//
 
 
-
 #ifndef _HVX_HEXAGON_PROTOS_H_
 #define _HVX_HEXAGON_PROTOS_H_ 1
 
@@ -28,7 +27,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_R_vextract_VR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_extractw)
+#define Q6_R_vextract_VR(Vu,Rs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_extractw)(Vu,Rs)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -39,7 +38,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_hi_W __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_hi)
+#define Q6_V_hi_W(Vss) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_hi)(Vss)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -50,7 +49,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_lo_W __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lo)
+#define Q6_V_lo_W(Vss) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lo)(Vss)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -61,7 +60,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_V_vsplat_R __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lvsplatw)
+#define Q6_V_vsplat_R(Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lvsplatw)(Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -72,7 +71,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_and_QQ __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_and)
+#define Q6_Q_and_QQ(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -83,7 +82,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_and_QQn __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_and_n)
+#define Q6_Q_and_QQn(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_and_n)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -94,7 +93,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_not_Q __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_not)
+#define Q6_Q_not_Q(Qs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_not)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -105,7 +104,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_or_QQ __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_or)
+#define Q6_Q_or_QQ(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -116,7 +115,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_or_QQn __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_or_n)
+#define Q6_Q_or_QQn(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_or_n)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -127,7 +126,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vsetq_R __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_scalar2)
+#define Q6_Q_vsetq_R(Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_scalar2)(Rt)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -138,7 +137,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_xor_QQ __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_xor)
+#define Q6_Q_xor_QQ(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -149,7 +148,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vmem_QnRIV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_nqpred_ai)
+#define Q6_vmem_QnRIV(Qv,Rt,Vs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_nqpred_ai)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Rt,Vs)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -160,7 +159,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vmem_QnRIV_nt __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_nt_nqpred_ai)
+#define Q6_vmem_QnRIV_nt(Qv,Rt,Vs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_nt_nqpred_ai)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Rt,Vs)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -171,7 +170,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vmem_QRIV_nt __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_nt_qpred_ai)
+#define Q6_vmem_QRIV_nt(Qv,Rt,Vs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_nt_qpred_ai)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Rt,Vs)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -182,7 +181,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vmem_QRIV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_qpred_ai)
+#define Q6_vmem_QRIV(Qv,Rt,Vs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vS32b_qpred_ai)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Rt,Vs)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -193,7 +192,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuh_vabsdiff_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffh)
+#define Q6_Vuh_vabsdiff_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -204,7 +203,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vub_vabsdiff_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffub)
+#define Q6_Vub_vabsdiff_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -215,7 +214,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuh_vabsdiff_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffuh)
+#define Q6_Vuh_vabsdiff_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -226,7 +225,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vabsdiff_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffw)
+#define Q6_Vuw_vabsdiff_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsdiffw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -237,7 +236,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vabs_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsh)
+#define Q6_Vh_vabs_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -248,7 +247,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vabs_Vh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsh_sat)
+#define Q6_Vh_vabs_Vh_sat(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsh_sat)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -259,7 +258,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vabs_Vw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsw)
+#define Q6_Vw_vabs_Vw(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsw)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -270,7 +269,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vabs_Vw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsw_sat)
+#define Q6_Vw_vabs_Vw_sat(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsw_sat)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -281,7 +280,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vadd_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddb)
+#define Q6_Vb_vadd_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -292,7 +291,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wb_vadd_WbWb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddb_dv)
+#define Q6_Wb_vadd_WbWb(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddb_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -303,7 +302,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_condacc_QnVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbnq)
+#define Q6_Vb_condacc_QnVbVb(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbnq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -314,7 +313,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_condacc_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbq)
+#define Q6_Vb_condacc_QVbVb(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -325,7 +324,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vadd_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddh)
+#define Q6_Vh_vadd_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -336,7 +335,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vadd_WhWh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddh_dv)
+#define Q6_Wh_vadd_WhWh(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddh_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -347,7 +346,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_condacc_QnVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhnq)
+#define Q6_Vh_condacc_QnVhVh(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhnq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -358,7 +357,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_condacc_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhq)
+#define Q6_Vh_condacc_QVhVh(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -369,7 +368,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vadd_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhsat)
+#define Q6_Vh_vadd_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -380,7 +379,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vadd_WhWh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhsat_dv)
+#define Q6_Wh_vadd_WhWh_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -391,7 +390,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vadd_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhw)
+#define Q6_Ww_vadd_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -402,7 +401,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vadd_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubh)
+#define Q6_Wh_vadd_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -413,7 +412,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vadd_VubVub_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubsat)
+#define Q6_Vub_vadd_VubVub_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -424,7 +423,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wub_vadd_WubWub_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubsat_dv)
+#define Q6_Wub_vadd_WubWub_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -435,7 +434,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vadd_VuhVuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhsat)
+#define Q6_Vuh_vadd_VuhVuh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -446,7 +445,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuh_vadd_WuhWuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhsat_dv)
+#define Q6_Wuh_vadd_WuhWuh_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -457,7 +456,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vadd_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhw)
+#define Q6_Ww_vadd_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -468,7 +467,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vadd_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddw)
+#define Q6_Vw_vadd_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -479,7 +478,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vadd_WwWw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddw_dv)
+#define Q6_Ww_vadd_WwWw(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddw_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -490,7 +489,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_condacc_QnVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwnq)
+#define Q6_Vw_condacc_QnVwVw(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwnq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -501,7 +500,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_condacc_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwq)
+#define Q6_Vw_condacc_QVwVw(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -512,7 +511,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vadd_VwVw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwsat)
+#define Q6_Vw_vadd_VwVw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -523,7 +522,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vadd_WwWw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwsat_dv)
+#define Q6_Ww_vadd_WwWw_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddwsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -534,7 +533,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_valign_VVR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_valignb)
+#define Q6_V_valign_VVR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_valignb)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -545,7 +544,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_valign_VVI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_valignbi)
+#define Q6_V_valign_VVI(Vu,Vv,Iu3) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_valignbi)(Vu,Vv,Iu3)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -556,7 +555,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vand_VV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vand)
+#define Q6_V_vand_VV(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vand)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -567,7 +566,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_V_vand_QR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)
+#define Q6_V_vand_QR(Qu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qu),-1),Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -578,7 +577,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_V_vandor_VQR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt_acc)
+#define Q6_V_vandor_VQR(Vx,Qu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt_acc)(Vx,__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qu),-1),Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -589,7 +588,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Q_vand_VR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)
+#define Q6_Q_vand_VR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)(Vu,Rt)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -600,7 +599,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Q_vandor_QVR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt_acc)
+#define Q6_Q_vandor_QVR(Qx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt_acc)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Rt)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -611,7 +610,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasl_VhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslh)
+#define Q6_Vh_vasl_VhR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslh)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -622,7 +621,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasl_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslhv)
+#define Q6_Vh_vasl_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslhv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -633,7 +632,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vasl_VwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslw)
+#define Q6_Vw_vasl_VwR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslw)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -644,7 +643,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vaslacc_VwVwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslw_acc)
+#define Q6_Vw_vaslacc_VwVwR(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslw_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -655,7 +654,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vasl_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslwv)
+#define Q6_Vw_vasl_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslwv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -666,7 +665,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasr_VhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrh)
+#define Q6_Vh_vasr_VhR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrh)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -677,7 +676,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vasr_VhVhR_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhbrndsat)
+#define Q6_Vb_vasr_VhVhR_rnd_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhbrndsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -688,7 +687,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vasr_VhVhR_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhubrndsat)
+#define Q6_Vub_vasr_VhVhR_rnd_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhubrndsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -699,7 +698,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vasr_VhVhR_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhubsat)
+#define Q6_Vub_vasr_VhVhR_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhubsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -710,7 +709,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasr_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhv)
+#define Q6_Vh_vasr_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -721,7 +720,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vasr_VwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrw)
+#define Q6_Vw_vasr_VwR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrw)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -732,7 +731,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vasracc_VwVwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrw_acc)
+#define Q6_Vw_vasracc_VwVwR(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrw_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -743,7 +742,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasr_VwVwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwh)
+#define Q6_Vh_vasr_VwVwR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwh)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -754,7 +753,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasr_VwVwR_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwhrndsat)
+#define Q6_Vh_vasr_VwVwR_rnd_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwhrndsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -765,7 +764,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasr_VwVwR_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwhsat)
+#define Q6_Vh_vasr_VwVwR_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwhsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -776,7 +775,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vasr_VwVwR_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwuhsat)
+#define Q6_Vuh_vasr_VwVwR_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwuhsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -787,7 +786,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vasr_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwv)
+#define Q6_Vw_vasr_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -798,7 +797,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_equals_V __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vassign)
+#define Q6_V_equals_V(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vassign)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -809,7 +808,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_W_equals_W __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vassignp)
+#define Q6_W_equals_W(Vuu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vassignp)(Vuu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -820,7 +819,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vavg_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgh)
+#define Q6_Vh_vavg_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -831,7 +830,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vavg_VhVh_rnd __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavghrnd)
+#define Q6_Vh_vavg_VhVh_rnd(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavghrnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -842,7 +841,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vavg_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgub)
+#define Q6_Vub_vavg_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -853,7 +852,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vavg_VubVub_rnd __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgubrnd)
+#define Q6_Vub_vavg_VubVub_rnd(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgubrnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -864,7 +863,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vavg_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguh)
+#define Q6_Vuh_vavg_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -875,7 +874,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vavg_VuhVuh_rnd __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguhrnd)
+#define Q6_Vuh_vavg_VuhVuh_rnd(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguhrnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -886,7 +885,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vavg_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgw)
+#define Q6_Vw_vavg_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -897,7 +896,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vavg_VwVw_rnd __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgwrnd)
+#define Q6_Vw_vavg_VwVw_rnd(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgwrnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -908,7 +907,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vcl0_Vuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcl0h)
+#define Q6_Vuh_vcl0_Vuh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcl0h)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -919,7 +918,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vcl0_Vuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcl0w)
+#define Q6_Vuw_vcl0_Vuw(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcl0w)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -930,7 +929,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_W_vcombine_VV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcombine)
+#define Q6_W_vcombine_VV(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcombine)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -941,7 +940,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vzero __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vd0)
+#define Q6_V_vzero() __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vd0)()
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -952,7 +951,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vdeal_Vb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealb)
+#define Q6_Vb_vdeal_Vb(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealb)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -963,7 +962,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vdeale_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealb4w)
+#define Q6_Vb_vdeale_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealb4w)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -974,7 +973,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vdeal_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealh)
+#define Q6_Vh_vdeal_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -985,7 +984,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_W_vdeal_VVR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealvdd)
+#define Q6_W_vdeal_VVR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdealvdd)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -996,7 +995,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vdelta_VV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdelta)
+#define Q6_V_vdelta_VV(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdelta)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1007,7 +1006,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vdmpy_VubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus)
+#define Q6_Vh_vdmpy_VubRb(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1018,7 +1017,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vdmpyacc_VhVubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus_acc)
+#define Q6_Vh_vdmpyacc_VhVubRb(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1029,7 +1028,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vdmpy_WubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus_dv)
+#define Q6_Wh_vdmpy_WubRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus_dv)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1040,7 +1039,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vdmpyacc_WhWubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus_dv_acc)
+#define Q6_Wh_vdmpyacc_WhWubRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpybus_dv_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1051,7 +1050,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpy_VhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb)
+#define Q6_Vw_vdmpy_VhRb(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1062,7 +1061,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpyacc_VwVhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb_acc)
+#define Q6_Vw_vdmpyacc_VwVhRb(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1073,7 +1072,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vdmpy_WhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb_dv)
+#define Q6_Ww_vdmpy_WhRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb_dv)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1084,7 +1083,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vdmpyacc_WwWhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb_dv_acc)
+#define Q6_Ww_vdmpyacc_WwWhRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhb_dv_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1095,7 +1094,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpy_WhRh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhisat)
+#define Q6_Vw_vdmpy_WhRh_sat(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhisat)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1106,29 +1105,29 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpyacc_VwWhRh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhisat_acc)
+#define Q6_Vw_vdmpyacc_VwWhRh_sat(Vx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhisat_acc)(Vx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vd32.w=vdmpy(Vu32.h,Rt32.h):sat
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vdmpy_VhRh_sat(HVX_Vector Vu, Word32 Rt)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpy_VhRh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsat)
+#define Q6_Vw_vdmpy_VhRh_sat(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsat)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vx32.w+=vdmpy(Vu32.h,Rt32.h):sat
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vdmpyacc_VwVhRh_sat(HVX_Vector Vx, HVX_Vector Vu, Word32 Rt)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpyacc_VwVhRh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsat_acc)
+#define Q6_Vw_vdmpyacc_VwVhRh_sat(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsat_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1139,7 +1138,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpy_WhRuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsuisat)
+#define Q6_Vw_vdmpy_WhRuh_sat(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsuisat)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1150,40 +1149,40 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpyacc_VwWhRuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsuisat_acc)
+#define Q6_Vw_vdmpyacc_VwWhRuh_sat(Vx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsuisat_acc)(Vx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vd32.w=vdmpy(Vu32.h,Rt32.uh):sat
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vdmpy_VhRuh_sat(HVX_Vector Vu, Word32 Rt)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpy_VhRuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsusat)
+#define Q6_Vw_vdmpy_VhRuh_sat(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsusat)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vx32.w+=vdmpy(Vu32.h,Rt32.uh):sat
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vdmpyacc_VwVhRuh_sat(HVX_Vector Vx, HVX_Vector Vu, Word32 Rt)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpyacc_VwVhRuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsusat_acc)
+#define Q6_Vw_vdmpyacc_VwVhRuh_sat(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhsusat_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vd32.w=vdmpy(Vu32.h,Vv32.h):sat
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vdmpy_VhVh_sat(HVX_Vector Vu, HVX_Vector Vv)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpy_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhvsat)
+#define Q6_Vw_vdmpy_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhvsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1194,7 +1193,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vdmpyacc_VwVhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhvsat_acc)
+#define Q6_Vw_vdmpyacc_VwVhVh_sat(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpyhvsat_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1205,7 +1204,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vdsad_WuhRuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdsaduh)
+#define Q6_Wuw_vdsad_WuhRuh(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdsaduh)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1216,7 +1215,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vdsadacc_WuwWuhRuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdsaduh_acc)
+#define Q6_Wuw_vdsadacc_WuwWuhRuh(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdsaduh_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1227,7 +1226,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eq_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb)
+#define Q6_Q_vcmp_eq_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1238,7 +1237,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqand_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb_and)
+#define Q6_Q_vcmp_eqand_QVbVb(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1249,7 +1248,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqor_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb_or)
+#define Q6_Q_vcmp_eqor_QVbVb(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1260,7 +1259,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqxacc_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb_xor)
+#define Q6_Q_vcmp_eqxacc_QVbVb(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqb_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1271,7 +1270,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eq_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh)
+#define Q6_Q_vcmp_eq_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1282,7 +1281,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqand_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh_and)
+#define Q6_Q_vcmp_eqand_QVhVh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1293,7 +1292,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqor_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh_or)
+#define Q6_Q_vcmp_eqor_QVhVh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1304,7 +1303,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqxacc_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh_xor)
+#define Q6_Q_vcmp_eqxacc_QVhVh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqh_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1315,7 +1314,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eq_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw)
+#define Q6_Q_vcmp_eq_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1326,7 +1325,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqand_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw_and)
+#define Q6_Q_vcmp_eqand_QVwVw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1337,7 +1336,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqor_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw_or)
+#define Q6_Q_vcmp_eqor_QVwVw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1348,7 +1347,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_eqxacc_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw_xor)
+#define Q6_Q_vcmp_eqxacc_QVwVw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_veqw_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1359,7 +1358,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gt_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb)
+#define Q6_Q_vcmp_gt_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1370,7 +1369,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtand_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb_and)
+#define Q6_Q_vcmp_gtand_QVbVb(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1381,7 +1380,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtor_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb_or)
+#define Q6_Q_vcmp_gtor_QVbVb(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1392,7 +1391,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtxacc_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb_xor)
+#define Q6_Q_vcmp_gtxacc_QVbVb(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtb_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1403,7 +1402,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gt_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth)
+#define Q6_Q_vcmp_gt_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1414,7 +1413,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtand_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth_and)
+#define Q6_Q_vcmp_gtand_QVhVh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1425,7 +1424,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtor_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth_or)
+#define Q6_Q_vcmp_gtor_QVhVh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1436,7 +1435,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtxacc_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth_xor)
+#define Q6_Q_vcmp_gtxacc_QVhVh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgth_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1447,7 +1446,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gt_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub)
+#define Q6_Q_vcmp_gt_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1458,7 +1457,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtand_QVubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub_and)
+#define Q6_Q_vcmp_gtand_QVubVub(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1469,7 +1468,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtor_QVubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub_or)
+#define Q6_Q_vcmp_gtor_QVubVub(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1480,7 +1479,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtxacc_QVubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub_xor)
+#define Q6_Q_vcmp_gtxacc_QVubVub(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtub_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1491,7 +1490,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gt_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh)
+#define Q6_Q_vcmp_gt_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1502,7 +1501,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtand_QVuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh_and)
+#define Q6_Q_vcmp_gtand_QVuhVuh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1513,7 +1512,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtor_QVuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh_or)
+#define Q6_Q_vcmp_gtor_QVuhVuh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1524,7 +1523,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtxacc_QVuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh_xor)
+#define Q6_Q_vcmp_gtxacc_QVuhVuh(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuh_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1535,7 +1534,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gt_VuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw)
+#define Q6_Q_vcmp_gt_VuwVuw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1546,7 +1545,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtand_QVuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw_and)
+#define Q6_Q_vcmp_gtand_QVuwVuw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1557,7 +1556,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtor_QVuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw_or)
+#define Q6_Q_vcmp_gtor_QVuwVuw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1568,7 +1567,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtxacc_QVuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw_xor)
+#define Q6_Q_vcmp_gtxacc_QVuwVuw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtuw_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1579,7 +1578,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gt_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw)
+#define Q6_Q_vcmp_gt_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1590,7 +1589,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtand_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw_and)
+#define Q6_Q_vcmp_gtand_QVwVw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1601,7 +1600,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtor_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw_or)
+#define Q6_Q_vcmp_gtor_QVwVw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1612,7 +1611,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vcmp_gtxacc_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw_xor)
+#define Q6_Q_vcmp_gtxacc_QVwVw(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtw_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1623,7 +1622,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vinsert_VwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vinsertwr)
+#define Q6_Vw_vinsert_VwR(Vx,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vinsertwr)(Vx,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1634,7 +1633,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vlalign_VVR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlalignb)
+#define Q6_V_vlalign_VVR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlalignb)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1645,7 +1644,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vlalign_VVI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlalignbi)
+#define Q6_V_vlalign_VVI(Vu,Vv,Iu3) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlalignbi)(Vu,Vv,Iu3)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1656,7 +1655,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vlsr_VuhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrh)
+#define Q6_Vuh_vlsr_VuhR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrh)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1667,7 +1666,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vlsr_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrhv)
+#define Q6_Vh_vlsr_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrhv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1678,7 +1677,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vlsr_VuwR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrw)
+#define Q6_Vuw_vlsr_VuwR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrw)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1689,7 +1688,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vlsr_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrwv)
+#define Q6_Vw_vlsr_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrwv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1700,7 +1699,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vlut32_VbVbR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb)
+#define Q6_Vb_vlut32_VbVbR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1711,7 +1710,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vlut32or_VbVbVbR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb_oracc)
+#define Q6_Vb_vlut32or_VbVbVbR(Vx,Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb_oracc)(Vx,Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1722,7 +1721,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vlut16_VbVhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh)
+#define Q6_Wh_vlut16_VbVhR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1733,7 +1732,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vlut16or_WhVbVhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh_oracc)
+#define Q6_Wh_vlut16or_WhVbVhR(Vxx,Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh_oracc)(Vxx,Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1744,7 +1743,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vmax_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxh)
+#define Q6_Vh_vmax_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1755,7 +1754,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vmax_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxub)
+#define Q6_Vub_vmax_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1766,7 +1765,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vmax_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxuh)
+#define Q6_Vuh_vmax_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1777,7 +1776,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vmax_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxw)
+#define Q6_Vw_vmax_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1788,7 +1787,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vmin_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminh)
+#define Q6_Vh_vmin_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1799,7 +1798,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vmin_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminub)
+#define Q6_Vub_vmin_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1810,7 +1809,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vmin_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminuh)
+#define Q6_Vuh_vmin_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1821,7 +1820,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vmin_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminw)
+#define Q6_Vw_vmin_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1832,7 +1831,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpa_WubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabus)
+#define Q6_Wh_vmpa_WubRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabus)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1843,7 +1842,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpaacc_WhWubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabus_acc)
+#define Q6_Wh_vmpaacc_WhWubRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabus_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1854,7 +1853,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpa_WubWb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabusv)
+#define Q6_Wh_vmpa_WubWb(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabusv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1865,7 +1864,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpa_WubWub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabuuv)
+#define Q6_Wh_vmpa_WubWub(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabuuv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1876,7 +1875,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpa_WhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpahb)
+#define Q6_Ww_vmpa_WhRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpahb)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1887,7 +1886,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpaacc_WwWhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpahb_acc)
+#define Q6_Ww_vmpaacc_WwWhRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpahb_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1898,7 +1897,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpy_VubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybus)
+#define Q6_Wh_vmpy_VubRb(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybus)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1909,7 +1908,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpyacc_WhVubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybus_acc)
+#define Q6_Wh_vmpyacc_WhVubRb(Vxx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybus_acc)(Vxx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1920,7 +1919,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpy_VubVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybusv)
+#define Q6_Wh_vmpy_VubVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybusv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1931,7 +1930,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpyacc_WhVubVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybusv_acc)
+#define Q6_Wh_vmpyacc_WhVubVb(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybusv_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1942,7 +1941,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpy_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybv)
+#define Q6_Wh_vmpy_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1953,7 +1952,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpyacc_WhVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybv_acc)
+#define Q6_Wh_vmpyacc_WhVbVb(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpybv_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1964,7 +1963,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpye_VwVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyewuh)
+#define Q6_Vw_vmpye_VwVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyewuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1975,7 +1974,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpy_VhRh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyh)
+#define Q6_Ww_vmpy_VhRh(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyh)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -1986,29 +1985,29 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpyacc_WwVhRh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhsat_acc)
+#define Q6_Ww_vmpyacc_WwVhRh_sat(Vxx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhsat_acc)(Vxx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vd32.h=vmpy(Vu32.h,Rt32.h):<<1:rnd:sat
    C Intrinsic Prototype: HVX_Vector Q6_Vh_vmpy_VhRh_s1_rnd_sat(HVX_Vector Vu, Word32 Rt)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpy_VhRh_s1_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhsrs)
+#define Q6_Vh_vmpy_VhRh_s1_rnd_sat(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhsrs)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vd32.h=vmpy(Vu32.h,Rt32.h):<<1:sat
    C Intrinsic Prototype: HVX_Vector Q6_Vh_vmpy_VhRh_s1_sat(HVX_Vector Vu, Word32 Rt)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpy_VhRh_s1_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhss)
+#define Q6_Vh_vmpy_VhRh_s1_sat(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhss)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2019,7 +2018,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpy_VhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhus)
+#define Q6_Ww_vmpy_VhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhus)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2030,7 +2029,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpyacc_WwVhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhus_acc)
+#define Q6_Ww_vmpyacc_WwVhVuh(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhus_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2041,7 +2040,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpy_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhv)
+#define Q6_Ww_vmpy_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2052,18 +2051,18 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpyacc_WwVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhv_acc)
+#define Q6_Ww_vmpyacc_WwVhVh(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhv_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vd32.h=vmpy(Vu32.h,Vv32.h):<<1:rnd:sat
    C Intrinsic Prototype: HVX_Vector Q6_Vh_vmpy_VhVh_s1_rnd_sat(HVX_Vector Vu, HVX_Vector Vv)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpy_VhVh_s1_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhvsrs)
+#define Q6_Vh_vmpy_VhVh_s1_rnd_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyhvsrs)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2074,7 +2073,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyieo_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyieoh)
+#define Q6_Vw_vmpyieo_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyieoh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2085,7 +2084,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyieacc_VwVwVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiewh_acc)
+#define Q6_Vw_vmpyieacc_VwVwVh(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiewh_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2096,7 +2095,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyie_VwVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiewuh)
+#define Q6_Vw_vmpyie_VwVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiewuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2107,7 +2106,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyieacc_VwVwVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiewuh_acc)
+#define Q6_Vw_vmpyieacc_VwVwVuh(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiewuh_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2118,7 +2117,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpyi_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyih)
+#define Q6_Vh_vmpyi_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyih)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2129,7 +2128,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpyiacc_VhVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyih_acc)
+#define Q6_Vh_vmpyiacc_VhVhVh(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyih_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2140,7 +2139,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpyi_VhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyihb)
+#define Q6_Vh_vmpyi_VhRb(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyihb)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2151,7 +2150,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vmpyiacc_VhVhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyihb_acc)
+#define Q6_Vh_vmpyiacc_VhVhRb(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyihb_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2162,7 +2161,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyio_VwVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiowh)
+#define Q6_Vw_vmpyio_VwVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiowh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2173,7 +2172,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyi_VwRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwb)
+#define Q6_Vw_vmpyi_VwRb(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwb)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2184,7 +2183,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyiacc_VwVwRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwb_acc)
+#define Q6_Vw_vmpyiacc_VwVwRb(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwb_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2195,7 +2194,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyi_VwRh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwh)
+#define Q6_Vw_vmpyi_VwRh(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwh)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2206,7 +2205,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyiacc_VwVwRh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwh_acc)
+#define Q6_Vw_vmpyiacc_VwVwRh(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwh_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2217,7 +2216,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyo_VwVh_s1_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh)
+#define Q6_Vw_vmpyo_VwVh_s1_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2228,7 +2227,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyo_VwVh_s1_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_rnd)
+#define Q6_Vw_vmpyo_VwVh_s1_rnd_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_rnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2239,7 +2238,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyoacc_VwVwVh_s1_rnd_sat_shift __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_rnd_sacc)
+#define Q6_Vw_vmpyoacc_VwVwVh_s1_rnd_sat_shift(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_rnd_sacc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2250,7 +2249,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyoacc_VwVwVh_s1_sat_shift __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_sacc)
+#define Q6_Vw_vmpyoacc_VwVwVh_s1_sat_shift(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_sacc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2261,7 +2260,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuh_vmpy_VubRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyub)
+#define Q6_Wuh_vmpy_VubRub(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyub)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2272,7 +2271,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuh_vmpyacc_WuhVubRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyub_acc)
+#define Q6_Wuh_vmpyacc_WuhVubRub(Vxx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyub_acc)(Vxx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2283,7 +2282,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuh_vmpy_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyubv)
+#define Q6_Wuh_vmpy_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyubv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2294,7 +2293,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuh_vmpyacc_WuhVubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyubv_acc)
+#define Q6_Wuh_vmpyacc_WuhVubVub(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyubv_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2305,7 +2304,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vmpy_VuhRuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuh)
+#define Q6_Wuw_vmpy_VuhRuh(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuh)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2316,7 +2315,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vmpyacc_WuwVuhRuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuh_acc)
+#define Q6_Wuw_vmpyacc_WuwVuhRuh(Vxx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuh_acc)(Vxx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2327,7 +2326,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vmpy_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhv)
+#define Q6_Wuw_vmpy_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2338,7 +2337,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vmpyacc_WuwVuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhv_acc)
+#define Q6_Wuw_vmpyacc_WuwVuhVuh(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhv_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2349,7 +2348,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vmux_QVV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmux)
+#define Q6_V_vmux_QVV(Qt,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmux)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1),Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2360,7 +2359,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vnavg_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgh)
+#define Q6_Vh_vnavg_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2371,7 +2370,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vnavg_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgub)
+#define Q6_Vb_vnavg_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2382,7 +2381,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vnavg_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgw)
+#define Q6_Vw_vnavg_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2393,7 +2392,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vnormamt_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnormamth)
+#define Q6_Vh_vnormamt_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnormamth)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2404,7 +2403,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vnormamt_Vw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnormamtw)
+#define Q6_Vw_vnormamt_Vw(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnormamtw)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2415,7 +2414,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vnot_V __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnot)
+#define Q6_V_vnot_V(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnot)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2426,7 +2425,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vor_VV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vor)
+#define Q6_V_vor_VV(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vor)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2437,7 +2436,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vpacke_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackeb)
+#define Q6_Vb_vpacke_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackeb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2448,7 +2447,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vpacke_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackeh)
+#define Q6_Vh_vpacke_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackeh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2459,7 +2458,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vpack_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackhb_sat)
+#define Q6_Vb_vpack_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackhb_sat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2470,7 +2469,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vpack_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackhub_sat)
+#define Q6_Vub_vpack_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackhub_sat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2481,7 +2480,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vpacko_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackob)
+#define Q6_Vb_vpacko_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackob)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2492,7 +2491,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vpacko_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackoh)
+#define Q6_Vh_vpacko_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackoh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2503,7 +2502,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vpack_VwVw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackwh_sat)
+#define Q6_Vh_vpack_VwVw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackwh_sat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2514,7 +2513,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vpack_VwVw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackwuh_sat)
+#define Q6_Vuh_vpack_VwVw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpackwuh_sat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2525,7 +2524,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vpopcount_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpopcounth)
+#define Q6_Vh_vpopcount_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vpopcounth)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2536,7 +2535,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vrdelta_VV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrdelta)
+#define Q6_V_vrdelta_VV(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrdelta)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2547,7 +2546,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vrmpy_VubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybus)
+#define Q6_Vw_vrmpy_VubRb(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybus)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2558,7 +2557,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vrmpyacc_VwVubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybus_acc)
+#define Q6_Vw_vrmpyacc_VwVubRb(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybus_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2569,7 +2568,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vrmpy_WubRbI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusi)
+#define Q6_Ww_vrmpy_WubRbI(Vuu,Rt,Iu1) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusi)(Vuu,Rt,Iu1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2580,7 +2579,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vrmpyacc_WwWubRbI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusi_acc)
+#define Q6_Ww_vrmpyacc_WwWubRbI(Vxx,Vuu,Rt,Iu1) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusi_acc)(Vxx,Vuu,Rt,Iu1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2591,18 +2590,18 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vrmpy_VubVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusv)
+#define Q6_Vw_vrmpy_VubVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vx32.w+=vrmpy(Vu32.ub,Vv32.b)
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vrmpyacc_VwVubVb(HVX_Vector Vx, HVX_Vector Vu, HVX_Vector Vv)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vrmpyacc_VwVubVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusv_acc)
+#define Q6_Vw_vrmpyacc_VwVubVb(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybusv_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2613,18 +2612,18 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vrmpy_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybv)
+#define Q6_Vw_vrmpy_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vx32.w+=vrmpy(Vu32.b,Vv32.b)
    C Intrinsic Prototype: HVX_Vector Q6_Vw_vrmpyacc_VwVbVb(HVX_Vector Vx, HVX_Vector Vu, HVX_Vector Vv)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vrmpyacc_VwVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybv_acc)
+#define Q6_Vw_vrmpyacc_VwVbVb(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpybv_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2635,7 +2634,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vrmpy_VubRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyub)
+#define Q6_Vuw_vrmpy_VubRub(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyub)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2646,7 +2645,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vrmpyacc_VuwVubRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyub_acc)
+#define Q6_Vuw_vrmpyacc_VuwVubRub(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyub_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2657,7 +2656,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vrmpy_WubRubI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubi)
+#define Q6_Wuw_vrmpy_WubRubI(Vuu,Rt,Iu1) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubi)(Vuu,Rt,Iu1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2668,7 +2667,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vrmpyacc_WuwWubRubI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubi_acc)
+#define Q6_Wuw_vrmpyacc_WuwWubRubI(Vxx,Vuu,Rt,Iu1) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubi_acc)(Vxx,Vuu,Rt,Iu1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2679,18 +2678,18 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vrmpy_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubv)
+#define Q6_Vuw_vrmpy_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubv)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
 /* ==========================================================================
    Assembly Syntax:       Vx32.uw+=vrmpy(Vu32.ub,Vv32.ub)
    C Intrinsic Prototype: HVX_Vector Q6_Vuw_vrmpyacc_VuwVubVub(HVX_Vector Vx, HVX_Vector Vu, HVX_Vector Vv)
-   Instruction Type:      CVI_VX_DV
+   Instruction Type:      CVI_VX
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vrmpyacc_VuwVubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubv_acc)
+#define Q6_Vuw_vrmpyacc_VuwVubVub(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrmpyubv_acc)(Vx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2701,7 +2700,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vror_VR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vror)
+#define Q6_V_vror_VR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vror)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2712,7 +2711,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vround_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundhb)
+#define Q6_Vb_vround_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundhb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2723,7 +2722,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vround_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundhub)
+#define Q6_Vub_vround_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundhub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2734,7 +2733,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vround_VwVw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundwh)
+#define Q6_Vh_vround_VwVw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundwh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2745,7 +2744,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vround_VwVw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundwuh)
+#define Q6_Vuh_vround_VwVw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vroundwuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2756,7 +2755,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vrsad_WubRubI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrsadubi)
+#define Q6_Wuw_vrsad_WubRubI(Vuu,Rt,Iu1) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrsadubi)(Vuu,Rt,Iu1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2767,7 +2766,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wuw_vrsadacc_WuwWubRubI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrsadubi_acc)
+#define Q6_Wuw_vrsadacc_WuwWubRubI(Vxx,Vuu,Rt,Iu1) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrsadubi_acc)(Vxx,Vuu,Rt,Iu1)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2778,7 +2777,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vsat_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsathub)
+#define Q6_Vub_vsat_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsathub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2789,7 +2788,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vsat_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsatwh)
+#define Q6_Vh_vsat_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsatwh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2800,7 +2799,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vsxt_Vb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsb)
+#define Q6_Wh_vsxt_Vb(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsb)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2811,7 +2810,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vsxt_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsh)
+#define Q6_Ww_vsxt_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2822,7 +2821,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vshuffe_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufeh)
+#define Q6_Vh_vshuffe_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufeh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2833,7 +2832,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vshuff_Vb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffb)
+#define Q6_Vb_vshuff_Vb(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffb)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2844,7 +2843,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vshuffe_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffeb)
+#define Q6_Vb_vshuffe_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffeb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2855,7 +2854,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vshuff_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffh)
+#define Q6_Vh_vshuff_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2866,7 +2865,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vshuffo_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffob)
+#define Q6_Vb_vshuffo_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffob)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2877,7 +2876,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_W_vshuff_VVR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffvdd)
+#define Q6_W_vshuff_VVR(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshuffvdd)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2888,7 +2887,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wb_vshuffoe_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufoeb)
+#define Q6_Wb_vshuffoe_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufoeb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2899,7 +2898,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vshuffoe_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufoeh)
+#define Q6_Wh_vshuffoe_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufoeh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2910,7 +2909,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vshuffo_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufoh)
+#define Q6_Vh_vshuffo_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vshufoh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2921,7 +2920,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vsub_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubb)
+#define Q6_Vb_vsub_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2932,7 +2931,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wb_vsub_WbWb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubb_dv)
+#define Q6_Wb_vsub_WbWb(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubb_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2943,7 +2942,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_condnac_QnVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbnq)
+#define Q6_Vb_condnac_QnVbVb(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbnq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2954,7 +2953,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_condnac_QVbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbq)
+#define Q6_Vb_condnac_QVbVb(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2965,7 +2964,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vsub_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubh)
+#define Q6_Vh_vsub_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2976,7 +2975,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vsub_WhWh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubh_dv)
+#define Q6_Wh_vsub_WhWh(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubh_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2987,7 +2986,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_condnac_QnVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhnq)
+#define Q6_Vh_condnac_QnVhVh(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhnq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -2998,7 +2997,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_condnac_QVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhq)
+#define Q6_Vh_condnac_QVhVh(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3009,7 +3008,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vsub_VhVh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhsat)
+#define Q6_Vh_vsub_VhVh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3020,7 +3019,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vsub_WhWh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhsat_dv)
+#define Q6_Wh_vsub_WhWh_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3031,7 +3030,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vsub_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhw)
+#define Q6_Ww_vsub_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubhw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3042,7 +3041,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vsub_VubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsububh)
+#define Q6_Wh_vsub_VubVub(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsububh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3053,7 +3052,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vsub_VubVub_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsububsat)
+#define Q6_Vub_vsub_VubVub_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsububsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3064,7 +3063,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wub_vsub_WubWub_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsububsat_dv)
+#define Q6_Wub_vsub_WubWub_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsububsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3075,7 +3074,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vsub_VuhVuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuhsat)
+#define Q6_Vuh_vsub_VuhVuh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuhsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3086,7 +3085,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuh_vsub_WuhWuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuhsat_dv)
+#define Q6_Wuh_vsub_WuhWuh_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuhsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3097,7 +3096,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vsub_VuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuhw)
+#define Q6_Ww_vsub_VuhVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuhw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3108,7 +3107,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vsub_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubw)
+#define Q6_Vw_vsub_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3119,7 +3118,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vsub_WwWw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubw_dv)
+#define Q6_Ww_vsub_WwWw(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubw_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3130,7 +3129,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_condnac_QnVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwnq)
+#define Q6_Vw_condnac_QnVwVw(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwnq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3141,7 +3140,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_condnac_QVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwq)
+#define Q6_Vw_condnac_QVwVw(Qv,Vx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3152,7 +3151,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vsub_VwVw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwsat)
+#define Q6_Vw_vsub_VwVw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3163,7 +3162,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vsub_WwWw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwsat_dv)
+#define Q6_Ww_vsub_WwWw_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubwsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3174,7 +3173,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_W_vswap_QVV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vswap)
+#define Q6_W_vswap_QVV(Qt,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vswap)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1),Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3185,7 +3184,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vtmpy_WbRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyb)
+#define Q6_Wh_vtmpy_WbRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyb)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3196,7 +3195,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vtmpyacc_WhWbRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyb_acc)
+#define Q6_Wh_vtmpyacc_WhWbRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyb_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3207,7 +3206,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vtmpy_WubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpybus)
+#define Q6_Wh_vtmpy_WubRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpybus)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3218,7 +3217,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vtmpyacc_WhWubRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpybus_acc)
+#define Q6_Wh_vtmpyacc_WhWubRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpybus_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3229,7 +3228,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vtmpy_WhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyhb)
+#define Q6_Ww_vtmpy_WhRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyhb)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3240,7 +3239,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vtmpyacc_WwWhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyhb_acc)
+#define Q6_Ww_vtmpyacc_WwWhRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vtmpyhb_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3251,7 +3250,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vunpack_Vb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackb)
+#define Q6_Wh_vunpack_Vb(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackb)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3262,7 +3261,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vunpack_Vh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackh)
+#define Q6_Ww_vunpack_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3273,7 +3272,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vunpackoor_WhVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackob)
+#define Q6_Wh_vunpackoor_WhVb(Vxx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackob)(Vxx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3284,7 +3283,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vunpackoor_WwVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackoh)
+#define Q6_Ww_vunpackoor_WwVh(Vxx,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackoh)(Vxx,Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3295,7 +3294,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuh_vunpack_Vub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackub)
+#define Q6_Wuh_vunpack_Vub(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackub)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3306,7 +3305,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuw_vunpack_Vuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackuh)
+#define Q6_Wuw_vunpack_Vuh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vunpackuh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3317,7 +3316,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vxor_VV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vxor)
+#define Q6_V_vxor_VV(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vxor)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3328,7 +3327,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuh_vzxt_Vub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vzb)
+#define Q6_Wuh_vzxt_Vub(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vzb)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 60
@@ -3339,7 +3338,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuw_vzxt_Vuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vzh)
+#define Q6_Wuw_vzxt_Vuh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vzh)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 60 */
 
 #if __HVX_ARCH__ >= 62
@@ -3350,7 +3349,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vb_vsplat_R __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lvsplatb)
+#define Q6_Vb_vsplat_R(Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lvsplatb)(Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3361,7 +3360,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vh_vsplat_R __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lvsplath)
+#define Q6_Vh_vsplat_R(Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_lvsplath)(Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3372,7 +3371,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Q_vsetq2_R __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_scalar2v2)
+#define Q6_Q_vsetq2_R(Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_pred_scalar2v2)(Rt)),-1)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3383,7 +3382,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Qb_vshuffe_QhQh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_shuffeqh)
+#define Q6_Qb_vshuffe_QhQh(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_shuffeqh)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3394,7 +3393,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Qh_vshuffe_QwQw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_shuffeqw)
+#define Q6_Qh_vshuffe_QwQw(Qs,Qt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_shuffeqw)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qt),-1))),-1)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3405,7 +3404,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vadd_VbVb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbsat)
+#define Q6_Vb_vadd_VbVb_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3416,7 +3415,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wb_vadd_WbWb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbsat_dv)
+#define Q6_Wb_vadd_WbWb_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddbsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3427,7 +3426,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vadd_VwVwQ_carry __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddcarry)
+#define Q6_Vw_vadd_VwVwQ_carry(Vu,Vv,Qx) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddcarry)(Vu,Vv,Qx)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3438,7 +3437,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vadd_vclb_VhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddclbh)
+#define Q6_Vh_vadd_vclb_VhVh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddclbh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3449,7 +3448,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vadd_vclb_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddclbw)
+#define Q6_Vw_vadd_vclb_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddclbw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3460,7 +3459,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vaddacc_WwVhVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhw_acc)
+#define Q6_Ww_vaddacc_WwVhVh(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddhw_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3471,7 +3470,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vaddacc_WhVubVub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubh_acc)
+#define Q6_Wh_vaddacc_WhVubVub(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddubh_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3482,7 +3481,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vadd_VubVb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddububb_sat)
+#define Q6_Vub_vadd_VubVb_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddububb_sat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3493,7 +3492,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vaddacc_WwVuhVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhw_acc)
+#define Q6_Ww_vaddacc_WwVuhVuh(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduhw_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3504,7 +3503,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vadd_VuwVuw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduwsat)
+#define Q6_Vuw_vadd_VuwVuw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduwsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3515,7 +3514,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuw_vadd_WuwWuw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduwsat_dv)
+#define Q6_Wuw_vadd_WuwWuw_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadduwsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3526,7 +3525,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_V_vand_QnR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandnqrt)
+#define Q6_V_vand_QnR(Qu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandnqrt)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qu),-1),Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3537,7 +3536,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_V_vandor_VQnR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandnqrt_acc)
+#define Q6_V_vandor_VQnR(Vx,Qu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandnqrt_acc)(Vx,__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qu),-1),Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3548,7 +3547,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vand_QnV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvnqv)
+#define Q6_V_vand_QnV(Qv,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvnqv)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vu)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3559,7 +3558,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_V_vand_QV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvqv)
+#define Q6_V_vand_QV(Qv,Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvqv)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1),Vu)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3570,7 +3569,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vasr_VhVhR_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhbsat)
+#define Q6_Vb_vasr_VhVhR_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrhbsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3581,7 +3580,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vasr_VuwVuwR_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruwuhrndsat)
+#define Q6_Vuh_vasr_VuwVuwR_rnd_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruwuhrndsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3592,7 +3591,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vasr_VwVwR_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwuhrndsat)
+#define Q6_Vuh_vasr_VwVwR_rnd_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrwuhrndsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3603,7 +3602,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vlsr_VubR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrb)
+#define Q6_Vub_vlsr_VubR(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlsrb)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3614,7 +3613,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vlut32_VbVbR_nomatch __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb_nm)
+#define Q6_Vb_vlut32_VbVbR_nomatch(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb_nm)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3625,7 +3624,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vlut32or_VbVbVbI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb_oracci)
+#define Q6_Vb_vlut32or_VbVbVbI(Vx,Vu,Vv,Iu3) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvb_oracci)(Vx,Vu,Vv,Iu3)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3636,7 +3635,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vlut32_VbVbI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvbi)
+#define Q6_Vb_vlut32_VbVbI(Vu,Vv,Iu3) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvvbi)(Vu,Vv,Iu3)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3647,7 +3646,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vlut16_VbVhR_nomatch __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh_nm)
+#define Q6_Wh_vlut16_VbVhR_nomatch(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh_nm)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3658,7 +3657,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vlut16or_WhVbVhI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh_oracci)
+#define Q6_Wh_vlut16or_WhVbVhI(Vxx,Vu,Vv,Iu3) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwh_oracci)(Vxx,Vu,Vv,Iu3)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3669,7 +3668,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wh_vlut16_VbVhI __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwhi)
+#define Q6_Wh_vlut16_VbVhI(Vu,Vv,Iu3) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlutvwhi)(Vu,Vv,Iu3)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3680,7 +3679,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vmax_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxb)
+#define Q6_Vb_vmax_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmaxb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3691,7 +3690,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vmin_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminb)
+#define Q6_Vb_vmin_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vminb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3702,7 +3701,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpa_WuhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpauhb)
+#define Q6_Ww_vmpa_WuhRb(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpauhb)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3713,7 +3712,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpaacc_WwWuhRb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpauhb_acc)
+#define Q6_Ww_vmpaacc_WwWuhRb(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpauhb_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3724,7 +3723,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_W_vmpye_VwVuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyewuh_64)
+#define Q6_W_vmpye_VwVuh(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyewuh_64)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3735,7 +3734,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyi_VwRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwub)
+#define Q6_Vw_vmpyi_VwRub(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwub)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3746,7 +3745,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vw_vmpyiacc_VwVwRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwub_acc)
+#define Q6_Vw_vmpyiacc_VwVwRub(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyiwub_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3757,7 +3756,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_W_vmpyoacc_WVwVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_64_acc)
+#define Q6_W_vmpyoacc_WVwVh(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyowh_64_acc)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3768,7 +3767,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vround_VuhVuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrounduhub)
+#define Q6_Vub_vround_VuhVuh_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrounduhub)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3779,7 +3778,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vround_VuwVuw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrounduwuh)
+#define Q6_Vuh_vround_VuwVuw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrounduwuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3790,7 +3789,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vsat_VuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsatuwuh)
+#define Q6_Vuh_vsat_VuwVuw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsatuwuh)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3801,7 +3800,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vsub_VbVb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbsat)
+#define Q6_Vb_vsub_VbVb_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3812,7 +3811,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wb_vsub_WbWb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbsat_dv)
+#define Q6_Wb_vsub_WbWb_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubbsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3823,7 +3822,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vsub_VwVwQ_carry __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubcarry)
+#define Q6_Vw_vsub_VwVwQ_carry(Vu,Vv,Qx) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubcarry)(Vu,Vv,Qx)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3834,7 +3833,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vsub_VubVb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubububb_sat)
+#define Q6_Vub_vsub_VubVb_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubububb_sat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3845,7 +3844,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vsub_VuwVuw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuwsat)
+#define Q6_Vuw_vsub_VuwVuw_sat(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuwsat)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 62
@@ -3856,7 +3855,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Wuw_vsub_WuwWuw_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuwsat_dv)
+#define Q6_Wuw_vsub_WuwWuw_sat(Vuu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsubuwsat_dv)(Vuu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 62 */
 
 #if __HVX_ARCH__ >= 65
@@ -3867,7 +3866,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vabs_Vb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsb)
+#define Q6_Vb_vabs_Vb(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsb)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3878,7 +3877,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vabs_Vb_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsb_sat)
+#define Q6_Vb_vabs_Vb_sat(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabsb_sat)(Vu)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3889,7 +3888,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vaslacc_VhVhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslh_acc)
+#define Q6_Vh_vaslacc_VhVhR(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaslh_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3900,7 +3899,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_vasracc_VhVhR __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrh_acc)
+#define Q6_Vh_vasracc_VhVhR(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrh_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3911,7 +3910,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vasr_VuhVuhR_rnd_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruhubrndsat)
+#define Q6_Vub_vasr_VuhVuhR_rnd_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruhubrndsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3922,7 +3921,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vub_vasr_VuhVuhR_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruhubsat)
+#define Q6_Vub_vasr_VuhVuhR_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruhubsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3933,7 +3932,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuh_vasr_VuwVuwR_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruwuhsat)
+#define Q6_Vuh_vasr_VuwVuwR_sat(Vu,Vv,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasruwuhsat)(Vu,Vv,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3944,7 +3943,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vavg_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgb)
+#define Q6_Vb_vavg_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3955,7 +3954,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vavg_VbVb_rnd __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgbrnd)
+#define Q6_Vb_vavg_VbVb_rnd(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavgbrnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3966,7 +3965,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vavg_VuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguw)
+#define Q6_Vuw_vavg_VuwVuw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3977,7 +3976,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vavg_VuwVuw_rnd __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguwrnd)
+#define Q6_Vuw_vavg_VuwVuw_rnd(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vavguwrnd)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3988,7 +3987,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_W_vzero __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdd0)
+#define Q6_W_vzero() __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdd0)()
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -3999,7 +3998,7 @@
    Execution Slots:       SLOT01
    ========================================================================== */
 
-#define Q6_vgather_ARMVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermh)
+#define Q6_vgather_ARMVh(Rs,Rt,Mu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermh)(Rs,Rt,Mu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4010,7 +4009,7 @@
    Execution Slots:       SLOT01
    ========================================================================== */
 
-#define Q6_vgather_AQRMVh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermhq)
+#define Q6_vgather_AQRMVh(Rs,Qs,Rt,Mu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermhq)(Rs,__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),Rt,Mu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4021,7 +4020,7 @@
    Execution Slots:       SLOT01
    ========================================================================== */
 
-#define Q6_vgather_ARMWw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermhw)
+#define Q6_vgather_ARMWw(Rs,Rt,Mu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermhw)(Rs,Rt,Mu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4032,7 +4031,7 @@
    Execution Slots:       SLOT01
    ========================================================================== */
 
-#define Q6_vgather_AQRMWw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermhwq)
+#define Q6_vgather_AQRMWw(Rs,Qs,Rt,Mu,Vvv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermhwq)(Rs,__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),Rt,Mu,Vvv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4043,7 +4042,7 @@
    Execution Slots:       SLOT01
    ========================================================================== */
 
-#define Q6_vgather_ARMVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermw)
+#define Q6_vgather_ARMVw(Rs,Rt,Mu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermw)(Rs,Rt,Mu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4054,7 +4053,7 @@
    Execution Slots:       SLOT01
    ========================================================================== */
 
-#define Q6_vgather_AQRMVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermwq)
+#define Q6_vgather_AQRMVw(Rs,Qs,Rt,Mu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgathermwq)(Rs,__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),Rt,Mu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4065,7 +4064,7 @@
    Execution Slots:       SLOT2
    ========================================================================== */
 
-#define Q6_Vh_vlut4_VuhPh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlut4)
+#define Q6_Vh_vlut4_VuhPh(Vu,Rtt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vlut4)(Vu,Rtt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4076,7 +4075,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpa_WubRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabuu)
+#define Q6_Wh_vmpa_WubRub(Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabuu)(Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4087,7 +4086,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Wh_vmpaacc_WhWubRub __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabuu_acc)
+#define Q6_Wh_vmpaacc_WhWubRub(Vxx,Vuu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpabuu_acc)(Vxx,Vuu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4098,7 +4097,7 @@
    Execution Slots:       SLOT2
    ========================================================================== */
 
-#define Q6_Vh_vmpa_VhVhVhPh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpahhsat)
+#define Q6_Vh_vmpa_VhVhVhPh_sat(Vx,Vu,Rtt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpahhsat)(Vx,Vu,Rtt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4109,7 +4108,7 @@
    Execution Slots:       SLOT2
    ========================================================================== */
 
-#define Q6_Vh_vmpa_VhVhVuhPuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpauhuhsat)
+#define Q6_Vh_vmpa_VhVhVuhPuh_sat(Vx,Vu,Rtt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpauhuhsat)(Vx,Vu,Rtt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4120,7 +4119,7 @@
    Execution Slots:       SLOT2
    ========================================================================== */
 
-#define Q6_Vh_vmps_VhVhVuhPuh_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpsuhuhsat)
+#define Q6_Vh_vmps_VhVhVuhPuh_sat(Vx,Vu,Rtt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpsuhuhsat)(Vx,Vu,Rtt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4131,7 +4130,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_vmpyacc_WwVhRh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyh_acc)
+#define Q6_Ww_vmpyacc_WwVhRh(Vxx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyh_acc)(Vxx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4142,7 +4141,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vmpye_VuhRuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhe)
+#define Q6_Vuw_vmpye_VuhRuh(Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhe)(Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4153,7 +4152,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Vuw_vmpyeacc_VuwVuhRuh __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhe_acc)
+#define Q6_Vuw_vmpyeacc_VuwVuhRuh(Vx,Vu,Rt) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhe_acc)(Vx,Vu,Rt)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4164,7 +4163,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_vnavg_VbVb __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgb)
+#define Q6_Vb_vnavg_VbVb(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vnavgb)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4175,7 +4174,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vb_prefixsum_Q __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vprefixqb)
+#define Q6_Vb_prefixsum_Q(Qv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vprefixqb)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1))
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4186,7 +4185,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vh_prefixsum_Q __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vprefixqh)
+#define Q6_Vh_prefixsum_Q(Qv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vprefixqh)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1))
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4197,7 +4196,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_prefixsum_Q __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vprefixqw)
+#define Q6_Vw_prefixsum_Q(Qv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vprefixqw)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qv),-1))
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4208,7 +4207,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatter_RMVhV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermh)
+#define Q6_vscatter_RMVhV(Rt,Mu,Vv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermh)(Rt,Mu,Vv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4219,7 +4218,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatteracc_RMVhV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermh_add)
+#define Q6_vscatteracc_RMVhV(Rt,Mu,Vv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermh_add)(Rt,Mu,Vv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4230,7 +4229,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatter_QRMVhV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhq)
+#define Q6_vscatter_QRMVhV(Qs,Rt,Mu,Vv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),Rt,Mu,Vv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4241,7 +4240,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatter_RMWwV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhw)
+#define Q6_vscatter_RMWwV(Rt,Mu,Vvv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhw)(Rt,Mu,Vvv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4252,7 +4251,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatteracc_RMWwV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhw_add)
+#define Q6_vscatteracc_RMWwV(Rt,Mu,Vvv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhw_add)(Rt,Mu,Vvv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4263,7 +4262,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatter_QRMWwV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhwq)
+#define Q6_vscatter_QRMWwV(Qs,Rt,Mu,Vvv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermhwq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),Rt,Mu,Vvv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4274,7 +4273,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatter_RMVwV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermw)
+#define Q6_vscatter_RMVwV(Rt,Mu,Vv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermw)(Rt,Mu,Vv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4285,7 +4284,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatteracc_RMVwV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermw_add)
+#define Q6_vscatteracc_RMVwV(Rt,Mu,Vv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermw_add)(Rt,Mu,Vv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 65
@@ -4296,7 +4295,7 @@
    Execution Slots:       SLOT0
    ========================================================================== */
 
-#define Q6_vscatter_QRMVwV __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermwq)
+#define Q6_vscatter_QRMVwV(Qs,Rt,Mu,Vv,Vw) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vscattermwq)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1),Rt,Mu,Vv,Vw)
 #endif /* __HEXAGON_ARCH___ >= 65 */
 
 #if __HVX_ARCH__ >= 66
@@ -4307,7 +4306,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vadd_VwVwQ_carry_sat __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddcarrysat)
+#define Q6_Vw_vadd_VwVwQ_carry_sat(Vu,Vv,Qs) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vaddcarrysat)(Vu,Vv,__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qs),-1))
 #endif /* __HEXAGON_ARCH___ >= 66 */
 
 #if __HVX_ARCH__ >= 66
@@ -4318,7 +4317,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Ww_vasrinto_WwVwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasr_into)
+#define Q6_Ww_vasrinto_WwVwVw(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasr_into)(Vxx,Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 66 */
 
 #if __HVX_ARCH__ >= 66
@@ -4329,7 +4328,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vuw_vrotr_VuwVuw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrotr)
+#define Q6_Vuw_vrotr_VuwVuw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vrotr)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 66 */
 
 #if __HVX_ARCH__ >= 66
@@ -4340,7 +4339,7 @@
    Execution Slots:       SLOT0123
    ========================================================================== */
 
-#define Q6_Vw_vsatdw_VwVw __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsatdw)
+#define Q6_Vw_vsatdw_VwVw(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsatdw)(Vu,Vv)
 #endif /* __HEXAGON_ARCH___ >= 66 */
 
 #if __HVX_ARCH__ >= 68
@@ -4351,7 +4350,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_v6mpy_WubWbI_h __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyhubs10)
+#define Q6_Ww_v6mpy_WubWbI_h(Vuu,Vvv,Iu2) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyhubs10)(Vuu,Vvv,Iu2)
 #endif /* __HEXAGON_ARCH___ >= 68 */
 
 #if __HVX_ARCH__ >= 68
@@ -4362,7 +4361,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_v6mpyacc_WwWubWbI_h __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyhubs10_vxx)
+#define Q6_Ww_v6mpyacc_WwWubWbI_h(Vxx,Vuu,Vvv,Iu2) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyhubs10_vxx)(Vxx,Vuu,Vvv,Iu2)
 #endif /* __HEXAGON_ARCH___ >= 68 */
 
 #if __HVX_ARCH__ >= 68
@@ -4373,7 +4372,7 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_v6mpy_WubWbI_v __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyvubs10)
+#define Q6_Ww_v6mpy_WubWbI_v(Vuu,Vvv,Iu2) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyvubs10)(Vuu,Vvv,Iu2)
 #endif /* __HEXAGON_ARCH___ >= 68 */
 
 #if __HVX_ARCH__ >= 68
@@ -4384,9 +4383,801 @@
    Execution Slots:       SLOT23
    ========================================================================== */
 
-#define Q6_Ww_v6mpyacc_WwWubWbI_v __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyvubs10_vxx)
+#define Q6_Ww_v6mpyacc_WwWubWbI_v(Vxx,Vuu,Vvv,Iu2) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_v6mpyvubs10_vxx)(Vxx,Vuu,Vvv,Iu2)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vabs(Vu32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vabs_Vhf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vabs_Vhf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabs_hf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vabs(Vu32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vabs_Vsf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vabs_Vsf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vabs_sf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vadd(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vadd_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf16_vadd_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vadd(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vadd_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vadd_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_hf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vadd(Vu32.qf16,Vv32.qf16)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vadd_Vqf16Vqf16(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf16_vadd_Vqf16Vqf16(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_qf16)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vadd(Vu32.qf16,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vadd_Vqf16Vhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf16_vadd_Vqf16Vhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_qf16_mix)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vadd(Vu32.qf32,Vv32.qf32)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vadd_Vqf32Vqf32(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf32_vadd_Vqf32Vqf32(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_qf32)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vadd(Vu32.qf32,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vadd_Vqf32Vsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf32_vadd_Vqf32Vsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_qf32_mix)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vadd(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vadd_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf32_vadd_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.sf=vadd(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wsf_vadd_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wsf_vadd_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_sf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vadd(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vadd_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vadd_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vadd_sf_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.w=vfmv(Vu32.w)
+   C Intrinsic Prototype: HVX_Vector Q6_Vw_vfmv_Vw(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vw_vfmv_Vw(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vassign_fp)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=Vu32.qf16
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_equals_Vqf16(HVX_Vector Vu)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vhf_equals_Vqf16(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vconv_hf_qf16)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=Vuu32.qf32
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_equals_Wqf32(HVX_VectorPair Vuu)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vhf_equals_Wqf32(Vuu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vconv_hf_qf32)(Vuu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=Vu32.qf32
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_equals_Vqf32(HVX_Vector Vu)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vsf_equals_Vqf32(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vconv_sf_qf32)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.b=vcvt(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vb_vcvt_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vb_vcvt_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_b_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.h=vcvt(Vu32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vh_vcvt_Vhf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vh_vcvt_Vhf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_h_hf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.hf=vcvt(Vu32.b)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Whf_vcvt_Vb(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Whf_vcvt_Vb(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_hf_b)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vcvt(Vu32.h)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vcvt_Vh(HVX_Vector Vu)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vcvt_Vh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_hf_h)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vcvt(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vcvt_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vcvt_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_hf_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.hf=vcvt(Vu32.ub)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Whf_vcvt_Vub(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Whf_vcvt_Vub(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_hf_ub)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vcvt(Vu32.uh)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vcvt_Vuh(HVX_Vector Vu)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vcvt_Vuh(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_hf_uh)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.sf=vcvt(Vu32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wsf_vcvt_Vhf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wsf_vcvt_Vhf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_sf_hf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.ub=vcvt(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vub_vcvt_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vub_vcvt_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_ub_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.uh=vcvt(Vu32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vuh_vcvt_Vhf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vuh_vcvt_Vhf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vcvt_uh_hf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vdmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vdmpy_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vdmpy_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpy_sf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vx32.sf+=vdmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vdmpyacc_VsfVhfVhf(HVX_Vector Vx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vdmpyacc_VsfVhfVhf(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vdmpy_sf_hf_acc)(Vx,Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vfmax(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vfmax_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vfmax_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vfmax_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vfmax(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vfmax_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vfmax_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vfmax_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vfmin(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vfmin_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vfmin_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vfmin_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vfmin(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vfmin_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vfmin_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vfmin_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vfneg(Vu32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vfneg_Vhf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vfneg_Vhf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vfneg_hf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vfneg(Vu32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vfneg_Vsf(HVX_Vector Vu)
+   Instruction Type:      CVI_VX_LATE
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vfneg_Vsf(Vu) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vfneg_sf)(Vu)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qd4=vcmp.gt(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gt_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gt_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgthf)(Vu,Vv)),-1)
 #endif /* __HEXAGON_ARCH___ >= 68 */
 
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qx4&=vcmp.gt(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gtand_QVhfVhf(HVX_VectorPred Qx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gtand_QVhfVhf(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgthf_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qx4|=vcmp.gt(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gtor_QVhfVhf(HVX_VectorPred Qx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gtor_QVhfVhf(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgthf_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qx4^=vcmp.gt(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gtxacc_QVhfVhf(HVX_VectorPred Qx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gtxacc_QVhfVhf(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgthf_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qd4=vcmp.gt(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gt_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gt_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtsf)(Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qx4&=vcmp.gt(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gtand_QVsfVsf(HVX_VectorPred Qx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gtand_QVsfVsf(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtsf_and)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qx4|=vcmp.gt(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gtor_QVsfVsf(HVX_VectorPred Qx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gtor_QVsfVsf(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtsf_or)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Qx4^=vcmp.gt(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_VectorPred Q6_Q_vcmp_gtxacc_QVsfVsf(HVX_VectorPred Qx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Q_vcmp_gtxacc_QVsfVsf(Qx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandqrt)((__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vgtsf_xor)(__BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vandvrt)((Qx),-1),Vu,Vv)),-1)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vmax(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vmax_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vhf_vmax_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmax_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vmax(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vmax_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vsf_vmax_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmax_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vmin(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vmin_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vhf_vmin_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmin_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vmin(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vmin_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VA
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vsf_vmin_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmin_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vmpy_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vmpy_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_hf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vx32.hf+=vmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vmpyacc_VhfVhfVhf(HVX_Vector Vx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vmpyacc_VhfVhfVhf(Vx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_hf_hf_acc)(Vx,Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vmpy(Vu32.qf16,Vv32.qf16)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vmpy_Vqf16Vqf16(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vqf16_vmpy_Vqf16Vqf16(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf16)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vmpy_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vqf16_vmpy_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf16_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vmpy(Vu32.qf16,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vmpy_Vqf16Vhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vqf16_vmpy_Vqf16Vhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf16_mix_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vmpy(Vu32.qf32,Vv32.qf32)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vmpy_Vqf32Vqf32(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vqf32_vmpy_Vqf32Vqf32(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf32)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.qf32=vmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wqf32_vmpy_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wqf32_vmpy_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf32_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.qf32=vmpy(Vu32.qf16,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wqf32_vmpy_Vqf16Vhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wqf32_vmpy_Vqf16Vhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf32_mix_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.qf32=vmpy(Vu32.qf16,Vv32.qf16)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wqf32_vmpy_Vqf16Vqf16(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wqf32_vmpy_Vqf16Vqf16(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf32_qf16)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vmpy(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vmpy_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vqf32_vmpy_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_qf32_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.sf=vmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wsf_vmpy_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wsf_vmpy_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_sf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vxx32.sf+=vmpy(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wsf_vmpyacc_WsfVhfVhf(HVX_VectorPair Vxx, HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wsf_vmpyacc_WsfVhfVhf(Vxx,Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_sf_hf_acc)(Vxx,Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vmpy(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vmpy_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vmpy_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpy_sf_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vsub(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vsub_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf16_vsub_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.hf=vsub(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vhf_vsub_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vhf_vsub_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_hf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vsub(Vu32.qf16,Vv32.qf16)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vsub_Vqf16Vqf16(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf16_vsub_Vqf16Vqf16(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_qf16)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf16=vsub(Vu32.qf16,Vv32.hf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf16_vsub_Vqf16Vhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf16_vsub_Vqf16Vhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_qf16_mix)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vsub(Vu32.qf32,Vv32.qf32)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vsub_Vqf32Vqf32(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf32_vsub_Vqf32Vqf32(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_qf32)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vsub(Vu32.qf32,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vsub_Vqf32Vsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf32_vsub_Vqf32Vsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_qf32_mix)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.qf32=vsub(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vqf32_vsub_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vqf32_vsub_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vdd32.sf=vsub(Vu32.hf,Vv32.hf)
+   C Intrinsic Prototype: HVX_VectorPair Q6_Wsf_vsub_VhfVhf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX_DV
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Wsf_vsub_VhfVhf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_sf_hf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 68
+/* ==========================================================================
+   Assembly Syntax:       Vd32.sf=vsub(Vu32.sf,Vv32.sf)
+   C Intrinsic Prototype: HVX_Vector Q6_Vsf_vsub_VsfVsf(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vsf_vsub_VsfVsf(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vsub_sf_sf)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 68 */
+
+#if __HVX_ARCH__ >= 69
+/* ==========================================================================
+   Assembly Syntax:       Vd32.ub=vasr(Vuu32.uh,Vv32.ub):rnd:sat
+   C Intrinsic Prototype: HVX_Vector Q6_Vub_vasr_WuhVub_rnd_sat(HVX_VectorPair Vuu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vub_vasr_WuhVub_rnd_sat(Vuu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrvuhubrndsat)(Vuu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 69 */
+
+#if __HVX_ARCH__ >= 69
+/* ==========================================================================
+   Assembly Syntax:       Vd32.ub=vasr(Vuu32.uh,Vv32.ub):sat
+   C Intrinsic Prototype: HVX_Vector Q6_Vub_vasr_WuhVub_sat(HVX_VectorPair Vuu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vub_vasr_WuhVub_sat(Vuu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrvuhubsat)(Vuu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 69 */
+
+#if __HVX_ARCH__ >= 69
+/* ==========================================================================
+   Assembly Syntax:       Vd32.uh=vasr(Vuu32.w,Vv32.uh):rnd:sat
+   C Intrinsic Prototype: HVX_Vector Q6_Vuh_vasr_WwVuh_rnd_sat(HVX_VectorPair Vuu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vuh_vasr_WwVuh_rnd_sat(Vuu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrvwuhrndsat)(Vuu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 69 */
+
+#if __HVX_ARCH__ >= 69
+/* ==========================================================================
+   Assembly Syntax:       Vd32.uh=vasr(Vuu32.w,Vv32.uh):sat
+   C Intrinsic Prototype: HVX_Vector Q6_Vuh_vasr_WwVuh_sat(HVX_VectorPair Vuu, HVX_Vector Vv)
+   Instruction Type:      CVI_VS
+   Execution Slots:       SLOT0123
+   ========================================================================== */
+
+#define Q6_Vuh_vasr_WwVuh_sat(Vuu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vasrvwuhsat)(Vuu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 69 */
+
+#if __HVX_ARCH__ >= 69
+/* ==========================================================================
+   Assembly Syntax:       Vd32.uh=vmpy(Vu32.uh,Vv32.uh):>>16
+   C Intrinsic Prototype: HVX_Vector Q6_Vuh_vmpy_VuhVuh_rs16(HVX_Vector Vu, HVX_Vector Vv)
+   Instruction Type:      CVI_VX
+   Execution Slots:       SLOT23
+   ========================================================================== */
+
+#define Q6_Vuh_vmpy_VuhVuh_rs16(Vu,Vv) __BUILTIN_VECTOR_WRAP(__builtin_HEXAGON_V6_vmpyuhvs)(Vu,Vv)
+#endif /* __HEXAGON_ARCH___ >= 69 */
+
 #endif /* __HVX__ */
 
 #endif
diff --git a/clang/lib/Headers/opencl-c.h b/clang/lib/Headers/opencl-c.h
index 32af848a94c4..77a7a8b9bb3a 100644
--- a/clang/lib/Headers/opencl-c.h
+++ b/clang/lib/Headers/opencl-c.h
@@ -11190,305 +11190,305 @@ half16 __ovld __cnfn select(half16 a, half16 b, ushort16 c);
  * 64-bit aligned if gentype is long, ulong, double.
  */
 
-char2 __ovld vload2(size_t offset, const __constant char *p);
-uchar2 __ovld vload2(size_t offset, const __constant uchar *p);
-short2 __ovld vload2(size_t offset, const __constant short *p);
-ushort2 __ovld vload2(size_t offset, const __constant ushort *p);
-int2 __ovld vload2(size_t offset, const __constant int *p);
-uint2 __ovld vload2(size_t offset, const __constant uint *p);
-long2 __ovld vload2(size_t offset, const __constant long *p);
-ulong2 __ovld vload2(size_t offset, const __constant ulong *p);
-float2 __ovld vload2(size_t offset, const __constant float *p);
-char3 __ovld vload3(size_t offset, const __constant char *p);
-uchar3 __ovld vload3(size_t offset, const __constant uchar *p);
-short3 __ovld vload3(size_t offset, const __constant short *p);
-ushort3 __ovld vload3(size_t offset, const __constant ushort *p);
-int3 __ovld vload3(size_t offset, const __constant int *p);
-uint3 __ovld vload3(size_t offset, const __constant uint *p);
-long3 __ovld vload3(size_t offset, const __constant long *p);
-ulong3 __ovld vload3(size_t offset, const __constant ulong *p);
-float3 __ovld vload3(size_t offset, const __constant float *p);
-char4 __ovld vload4(size_t offset, const __constant char *p);
-uchar4 __ovld vload4(size_t offset, const __constant uchar *p);
-short4 __ovld vload4(size_t offset, const __constant short *p);
-ushort4 __ovld vload4(size_t offset, const __constant ushort *p);
-int4 __ovld vload4(size_t offset, const __constant int *p);
-uint4 __ovld vload4(size_t offset, const __constant uint *p);
-long4 __ovld vload4(size_t offset, const __constant long *p);
-ulong4 __ovld vload4(size_t offset, const __constant ulong *p);
-float4 __ovld vload4(size_t offset, const __constant float *p);
-char8 __ovld vload8(size_t offset, const __constant char *p);
-uchar8 __ovld vload8(size_t offset, const __constant uchar *p);
-short8 __ovld vload8(size_t offset, const __constant short *p);
-ushort8 __ovld vload8(size_t offset, const __constant ushort *p);
-int8 __ovld vload8(size_t offset, const __constant int *p);
-uint8 __ovld vload8(size_t offset, const __constant uint *p);
-long8 __ovld vload8(size_t offset, const __constant long *p);
-ulong8 __ovld vload8(size_t offset, const __constant ulong *p);
-float8 __ovld vload8(size_t offset, const __constant float *p);
-char16 __ovld vload16(size_t offset, const __constant char *p);
-uchar16 __ovld vload16(size_t offset, const __constant uchar *p);
-short16 __ovld vload16(size_t offset, const __constant short *p);
-ushort16 __ovld vload16(size_t offset, const __constant ushort *p);
-int16 __ovld vload16(size_t offset, const __constant int *p);
-uint16 __ovld vload16(size_t offset, const __constant uint *p);
-long16 __ovld vload16(size_t offset, const __constant long *p);
-ulong16 __ovld vload16(size_t offset, const __constant ulong *p);
-float16 __ovld vload16(size_t offset, const __constant float *p);
+char2 __ovld __purefn vload2(size_t offset, const __constant char *p);
+uchar2 __ovld __purefn vload2(size_t offset, const __constant uchar *p);
+short2 __ovld __purefn vload2(size_t offset, const __constant short *p);
+ushort2 __ovld __purefn vload2(size_t offset, const __constant ushort *p);
+int2 __ovld __purefn vload2(size_t offset, const __constant int *p);
+uint2 __ovld __purefn vload2(size_t offset, const __constant uint *p);
+long2 __ovld __purefn vload2(size_t offset, const __constant long *p);
+ulong2 __ovld __purefn vload2(size_t offset, const __constant ulong *p);
+float2 __ovld __purefn vload2(size_t offset, const __constant float *p);
+char3 __ovld __purefn vload3(size_t offset, const __constant char *p);
+uchar3 __ovld __purefn vload3(size_t offset, const __constant uchar *p);
+short3 __ovld __purefn vload3(size_t offset, const __constant short *p);
+ushort3 __ovld __purefn vload3(size_t offset, const __constant ushort *p);
+int3 __ovld __purefn vload3(size_t offset, const __constant int *p);
+uint3 __ovld __purefn vload3(size_t offset, const __constant uint *p);
+long3 __ovld __purefn vload3(size_t offset, const __constant long *p);
+ulong3 __ovld __purefn vload3(size_t offset, const __constant ulong *p);
+float3 __ovld __purefn vload3(size_t offset, const __constant float *p);
+char4 __ovld __purefn vload4(size_t offset, const __constant char *p);
+uchar4 __ovld __purefn vload4(size_t offset, const __constant uchar *p);
+short4 __ovld __purefn vload4(size_t offset, const __constant short *p);
+ushort4 __ovld __purefn vload4(size_t offset, const __constant ushort *p);
+int4 __ovld __purefn vload4(size_t offset, const __constant int *p);
+uint4 __ovld __purefn vload4(size_t offset, const __constant uint *p);
+long4 __ovld __purefn vload4(size_t offset, const __constant long *p);
+ulong4 __ovld __purefn vload4(size_t offset, const __constant ulong *p);
+float4 __ovld __purefn vload4(size_t offset, const __constant float *p);
+char8 __ovld __purefn vload8(size_t offset, const __constant char *p);
+uchar8 __ovld __purefn vload8(size_t offset, const __constant uchar *p);
+short8 __ovld __purefn vload8(size_t offset, const __constant short *p);
+ushort8 __ovld __purefn vload8(size_t offset, const __constant ushort *p);
+int8 __ovld __purefn vload8(size_t offset, const __constant int *p);
+uint8 __ovld __purefn vload8(size_t offset, const __constant uint *p);
+long8 __ovld __purefn vload8(size_t offset, const __constant long *p);
+ulong8 __ovld __purefn vload8(size_t offset, const __constant ulong *p);
+float8 __ovld __purefn vload8(size_t offset, const __constant float *p);
+char16 __ovld __purefn vload16(size_t offset, const __constant char *p);
+uchar16 __ovld __purefn vload16(size_t offset, const __constant uchar *p);
+short16 __ovld __purefn vload16(size_t offset, const __constant short *p);
+ushort16 __ovld __purefn vload16(size_t offset, const __constant ushort *p);
+int16 __ovld __purefn vload16(size_t offset, const __constant int *p);
+uint16 __ovld __purefn vload16(size_t offset, const __constant uint *p);
+long16 __ovld __purefn vload16(size_t offset, const __constant long *p);
+ulong16 __ovld __purefn vload16(size_t offset, const __constant ulong *p);
+float16 __ovld __purefn vload16(size_t offset, const __constant float *p);
 #ifdef cl_khr_fp64
-double2 __ovld vload2(size_t offset, const __constant double *p);
-double3 __ovld vload3(size_t offset, const __constant double *p);
-double4 __ovld vload4(size_t offset, const __constant double *p);
-double8 __ovld vload8(size_t offset, const __constant double *p);
-double16 __ovld vload16(size_t offset, const __constant double *p);
+double2 __ovld __purefn vload2(size_t offset, const __constant double *p);
+double3 __ovld __purefn vload3(size_t offset, const __constant double *p);
+double4 __ovld __purefn vload4(size_t offset, const __constant double *p);
+double8 __ovld __purefn vload8(size_t offset, const __constant double *p);
+double16 __ovld __purefn vload16(size_t offset, const __constant double *p);
 #endif //cl_khr_fp64
 
 #ifdef cl_khr_fp16
-half __ovld vload(size_t offset, const __constant half *p);
-half2 __ovld vload2(size_t offset, const __constant half *p);
-half3 __ovld vload3(size_t offset, const __constant half *p);
-half4 __ovld vload4(size_t offset, const __constant half *p);
-half8 __ovld vload8(size_t offset, const __constant half *p);
-half16 __ovld vload16(size_t offset, const __constant half *p);
+half __ovld __purefn vload(size_t offset, const __constant half *p);
+half2 __ovld __purefn vload2(size_t offset, const __constant half *p);
+half3 __ovld __purefn vload3(size_t offset, const __constant half *p);
+half4 __ovld __purefn vload4(size_t offset, const __constant half *p);
+half8 __ovld __purefn vload8(size_t offset, const __constant half *p);
+half16 __ovld __purefn vload16(size_t offset, const __constant half *p);
 #endif //cl_khr_fp16
 
 #if defined(__opencl_c_generic_address_space)
-char2 __ovld vload2(size_t offset, const char *p);
-uchar2 __ovld vload2(size_t offset, const uchar *p);
-short2 __ovld vload2(size_t offset, const short *p);
-ushort2 __ovld vload2(size_t offset, const ushort *p);
-int2 __ovld vload2(size_t offset, const int *p);
-uint2 __ovld vload2(size_t offset, const uint *p);
-long2 __ovld vload2(size_t offset, const long *p);
-ulong2 __ovld vload2(size_t offset, const ulong *p);
-float2 __ovld vload2(size_t offset, const float *p);
-char3 __ovld vload3(size_t offset, const char *p);
-uchar3 __ovld vload3(size_t offset, const uchar *p);
-short3 __ovld vload3(size_t offset, const short *p);
-ushort3 __ovld vload3(size_t offset, const ushort *p);
-int3 __ovld vload3(size_t offset, const int *p);
-uint3 __ovld vload3(size_t offset, const uint *p);
-long3 __ovld vload3(size_t offset, const long *p);
-ulong3 __ovld vload3(size_t offset, const ulong *p);
-float3 __ovld vload3(size_t offset, const float *p);
-char4 __ovld vload4(size_t offset, const char *p);
-uchar4 __ovld vload4(size_t offset, const uchar *p);
-short4 __ovld vload4(size_t offset, const short *p);
-ushort4 __ovld vload4(size_t offset, const ushort *p);
-int4 __ovld vload4(size_t offset, const int *p);
-uint4 __ovld vload4(size_t offset, const uint *p);
-long4 __ovld vload4(size_t offset, const long *p);
-ulong4 __ovld vload4(size_t offset, const ulong *p);
-float4 __ovld vload4(size_t offset, const float *p);
-char8 __ovld vload8(size_t offset, const char *p);
-uchar8 __ovld vload8(size_t offset, const uchar *p);
-short8 __ovld vload8(size_t offset, const short *p);
-ushort8 __ovld vload8(size_t offset, const ushort *p);
-int8 __ovld vload8(size_t offset, const int *p);
-uint8 __ovld vload8(size_t offset, const uint *p);
-long8 __ovld vload8(size_t offset, const long *p);
-ulong8 __ovld vload8(size_t offset, const ulong *p);
-float8 __ovld vload8(size_t offset, const float *p);
-char16 __ovld vload16(size_t offset, const char *p);
-uchar16 __ovld vload16(size_t offset, const uchar *p);
-short16 __ovld vload16(size_t offset, const short *p);
-ushort16 __ovld vload16(size_t offset, const ushort *p);
-int16 __ovld vload16(size_t offset, const int *p);
-uint16 __ovld vload16(size_t offset, const uint *p);
-long16 __ovld vload16(size_t offset, const long *p);
-ulong16 __ovld vload16(size_t offset, const ulong *p);
-float16 __ovld vload16(size_t offset, const float *p);
+char2 __ovld __purefn vload2(size_t offset, const char *p);
+uchar2 __ovld __purefn vload2(size_t offset, const uchar *p);
+short2 __ovld __purefn vload2(size_t offset, const short *p);
+ushort2 __ovld __purefn vload2(size_t offset, const ushort *p);
+int2 __ovld __purefn vload2(size_t offset, const int *p);
+uint2 __ovld __purefn vload2(size_t offset, const uint *p);
+long2 __ovld __purefn vload2(size_t offset, const long *p);
+ulong2 __ovld __purefn vload2(size_t offset, const ulong *p);
+float2 __ovld __purefn vload2(size_t offset, const float *p);
+char3 __ovld __purefn vload3(size_t offset, const char *p);
+uchar3 __ovld __purefn vload3(size_t offset, const uchar *p);
+short3 __ovld __purefn vload3(size_t offset, const short *p);
+ushort3 __ovld __purefn vload3(size_t offset, const ushort *p);
+int3 __ovld __purefn vload3(size_t offset, const int *p);
+uint3 __ovld __purefn vload3(size_t offset, const uint *p);
+long3 __ovld __purefn vload3(size_t offset, const long *p);
+ulong3 __ovld __purefn vload3(size_t offset, const ulong *p);
+float3 __ovld __purefn vload3(size_t offset, const float *p);
+char4 __ovld __purefn vload4(size_t offset, const char *p);
+uchar4 __ovld __purefn vload4(size_t offset, const uchar *p);
+short4 __ovld __purefn vload4(size_t offset, const short *p);
+ushort4 __ovld __purefn vload4(size_t offset, const ushort *p);
+int4 __ovld __purefn vload4(size_t offset, const int *p);
+uint4 __ovld __purefn vload4(size_t offset, const uint *p);
+long4 __ovld __purefn vload4(size_t offset, const long *p);
+ulong4 __ovld __purefn vload4(size_t offset, const ulong *p);
+float4 __ovld __purefn vload4(size_t offset, const float *p);
+char8 __ovld __purefn vload8(size_t offset, const char *p);
+uchar8 __ovld __purefn vload8(size_t offset, const uchar *p);
+short8 __ovld __purefn vload8(size_t offset, const short *p);
+ushort8 __ovld __purefn vload8(size_t offset, const ushort *p);
+int8 __ovld __purefn vload8(size_t offset, const int *p);
+uint8 __ovld __purefn vload8(size_t offset, const uint *p);
+long8 __ovld __purefn vload8(size_t offset, const long *p);
+ulong8 __ovld __purefn vload8(size_t offset, const ulong *p);
+float8 __ovld __purefn vload8(size_t offset, const float *p);
+char16 __ovld __purefn vload16(size_t offset, const char *p);
+uchar16 __ovld __purefn vload16(size_t offset, const uchar *p);
+short16 __ovld __purefn vload16(size_t offset, const short *p);
+ushort16 __ovld __purefn vload16(size_t offset, const ushort *p);
+int16 __ovld __purefn vload16(size_t offset, const int *p);
+uint16 __ovld __purefn vload16(size_t offset, const uint *p);
+long16 __ovld __purefn vload16(size_t offset, const long *p);
+ulong16 __ovld __purefn vload16(size_t offset, const ulong *p);
+float16 __ovld __purefn vload16(size_t offset, const float *p);
 
 #ifdef cl_khr_fp64
-double2 __ovld vload2(size_t offset, const double *p);
-double3 __ovld vload3(size_t offset, const double *p);
-double4 __ovld vload4(size_t offset, const double *p);
-double8 __ovld vload8(size_t offset, const double *p);
-double16 __ovld vload16(size_t offset, const double *p);
+double2 __ovld __purefn vload2(size_t offset, const double *p);
+double3 __ovld __purefn vload3(size_t offset, const double *p);
+double4 __ovld __purefn vload4(size_t offset, const double *p);
+double8 __ovld __purefn vload8(size_t offset, const double *p);
+double16 __ovld __purefn vload16(size_t offset, const double *p);
 #endif //cl_khr_fp64
 
 #ifdef cl_khr_fp16
-half __ovld vload(size_t offset, const half *p);
-half2 __ovld vload2(size_t offset, const half *p);
-half3 __ovld vload3(size_t offset, const half *p);
-half4 __ovld vload4(size_t offset, const half *p);
-half8 __ovld vload8(size_t offset, const half *p);
-half16 __ovld vload16(size_t offset, const half *p);
+half __ovld __purefn vload(size_t offset, const half *p);
+half2 __ovld __purefn vload2(size_t offset, const half *p);
+half3 __ovld __purefn vload3(size_t offset, const half *p);
+half4 __ovld __purefn vload4(size_t offset, const half *p);
+half8 __ovld __purefn vload8(size_t offset, const half *p);
+half16 __ovld __purefn vload16(size_t offset, const half *p);
 #endif //cl_khr_fp16
 #else
-char2 __ovld vload2(size_t offset, const __global char *p);
-uchar2 __ovld vload2(size_t offset, const __global uchar *p);
-short2 __ovld vload2(size_t offset, const __global short *p);
-ushort2 __ovld vload2(size_t offset, const __global ushort *p);
-int2 __ovld vload2(size_t offset, const __global int *p);
-uint2 __ovld vload2(size_t offset, const __global uint *p);
-long2 __ovld vload2(size_t offset, const __global long *p);
-ulong2 __ovld vload2(size_t offset, const __global ulong *p);
-float2 __ovld vload2(size_t offset, const __global float *p);
-char3 __ovld vload3(size_t offset, const __global char *p);
-uchar3 __ovld vload3(size_t offset, const __global uchar *p);
-short3 __ovld vload3(size_t offset, const __global short *p);
-ushort3 __ovld vload3(size_t offset, const __global ushort *p);
-int3 __ovld vload3(size_t offset, const __global int *p);
-uint3 __ovld vload3(size_t offset, const __global uint *p);
-long3 __ovld vload3(size_t offset, const __global long *p);
-ulong3 __ovld vload3(size_t offset, const __global ulong *p);
-float3 __ovld vload3(size_t offset, const __global float *p);
-char4 __ovld vload4(size_t offset, const __global char *p);
-uchar4 __ovld vload4(size_t offset, const __global uchar *p);
-short4 __ovld vload4(size_t offset, const __global short *p);
-ushort4 __ovld vload4(size_t offset, const __global ushort *p);
-int4 __ovld vload4(size_t offset, const __global int *p);
-uint4 __ovld vload4(size_t offset, const __global uint *p);
-long4 __ovld vload4(size_t offset, const __global long *p);
-ulong4 __ovld vload4(size_t offset, const __global ulong *p);
-float4 __ovld vload4(size_t offset, const __global float *p);
-char8 __ovld vload8(size_t offset, const __global char *p);
-uchar8 __ovld vload8(size_t offset, const __global uchar *p);
-short8 __ovld vload8(size_t offset, const __global short *p);
-ushort8 __ovld vload8(size_t offset, const __global ushort *p);
-int8 __ovld vload8(size_t offset, const __global int *p);
-uint8 __ovld vload8(size_t offset, const __global uint *p);
-long8 __ovld vload8(size_t offset, const __global long *p);
-ulong8 __ovld vload8(size_t offset, const __global ulong *p);
-float8 __ovld vload8(size_t offset, const __global float *p);
-char16 __ovld vload16(size_t offset, const __global char *p);
-uchar16 __ovld vload16(size_t offset, const __global uchar *p);
-short16 __ovld vload16(size_t offset, const __global short *p);
-ushort16 __ovld vload16(size_t offset, const __global ushort *p);
-int16 __ovld vload16(size_t offset, const __global int *p);
-uint16 __ovld vload16(size_t offset, const __global uint *p);
-long16 __ovld vload16(size_t offset, const __global long *p);
-ulong16 __ovld vload16(size_t offset, const __global ulong *p);
-float16 __ovld vload16(size_t offset, const __global float *p);
-char2 __ovld vload2(size_t offset, const __local char *p);
-uchar2 __ovld vload2(size_t offset, const __local uchar *p);
-short2 __ovld vload2(size_t offset, const __local short *p);
-ushort2 __ovld vload2(size_t offset, const __local ushort *p);
-int2 __ovld vload2(size_t offset, const __local int *p);
-uint2 __ovld vload2(size_t offset, const __local uint *p);
-long2 __ovld vload2(size_t offset, const __local long *p);
-ulong2 __ovld vload2(size_t offset, const __local ulong *p);
-float2 __ovld vload2(size_t offset, const __local float *p);
-char3 __ovld vload3(size_t offset, const __local char *p);
-uchar3 __ovld vload3(size_t offset, const __local uchar *p);
-short3 __ovld vload3(size_t offset, const __local short *p);
-ushort3 __ovld vload3(size_t offset, const __local ushort *p);
-int3 __ovld vload3(size_t offset, const __local int *p);
-uint3 __ovld vload3(size_t offset, const __local uint *p);
-long3 __ovld vload3(size_t offset, const __local long *p);
-ulong3 __ovld vload3(size_t offset, const __local ulong *p);
-float3 __ovld vload3(size_t offset, const __local float *p);
-char4 __ovld vload4(size_t offset, const __local char *p);
-uchar4 __ovld vload4(size_t offset, const __local uchar *p);
-short4 __ovld vload4(size_t offset, const __local short *p);
-ushort4 __ovld vload4(size_t offset, const __local ushort *p);
-int4 __ovld vload4(size_t offset, const __local int *p);
-uint4 __ovld vload4(size_t offset, const __local uint *p);
-long4 __ovld vload4(size_t offset, const __local long *p);
-ulong4 __ovld vload4(size_t offset, const __local ulong *p);
-float4 __ovld vload4(size_t offset, const __local float *p);
-char8 __ovld vload8(size_t offset, const __local char *p);
-uchar8 __ovld vload8(size_t offset, const __local uchar *p);
-short8 __ovld vload8(size_t offset, const __local short *p);
-ushort8 __ovld vload8(size_t offset, const __local ushort *p);
-int8 __ovld vload8(size_t offset, const __local int *p);
-uint8 __ovld vload8(size_t offset, const __local uint *p);
-long8 __ovld vload8(size_t offset, const __local long *p);
-ulong8 __ovld vload8(size_t offset, const __local ulong *p);
-float8 __ovld vload8(size_t offset, const __local float *p);
-char16 __ovld vload16(size_t offset, const __local char *p);
-uchar16 __ovld vload16(size_t offset, const __local uchar *p);
-short16 __ovld vload16(size_t offset, const __local short *p);
-ushort16 __ovld vload16(size_t offset, const __local ushort *p);
-int16 __ovld vload16(size_t offset, const __local int *p);
-uint16 __ovld vload16(size_t offset, const __local uint *p);
-long16 __ovld vload16(size_t offset, const __local long *p);
-ulong16 __ovld vload16(size_t offset, const __local ulong *p);
-float16 __ovld vload16(size_t offset, const __local float *p);
-char2 __ovld vload2(size_t offset, const __private char *p);
-uchar2 __ovld vload2(size_t offset, const __private uchar *p);
-short2 __ovld vload2(size_t offset, const __private short *p);
-ushort2 __ovld vload2(size_t offset, const __private ushort *p);
-int2 __ovld vload2(size_t offset, const __private int *p);
-uint2 __ovld vload2(size_t offset, const __private uint *p);
-long2 __ovld vload2(size_t offset, const __private long *p);
-ulong2 __ovld vload2(size_t offset, const __private ulong *p);
-float2 __ovld vload2(size_t offset, const __private float *p);
-char3 __ovld vload3(size_t offset, const __private char *p);
-uchar3 __ovld vload3(size_t offset, const __private uchar *p);
-short3 __ovld vload3(size_t offset, const __private short *p);
-ushort3 __ovld vload3(size_t offset, const __private ushort *p);
-int3 __ovld vload3(size_t offset, const __private int *p);
-uint3 __ovld vload3(size_t offset, const __private uint *p);
-long3 __ovld vload3(size_t offset, const __private long *p);
-ulong3 __ovld vload3(size_t offset, const __private ulong *p);
-float3 __ovld vload3(size_t offset, const __private float *p);
-char4 __ovld vload4(size_t offset, const __private char *p);
-uchar4 __ovld vload4(size_t offset, const __private uchar *p);
-short4 __ovld vload4(size_t offset, const __private short *p);
-ushort4 __ovld vload4(size_t offset, const __private ushort *p);
-int4 __ovld vload4(size_t offset, const __private int *p);
-uint4 __ovld vload4(size_t offset, const __private uint *p);
-long4 __ovld vload4(size_t offset, const __private long *p);
-ulong4 __ovld vload4(size_t offset, const __private ulong *p);
-float4 __ovld vload4(size_t offset, const __private float *p);
-char8 __ovld vload8(size_t offset, const __private char *p);
-uchar8 __ovld vload8(size_t offset, const __private uchar *p);
-short8 __ovld vload8(size_t offset, const __private short *p);
-ushort8 __ovld vload8(size_t offset, const __private ushort *p);
-int8 __ovld vload8(size_t offset, const __private int *p);
-uint8 __ovld vload8(size_t offset, const __private uint *p);
-long8 __ovld vload8(size_t offset, const __private long *p);
-ulong8 __ovld vload8(size_t offset, const __private ulong *p);
-float8 __ovld vload8(size_t offset, const __private float *p);
-char16 __ovld vload16(size_t offset, const __private char *p);
-uchar16 __ovld vload16(size_t offset, const __private uchar *p);
-short16 __ovld vload16(size_t offset, const __private short *p);
-ushort16 __ovld vload16(size_t offset, const __private ushort *p);
-int16 __ovld vload16(size_t offset, const __private int *p);
-uint16 __ovld vload16(size_t offset, const __private uint *p);
-long16 __ovld vload16(size_t offset, const __private long *p);
-ulong16 __ovld vload16(size_t offset, const __private ulong *p);
-float16 __ovld vload16(size_t offset, const __private float *p);
+char2 __ovld __purefn vload2(size_t offset, const __global char *p);
+uchar2 __ovld __purefn vload2(size_t offset, const __global uchar *p);
+short2 __ovld __purefn vload2(size_t offset, const __global short *p);
+ushort2 __ovld __purefn vload2(size_t offset, const __global ushort *p);
+int2 __ovld __purefn vload2(size_t offset, const __global int *p);
+uint2 __ovld __purefn vload2(size_t offset, const __global uint *p);
+long2 __ovld __purefn vload2(size_t offset, const __global long *p);
+ulong2 __ovld __purefn vload2(size_t offset, const __global ulong *p);
+float2 __ovld __purefn vload2(size_t offset, const __global float *p);
+char3 __ovld __purefn vload3(size_t offset, const __global char *p);
+uchar3 __ovld __purefn vload3(size_t offset, const __global uchar *p);
+short3 __ovld __purefn vload3(size_t offset, const __global short *p);
+ushort3 __ovld __purefn vload3(size_t offset, const __global ushort *p);
+int3 __ovld __purefn vload3(size_t offset, const __global int *p);
+uint3 __ovld __purefn vload3(size_t offset, const __global uint *p);
+long3 __ovld __purefn vload3(size_t offset, const __global long *p);
+ulong3 __ovld __purefn vload3(size_t offset, const __global ulong *p);
+float3 __ovld __purefn vload3(size_t offset, const __global float *p);
+char4 __ovld __purefn vload4(size_t offset, const __global char *p);
+uchar4 __ovld __purefn vload4(size_t offset, const __global uchar *p);
+short4 __ovld __purefn vload4(size_t offset, const __global short *p);
+ushort4 __ovld __purefn vload4(size_t offset, const __global ushort *p);
+int4 __ovld __purefn vload4(size_t offset, const __global int *p);
+uint4 __ovld __purefn vload4(size_t offset, const __global uint *p);
+long4 __ovld __purefn vload4(size_t offset, const __global long *p);
+ulong4 __ovld __purefn vload4(size_t offset, const __global ulong *p);
+float4 __ovld __purefn vload4(size_t offset, const __global float *p);
+char8 __ovld __purefn vload8(size_t offset, const __global char *p);
+uchar8 __ovld __purefn vload8(size_t offset, const __global uchar *p);
+short8 __ovld __purefn vload8(size_t offset, const __global short *p);
+ushort8 __ovld __purefn vload8(size_t offset, const __global ushort *p);
+int8 __ovld __purefn vload8(size_t offset, const __global int *p);
+uint8 __ovld __purefn vload8(size_t offset, const __global uint *p);
+long8 __ovld __purefn vload8(size_t offset, const __global long *p);
+ulong8 __ovld __purefn vload8(size_t offset, const __global ulong *p);
+float8 __ovld __purefn vload8(size_t offset, const __global float *p);
+char16 __ovld __purefn vload16(size_t offset, const __global char *p);
+uchar16 __ovld __purefn vload16(size_t offset, const __global uchar *p);
+short16 __ovld __purefn vload16(size_t offset, const __global short *p);
+ushort16 __ovld __purefn vload16(size_t offset, const __global ushort *p);
+int16 __ovld __purefn vload16(size_t offset, const __global int *p);
+uint16 __ovld __purefn vload16(size_t offset, const __global uint *p);
+long16 __ovld __purefn vload16(size_t offset, const __global long *p);
+ulong16 __ovld __purefn vload16(size_t offset, const __global ulong *p);
+float16 __ovld __purefn vload16(size_t offset, const __global float *p);
+char2 __ovld __purefn vload2(size_t offset, const __local char *p);
+uchar2 __ovld __purefn vload2(size_t offset, const __local uchar *p);
+short2 __ovld __purefn vload2(size_t offset, const __local short *p);
+ushort2 __ovld __purefn vload2(size_t offset, const __local ushort *p);
+int2 __ovld __purefn vload2(size_t offset, const __local int *p);
+uint2 __ovld __purefn vload2(size_t offset, const __local uint *p);
+long2 __ovld __purefn vload2(size_t offset, const __local long *p);
+ulong2 __ovld __purefn vload2(size_t offset, const __local ulong *p);
+float2 __ovld __purefn vload2(size_t offset, const __local float *p);
+char3 __ovld __purefn vload3(size_t offset, const __local char *p);
+uchar3 __ovld __purefn vload3(size_t offset, const __local uchar *p);
+short3 __ovld __purefn vload3(size_t offset, const __local short *p);
+ushort3 __ovld __purefn vload3(size_t offset, const __local ushort *p);
+int3 __ovld __purefn vload3(size_t offset, const __local int *p);
+uint3 __ovld __purefn vload3(size_t offset, const __local uint *p);
+long3 __ovld __purefn vload3(size_t offset, const __local long *p);
+ulong3 __ovld __purefn vload3(size_t offset, const __local ulong *p);
+float3 __ovld __purefn vload3(size_t offset, const __local float *p);
+char4 __ovld __purefn vload4(size_t offset, const __local char *p);
+uchar4 __ovld __purefn vload4(size_t offset, const __local uchar *p);
+short4 __ovld __purefn vload4(size_t offset, const __local short *p);
+ushort4 __ovld __purefn vload4(size_t offset, const __local ushort *p);
+int4 __ovld __purefn vload4(size_t offset, const __local int *p);
+uint4 __ovld __purefn vload4(size_t offset, const __local uint *p);
+long4 __ovld __purefn vload4(size_t offset, const __local long *p);
+ulong4 __ovld __purefn vload4(size_t offset, const __local ulong *p);
+float4 __ovld __purefn vload4(size_t offset, const __local float *p);
+char8 __ovld __purefn vload8(size_t offset, const __local char *p);
+uchar8 __ovld __purefn vload8(size_t offset, const __local uchar *p);
+short8 __ovld __purefn vload8(size_t offset, const __local short *p);
+ushort8 __ovld __purefn vload8(size_t offset, const __local ushort *p);
+int8 __ovld __purefn vload8(size_t offset, const __local int *p);
+uint8 __ovld __purefn vload8(size_t offset, const __local uint *p);
+long8 __ovld __purefn vload8(size_t offset, const __local long *p);
+ulong8 __ovld __purefn vload8(size_t offset, const __local ulong *p);
+float8 __ovld __purefn vload8(size_t offset, const __local float *p);
+char16 __ovld __purefn vload16(size_t offset, const __local char *p);
+uchar16 __ovld __purefn vload16(size_t offset, const __local uchar *p);
+short16 __ovld __purefn vload16(size_t offset, const __local short *p);
+ushort16 __ovld __purefn vload16(size_t offset, const __local ushort *p);
+int16 __ovld __purefn vload16(size_t offset, const __local int *p);
+uint16 __ovld __purefn vload16(size_t offset, const __local uint *p);
+long16 __ovld __purefn vload16(size_t offset, const __local long *p);
+ulong16 __ovld __purefn vload16(size_t offset, const __local ulong *p);
+float16 __ovld __purefn vload16(size_t offset, const __local float *p);
+char2 __ovld __purefn vload2(size_t offset, const __private char *p);
+uchar2 __ovld __purefn vload2(size_t offset, const __private uchar *p);
+short2 __ovld __purefn vload2(size_t offset, const __private short *p);
+ushort2 __ovld __purefn vload2(size_t offset, const __private ushort *p);
+int2 __ovld __purefn vload2(size_t offset, const __private int *p);
+uint2 __ovld __purefn vload2(size_t offset, const __private uint *p);
+long2 __ovld __purefn vload2(size_t offset, const __private long *p);
+ulong2 __ovld __purefn vload2(size_t offset, const __private ulong *p);
+float2 __ovld __purefn vload2(size_t offset, const __private float *p);
+char3 __ovld __purefn vload3(size_t offset, const __private char *p);
+uchar3 __ovld __purefn vload3(size_t offset, const __private uchar *p);
+short3 __ovld __purefn vload3(size_t offset, const __private short *p);
+ushort3 __ovld __purefn vload3(size_t offset, const __private ushort *p);
+int3 __ovld __purefn vload3(size_t offset, const __private int *p);
+uint3 __ovld __purefn vload3(size_t offset, const __private uint *p);
+long3 __ovld __purefn vload3(size_t offset, const __private long *p);
+ulong3 __ovld __purefn vload3(size_t offset, const __private ulong *p);
+float3 __ovld __purefn vload3(size_t offset, const __private float *p);
+char4 __ovld __purefn vload4(size_t offset, const __private char *p);
+uchar4 __ovld __purefn vload4(size_t offset, const __private uchar *p);
+short4 __ovld __purefn vload4(size_t offset, const __private short *p);
+ushort4 __ovld __purefn vload4(size_t offset, const __private ushort *p);
+int4 __ovld __purefn vload4(size_t offset, const __private int *p);
+uint4 __ovld __purefn vload4(size_t offset, const __private uint *p);
+long4 __ovld __purefn vload4(size_t offset, const __private long *p);
+ulong4 __ovld __purefn vload4(size_t offset, const __private ulong *p);
+float4 __ovld __purefn vload4(size_t offset, const __private float *p);
+char8 __ovld __purefn vload8(size_t offset, const __private char *p);
+uchar8 __ovld __purefn vload8(size_t offset, const __private uchar *p);
+short8 __ovld __purefn vload8(size_t offset, const __private short *p);
+ushort8 __ovld __purefn vload8(size_t offset, const __private ushort *p);
+int8 __ovld __purefn vload8(size_t offset, const __private int *p);
+uint8 __ovld __purefn vload8(size_t offset, const __private uint *p);
+long8 __ovld __purefn vload8(size_t offset, const __private long *p);
+ulong8 __ovld __purefn vload8(size_t offset, const __private ulong *p);
+float8 __ovld __purefn vload8(size_t offset, const __private float *p);
+char16 __ovld __purefn vload16(size_t offset, const __private char *p);
+uchar16 __ovld __purefn vload16(size_t offset, const __private uchar *p);
+short16 __ovld __purefn vload16(size_t offset, const __private short *p);
+ushort16 __ovld __purefn vload16(size_t offset, const __private ushort *p);
+int16 __ovld __purefn vload16(size_t offset, const __private int *p);
+uint16 __ovld __purefn vload16(size_t offset, const __private uint *p);
+long16 __ovld __purefn vload16(size_t offset, const __private long *p);
+ulong16 __ovld __purefn vload16(size_t offset, const __private ulong *p);
+float16 __ovld __purefn vload16(size_t offset, const __private float *p);
 
 #ifdef cl_khr_fp64
-double2 __ovld vload2(size_t offset, const __global double *p);
-double3 __ovld vload3(size_t offset, const __global double *p);
-double4 __ovld vload4(size_t offset, const __global double *p);
-double8 __ovld vload8(size_t offset, const __global double *p);
-double16 __ovld vload16(size_t offset, const __global double *p);
-double2 __ovld vload2(size_t offset, const __local double *p);
-double3 __ovld vload3(size_t offset, const __local double *p);
-double4 __ovld vload4(size_t offset, const __local double *p);
-double8 __ovld vload8(size_t offset, const __local double *p);
-double16 __ovld vload16(size_t offset, const __local double *p);
-double2 __ovld vload2(size_t offset, const __private double *p);
-double3 __ovld vload3(size_t offset, const __private double *p);
-double4 __ovld vload4(size_t offset, const __private double *p);
-double8 __ovld vload8(size_t offset, const __private double *p);
-double16 __ovld vload16(size_t offset, const __private double *p);
+double2 __ovld __purefn vload2(size_t offset, const __global double *p);
+double3 __ovld __purefn vload3(size_t offset, const __global double *p);
+double4 __ovld __purefn vload4(size_t offset, const __global double *p);
+double8 __ovld __purefn vload8(size_t offset, const __global double *p);
+double16 __ovld __purefn vload16(size_t offset, const __global double *p);
+double2 __ovld __purefn vload2(size_t offset, const __local double *p);
+double3 __ovld __purefn vload3(size_t offset, const __local double *p);
+double4 __ovld __purefn vload4(size_t offset, const __local double *p);
+double8 __ovld __purefn vload8(size_t offset, const __local double *p);
+double16 __ovld __purefn vload16(size_t offset, const __local double *p);
+double2 __ovld __purefn vload2(size_t offset, const __private double *p);
+double3 __ovld __purefn vload3(size_t offset, const __private double *p);
+double4 __ovld __purefn vload4(size_t offset, const __private double *p);
+double8 __ovld __purefn vload8(size_t offset, const __private double *p);
+double16 __ovld __purefn vload16(size_t offset, const __private double *p);
 #endif //cl_khr_fp64
 
 #ifdef cl_khr_fp16
-half __ovld vload(size_t offset, const __global half *p);
-half2 __ovld vload2(size_t offset, const __global half *p);
-half3 __ovld vload3(size_t offset, const __global half *p);
-half4 __ovld vload4(size_t offset, const __global half *p);
-half8 __ovld vload8(size_t offset, const __global half *p);
-half16 __ovld vload16(size_t offset, const __global half *p);
-half __ovld vload(size_t offset, const __local half *p);
-half2 __ovld vload2(size_t offset, const __local half *p);
-half3 __ovld vload3(size_t offset, const __local half *p);
-half4 __ovld vload4(size_t offset, const __local half *p);
-half8 __ovld vload8(size_t offset, const __local half *p);
-half16 __ovld vload16(size_t offset, const __local half *p);
-half __ovld vload(size_t offset, const __private half *p);
-half2 __ovld vload2(size_t offset, const __private half *p);
-half3 __ovld vload3(size_t offset, const __private half *p);
-half4 __ovld vload4(size_t offset, const __private half *p);
-half8 __ovld vload8(size_t offset, const __private half *p);
-half16 __ovld vload16(size_t offset, const __private half *p);
+half __ovld __purefn vload(size_t offset, const __global half *p);
+half2 __ovld __purefn vload2(size_t offset, const __global half *p);
+half3 __ovld __purefn vload3(size_t offset, const __global half *p);
+half4 __ovld __purefn vload4(size_t offset, const __global half *p);
+half8 __ovld __purefn vload8(size_t offset, const __global half *p);
+half16 __ovld __purefn vload16(size_t offset, const __global half *p);
+half __ovld __purefn vload(size_t offset, const __local half *p);
+half2 __ovld __purefn vload2(size_t offset, const __local half *p);
+half3 __ovld __purefn vload3(size_t offset, const __local half *p);
+half4 __ovld __purefn vload4(size_t offset, const __local half *p);
+half8 __ovld __purefn vload8(size_t offset, const __local half *p);
+half16 __ovld __purefn vload16(size_t offset, const __local half *p);
+half __ovld __purefn vload(size_t offset, const __private half *p);
+half2 __ovld __purefn vload2(size_t offset, const __private half *p);
+half3 __ovld __purefn vload3(size_t offset, const __private half *p);
+half4 __ovld __purefn vload4(size_t offset, const __private half *p);
+half8 __ovld __purefn vload8(size_t offset, const __private half *p);
+half16 __ovld __purefn vload16(size_t offset, const __private half *p);
 #endif //cl_khr_fp16
 #endif //defined(__opencl_c_generic_address_space)
 
@@ -11736,13 +11736,13 @@ void __ovld vstore16(half16 data, size_t offset, __private half *p);
  * The read address computed as (p + offset)
  * must be 16-bit aligned.
  */
-float __ovld vload_half(size_t offset, const __constant half *p);
+float __ovld __purefn vload_half(size_t offset, const __constant half *p);
 #if defined(__opencl_c_generic_address_space)
-float __ovld vload_half(size_t offset, const half *p);
+float __ovld __purefn vload_half(size_t offset, const half *p);
 #else
-float __ovld vload_half(size_t offset, const __global half *p);
-float __ovld vload_half(size_t offset, const __local half *p);
-float __ovld vload_half(size_t offset, const __private half *p);
+float __ovld __purefn vload_half(size_t offset, const __global half *p);
+float __ovld __purefn vload_half(size_t offset, const __local half *p);
+float __ovld __purefn vload_half(size_t offset, const __private half *p);
 #endif //defined(__opencl_c_generic_address_space)
 
 /**
@@ -11753,33 +11753,33 @@ float __ovld vload_half(size_t offset, const __private half *p);
  * value is returned. The read address computed
  * as (p + (offset * n)) must be 16-bit aligned.
  */
-float2 __ovld vload_half2(size_t offset, const __constant half *p);
-float3 __ovld vload_half3(size_t offset, const __constant half *p);
-float4 __ovld vload_half4(size_t offset, const __constant half *p);
-float8 __ovld vload_half8(size_t offset, const __constant half *p);
-float16 __ovld vload_half16(size_t offset, const __constant half *p);
+float2 __ovld __purefn vload_half2(size_t offset, const __constant half *p);
+float3 __ovld __purefn vload_half3(size_t offset, const __constant half *p);
+float4 __ovld __purefn vload_half4(size_t offset, const __constant half *p);
+float8 __ovld __purefn vload_half8(size_t offset, const __constant half *p);
+float16 __ovld __purefn vload_half16(size_t offset, const __constant half *p);
 #if defined(__opencl_c_generic_address_space)
-float2 __ovld vload_half2(size_t offset, const half *p);
-float3 __ovld vload_half3(size_t offset, const half *p);
-float4 __ovld vload_half4(size_t offset, const half *p);
-float8 __ovld vload_half8(size_t offset, const half *p);
-float16 __ovld vload_half16(size_t offset, const half *p);
+float2 __ovld __purefn vload_half2(size_t offset, const half *p);
+float3 __ovld __purefn vload_half3(size_t offset, const half *p);
+float4 __ovld __purefn vload_half4(size_t offset, const half *p);
+float8 __ovld __purefn vload_half8(size_t offset, const half *p);
+float16 __ovld __purefn vload_half16(size_t offset, const half *p);
 #else
-float2 __ovld vload_half2(size_t offset, const __global half *p);
-float3 __ovld vload_half3(size_t offset, const __global half *p);
-float4 __ovld vload_half4(size_t offset, const __global half *p);
-float8 __ovld vload_half8(size_t offset, const __global half *p);
-float16 __ovld vload_half16(size_t offset, const __global half *p);
-float2 __ovld vload_half2(size_t offset, const __local half *p);
-float3 __ovld vload_half3(size_t offset, const __local half *p);
-float4 __ovld vload_half4(size_t offset, const __local half *p);
-float8 __ovld vload_half8(size_t offset, const __local half *p);
-float16 __ovld vload_half16(size_t offset, const __local half *p);
-float2 __ovld vload_half2(size_t offset, const __private half *p);
-float3 __ovld vload_half3(size_t offset, const __private half *p);
-float4 __ovld vload_half4(size_t offset, const __private half *p);
-float8 __ovld vload_half8(size_t offset, const __private half *p);
-float16 __ovld vload_half16(size_t offset, const __private half *p);
+float2 __ovld __purefn vload_half2(size_t offset, const __global half *p);
+float3 __ovld __purefn vload_half3(size_t offset, const __global half *p);
+float4 __ovld __purefn vload_half4(size_t offset, const __global half *p);
+float8 __ovld __purefn vload_half8(size_t offset, const __global half *p);
+float16 __ovld __purefn vload_half16(size_t offset, const __global half *p);
+float2 __ovld __purefn vload_half2(size_t offset, const __local half *p);
+float3 __ovld __purefn vload_half3(size_t offset, const __local half *p);
+float4 __ovld __purefn vload_half4(size_t offset, const __local half *p);
+float8 __ovld __purefn vload_half8(size_t offset, const __local half *p);
+float16 __ovld __purefn vload_half16(size_t offset, const __local half *p);
+float2 __ovld __purefn vload_half2(size_t offset, const __private half *p);
+float3 __ovld __purefn vload_half3(size_t offset, const __private half *p);
+float4 __ovld __purefn vload_half4(size_t offset, const __private half *p);
+float8 __ovld __purefn vload_half8(size_t offset, const __private half *p);
+float16 __ovld __purefn vload_half16(size_t offset, const __private half *p);
 #endif //defined(__opencl_c_generic_address_space)
 
 /**
@@ -12073,33 +12073,33 @@ void __ovld vstore_half16_rtn(double16 data, size_t offset, __private half *p);
  * The address computed as (p + (offset * 4))
  * must be aligned to sizeof (half) * 4 bytes.
  */
-float2 __ovld vloada_half2(size_t offset, const __constant half *p);
-float3 __ovld vloada_half3(size_t offset, const __constant half *p);
-float4 __ovld vloada_half4(size_t offset, const __constant half *p);
-float8 __ovld vloada_half8(size_t offset, const __constant half *p);
-float16 __ovld vloada_half16(size_t offset, const __constant half *p);
+float2 __ovld __purefn vloada_half2(size_t offset, const __constant half *p);
+float3 __ovld __purefn vloada_half3(size_t offset, const __constant half *p);
+float4 __ovld __purefn vloada_half4(size_t offset, const __constant half *p);
+float8 __ovld __purefn vloada_half8(size_t offset, const __constant half *p);
+float16 __ovld __purefn vloada_half16(size_t offset, const __constant half *p);
 #if defined(__opencl_c_generic_address_space)
-float2 __ovld vloada_half2(size_t offset, const half *p);
-float3 __ovld vloada_half3(size_t offset, const half *p);
-float4 __ovld vloada_half4(size_t offset, const half *p);
-float8 __ovld vloada_half8(size_t offset, const half *p);
-float16 __ovld vloada_half16(size_t offset, const half *p);
+float2 __ovld __purefn vloada_half2(size_t offset, const half *p);
+float3 __ovld __purefn vloada_half3(size_t offset, const half *p);
+float4 __ovld __purefn vloada_half4(size_t offset, const half *p);
+float8 __ovld __purefn vloada_half8(size_t offset, const half *p);
+float16 __ovld __purefn vloada_half16(size_t offset, const half *p);
 #else
-float2 __ovld vloada_half2(size_t offset, const __global half *p);
-float3 __ovld vloada_half3(size_t offset, const __global half *p);
-float4 __ovld vloada_half4(size_t offset, const __global half *p);
-float8 __ovld vloada_half8(size_t offset, const __global half *p);
-float16 __ovld vloada_half16(size_t offset, const __global half *p);
-float2 __ovld vloada_half2(size_t offset, const __local half *p);
-float3 __ovld vloada_half3(size_t offset, const __local half *p);
-float4 __ovld vloada_half4(size_t offset, const __local half *p);
-float8 __ovld vloada_half8(size_t offset, const __local half *p);
-float16 __ovld vloada_half16(size_t offset, const __local half *p);
-float2 __ovld vloada_half2(size_t offset, const __private half *p);
-float3 __ovld vloada_half3(size_t offset, const __private half *p);
-float4 __ovld vloada_half4(size_t offset, const __private half *p);
-float8 __ovld vloada_half8(size_t offset, const __private half *p);
-float16 __ovld vloada_half16(size_t offset, const __private half *p);
+float2 __ovld __purefn vloada_half2(size_t offset, const __global half *p);
+float3 __ovld __purefn vloada_half3(size_t offset, const __global half *p);
+float4 __ovld __purefn vloada_half4(size_t offset, const __global half *p);
+float8 __ovld __purefn vloada_half8(size_t offset, const __global half *p);
+float16 __ovld __purefn vloada_half16(size_t offset, const __global half *p);
+float2 __ovld __purefn vloada_half2(size_t offset, const __local half *p);
+float3 __ovld __purefn vloada_half3(size_t offset, const __local half *p);
+float4 __ovld __purefn vloada_half4(size_t offset, const __local half *p);
+float8 __ovld __purefn vloada_half8(size_t offset, const __local half *p);
+float16 __ovld __purefn vloada_half16(size_t offset, const __local half *p);
+float2 __ovld __purefn vloada_half2(size_t offset, const __private half *p);
+float3 __ovld __purefn vloada_half3(size_t offset, const __private half *p);
+float4 __ovld __purefn vloada_half4(size_t offset, const __private half *p);
+float8 __ovld __purefn vloada_half8(size_t offset, const __private half *p);
+float16 __ovld __purefn vloada_half16(size_t offset, const __private half *p);
 #endif //defined(__opencl_c_generic_address_space)
 
 /**
diff --git a/clang/lib/Headers/unwind.h b/clang/lib/Headers/unwind.h
index 029524b7bc84..6e069798f02d 100644
--- a/clang/lib/Headers/unwind.h
+++ b/clang/lib/Headers/unwind.h
@@ -172,7 +172,8 @@ typedef enum {
   _UVRSC_CORE = 0,        /* integer register */
   _UVRSC_VFP = 1,         /* vfp */
   _UVRSC_WMMXD = 3,       /* Intel WMMX data register */
-  _UVRSC_WMMXC = 4        /* Intel WMMX control register */
+  _UVRSC_WMMXC = 4,       /* Intel WMMX control register */
+  _UVRSC_PSEUDO = 5       /* Special purpose pseudo register */
 } _Unwind_VRS_RegClass;
 
 typedef enum {
diff --git a/clang/lib/Lex/ModuleMap.cpp b/clang/lib/Lex/ModuleMap.cpp
index 9fa170410da3..0b136aeb580f 100644
--- a/clang/lib/Lex/ModuleMap.cpp
+++ b/clang/lib/Lex/ModuleMap.cpp
@@ -832,12 +832,16 @@ std::pair<Module *, bool> ModuleMap::findOrCreateModule(StringRef Name,
   return std::make_pair(Result, true);
 }
 
-Module *ModuleMap::createGlobalModuleFragmentForModuleUnit(SourceLocation Loc) {
-  PendingSubmodules.emplace_back(
-      new Module("<global>", Loc, nullptr, /*IsFramework*/ false,
-                 /*IsExplicit*/ true, NumCreatedModules++));
-  PendingSubmodules.back()->Kind = Module::GlobalModuleFragment;
-  return PendingSubmodules.back().get();
+Module *ModuleMap::createGlobalModuleFragmentForModuleUnit(SourceLocation Loc,
+                                                           Module *Parent) {
+  auto *Result = new Module("<global>", Loc, Parent, /*IsFramework*/ false,
+                            /*IsExplicit*/ true, NumCreatedModules++);
+  Result->Kind = Module::GlobalModuleFragment;
+  // If the created module isn't owned by a parent, send it to PendingSubmodules
+  // to wait for its parent.
+  if (!Result->Parent)
+    PendingSubmodules.emplace_back(Result);
+  return Result;
 }
 
 Module *
diff --git a/clang/lib/Lex/TokenLexer.cpp b/clang/lib/Lex/TokenLexer.cpp
index 41e7f3f1dccb..e71a65f031e4 100644
--- a/clang/lib/Lex/TokenLexer.cpp
+++ b/clang/lib/Lex/TokenLexer.cpp
@@ -472,11 +472,9 @@ void TokenLexer::ExpandFunctionArguments() {
 
         // If the '##' came from expanding an argument, turn it into 'unknown'
         // to avoid pasting.
-        for (Token &Tok : llvm::make_range(ResultToks.begin() + FirstResult,
-                                           ResultToks.end())) {
+        for (Token &Tok : llvm::drop_begin(ResultToks, FirstResult))
           if (Tok.is(tok::hashhash))
             Tok.setKind(tok::unknown);
-        }
 
         if(ExpandLocStart.isValid()) {
           updateLocForMacroArgTokens(CurTok.getLocation(),
diff --git a/clang/lib/Parse/ParseCXXInlineMethods.cpp b/clang/lib/Parse/ParseCXXInlineMethods.cpp
index 116724a0d50b..19cddc69ebfc 100644
--- a/clang/lib/Parse/ParseCXXInlineMethods.cpp
+++ b/clang/lib/Parse/ParseCXXInlineMethods.cpp
@@ -452,13 +452,14 @@ void Parser::ParseLexedMethodDeclaration(LateParsedMethodDeclaration &LM) {
     CXXMethodDecl *Method;
     if (FunctionTemplateDecl *FunTmpl
           = dyn_cast<FunctionTemplateDecl>(LM.Method))
-      Method = cast<CXXMethodDecl>(FunTmpl->getTemplatedDecl());
+      Method = dyn_cast<CXXMethodDecl>(FunTmpl->getTemplatedDecl());
     else
-      Method = cast<CXXMethodDecl>(LM.Method);
+      Method = dyn_cast<CXXMethodDecl>(LM.Method);
 
-    Sema::CXXThisScopeRAII ThisScope(Actions, Method->getParent(),
-                                     Method->getMethodQualifiers(),
-                                     getLangOpts().CPlusPlus11);
+    Sema::CXXThisScopeRAII ThisScope(
+        Actions, Method ? Method->getParent() : nullptr,
+        Method ? Method->getMethodQualifiers() : Qualifiers{},
+        Method && getLangOpts().CPlusPlus11);
 
     // Parse the exception-specification.
     SourceRange SpecificationRange;
diff --git a/clang/lib/Parse/ParseDecl.cpp b/clang/lib/Parse/ParseDecl.cpp
index 1bdeccc4cbf5..0c1f88bc51d1 100644
--- a/clang/lib/Parse/ParseDecl.cpp
+++ b/clang/lib/Parse/ParseDecl.cpp
@@ -2891,7 +2891,8 @@ void Parser::ParseAlignmentSpecifier(ParsedAttributes &Attrs,
 }
 
 ExprResult Parser::ParseExtIntegerArgument() {
-  assert(Tok.is(tok::kw__ExtInt) && "Not an extended int type");
+  assert(Tok.isOneOf(tok::kw__ExtInt, tok::kw__BitInt) &&
+         "Not an extended int type");
   ConsumeToken();
 
   BalancedDelimiterTracker T(*this, tok::l_paren);
@@ -3882,11 +3883,13 @@ void Parser::ParseDeclarationSpecifiers(DeclSpec &DS,
       isInvalid = DS.SetTypeSpecType(DeclSpec::TST_int, Loc, PrevSpec,
                                      DiagID, Policy);
       break;
-    case tok::kw__ExtInt: {
+    case tok::kw__ExtInt:
+    case tok::kw__BitInt: {
+      DiagnoseBitIntUse(Tok);
       ExprResult ER = ParseExtIntegerArgument();
       if (ER.isInvalid())
         continue;
-      isInvalid = DS.SetExtIntType(Loc, ER.get(), PrevSpec, DiagID, Policy);
+      isInvalid = DS.SetBitIntType(Loc, ER.get(), PrevSpec, DiagID, Policy);
       ConsumedEnd = PrevTokLocation;
       break;
     }
@@ -5015,6 +5018,7 @@ bool Parser::isKnownToBeTypeSpecifier(const Token &Tok) const {
   case tok::kw_char32_t:
   case tok::kw_int:
   case tok::kw__ExtInt:
+  case tok::kw__BitInt:
   case tok::kw___bf16:
   case tok::kw_half:
   case tok::kw_float:
@@ -5097,6 +5101,7 @@ bool Parser::isTypeSpecifierQualifier() {
   case tok::kw_char32_t:
   case tok::kw_int:
   case tok::kw__ExtInt:
+  case tok::kw__BitInt:
   case tok::kw_half:
   case tok::kw___bf16:
   case tok::kw_float:
@@ -5268,6 +5273,7 @@ bool Parser::isDeclarationSpecifier(bool DisambiguatingWithExpression) {
 
   case tok::kw_int:
   case tok::kw__ExtInt:
+  case tok::kw__BitInt:
   case tok::kw_half:
   case tok::kw___bf16:
   case tok::kw_float:
@@ -7476,3 +7482,24 @@ bool Parser::TryAltiVecTokenOutOfLine(DeclSpec &DS, SourceLocation Loc,
   }
   return false;
 }
+
+void Parser::DiagnoseBitIntUse(const Token &Tok) {
+  // If the token is for _ExtInt, diagnose it as being deprecated. Otherwise,
+  // the token is about _BitInt and gets (potentially) diagnosed as use of an
+  // extension.
+  assert(Tok.isOneOf(tok::kw__ExtInt, tok::kw__BitInt) &&
+         "expected either an _ExtInt or _BitInt token!");
+
+  SourceLocation Loc = Tok.getLocation();
+  if (Tok.is(tok::kw__ExtInt)) {
+    Diag(Loc, diag::warn_ext_int_deprecated)
+        << FixItHint::CreateReplacement(Loc, "_BitInt");
+  } else {
+    // In C2x mode, diagnose that the use is not compatible with pre-C2x modes.
+    // Otherwise, diagnose that the use is a Clang extension.
+    if (getLangOpts().C2x)
+      Diag(Loc, diag::warn_c17_compat_bit_int);
+    else
+      Diag(Loc, diag::ext_bit_int) << getLangOpts().CPlusPlus;
+  }
+}
diff --git a/clang/lib/Parse/ParseExpr.cpp b/clang/lib/Parse/ParseExpr.cpp
index 2c8b4f9f441f..09a3842f5809 100644
--- a/clang/lib/Parse/ParseExpr.cpp
+++ b/clang/lib/Parse/ParseExpr.cpp
@@ -1513,6 +1513,7 @@ ExprResult Parser::ParseCastExpression(CastParseKind ParseKind,
   case tok::kw___int64:
   case tok::kw___int128:
   case tok::kw__ExtInt:
+  case tok::kw__BitInt:
   case tok::kw_signed:
   case tok::kw_unsigned:
   case tok::kw_half:
diff --git a/clang/lib/Parse/ParseExprCXX.cpp b/clang/lib/Parse/ParseExprCXX.cpp
index 4e5c0ac6c1c1..76c510ddd36c 100644
--- a/clang/lib/Parse/ParseExprCXX.cpp
+++ b/clang/lib/Parse/ParseExprCXX.cpp
@@ -2191,12 +2191,14 @@ void Parser::ParseCXXSimpleTypeSpecifier(DeclSpec &DS) {
     return;
   }
 
-  case tok::kw__ExtInt: {
+  case tok::kw__ExtInt:
+  case tok::kw__BitInt: {
+    DiagnoseBitIntUse(Tok);
     ExprResult ER = ParseExtIntegerArgument();
     if (ER.isInvalid())
       DS.SetTypeSpecError();
     else
-      DS.SetExtIntType(Loc, ER.get(), PrevSpec, DiagID, Policy);
+      DS.SetBitIntType(Loc, ER.get(), PrevSpec, DiagID, Policy);
 
     // Do this here because we have already consumed the close paren.
     DS.SetRangeEnd(PrevTokLocation);
diff --git a/clang/lib/Parse/ParseOpenMP.cpp b/clang/lib/Parse/ParseOpenMP.cpp
index 613ad742c93f..300b022d83b9 100644
--- a/clang/lib/Parse/ParseOpenMP.cpp
+++ b/clang/lib/Parse/ParseOpenMP.cpp
@@ -3192,6 +3192,7 @@ OMPClause *Parser::ParseOpenMPClause(OpenMPDirectiveKind DKind,
   case OMPC_read:
   case OMPC_write:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
diff --git a/clang/lib/Parse/ParseTentative.cpp b/clang/lib/Parse/ParseTentative.cpp
index be3823ecda01..35c9036fb27e 100644
--- a/clang/lib/Parse/ParseTentative.cpp
+++ b/clang/lib/Parse/ParseTentative.cpp
@@ -1690,6 +1690,7 @@ Parser::isCXXDeclarationSpecifier(Parser::TPResult BracedCastResult,
   case tok::kw__Atomic:
     return TPResult::True;
 
+  case tok::kw__BitInt:
   case tok::kw__ExtInt: {
     if (NextToken().isNot(tok::l_paren))
       return TPResult::Error;
@@ -1741,6 +1742,7 @@ bool Parser::isCXXDeclarationSpecifierAType() {
   case tok::kw_short:
   case tok::kw_int:
   case tok::kw__ExtInt:
+  case tok::kw__BitInt:
   case tok::kw_long:
   case tok::kw___int64:
   case tok::kw___int128:
diff --git a/clang/lib/Rewrite/HTMLRewrite.cpp b/clang/lib/Rewrite/HTMLRewrite.cpp
index 371557a624c9..e9b678b69594 100644
--- a/clang/lib/Rewrite/HTMLRewrite.cpp
+++ b/clang/lib/Rewrite/HTMLRewrite.cpp
@@ -203,7 +203,7 @@ std::string html::EscapeText(StringRef s, bool EscapeSpaces, bool ReplaceTabs) {
     }
   }
 
-  return os.str();
+  return Str;
 }
 
 static void AddLineNumber(RewriteBuffer &RB, unsigned LineNo,
diff --git a/clang/lib/Sema/CodeCompleteConsumer.cpp b/clang/lib/Sema/CodeCompleteConsumer.cpp
index 6c47cedfccf3..0a2ca54e244a 100644
--- a/clang/lib/Sema/CodeCompleteConsumer.cpp
+++ b/clang/lib/Sema/CodeCompleteConsumer.cpp
@@ -335,7 +335,7 @@ std::string CodeCompletionString::getAsString() const {
       break;
     }
   }
-  return OS.str();
+  return Result;
 }
 
 const char *CodeCompletionString::getTypedText() const {
@@ -640,7 +640,7 @@ static std::string getOverloadAsString(const CodeCompletionString &CCS) {
       break;
     }
   }
-  return OS.str();
+  return Result;
 }
 
 void PrintingCodeCompleteConsumer::ProcessOverloadCandidates(
diff --git a/clang/lib/Sema/DeclSpec.cpp b/clang/lib/Sema/DeclSpec.cpp
index 4405f29f3d99..d4dc790c008a 100644
--- a/clang/lib/Sema/DeclSpec.cpp
+++ b/clang/lib/Sema/DeclSpec.cpp
@@ -365,7 +365,7 @@ bool Declarator::isDeclarationOfFunction() const {
     case TST_half:
     case TST_int:
     case TST_int128:
-    case TST_extint:
+    case TST_bitint:
     case TST_struct:
     case TST_interface:
     case TST_union:
@@ -551,7 +551,7 @@ const char *DeclSpec::getSpecifierName(DeclSpec::TST T,
   case DeclSpec::TST_char32:      return "char32_t";
   case DeclSpec::TST_int:         return "int";
   case DeclSpec::TST_int128:      return "__int128";
-  case DeclSpec::TST_extint:      return "_ExtInt";
+  case DeclSpec::TST_bitint:      return "_BitInt";
   case DeclSpec::TST_half:        return "half";
   case DeclSpec::TST_float:       return "float";
   case DeclSpec::TST_double:      return "double";
@@ -932,7 +932,7 @@ bool DeclSpec::SetTypeSpecError() {
   return false;
 }
 
-bool DeclSpec::SetExtIntType(SourceLocation KWLoc, Expr *BitsExpr,
+bool DeclSpec::SetBitIntType(SourceLocation KWLoc, Expr *BitsExpr,
                              const char *&PrevSpec, unsigned &DiagID,
                              const PrintingPolicy &Policy) {
   assert(BitsExpr && "no expression provided!");
@@ -945,7 +945,7 @@ bool DeclSpec::SetExtIntType(SourceLocation KWLoc, Expr *BitsExpr,
     return true;
   }
 
-  TypeSpecType = TST_extint;
+  TypeSpecType = TST_bitint;
   ExprRep = BitsExpr;
   TSTLoc = KWLoc;
   TSTNameLoc = KWLoc;
@@ -1252,7 +1252,7 @@ void DeclSpec::Finish(Sema &S, const PrintingPolicy &Policy) {
       TypeSpecType = TST_int; // unsigned -> unsigned int, signed -> signed int.
     else if (TypeSpecType != TST_int && TypeSpecType != TST_int128 &&
              TypeSpecType != TST_char && TypeSpecType != TST_wchar &&
-             !IsFixedPointType && TypeSpecType != TST_extint) {
+             !IsFixedPointType && TypeSpecType != TST_bitint) {
       S.Diag(TSSLoc, diag::err_invalid_sign_spec)
         << getSpecifierName((TST)TypeSpecType, Policy);
       // signed double -> double.
@@ -1302,7 +1302,7 @@ void DeclSpec::Finish(Sema &S, const PrintingPolicy &Policy) {
                                                  " double");
       TypeSpecType = TST_double;   // _Complex -> _Complex double.
     } else if (TypeSpecType == TST_int || TypeSpecType == TST_char ||
-               TypeSpecType == TST_extint) {
+               TypeSpecType == TST_bitint) {
       // Note that this intentionally doesn't include _Complex _Bool.
       if (!S.getLangOpts().CPlusPlus)
         S.Diag(TSTLoc, diag::ext_integer_complex);
diff --git a/clang/lib/Sema/OpenCLBuiltins.td b/clang/lib/Sema/OpenCLBuiltins.td
index 8cf7ec58eff5..38debc5aa9fc 100644
--- a/clang/lib/Sema/OpenCLBuiltins.td
+++ b/clang/lib/Sema/OpenCLBuiltins.td
@@ -806,17 +806,17 @@ multiclass VloadVstore<list<AddressSpace> addrspaces, bit defStores> {
   foreach AS = addrspaces in {
     foreach VSize = [2, 3, 4, 8, 16] in {
       foreach name = ["vload" # VSize] in {
-        def : Builtin<name, [VectorType<Char, VSize>, Size, PointerType<ConstType<Char>, AS>]>;
-        def : Builtin<name, [VectorType<UChar, VSize>, Size, PointerType<ConstType<UChar>, AS>]>;
-        def : Builtin<name, [VectorType<Short, VSize>, Size, PointerType<ConstType<Short>, AS>]>;
-        def : Builtin<name, [VectorType<UShort, VSize>, Size, PointerType<ConstType<UShort>, AS>]>;
-        def : Builtin<name, [VectorType<Int, VSize>, Size, PointerType<ConstType<Int>, AS>]>;
-        def : Builtin<name, [VectorType<UInt, VSize>, Size, PointerType<ConstType<UInt>, AS>]>;
-        def : Builtin<name, [VectorType<Long, VSize>, Size, PointerType<ConstType<Long>, AS>]>;
-        def : Builtin<name, [VectorType<ULong, VSize>, Size, PointerType<ConstType<ULong>, AS>]>;
-        def : Builtin<name, [VectorType<Float, VSize>, Size, PointerType<ConstType<Float>, AS>]>;
-        def : Builtin<name, [VectorType<Double, VSize>, Size, PointerType<ConstType<Double>, AS>]>;
-        def : Builtin<name, [VectorType<Half, VSize>, Size, PointerType<ConstType<Half>, AS>]>;
+        def : Builtin<name, [VectorType<Char, VSize>, Size, PointerType<ConstType<Char>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<UChar, VSize>, Size, PointerType<ConstType<UChar>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<Short, VSize>, Size, PointerType<ConstType<Short>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<UShort, VSize>, Size, PointerType<ConstType<UShort>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<Int, VSize>, Size, PointerType<ConstType<Int>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<UInt, VSize>, Size, PointerType<ConstType<UInt>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<Long, VSize>, Size, PointerType<ConstType<Long>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<ULong, VSize>, Size, PointerType<ConstType<ULong>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<Float, VSize>, Size, PointerType<ConstType<Float>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<Double, VSize>, Size, PointerType<ConstType<Double>, AS>], Attr.Pure>;
+        def : Builtin<name, [VectorType<Half, VSize>, Size, PointerType<ConstType<Half>, AS>], Attr.Pure>;
       }
       if defStores then {
         foreach name = ["vstore" # VSize] in {
@@ -848,10 +848,10 @@ defm : VloadVstore<[ConstantAS], 0>;
 
 multiclass VloadVstoreHalf<list<AddressSpace> addrspaces, bit defStores> {
   foreach AS = addrspaces in {
-    def : Builtin<"vload_half", [Float, Size, PointerType<ConstType<Half>, AS>]>;
+    def : Builtin<"vload_half", [Float, Size, PointerType<ConstType<Half>, AS>], Attr.Pure>;
     foreach VSize = [2, 3, 4, 8, 16] in {
       foreach name = ["vload_half" # VSize, "vloada_half" # VSize] in {
-        def : Builtin<name, [VectorType<Float, VSize>, Size, PointerType<ConstType<Half>, AS>]>;
+        def : Builtin<name, [VectorType<Float, VSize>, Size, PointerType<ConstType<Half>, AS>], Attr.Pure>;
       }
     }
     if defStores then {
@@ -877,7 +877,7 @@ let MaxVersion = CL20 in {
 let MinVersion = CL20 in {
   defm : VloadVstoreHalf<[GenericAS], 1>;
 }
-// vload with constant address space is available regardless of version.
+// vload_half and vloada_half with constant address space are available regardless of version.
 defm : VloadVstoreHalf<[ConstantAS], 0>;
 
 // OpenCL v3.0 s6.15.8 - Synchronization Functions.
diff --git a/clang/lib/Sema/Sema.cpp b/clang/lib/Sema/Sema.cpp
index a2b8f475aa8c..734ed0f62ec6 100644
--- a/clang/lib/Sema/Sema.cpp
+++ b/clang/lib/Sema/Sema.cpp
@@ -1881,8 +1881,8 @@ void Sema::checkTypeSupport(QualType Ty, SourceLocation Loc, ValueDecl *D) {
     if (Ty->isDependentType())
       return;
 
-    if (Ty->isExtIntType()) {
-      if (!Context.getTargetInfo().hasExtIntType()) {
+    if (Ty->isBitIntType()) {
+      if (!Context.getTargetInfo().hasBitIntType()) {
         PartialDiagnostic PD = PDiag(diag::err_target_unsupported_type);
         if (D)
           PD << D;
diff --git a/clang/lib/Sema/SemaAttr.cpp b/clang/lib/Sema/SemaAttr.cpp
index 100f8e36a9b8..b69492768848 100644
--- a/clang/lib/Sema/SemaAttr.cpp
+++ b/clang/lib/Sema/SemaAttr.cpp
@@ -792,7 +792,7 @@ attrMatcherRuleListToString(ArrayRef<attr::SubjectMatchRule> Rules) {
       OS << (I.index() == Rules.size() - 1 ? ", and " : ", ");
     OS << "'" << attr::getSubjectMatchRuleSpelling(I.value()) << "'";
   }
-  return OS.str();
+  return Result;
 }
 
 } // end anonymous namespace
diff --git a/clang/lib/Sema/SemaCUDA.cpp b/clang/lib/Sema/SemaCUDA.cpp
index 840b3daae63c..59601c5ce79d 100644
--- a/clang/lib/Sema/SemaCUDA.cpp
+++ b/clang/lib/Sema/SemaCUDA.cpp
@@ -886,7 +886,6 @@ void Sema::CUDACheckLambdaCapture(CXXMethodDecl *Callee,
                           diag::warn_maybe_capture_bad_target_this_ptr, Callee,
                           *this);
   }
-  return;
 }
 
 void Sema::CUDASetLambdaAttrs(CXXMethodDecl *Method) {
diff --git a/clang/lib/Sema/SemaCXXScopeSpec.cpp b/clang/lib/Sema/SemaCXXScopeSpec.cpp
index c4826b5a6e8f..8cecf6c6ab4f 100644
--- a/clang/lib/Sema/SemaCXXScopeSpec.cpp
+++ b/clang/lib/Sema/SemaCXXScopeSpec.cpp
@@ -736,8 +736,15 @@ bool Sema::BuildCXXNestedNameSpecifier(Scope *S, NestedNameSpecInfo &IdInfo,
 
     QualType T =
         Context.getTypeDeclType(cast<TypeDecl>(SD->getUnderlyingDecl()));
+
+    if (T->isEnumeralType())
+      Diag(IdInfo.IdentifierLoc, diag::warn_cxx98_compat_enum_nested_name_spec);
+
     TypeLocBuilder TLB;
-    if (isa<InjectedClassNameType>(T)) {
+    if (const auto *USD = dyn_cast<UsingShadowDecl>(SD)) {
+      T = Context.getUsingType(USD, T);
+      TLB.pushTypeSpec(T).setNameLoc(IdInfo.IdentifierLoc);
+    } else if (isa<InjectedClassNameType>(T)) {
       InjectedClassNameTypeLoc InjectedTL
         = TLB.push<InjectedClassNameTypeLoc>(T);
       InjectedTL.setNameLoc(IdInfo.IdentifierLoc);
@@ -770,9 +777,6 @@ bool Sema::BuildCXXNestedNameSpecifier(Scope *S, NestedNameSpecInfo &IdInfo,
       llvm_unreachable("Unhandled TypeDecl node in nested-name-specifier");
     }
 
-    if (T->isEnumeralType())
-      Diag(IdInfo.IdentifierLoc, diag::warn_cxx98_compat_enum_nested_name_spec);
-
     SS.Extend(Context, SourceLocation(), TLB.getTypeLocInContext(Context, T),
               IdInfo.CCLoc);
     return false;
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 33e2b3b5027d..4e83fa1fffca 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -195,6 +195,29 @@ static bool SemaBuiltinAddressof(Sema &S, CallExpr *TheCall) {
   return false;
 }
 
+/// Check that the argument to __builtin_function_start is a function.
+static bool SemaBuiltinFunctionStart(Sema &S, CallExpr *TheCall) {
+  if (checkArgCount(S, TheCall, 1))
+    return true;
+
+  ExprResult Arg = S.DefaultFunctionArrayLvalueConversion(TheCall->getArg(0));
+  if (Arg.isInvalid())
+    return true;
+
+  TheCall->setArg(0, Arg.get());
+  const FunctionDecl *FD = dyn_cast_or_null<FunctionDecl>(
+      Arg.get()->getAsBuiltinConstantDeclRef(S.getASTContext()));
+
+  if (!FD) {
+    S.Diag(TheCall->getBeginLoc(), diag::err_function_start_invalid_type)
+        << TheCall->getSourceRange();
+    return true;
+  }
+
+  return !S.checkAddressOfFunctionIsAvailable(FD, /*Complain=*/true,
+                                              TheCall->getBeginLoc());
+}
+
 /// Check the number of arguments and set the result type to
 /// the argument type.
 static bool SemaBuiltinPreserveAI(Sema &S, CallExpr *TheCall) {
@@ -325,17 +348,17 @@ static bool SemaBuiltinOverflow(Sema &S, CallExpr *TheCall,
     }
   }
 
-  // Disallow signed ExtIntType args larger than 128 bits to mul function until
-  // we improve backend support.
+  // Disallow signed bit-precise integer args larger than 128 bits to mul
+  // function until we improve backend support.
   if (BuiltinID == Builtin::BI__builtin_mul_overflow) {
     for (unsigned I = 0; I < 3; ++I) {
       const auto Arg = TheCall->getArg(I);
       // Third argument will be a pointer.
       auto Ty = I < 2 ? Arg->getType() : Arg->getType()->getPointeeType();
-      if (Ty->isExtIntType() && Ty->isSignedIntegerType() &&
+      if (Ty->isBitIntType() && Ty->isSignedIntegerType() &&
           S.getASTContext().getIntWidth(Ty) > 128)
         return S.Diag(Arg->getBeginLoc(),
-                      diag::err_overflow_builtin_ext_int_max_size)
+                      diag::err_overflow_builtin_bit_int_max_size)
                << 128;
     }
   }
@@ -446,14 +469,14 @@ public:
       break;
     }
 
-    auto OptionalFW = FS.getFieldWidth();
-    if (OptionalFW.getHowSpecified() !=
+    analyze_format_string::OptionalAmount FW = FS.getFieldWidth();
+    if (FW.getHowSpecified() !=
         analyze_format_string::OptionalAmount::HowSpecified::Constant)
       return true;
 
-    unsigned SourceSize = OptionalFW.getConstantAmount() + NulByte;
+    unsigned SourceSize = FW.getConstantAmount() + NulByte;
 
-    auto DestSizeAPS = ComputeSizeArgument(FS.getArgIndex());
+    Optional<llvm::APSInt> DestSizeAPS = ComputeSizeArgument(FS.getArgIndex());
     if (!DestSizeAPS)
       return true;
 
@@ -652,20 +675,53 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD,
       isConstantEvaluated())
     return;
 
-  unsigned BuiltinID = FD->getBuiltinID(/*ConsiderWrappers=*/true);
+  bool UseDABAttr = false;
+  const FunctionDecl *UseDecl = FD;
+
+  const auto *DABAttr = FD->getAttr<DiagnoseAsBuiltinAttr>();
+  if (DABAttr) {
+    UseDecl = DABAttr->getFunction();
+    assert(UseDecl && "Missing FunctionDecl in DiagnoseAsBuiltin attribute!");
+    UseDABAttr = true;
+  }
+
+  unsigned BuiltinID = UseDecl->getBuiltinID(/*ConsiderWrappers=*/true);
+
   if (!BuiltinID)
     return;
 
   const TargetInfo &TI = getASTContext().getTargetInfo();
   unsigned SizeTypeWidth = TI.getTypeWidth(TI.getSizeType());
 
+  auto TranslateIndex = [&](unsigned Index) -> Optional<unsigned> {
+    // If we refer to a diagnose_as_builtin attribute, we need to change the
+    // argument index to refer to the arguments of the called function. Unless
+    // the index is out of bounds, which presumably means it's a variadic
+    // function.
+    if (!UseDABAttr)
+      return Index;
+    unsigned DABIndices = DABAttr->argIndices_size();
+    unsigned NewIndex = Index < DABIndices
+                            ? DABAttr->argIndices_begin()[Index]
+                            : Index - DABIndices + FD->getNumParams();
+    if (NewIndex >= TheCall->getNumArgs())
+      return llvm::None;
+    return NewIndex;
+  };
+
   auto ComputeExplicitObjectSizeArgument =
       [&](unsigned Index) -> Optional<llvm::APSInt> {
+    Optional<unsigned> IndexOptional = TranslateIndex(Index);
+    if (!IndexOptional)
+      return llvm::None;
+    unsigned NewIndex = IndexOptional.getValue();
     Expr::EvalResult Result;
-    Expr *SizeArg = TheCall->getArg(Index);
+    Expr *SizeArg = TheCall->getArg(NewIndex);
     if (!SizeArg->EvaluateAsInt(Result, getASTContext()))
       return llvm::None;
-    return Result.Val.getInt();
+    llvm::APSInt Integer = Result.Val.getInt();
+    Integer.setIsUnsigned(true);
+    return Integer;
   };
 
   auto ComputeSizeArgument = [&](unsigned Index) -> Optional<llvm::APSInt> {
@@ -680,7 +736,12 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD,
         BOSType = POS->getType();
     }
 
-    const Expr *ObjArg = TheCall->getArg(Index);
+    Optional<unsigned> IndexOptional = TranslateIndex(Index);
+    if (!IndexOptional)
+      return llvm::None;
+    unsigned NewIndex = IndexOptional.getValue();
+
+    const Expr *ObjArg = TheCall->getArg(NewIndex);
     uint64_t Result;
     if (!ObjArg->tryEvaluateObjectSize(Result, getASTContext(), BOSType))
       return llvm::None;
@@ -690,7 +751,12 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD,
   };
 
   auto ComputeStrLenArgument = [&](unsigned Index) -> Optional<llvm::APSInt> {
-    Expr *ObjArg = TheCall->getArg(Index);
+    Optional<unsigned> IndexOptional = TranslateIndex(Index);
+    if (!IndexOptional)
+      return llvm::None;
+    unsigned NewIndex = IndexOptional.getValue();
+
+    const Expr *ObjArg = TheCall->getArg(NewIndex);
     uint64_t Result;
     if (!ObjArg->tryEvaluateStrLen(Result, getASTContext()))
       return llvm::None;
@@ -898,7 +964,8 @@ void Sema::checkFortifiedBuiltinMemoryFunction(FunctionDecl *FD,
   }
 
   if (!SourceSize || !DestinationSize ||
-      SourceSize.getValue().ule(DestinationSize.getValue()))
+      llvm::APSInt::compareValues(SourceSize.getValue(),
+                                  DestinationSize.getValue()) <= 0)
     return;
 
   StringRef FunctionName = GetFunctionName();
@@ -1874,6 +1941,10 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl, unsigned BuiltinID,
     if (SemaBuiltinAddressof(*this, TheCall))
       return ExprError();
     break;
+  case Builtin::BI__builtin_function_start:
+    if (SemaBuiltinFunctionStart(*this, TheCall))
+      return ExprError();
+    break;
   case Builtin::BI__builtin_is_aligned:
   case Builtin::BI__builtin_align_up:
   case Builtin::BI__builtin_align_down:
@@ -2098,20 +2169,85 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl, unsigned BuiltinID,
     break;
   }
 
-  case Builtin::BI__builtin_elementwise_abs:
-    if (SemaBuiltinElementwiseMathOneArg(TheCall))
+  // __builtin_elementwise_abs restricts the element type to signed integers or
+  // floating point types only.
+  case Builtin::BI__builtin_elementwise_abs: {
+    if (PrepareBuiltinElementwiseMathOneArgCall(TheCall))
+      return ExprError();
+
+    QualType ArgTy = TheCall->getArg(0)->getType();
+    QualType EltTy = ArgTy;
+
+    if (auto *VecTy = EltTy->getAs<VectorType>())
+      EltTy = VecTy->getElementType();
+    if (EltTy->isUnsignedIntegerType()) {
+      Diag(TheCall->getArg(0)->getBeginLoc(),
+           diag::err_builtin_invalid_arg_type)
+          << 1 << /* signed integer or float ty*/ 3 << ArgTy;
+      return ExprError();
+    }
+    break;
+  }
+
+  // __builtin_elementwise_ceil restricts the element type to floating point
+  // types only.
+  case Builtin::BI__builtin_elementwise_ceil: {
+    if (PrepareBuiltinElementwiseMathOneArgCall(TheCall))
+      return ExprError();
+
+    QualType ArgTy = TheCall->getArg(0)->getType();
+    QualType EltTy = ArgTy;
+
+    if (auto *VecTy = EltTy->getAs<VectorType>())
+      EltTy = VecTy->getElementType();
+    if (!EltTy->isFloatingType()) {
+      Diag(TheCall->getArg(0)->getBeginLoc(),
+           diag::err_builtin_invalid_arg_type)
+          << 1 << /* float ty*/ 5 << ArgTy;
+
       return ExprError();
+    }
     break;
+  }
+
   case Builtin::BI__builtin_elementwise_min:
   case Builtin::BI__builtin_elementwise_max:
     if (SemaBuiltinElementwiseMath(TheCall))
       return ExprError();
     break;
   case Builtin::BI__builtin_reduce_max:
-  case Builtin::BI__builtin_reduce_min:
-    if (SemaBuiltinReduceMath(TheCall))
+  case Builtin::BI__builtin_reduce_min: {
+    if (PrepareBuiltinReduceMathOneArgCall(TheCall))
+      return ExprError();
+
+    const Expr *Arg = TheCall->getArg(0);
+    const auto *TyA = Arg->getType()->getAs<VectorType>();
+    if (!TyA) {
+      Diag(Arg->getBeginLoc(), diag::err_builtin_invalid_arg_type)
+          << 1 << /* vector ty*/ 4 << Arg->getType();
+      return ExprError();
+    }
+
+    TheCall->setType(TyA->getElementType());
+    break;
+  }
+
+  // __builtin_reduce_xor supports vector of integers only.
+  case Builtin::BI__builtin_reduce_xor: {
+    if (PrepareBuiltinReduceMathOneArgCall(TheCall))
+      return ExprError();
+
+    const Expr *Arg = TheCall->getArg(0);
+    const auto *TyA = Arg->getType()->getAs<VectorType>();
+    if (!TyA || !TyA->getElementType()->isIntegerType()) {
+      Diag(Arg->getBeginLoc(), diag::err_builtin_invalid_arg_type)
+          << 1  << /* vector of integers */ 6 << Arg->getType();
       return ExprError();
+    }
+    TheCall->setType(TyA->getElementType());
     break;
+  }
+
   case Builtin::BI__builtin_matrix_transpose:
     return SemaBuiltinMatrixTranspose(TheCall, TheCallResult);
 
@@ -3496,14 +3632,43 @@ bool Sema::CheckPPCBuiltinFunctionCall(const TargetInfo &TI, unsigned BuiltinID,
   case PPC::BI__builtin_altivec_dss:
     return SemaBuiltinConstantArgRange(TheCall, 0, 0, 3);
   case PPC::BI__builtin_tbegin:
-  case PPC::BI__builtin_tend: i = 0; l = 0; u = 1; break;
-  case PPC::BI__builtin_tsr: i = 0; l = 0; u = 7; break;
+  case PPC::BI__builtin_tend:
+    return SemaBuiltinConstantArgRange(TheCall, 0, 0, 1) ||
+           SemaFeatureCheck(*this, TheCall, "htm",
+                            diag::err_ppc_builtin_requires_htm);
+  case PPC::BI__builtin_tsr:
+    return SemaBuiltinConstantArgRange(TheCall, 0, 0, 7) ||
+           SemaFeatureCheck(*this, TheCall, "htm",
+                            diag::err_ppc_builtin_requires_htm);
   case PPC::BI__builtin_tabortwc:
-  case PPC::BI__builtin_tabortdc: i = 0; l = 0; u = 31; break;
+  case PPC::BI__builtin_tabortdc:
+    return SemaBuiltinConstantArgRange(TheCall, 0, 0, 31) ||
+           SemaFeatureCheck(*this, TheCall, "htm",
+                            diag::err_ppc_builtin_requires_htm);
   case PPC::BI__builtin_tabortwci:
   case PPC::BI__builtin_tabortdci:
-    return SemaBuiltinConstantArgRange(TheCall, 0, 0, 31) ||
-           SemaBuiltinConstantArgRange(TheCall, 2, 0, 31);
+    return SemaFeatureCheck(*this, TheCall, "htm",
+                            diag::err_ppc_builtin_requires_htm) ||
+           (SemaBuiltinConstantArgRange(TheCall, 0, 0, 31) ||
+            SemaBuiltinConstantArgRange(TheCall, 2, 0, 31));
+  case PPC::BI__builtin_tabort:
+  case PPC::BI__builtin_tcheck:
+  case PPC::BI__builtin_treclaim:
+  case PPC::BI__builtin_trechkpt:
+  case PPC::BI__builtin_tendall:
+  case PPC::BI__builtin_tresume:
+  case PPC::BI__builtin_tsuspend:
+  case PPC::BI__builtin_get_texasr:
+  case PPC::BI__builtin_get_texasru:
+  case PPC::BI__builtin_get_tfhar:
+  case PPC::BI__builtin_get_tfiar:
+  case PPC::BI__builtin_set_texasr:
+  case PPC::BI__builtin_set_texasru:
+  case PPC::BI__builtin_set_tfhar:
+  case PPC::BI__builtin_set_tfiar:
+  case PPC::BI__builtin_ttest:
+    return SemaFeatureCheck(*this, TheCall, "htm",
+                            diag::err_ppc_builtin_requires_htm);
   // According to GCC 'Basic PowerPC Built-in Functions Available on ISA 2.05',
   // __builtin_(un)pack_longdouble are available only if long double uses IBM
   // extended double representation.
@@ -5819,8 +5984,8 @@ ExprResult Sema::BuildAtomicExpr(SourceRange CallRange, SourceRange ExprRange,
                 ? 0
                 : 1);
 
-  if (ValType->isExtIntType()) {
-    Diag(Ptr->getExprLoc(), diag::err_atomic_builtin_ext_int_prohibit);
+  if (ValType->isBitIntType()) {
+    Diag(Ptr->getExprLoc(), diag::err_atomic_builtin_bit_int_prohibit);
     return ExprError();
   }
 
@@ -6217,11 +6382,11 @@ Sema::SemaBuiltinAtomicOverloaded(ExprResult TheCallResult) {
   // gracefully.
   TheCall->setType(ResultType);
 
-  // Prohibit use of _ExtInt with atomic builtins.
-  // The arguments would have already been converted to the first argument's
-  // type, so only need to check the first argument.
-  const auto *ExtIntValType = ValType->getAs<ExtIntType>();
-  if (ExtIntValType && !llvm::isPowerOf2_64(ExtIntValType->getNumBits())) {
+  // Prohibit problematic uses of bit-precise integer types with atomic
+  // builtins. The arguments would have already been converted to the first
+  // argument's type, so only need to check the first argument.
+  const auto *BitIntValType = ValType->getAs<BitIntType>();
+  if (BitIntValType && !llvm::isPowerOf2_64(BitIntValType->getNumBits())) {
     Diag(FirstArg->getExprLoc(), diag::err_atomic_builtin_ext_int_size);
     return ExprError();
   }
@@ -11249,7 +11414,7 @@ struct IntRange {
                         false/*NonNegative*/);
     }
 
-    if (const auto *EIT = dyn_cast<ExtIntType>(T))
+    if (const auto *EIT = dyn_cast<BitIntType>(T))
       return IntRange(EIT->getNumBits(), EIT->isUnsigned());
 
     const BuiltinType *BT = cast<BuiltinType>(T);
@@ -11275,7 +11440,7 @@ struct IntRange {
     if (const EnumType *ET = dyn_cast<EnumType>(T))
       T = C.getCanonicalType(ET->getDecl()->getIntegerType()).getTypePtr();
 
-    if (const auto *EIT = dyn_cast<ExtIntType>(T))
+    if (const auto *EIT = dyn_cast<BitIntType>(T))
       return IntRange(EIT->getNumBits(), EIT->isUnsigned());
 
     const BuiltinType *BT = cast<BuiltinType>(T);
@@ -16697,26 +16862,19 @@ static bool checkMathBuiltinElementType(Sema &S, SourceLocation Loc,
   return false;
 }
 
-bool Sema::SemaBuiltinElementwiseMathOneArg(CallExpr *TheCall) {
+bool Sema::PrepareBuiltinElementwiseMathOneArgCall(CallExpr *TheCall) {
   if (checkArgCount(*this, TheCall, 1))
     return true;
 
   ExprResult A = UsualUnaryConversions(TheCall->getArg(0));
-  SourceLocation ArgLoc = TheCall->getArg(0)->getBeginLoc();
   if (A.isInvalid())
     return true;
 
   TheCall->setArg(0, A.get());
   QualType TyA = A.get()->getType();
-  if (checkMathBuiltinElementType(*this, ArgLoc, TyA))
-    return true;
 
-  QualType EltTy = TyA;
-  if (auto *VecTy = EltTy->getAs<VectorType>())
-    EltTy = VecTy->getElementType();
-  if (EltTy->isUnsignedIntegerType())
-    return Diag(ArgLoc, diag::err_builtin_invalid_arg_type)
-           << 1 << /*signed integer or float ty*/ 3 << TyA;
+  if (checkMathBuiltinElementType(*this, A.get()->getBeginLoc(), TyA))
+    return true;
 
   TheCall->setType(TyA);
   return false;
@@ -16752,7 +16910,7 @@ bool Sema::SemaBuiltinElementwiseMath(CallExpr *TheCall) {
   return false;
 }
 
-bool Sema::SemaBuiltinReduceMath(CallExpr *TheCall) {
+bool Sema::PrepareBuiltinReduceMathOneArgCall(CallExpr *TheCall) {
   if (checkArgCount(*this, TheCall, 1))
     return true;
 
@@ -16761,14 +16919,6 @@ bool Sema::SemaBuiltinReduceMath(CallExpr *TheCall) {
     return true;
 
   TheCall->setArg(0, A.get());
-  const VectorType *TyA = A.get()->getType()->getAs<VectorType>();
-  if (!TyA) {
-    SourceLocation ArgLoc = TheCall->getArg(0)->getBeginLoc();
-    return Diag(ArgLoc, diag::err_builtin_invalid_arg_type)
-           << 1 << /* vector ty*/ 4 << A.get()->getType();
-  }
-
-  TheCall->setType(TyA->getElementType());
   return false;
 }
 
diff --git a/clang/lib/Sema/SemaCodeComplete.cpp b/clang/lib/Sema/SemaCodeComplete.cpp
index 083a67db7a91..93c07ccc891f 100644
--- a/clang/lib/Sema/SemaCodeComplete.cpp
+++ b/clang/lib/Sema/SemaCodeComplete.cpp
@@ -569,7 +569,6 @@ void PreferredTypeBuilder::enterMemAccess(Sema &S, SourceLocation Tok,
     return;
   // Keep the expected type, only update the location.
   ExpectedLoc = Tok;
-  return;
 }
 
 void PreferredTypeBuilder::enterUnary(Sema &S, SourceLocation Tok,
diff --git a/clang/lib/Sema/SemaCoroutine.cpp b/clang/lib/Sema/SemaCoroutine.cpp
index b999b08d1662..e89cecd08cca 100644
--- a/clang/lib/Sema/SemaCoroutine.cpp
+++ b/clang/lib/Sema/SemaCoroutine.cpp
@@ -237,9 +237,9 @@ static bool isValidCoroutineContext(Sema &S, SourceLocation Loc,
   // placeholder type shall not be a coroutine."
   if (FD->getReturnType()->isUndeducedType())
     DiagInvalid(DiagAutoRet);
-  // [dcl.fct.def.coroutine]p1: "The parameter-declaration-clause of the
-  // coroutine shall not terminate with an ellipsis that is not part of a
-  // parameter-declaration."
+  // [dcl.fct.def.coroutine]p1
+  // The parameter-declaration-clause of the coroutine shall not terminate with
+  // an ellipsis that is not part of a parameter-declaration.
   if (FD->isVariadic())
     DiagInvalid(DiagVarargs);
 
@@ -579,8 +579,12 @@ VarDecl *Sema::buildCoroutinePromise(SourceLocation Loc) {
                                    /*TopLevelOfInitList=*/false,
                                    /*TreatUnavailableAsInvalid=*/false);
 
-    // Attempt to initialize the promise type with the arguments.
-    // If that fails, fall back to the promise type's default constructor.
+    // [dcl.fct.def.coroutine]5.7
+    // promise-constructor-arguments is determined as follows: overload
+    // resolution is performed on a promise constructor call created by
+    // assembling an argument list  q_1 ... q_n . If a viable constructor is
+    // found ([over.match.viable]), then promise-constructor-arguments is ( q_1
+    // , ...,  q_n ), otherwise promise-constructor-arguments is empty.
     if (InitSeq) {
       ExprResult Result = InitSeq.Perform(*this, Entity, Kind, CtorArgExprs);
       if (Result.isInvalid()) {
@@ -648,6 +652,10 @@ static void checkNoThrow(Sema &S, const Stmt *E,
           return;
       }
       if (ThrowingDecls.empty()) {
+        // [dcl.fct.def.coroutine]p15
+        //   The expression co_­await promise.final_­suspend() shall not be
+        //   potentially-throwing ([except.spec]).
+        //
         // First time seeing an error, emit the error message.
         S.Diag(cast<FunctionDecl>(S.CurContext)->getLocation(),
                diag::err_coroutine_promise_final_suspend_requires_nothrow);
@@ -995,9 +1003,8 @@ static Expr *buildStdNoThrowDeclRef(Sema &S, SourceLocation Loc) {
   LookupResult Result(S, &S.PP.getIdentifierTable().get("nothrow"), Loc,
                       Sema::LookupOrdinaryName);
   if (!S.LookupQualifiedName(Result, Std)) {
-    // FIXME: <coroutine> should have been included already.
-    // If we require it to include <new> then this diagnostic is no longer
-    // needed.
+    // <coroutine> is not requred to include <new>, so we couldn't omit
+    // the check here.
     S.Diag(Loc, diag::err_implicit_coroutine_std_nothrow_type_not_found);
     return nullptr;
   }
@@ -1029,9 +1036,21 @@ static FunctionDecl *findDeleteForPromise(Sema &S, SourceLocation Loc,
   auto *PointeeRD = PromiseType->getAsCXXRecordDecl();
   assert(PointeeRD && "PromiseType must be a CxxRecordDecl type");
 
+  // [dcl.fct.def.coroutine]p12
+  // The deallocation function's name is looked up by searching for it in the
+  // scope of the promise type. If nothing is found, a search is performed in
+  // the global scope.
   if (S.FindDeallocationFunction(Loc, PointeeRD, DeleteName, OperatorDelete))
     return nullptr;
 
+  // FIXME: We didn't implement following selection:
+  // [dcl.fct.def.coroutine]p12
+  //   If both a usual deallocation function with only a pointer parameter and a
+  //   usual deallocation function with both a pointer parameter and a size
+  //   parameter are found, then the selected deallocation function shall be the
+  //   one with two parameters. Otherwise, the selected deallocation function
+  //   shall be the function with one parameter.
+
   if (!OperatorDelete) {
     // Look for a global declaration.
     const bool CanProvideSize = S.isCompleteType(Loc, PromiseType);
@@ -1062,8 +1081,8 @@ void Sema::CheckCompletedCoroutineBody(FunctionDecl *FD, Stmt *&Body) {
     return;
   }
 
-  // Coroutines [stmt.return]p1:
-  //   A return statement shall not appear in a coroutine.
+  // [stmt.return.coroutine]p1:
+  //   A coroutine shall not enclose a return statement ([stmt.return]).
   if (Fn->FirstReturnLoc.isValid()) {
     assert(Fn->FirstCoroutineStmtLoc.isValid() &&
                    "first coroutine location not set");
@@ -1164,12 +1183,15 @@ bool CoroutineStmtBuilder::makeReturnOnAllocFailure() {
   assert(!IsPromiseDependentType &&
          "cannot make statement while the promise type is dependent");
 
-  // [dcl.fct.def.coroutine]/8
-  // The unqualified-id get_return_object_on_allocation_failure is looked up in
-  // the scope of class P by class member access lookup (3.4.5). ...
-  // If an allocation function returns nullptr, ... the coroutine return value
-  // is obtained by a call to ... get_return_object_on_allocation_failure().
-
+  // [dcl.fct.def.coroutine]p10
+  //   If a search for the name get_­return_­object_­on_­allocation_­failure in
+  // the scope of the promise type ([class.member.lookup]) finds any
+  // declarations, then the result of a call to an allocation function used to
+  // obtain storage for the coroutine state is assumed to return nullptr if it
+  // fails to obtain storage, ... If the allocation function returns nullptr,
+  // ... and the return value is obtained by a call to
+  // T::get_­return_­object_­on_­allocation_­failure(), where T is the
+  // promise type.
   DeclarationName DN =
       S.PP.getIdentifierInfo("get_return_object_on_allocation_failure");
   LookupResult Found(S, DN, Loc, Sema::LookupMemberName);
@@ -1215,12 +1237,11 @@ bool CoroutineStmtBuilder::makeNewAndDeleteExpr() {
 
   const bool RequiresNoThrowAlloc = ReturnStmtOnAllocFailure != nullptr;
 
-  // [dcl.fct.def.coroutine]/7
-  // Lookup allocation functions using a parameter list composed of the
-  // requested size of the coroutine state being allocated, followed by
-  // the coroutine function's arguments. If a matching allocation function
-  // exists, use it. Otherwise, use an allocation function that just takes
-  // the requested size.
+  // According to [dcl.fct.def.coroutine]p9, Lookup allocation functions using a
+  // parameter list composed of the requested size of the coroutine state being
+  // allocated, followed by the coroutine function's arguments. If a matching
+  // allocation function exists, use it. Otherwise, use an allocation function
+  // that just takes the requested size.
 
   FunctionDecl *OperatorNew = nullptr;
   FunctionDecl *OperatorDelete = nullptr;
@@ -1228,21 +1249,32 @@ bool CoroutineStmtBuilder::makeNewAndDeleteExpr() {
   bool PassAlignment = false;
   SmallVector<Expr *, 1> PlacementArgs;
 
-  // [dcl.fct.def.coroutine]/7
-  // "The allocation function’s name is looked up in the scope of P.
-  // [...] If the lookup finds an allocation function in the scope of P,
-  // overload resolution is performed on a function call created by assembling
-  // an argument list. The first argument is the amount of space requested,
-  // and has type std::size_t. The lvalues p1 ... pn are the succeeding
-  // arguments."
+  // [dcl.fct.def.coroutine]p9
+  //   An implementation may need to allocate additional storage for a
+  //   coroutine.
+  // This storage is known as the coroutine state and is obtained by calling a
+  // non-array allocation function ([basic.stc.dynamic.allocation]). The
+  // allocation function's name is looked up by searching for it in the scope of
+  // the promise type.
+  // - If any declarations are found, overload resolution is performed on a
+  // function call created by assembling an argument list. The first argument is
+  // the amount of space requested, and has type std::size_t. The
+  // lvalues p1 ... pn are the succeeding arguments.
   //
   // ...where "p1 ... pn" are defined earlier as:
   //
-  // [dcl.fct.def.coroutine]/3
-  // "For a coroutine f that is a non-static member function, let P1 denote the
-  // type of the implicit object parameter (13.3.1) and P2 ... Pn be the types
-  // of the function parameters; otherwise let P1 ... Pn be the types of the
-  // function parameters. Let p1 ... pn be lvalues denoting those objects."
+  // [dcl.fct.def.coroutine]p3
+  //   The promise type of a coroutine is `std::coroutine_traits<R, P1, ...,
+  //   Pn>`
+  // , where R is the return type of the function, and `P1, ..., Pn` are the
+  // sequence of types of the non-object function parameters, preceded by the
+  // type of the object parameter ([dcl.fct]) if the coroutine is a non-static
+  // member function. [dcl.fct.def.coroutine]p4 In the following, p_i is an
+  // lvalue of type P_i, where p1 denotes the object parameter and p_i+1 denotes
+  // the i-th non-object function parameter for a non-static member function,
+  // and p_i denotes the i-th function parameter otherwise. For a non-static
+  // member function, q_1 is an lvalue that denotes *this; any other q_i is an
+  // lvalue that denotes the parameter copy corresponding to p_i.
   if (auto *MD = dyn_cast<CXXMethodDecl>(&FD)) {
     if (MD->isInstance() && !isLambdaCallOperator(MD)) {
       ExprResult ThisExpr = S.ActOnCXXThis(Loc);
@@ -1273,10 +1305,10 @@ bool CoroutineStmtBuilder::makeNewAndDeleteExpr() {
                             /*isArray*/ false, PassAlignment, PlacementArgs,
                             OperatorNew, UnusedResult, /*Diagnose*/ false);
 
-  // [dcl.fct.def.coroutine]/7
-  // "If no matching function is found, overload resolution is performed again
-  // on a function call created by passing just the amount of space required as
-  // an argument of type std::size_t."
+  // [dcl.fct.def.coroutine]p9
+  //   If no viable function is found ([over.match.viable]), overload resolution
+  // is performed again on a function call created by passing just the amount of
+  // space required as an argument of type std::size_t.
   if (!OperatorNew && !PlacementArgs.empty()) {
     PlacementArgs.clear();
     S.FindAllocationFunctions(Loc, SourceRange(), /*NewScope*/ Sema::AFS_Class,
@@ -1285,10 +1317,11 @@ bool CoroutineStmtBuilder::makeNewAndDeleteExpr() {
                               OperatorNew, UnusedResult, /*Diagnose*/ false);
   }
 
-  // [dcl.fct.def.coroutine]/7
-  // "The allocation function’s name is looked up in the scope of P. If this
-  // lookup fails, the allocation function’s name is looked up in the global
-  // scope."
+  // [dcl.fct.def.coroutine]p9
+  //   The allocation function's name is looked up by searching for it in the
+  // scope of the promise type.
+  // - If any declarations are found, ...
+  // - Otherwise, a search is performed in the global scope.
   if (!OperatorNew) {
     S.FindAllocationFunctions(Loc, SourceRange(), /*NewScope*/ Sema::AFS_Global,
                               /*DeleteScope*/ Sema::AFS_Both, PromiseType,
@@ -1328,8 +1361,12 @@ bool CoroutineStmtBuilder::makeNewAndDeleteExpr() {
     }
   }
 
-  if ((OperatorDelete = findDeleteForPromise(S, Loc, PromiseType)) == nullptr)
+  if ((OperatorDelete = findDeleteForPromise(S, Loc, PromiseType)) == nullptr) {
+    // FIXME: We should add an error here. According to:
+    // [dcl.fct.def.coroutine]p12
+    //   If no usual deallocation function is found, the program is ill-formed.
     return false;
+  }
 
   Expr *FramePtr =
       S.BuildBuiltinCallExpr(Loc, Builtin::BI__builtin_coro_frame, {});
@@ -1368,7 +1405,11 @@ bool CoroutineStmtBuilder::makeNewAndDeleteExpr() {
 
   SmallVector<Expr *, 2> DeleteArgs{CoroFree};
 
-  // Check if we need to pass the size.
+  // [dcl.fct.def.coroutine]p12
+  //   The selected deallocation function shall be called with the address of
+  //   the block of storage to be reclaimed as its first argument. If a
+  //   deallocation function with a parameter of type std::size_t is
+  //   used, the size of the block is passed as the corresponding argument.
   const auto *OpDeleteType =
       OpDeleteQualType.getTypePtr()->castAs<FunctionProtoType>();
   if (OpDeleteType->getNumParams() > 1)
@@ -1391,9 +1432,13 @@ bool CoroutineStmtBuilder::makeOnFallthrough() {
   assert(!IsPromiseDependentType &&
          "cannot make statement while the promise type is dependent");
 
-  // [dcl.fct.def.coroutine]/4
-  // The unqualified-ids 'return_void' and 'return_value' are looked up in
-  // the scope of class P. If both are found, the program is ill-formed.
+  // [dcl.fct.def.coroutine]/p6
+  // If searches for the names return_­void and return_­value in the scope of
+  // the promise type each find any declarations, the program is ill-formed.
+  // [Note 1: If return_­void is found, flowing off the end of a coroutine is
+  // equivalent to a co_­return with no operand. Otherwise, flowing off the end
+  // of a coroutine results in undefined behavior ([stmt.return.coroutine]). —
+  // end note]
   bool HasRVoid, HasRValue;
   LookupResult LRVoid =
       lookupMember(S, "return_void", PromiseRecordDecl, Loc, HasRVoid);
@@ -1414,18 +1459,20 @@ bool CoroutineStmtBuilder::makeOnFallthrough() {
         << LRValue.getLookupName();
     return false;
   } else if (!HasRVoid && !HasRValue) {
-    // FIXME: The PDTS currently specifies this case as UB, not ill-formed.
-    // However we still diagnose this as an error since until the PDTS is fixed.
-    S.Diag(FD.getLocation(),
-           diag::err_coroutine_promise_requires_return_function)
-        << PromiseRecordDecl;
-    S.Diag(PromiseRecordDecl->getLocation(), diag::note_defined_here)
-        << PromiseRecordDecl;
-    return false;
+    // We need to set 'Fallthrough'. Otherwise the other analysis part might
+    // think the coroutine has defined a return_value method. So it might emit
+    // **false** positive warning. e.g.,
+    //
+    //    promise_without_return_func foo() {
+    //        co_await something();
+    //    }
+    //
+    // Then AnalysisBasedWarning would emit a warning about `foo()` lacking a
+    // co_return statements, which isn't correct.
+    Fallthrough = S.ActOnNullStmt(PromiseRecordDecl->getLocation());
+    if (Fallthrough.isInvalid())
+      return false;
   } else if (HasRVoid) {
-    // If the unqualified-id return_void is found, flowing off the end of a
-    // coroutine is equivalent to a co_return with no operand. Otherwise,
-    // flowing off the end of a coroutine results in undefined behavior.
     Fallthrough = S.BuildCoreturnStmt(FD.getLocation(), nullptr,
                                       /*IsImplicit*/false);
     Fallthrough = S.ActOnFinishFullStmt(Fallthrough.get());
@@ -1481,8 +1528,9 @@ bool CoroutineStmtBuilder::makeOnException() {
 }
 
 bool CoroutineStmtBuilder::makeReturnObject() {
-  // Build implicit 'p.get_return_object()' expression and form initialization
-  // of return type from it.
+  // [dcl.fct.def.coroutine]p7
+  // The expression promise.get_­return_­object() is used to initialize the
+  // returned reference or prvalue result object of a call to a coroutine.
   ExprResult ReturnObject =
       buildPromiseCall(S, Fn.CoroutinePromise, Loc, "get_return_object", None);
   if (ReturnObject.isInvalid())
@@ -1620,6 +1668,12 @@ bool Sema::buildCoroutineParameterMoves(SourceLocation Loc) {
   if (!ScopeInfo->CoroutineParameterMoves.empty())
     return false;
 
+  // [dcl.fct.def.coroutine]p13
+  //   When a coroutine is invoked, after initializing its parameters
+  //   ([expr.call]), a copy is created for each coroutine parameter. For a
+  //   parameter of type cv T, the copy is a variable of type cv T with
+  //   automatic storage duration that is direct-initialized from an xvalue of
+  //   type T referring to the parameter.
   for (auto *PD : FD->parameters()) {
     if (PD->getType()->isDependentType())
       continue;
@@ -1636,7 +1690,9 @@ bool Sema::buildCoroutineParameterMoves(SourceLocation Loc) {
       CExpr = castForMoving(*this, PDRefExpr.get());
     else
       CExpr = PDRefExpr.get();
-
+    // [dcl.fct.def.coroutine]p13
+    //   The initialization and destruction of each parameter copy occurs in the
+    //   context of the called coroutine.
     auto D = buildVarDecl(*this, Loc, PD->getType(), PD->getIdentifier());
     AddInitializerToDecl(D, CExpr, /*DirectInit=*/true);
 
@@ -1661,43 +1717,53 @@ ClassTemplateDecl *Sema::lookupCoroutineTraits(SourceLocation KwLoc,
                                                SourceLocation FuncLoc,
                                                NamespaceDecl *&Namespace) {
   if (!StdCoroutineTraitsCache) {
-    NamespaceDecl *CoroNamespace = getStdNamespace();
-    LookupResult Result(*this, &PP.getIdentifierTable().get("coroutine_traits"),
-                        FuncLoc, LookupOrdinaryName);
-
-    if (!CoroNamespace || !LookupQualifiedName(Result, CoroNamespace)) {
-      /// Look up in namespace std::experimental, for compatibility.
-      /// TODO: Remove this extra lookup when <experimental/coroutine> is
-      /// removed.
-      CoroNamespace = lookupStdExperimentalNamespace();
-      if (!CoroNamespace || !LookupQualifiedName(Result, CoroNamespace)) {
-        Diag(KwLoc, diag::err_implied_coroutine_type_not_found)
-            << "std::coroutine_traits";
-        return nullptr;
-      }
+    // Because coroutines moved from std::experimental in the TS to std in
+    // C++20, we look in both places to give users time to transition their
+    // TS-specific code to C++20.  Diagnostics are given when the TS usage is
+    // discovered.
+    // TODO: Become stricter when <experimental/coroutine> is removed.
+
+    auto const &TraitIdent = PP.getIdentifierTable().get("coroutine_traits");
+
+    NamespaceDecl *StdSpace = getStdNamespace();
+    LookupResult ResStd(*this, &TraitIdent, FuncLoc, LookupOrdinaryName);
+    bool InStd = StdSpace && LookupQualifiedName(ResStd, StdSpace);
+
+    NamespaceDecl *ExpSpace = lookupStdExperimentalNamespace();
+    LookupResult ResExp(*this, &TraitIdent, FuncLoc, LookupOrdinaryName);
+    bool InExp = ExpSpace && LookupQualifiedName(ResExp, ExpSpace);
+
+    if (!InStd && !InExp) {
+      // The goggles, they found nothing!
+      Diag(KwLoc, diag::err_implied_coroutine_type_not_found)
+          << "std::coroutine_traits";
+      return nullptr;
+    }
+
+    if (!InStd) {
+      // Found only in std::experimental.
       Diag(KwLoc, diag::warn_deprecated_coroutine_namespace)
           << "coroutine_traits";
-    } else {
-      /// When we found coroutine_traits in std namespace. Make sure there is no
-      /// misleading definition in std::experimental namespace.
-      NamespaceDecl *ExpNamespace = lookupStdExperimentalNamespace();
-      LookupResult ExpResult(*this,
-                             &PP.getIdentifierTable().get("coroutine_traits"),
-                             FuncLoc, LookupOrdinaryName);
-      if (ExpNamespace && LookupQualifiedName(ExpResult, ExpNamespace)) {
-        Diag(KwLoc,
-             diag::err_mixed_use_std_and_experimental_namespace_for_coroutine);
-        return nullptr;
-      }
+    } else if (InExp) {
+      // Found in std and std::experimental.
+      Diag(KwLoc,
+           diag::err_mixed_use_std_and_experimental_namespace_for_coroutine);
+      Diag(KwLoc, diag::warn_deprecated_coroutine_namespace)
+          << "coroutine_traits";
+      return nullptr;
     }
 
+    // Prefer ::std to std::experimental.
+    auto &Result = InStd ? ResStd : ResExp;
+    CoroTraitsNamespaceCache = InStd ? StdSpace : ExpSpace;
+
+    // coroutine_traits is required to be a class template.
     if (!(StdCoroutineTraitsCache = Result.getAsSingle<ClassTemplateDecl>())) {
       Result.suppressDiagnostics();
       NamedDecl *Found = *Result.begin();
       Diag(Found->getLocation(), diag::err_malformed_std_coroutine_traits);
       return nullptr;
     }
-    CoroTraitsNamespaceCache = CoroNamespace;
   }
   Namespace = CoroTraitsNamespaceCache;
   return StdCoroutineTraitsCache;
diff --git a/clang/lib/Sema/SemaDecl.cpp b/clang/lib/Sema/SemaDecl.cpp
index 7be71ca49ea2..3c58f1d19c04 100644
--- a/clang/lib/Sema/SemaDecl.cpp
+++ b/clang/lib/Sema/SemaDecl.cpp
@@ -372,6 +372,7 @@ ParsedType Sema::getTypeName(const IdentifierInfo &II, SourceLocation NameLoc,
   }
 
   NamedDecl *IIDecl = nullptr;
+  UsingShadowDecl *FoundUsingShadow = nullptr;
   switch (Result.getResultKind()) {
   case LookupResult::NotFound:
   case LookupResult::NotFoundInCurrentInstantiation:
@@ -441,8 +442,10 @@ ParsedType Sema::getTypeName(const IdentifierInfo &II, SourceLocation NameLoc,
           (AllowDeducedTemplate && getAsTypeTemplateDecl(RealRes))) {
         if (!IIDecl ||
             // Make the selection of the recovery decl deterministic.
-            RealRes->getLocation() < IIDecl->getLocation())
+            RealRes->getLocation() < IIDecl->getLocation()) {
           IIDecl = RealRes;
+          FoundUsingShadow = dyn_cast<UsingShadowDecl>(*Res);
+        }
       }
     }
 
@@ -465,6 +468,7 @@ ParsedType Sema::getTypeName(const IdentifierInfo &II, SourceLocation NameLoc,
 
   case LookupResult::Found:
     IIDecl = Result.getFoundDecl();
+    FoundUsingShadow = dyn_cast<UsingShadowDecl>(*Result.begin());
     break;
   }
 
@@ -491,14 +495,20 @@ ParsedType Sema::getTypeName(const IdentifierInfo &II, SourceLocation NameLoc,
     (void)DiagnoseUseOfDecl(IDecl, NameLoc);
     if (!HasTrailingDot)
       T = Context.getObjCInterfaceType(IDecl);
+    FoundUsingShadow = nullptr; // FIXME: Target must be a TypeDecl.
   } else if (auto *UD = dyn_cast<UnresolvedUsingIfExistsDecl>(IIDecl)) {
     (void)DiagnoseUseOfDecl(UD, NameLoc);
     // Recover with 'int'
     T = Context.IntTy;
+    FoundUsingShadow = nullptr;
   } else if (AllowDeducedTemplate) {
-    if (auto *TD = getAsTypeTemplateDecl(IIDecl))
+    if (auto *TD = getAsTypeTemplateDecl(IIDecl)) {
+      // FIXME: TemplateName should include FoundUsingShadow sugar.
       T = Context.getDeducedTemplateSpecializationType(TemplateName(TD),
                                                        QualType(), false);
+      // Don't wrap in a further UsingType.
+      FoundUsingShadow = nullptr;
+    }
   }
 
   if (T.isNull()) {
@@ -507,6 +517,9 @@ ParsedType Sema::getTypeName(const IdentifierInfo &II, SourceLocation NameLoc,
     return nullptr;
   }
 
+  if (FoundUsingShadow)
+    T = Context.getUsingType(FoundUsingShadow, T);
+
   // NOTE: avoid constructing an ElaboratedType(Loc) if this is a
   // constructor or destructor name (in such a case, the scope specifier
   // will be attached to the enclosing Expr or Decl node).
@@ -843,21 +856,6 @@ static bool isTagTypeWithMissingTag(Sema &SemaRef, LookupResult &Result,
   return false;
 }
 
-/// Build a ParsedType for a simple-type-specifier with a nested-name-specifier.
-static ParsedType buildNestedType(Sema &S, CXXScopeSpec &SS,
-                                  QualType T, SourceLocation NameLoc) {
-  ASTContext &Context = S.Context;
-
-  TypeLocBuilder Builder;
-  Builder.pushTypeSpec(T).setNameLoc(NameLoc);
-
-  T = S.getElaboratedType(ETK_None, SS, T);
-  ElaboratedTypeLoc ElabTL = Builder.push<ElaboratedTypeLoc>(T);
-  ElabTL.setElaboratedKeywordLoc(SourceLocation());
-  ElabTL.setQualifierLoc(SS.getWithLocInContext(Context));
-  return S.CreateParsedType(T, Builder.getTypeSourceInfo(Context, T));
-}
-
 Sema::NameClassification Sema::ClassifyName(Scope *S, CXXScopeSpec &SS,
                                             IdentifierInfo *&Name,
                                             SourceLocation NameLoc,
@@ -1134,14 +1132,28 @@ Corrected:
                          : NameClassification::TypeTemplate(Template);
   }
 
+  auto BuildTypeFor = [&](TypeDecl *Type, NamedDecl *Found) {
+    QualType T = Context.getTypeDeclType(Type);
+    if (const auto *USD = dyn_cast<UsingShadowDecl>(Found))
+      T = Context.getUsingType(USD, T);
+
+    if (SS.isEmpty()) // No elaborated type, trivial location info
+      return ParsedType::make(T);
+
+    TypeLocBuilder Builder;
+    Builder.pushTypeSpec(T).setNameLoc(NameLoc);
+    T = getElaboratedType(ETK_None, SS, T);
+    ElaboratedTypeLoc ElabTL = Builder.push<ElaboratedTypeLoc>(T);
+    ElabTL.setElaboratedKeywordLoc(SourceLocation());
+    ElabTL.setQualifierLoc(SS.getWithLocInContext(Context));
+    return CreateParsedType(T, Builder.getTypeSourceInfo(Context, T));
+  };
+
   NamedDecl *FirstDecl = (*Result.begin())->getUnderlyingDecl();
   if (TypeDecl *Type = dyn_cast<TypeDecl>(FirstDecl)) {
     DiagnoseUseOfDecl(Type, NameLoc);
     MarkAnyDeclReferenced(Type->getLocation(), Type, /*OdrUse=*/false);
-    QualType T = Context.getTypeDeclType(Type);
-    if (SS.isNotEmpty())
-      return buildNestedType(*this, SS, T, NameLoc);
-    return ParsedType::make(T);
+    return BuildTypeFor(Type, *Result.begin());
   }
 
   ObjCInterfaceDecl *Class = dyn_cast<ObjCInterfaceDecl>(FirstDecl);
@@ -1190,10 +1202,7 @@ Corrected:
       isTagTypeWithMissingTag(*this, Result, S, SS, Name, NameLoc)) {
     TypeDecl *Type = Result.getAsSingle<TypeDecl>();
     DiagnoseUseOfDecl(Type, NameLoc);
-    QualType T = Context.getTypeDeclType(Type);
-    if (SS.isNotEmpty())
-      return buildNestedType(*this, SS, T, NameLoc);
-    return ParsedType::make(T);
+    return BuildTypeFor(Type, *Result.begin());
   }
 
   // If we already know which single declaration is referenced, just annotate
@@ -8504,7 +8513,14 @@ static NamedDecl *DiagnoseInvalidRedeclaration(
         << NewFD->getParamDecl(Idx - 1)->getType();
     } else if (FDisConst != NewFDisConst) {
       SemaRef.Diag(FD->getLocation(), diag::note_member_def_close_const_match)
-          << NewFDisConst << FD->getSourceRange().getEnd();
+          << NewFDisConst << FD->getSourceRange().getEnd()
+          << (NewFDisConst
+                  ? FixItHint::CreateRemoval(ExtraArgs.D.getFunctionTypeInfo()
+                                                 .getConstQualifierLoc())
+                  : FixItHint::CreateInsertion(ExtraArgs.D.getFunctionTypeInfo()
+                                                   .getRParenLoc()
+                                                   .getLocWithOffset(1),
+                                               " const"));
     } else
       SemaRef.Diag(FD->getLocation(),
                    IsMember ? diag::note_member_def_close_match
@@ -9195,6 +9211,7 @@ Sema::ActOnFunctionDeclarator(Scope *S, Declarator &D, DeclContext *DC,
             << Name << RemoveRange
             << FixItHint::CreateRemoval(RemoveRange)
             << FixItHint::CreateInsertion(InsertLoc, "<>");
+          Invalid = true;
         }
       }
     } else {
@@ -15359,7 +15376,7 @@ bool Sema::CheckEnumUnderlyingType(TypeSourceInfo *TI) {
     if (BT->isInteger())
       return false;
 
-  if (T->isExtIntType())
+  if (T->isBitIntType())
     return false;
 
   return Diag(UnderlyingLoc, diag::err_enum_invalid_underlying) << T;
@@ -18277,7 +18294,7 @@ static void CheckForDuplicateEnumValues(Sema &S, ArrayRef<Decl *> Elements,
 
     // Emit one note for each of the remaining enum constants with
     // the same value.
-    for (auto *ECD : llvm::make_range(Vec->begin() + 1, Vec->end()))
+    for (auto *ECD : llvm::drop_begin(*Vec))
       S.Diag(ECD->getLocation(), diag::note_duplicate_element)
         << ECD << toString(ECD->getInitVal(), 10)
         << ECD->getSourceRange();
diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
index 4df8687aff89..b6bd2e69629d 100644
--- a/clang/lib/Sema/SemaDeclAttr.cpp
+++ b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -1001,6 +1001,84 @@ public:
 };
 }
 
+static void handleDiagnoseAsBuiltinAttr(Sema &S, Decl *D,
+                                        const ParsedAttr &AL) {
+  const auto *DeclFD = cast<FunctionDecl>(D);
+
+  if (const auto *MethodDecl = dyn_cast<CXXMethodDecl>(DeclFD))
+    if (!MethodDecl->isStatic()) {
+      S.Diag(AL.getLoc(), diag::err_attribute_no_member_function) << AL;
+      return;
+    }
+
+  auto DiagnoseType = [&](unsigned Index, AttributeArgumentNType T) {
+    SourceLocation Loc = [&]() {
+      auto Union = AL.getArg(Index - 1);
+      if (Union.is<Expr *>())
+        return Union.get<Expr *>()->getBeginLoc();
+      return Union.get<IdentifierLoc *>()->Loc;
+    }();
+
+    S.Diag(Loc, diag::err_attribute_argument_n_type) << AL << Index << T;
+  };
+
+  FunctionDecl *AttrFD = [&]() -> FunctionDecl * {
+    if (!AL.isArgExpr(0))
+      return nullptr;
+    auto *F = dyn_cast_or_null<DeclRefExpr>(AL.getArgAsExpr(0));
+    if (!F)
+      return nullptr;
+    return dyn_cast_or_null<FunctionDecl>(F->getFoundDecl());
+  }();
+
+  if (!AttrFD || !AttrFD->getBuiltinID(true)) {
+    DiagnoseType(1, AANT_ArgumentBuiltinFunction);
+    return;
+  }
+
+  if (AttrFD->getNumParams() != AL.getNumArgs() - 1) {
+    S.Diag(AL.getLoc(), diag::err_attribute_wrong_number_arguments_for)
+        << AL << AttrFD << AttrFD->getNumParams();
+    return;
+  }
+
+  SmallVector<unsigned, 8> Indices;
+
+  for (unsigned I = 1; I < AL.getNumArgs(); ++I) {
+    if (!AL.isArgExpr(I)) {
+      DiagnoseType(I + 1, AANT_ArgumentIntegerConstant);
+      return;
+    }
+
+    const Expr *IndexExpr = AL.getArgAsExpr(I);
+    uint32_t Index;
+
+    if (!checkUInt32Argument(S, AL, IndexExpr, Index, I + 1, false))
+      return;
+
+    if (Index > DeclFD->getNumParams()) {
+      S.Diag(AL.getLoc(), diag::err_attribute_bounds_for_function)
+          << AL << Index << DeclFD << DeclFD->getNumParams();
+      return;
+    }
+
+    QualType T1 = AttrFD->getParamDecl(I - 1)->getType();
+    QualType T2 = DeclFD->getParamDecl(Index - 1)->getType();
+
+    if (T1.getCanonicalType().getUnqualifiedType() !=
+        T2.getCanonicalType().getUnqualifiedType()) {
+      S.Diag(IndexExpr->getBeginLoc(), diag::err_attribute_parameter_types)
+          << AL << Index << DeclFD << T2 << I << AttrFD << T1;
+      return;
+    }
+
+    Indices.push_back(Index - 1);
+  }
+
+  D->addAttr(::new (S.Context) DiagnoseAsBuiltinAttr(
+      S.Context, AL, AttrFD, Indices.data(), Indices.size()));
+}
+
 static void handleDiagnoseIfAttr(Sema &S, Decl *D, const ParsedAttr &AL) {
   S.Diag(AL.getLoc(), diag::ext_clang_diagnose_if);
 
@@ -4502,7 +4580,7 @@ void Sema::AddModeAttr(Decl *D, const AttributeCommonInfo &CI,
     return;
   }
   bool IntegralOrAnyEnumType = (OldElemTy->isIntegralOrEnumerationType() &&
-                                !OldElemTy->isExtIntType()) ||
+                                !OldElemTy->isBitIntType()) ||
                                OldElemTy->getAs<EnumType>();
 
   if (!OldElemTy->getAs<BuiltinType>() && !OldElemTy->isComplexType() &&
@@ -8159,6 +8237,9 @@ static void ProcessDeclAttribute(Sema &S, Scope *scope, Decl *D,
   case ParsedAttr::AT_DiagnoseIf:
     handleDiagnoseIfAttr(S, D, AL);
     break;
+  case ParsedAttr::AT_DiagnoseAsBuiltin:
+    handleDiagnoseAsBuiltinAttr(S, D, AL);
+    break;
   case ParsedAttr::AT_NoBuiltin:
     handleNoBuiltinAttr(S, D, AL);
     break;
diff --git a/clang/lib/Sema/SemaDeclCXX.cpp b/clang/lib/Sema/SemaDeclCXX.cpp
index 2658e9698688..01f0079198c7 100644
--- a/clang/lib/Sema/SemaDeclCXX.cpp
+++ b/clang/lib/Sema/SemaDeclCXX.cpp
@@ -8435,9 +8435,6 @@ bool Sema::CheckExplicitlyDefaultedComparison(Scope *S, FunctionDecl *FD,
                                               DefaultedComparisonKind DCK) {
   assert(DCK != DefaultedComparisonKind::None && "not a defaulted comparison");
 
-  CXXRecordDecl *RD = dyn_cast<CXXRecordDecl>(FD->getLexicalDeclContext());
-  assert(RD && "defaulted comparison is not defaulted in a class");
-
   // Perform any unqualified lookups we're going to need to default this
   // function.
   if (S) {
@@ -8455,43 +8452,17 @@ bool Sema::CheckExplicitlyDefaultedComparison(Scope *S, FunctionDecl *FD,
   //       const C&, or
   //    -- a friend of C having two parameters of type const C& or two
   //       parameters of type C.
-  QualType ExpectedParmType1 = Context.getRecordType(RD);
-  QualType ExpectedParmType2 =
-      Context.getLValueReferenceType(ExpectedParmType1.withConst());
-  if (isa<CXXMethodDecl>(FD))
-    ExpectedParmType1 = ExpectedParmType2;
-  for (const ParmVarDecl *Param : FD->parameters()) {
-    if (!Param->getType()->isDependentType() &&
-        !Context.hasSameType(Param->getType(), ExpectedParmType1) &&
-        !Context.hasSameType(Param->getType(), ExpectedParmType2)) {
-      // Don't diagnose an implicit 'operator=='; we will have diagnosed the
-      // corresponding defaulted 'operator<=>' already.
-      if (!FD->isImplicit()) {
-        Diag(FD->getLocation(), diag::err_defaulted_comparison_param)
-            << (int)DCK << Param->getType() << ExpectedParmType1
-            << !isa<CXXMethodDecl>(FD)
-            << ExpectedParmType2 << Param->getSourceRange();
-      }
-      return true;
-    }
-  }
-  if (FD->getNumParams() == 2 &&
-      !Context.hasSameType(FD->getParamDecl(0)->getType(),
-                           FD->getParamDecl(1)->getType())) {
-    if (!FD->isImplicit()) {
-      Diag(FD->getLocation(), diag::err_defaulted_comparison_param_mismatch)
-          << (int)DCK
-          << FD->getParamDecl(0)->getType()
-          << FD->getParamDecl(0)->getSourceRange()
-          << FD->getParamDecl(1)->getType()
-          << FD->getParamDecl(1)->getSourceRange();
-    }
-    return true;
-  }
 
-  // ... non-static const member ...
-  if (auto *MD = dyn_cast<CXXMethodDecl>(FD)) {
+  CXXRecordDecl *RD = dyn_cast<CXXRecordDecl>(FD->getLexicalDeclContext());
+  bool IsMethod = isa<CXXMethodDecl>(FD);
+  if (IsMethod) {
+    auto *MD = cast<CXXMethodDecl>(FD);
     assert(!MD->isStatic() && "comparison function cannot be a static member");
+
+    // If we're out-of-class, this is the class we're comparing.
+    if (!RD)
+      RD = MD->getParent();
+
     if (!MD->isConst()) {
       SourceLocation InsertLoc;
       if (FunctionTypeLoc Loc = MD->getFunctionTypeLoc())
@@ -8500,7 +8471,7 @@ bool Sema::CheckExplicitlyDefaultedComparison(Scope *S, FunctionDecl *FD,
       // corresponding defaulted 'operator<=>' already.
       if (!MD->isImplicit()) {
         Diag(MD->getLocation(), diag::err_defaulted_comparison_non_const)
-          << (int)DCK << FixItHint::CreateInsertion(InsertLoc, " const");
+            << (int)DCK << FixItHint::CreateInsertion(InsertLoc, " const");
       }
 
       // Add the 'const' to the type to recover.
@@ -8510,9 +8481,98 @@ bool Sema::CheckExplicitlyDefaultedComparison(Scope *S, FunctionDecl *FD,
       MD->setType(Context.getFunctionType(FPT->getReturnType(),
                                           FPT->getParamTypes(), EPI));
     }
-  } else {
-    // A non-member function declared in a class must be a friend.
+  }
+
+  if (FD->getNumParams() != (IsMethod ? 1 : 2)) {
+    // Let's not worry about using a variadic template pack here -- who would do
+    // such a thing?
+    Diag(FD->getLocation(), diag::err_defaulted_comparison_num_args)
+        << int(IsMethod) << int(DCK);
+    return true;
+  }
+
+  const ParmVarDecl *KnownParm = nullptr;
+  for (const ParmVarDecl *Param : FD->parameters()) {
+    QualType ParmTy = Param->getType();
+    if (ParmTy->isDependentType())
+      continue;
+    if (!KnownParm) {
+      auto CTy = ParmTy;
+      // Is it `T const &`?
+      bool Ok = !IsMethod;
+      QualType ExpectedTy;
+      if (RD)
+        ExpectedTy = Context.getRecordType(RD);
+      if (auto *Ref = CTy->getAs<ReferenceType>()) {
+        CTy = Ref->getPointeeType();
+        if (RD)
+          ExpectedTy.addConst();
+        Ok = true;
+      }
+
+      // Is T a class?
+      if (!Ok) {
+      } else if (RD) {
+        if (!RD->isDependentType() && !Context.hasSameType(CTy, ExpectedTy))
+          Ok = false;
+      } else if (auto *CRD = CTy->getAsRecordDecl()) {
+        RD = cast<CXXRecordDecl>(CRD);
+      } else {
+        Ok = false;
+      }
+
+      if (Ok) {
+        KnownParm = Param;
+      } else {
+        // Don't diagnose an implicit 'operator=='; we will have diagnosed the
+        // corresponding defaulted 'operator<=>' already.
+        if (!FD->isImplicit()) {
+          if (RD) {
+            QualType PlainTy = Context.getRecordType(RD);
+            QualType RefTy =
+                Context.getLValueReferenceType(PlainTy.withConst());
+            Diag(FD->getLocation(), diag::err_defaulted_comparison_param)
+                << int(DCK) << ParmTy << RefTy << int(!IsMethod) << PlainTy
+                << Param->getSourceRange();
+          } else {
+            assert(!IsMethod && "should know expected type for method");
+            Diag(FD->getLocation(),
+                 diag::err_defaulted_comparison_param_unknown)
+                << int(DCK) << ParmTy << Param->getSourceRange();
+          }
+        }
+        return true;
+      }
+    } else if (!Context.hasSameType(KnownParm->getType(), ParmTy)) {
+      Diag(FD->getLocation(), diag::err_defaulted_comparison_param_mismatch)
+          << int(DCK) << KnownParm->getType() << KnownParm->getSourceRange()
+          << ParmTy << Param->getSourceRange();
+      return true;
+    }
+  }
+
+  assert(RD && "must have determined class");
+  if (IsMethod) {
+  } else if (isa<CXXRecordDecl>(FD->getLexicalDeclContext())) {
+    // In-class, must be a friend decl.
     assert(FD->getFriendObjectKind() && "expected a friend declaration");
+  } else {
+    // Out of class, require the defaulted comparison to be a friend (of a
+    // complete type).
+    if (RequireCompleteType(FD->getLocation(), Context.getRecordType(RD),
+                            diag::err_defaulted_comparison_not_friend, int(DCK),
+                            int(1)))
+      return true;
+
+    if (llvm::find_if(RD->friends(), [&](const FriendDecl *F) {
+          return FD->getCanonicalDecl() ==
+                 F->getFriendDecl()->getCanonicalDecl();
+        }) == RD->friends().end()) {
+      Diag(FD->getLocation(), diag::err_defaulted_comparison_not_friend)
+          << int(DCK) << int(0) << RD;
+      Diag(RD->getCanonicalDecl()->getLocation(), diag::note_declared_at);
+      return true;
+    }
   }
 
   // C++2a [class.eq]p1, [class.rel]p1:
@@ -8670,7 +8730,10 @@ void Sema::DefineDefaultedComparison(SourceLocation UseLoc, FunctionDecl *FD,
 
   {
     // Build and set up the function body.
-    CXXRecordDecl *RD = cast<CXXRecordDecl>(FD->getLexicalParent());
+    // The first parameter has type maybe-ref-to maybe-const T, use that to get
+    // the type of the class being compared.
+    auto PT = FD->getParamDecl(0)->getType();
+    CXXRecordDecl *RD = PT.getNonReferenceType()->getAsCXXRecordDecl();
     SourceLocation BodyLoc =
         FD->getEndLoc().isValid() ? FD->getEndLoc() : FD->getLocation();
     StmtResult Body =
@@ -16146,6 +16209,23 @@ Decl *Sema::ActOnStartLinkageSpecification(Scope *S, SourceLocation ExternLoc,
   LinkageSpecDecl *D = LinkageSpecDecl::Create(Context, CurContext, ExternLoc,
                                                LangStr->getExprLoc(), Language,
                                                LBraceLoc.isValid());
+
+  /// C++ [module.unit]p7.2.3
+  /// - Otherwise, if the declaration
+  ///   - ...
+  ///   - ...
+  ///   - appears within a linkage-specification,
+  ///   it is attached to the global module.
+  ///
+  /// If the declaration is already in global module fragment, we don't
+  /// need to attach it again.
+  if (getLangOpts().CPlusPlusModules && isCurrentModulePurview()) {
+    Module *GlobalModule =
+        PushGlobalModuleFragment(ExternLoc, /*IsImplicit=*/true);
+    D->setModuleOwnershipKind(Decl::ModuleOwnershipKind::ModulePrivate);
+    D->setLocalOwningModule(GlobalModule);
+  }
+
   CurContext->addDecl(D);
   PushDeclContext(S, D);
   return D;
@@ -16162,6 +16242,14 @@ Decl *Sema::ActOnFinishLinkageSpecification(Scope *S,
     LinkageSpecDecl* LSDecl = cast<LinkageSpecDecl>(LinkageSpec);
     LSDecl->setRBraceLoc(RBraceLoc);
   }
+
+  // If the current module doesn't has Parent, it implies that the
+  // LinkageSpec isn't in the module created by itself. So we don't
+  // need to pop it.
+  if (getLangOpts().CPlusPlusModules && getCurrentModule() &&
+      getCurrentModule()->isGlobalModule() && getCurrentModule()->Parent)
+    PopGlobalModuleFragment();
+
   PopDeclContext();
   return LinkageSpec;
 }
@@ -17155,13 +17243,6 @@ void Sema::SetDeclDefaulted(Decl *Dcl, SourceLocation DefaultLoc) {
     return;
   }
 
-  if (DefKind.isComparison() &&
-      !isa<CXXRecordDecl>(FD->getLexicalDeclContext())) {
-    Diag(FD->getLocation(), diag::err_defaulted_comparison_out_of_class)
-        << (int)DefKind.asComparison();
-    return;
-  }
-
   // Issue compatibility warning. We already warned if the operator is
   // 'operator<=>' when parsing the '<=>' token.
   if (DefKind.isComparison() &&
@@ -17183,31 +17264,40 @@ void Sema::SetDeclDefaulted(Decl *Dcl, SourceLocation DefaultLoc) {
   // that we've marked it as defaulted.
   FD->setWillHaveBody(false);
 
-  // If this definition appears within the record, do the checking when
-  // the record is complete. This is always the case for a defaulted
-  // comparison.
-  if (DefKind.isComparison())
-    return;
-  auto *MD = cast<CXXMethodDecl>(FD);
-
-  const FunctionDecl *Primary = FD;
-  if (const FunctionDecl *Pattern = FD->getTemplateInstantiationPattern())
-    // Ask the template instantiation pattern that actually had the
-    // '= default' on it.
-    Primary = Pattern;
+  if (DefKind.isComparison()) {
+    // If this comparison's defaulting occurs within the definition of its
+    // lexical class context, we have to do the checking when complete.
+    if (auto const *RD = dyn_cast<CXXRecordDecl>(FD->getLexicalDeclContext()))
+      if (!RD->isCompleteDefinition())
+        return;
+  }
 
-  // If the method was defaulted on its first declaration, we will have
+  // If this member fn was defaulted on its first declaration, we will have
   // already performed the checking in CheckCompletedCXXClass. Such a
   // declaration doesn't trigger an implicit definition.
-  if (Primary->getCanonicalDecl()->isDefaulted())
-    return;
+  if (isa<CXXMethodDecl>(FD)) {
+    const FunctionDecl *Primary = FD;
+    if (const FunctionDecl *Pattern = FD->getTemplateInstantiationPattern())
+      // Ask the template instantiation pattern that actually had the
+      // '= default' on it.
+      Primary = Pattern;
+    if (Primary->getCanonicalDecl()->isDefaulted())
+      return;
+  }
 
-  // FIXME: Once we support defining comparisons out of class, check for a
-  // defaulted comparison here.
-  if (CheckExplicitlyDefaultedSpecialMember(MD, DefKind.asSpecialMember()))
-    MD->setInvalidDecl();
-  else
-    DefineDefaultedFunction(*this, MD, DefaultLoc);
+  if (DefKind.isComparison()) {
+    if (CheckExplicitlyDefaultedComparison(nullptr, FD, DefKind.asComparison()))
+      FD->setInvalidDecl();
+    else
+      DefineDefaultedComparison(DefaultLoc, FD, DefKind.asComparison());
+  } else {
+    auto *MD = cast<CXXMethodDecl>(FD);
+
+    if (CheckExplicitlyDefaultedSpecialMember(MD, DefKind.asSpecialMember()))
+      MD->setInvalidDecl();
+    else
+      DefineDefaultedFunction(*this, MD, DefaultLoc);
+  }
 }
 
 static void SearchForReturnInStmt(Sema &Self, Stmt *S) {
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
index b305d4e5b92f..d32b3f217aa0 100644
--- a/clang/lib/Sema/SemaExpr.cpp
+++ b/clang/lib/Sema/SemaExpr.cpp
@@ -4384,7 +4384,7 @@ static void captureVariablyModifiedType(ASTContext &Context, QualType T,
     case Type::ObjCObjectPointer:
     case Type::ObjCTypeParam:
     case Type::Pipe:
-    case Type::ExtInt:
+    case Type::BitInt:
       llvm_unreachable("type class is never variably-modified!");
     case Type::Adjusted:
       T = cast<AdjustedType>(Ty)->getOriginalType();
@@ -4443,6 +4443,9 @@ static void captureVariablyModifiedType(ASTContext &Context, QualType T,
     case Type::Decltype:
       T = cast<DecltypeType>(Ty)->desugar();
       break;
+    case Type::Using:
+      T = cast<UsingType>(Ty)->desugar();
+      break;
     case Type::Auto:
     case Type::DeducedTemplateSpecialization:
       T = cast<DeducedType>(Ty)->getDeducedType();
@@ -8388,9 +8391,10 @@ QualType Sema::CheckConditionalOperands(ExprResult &Cond, ExprResult &LHS,
   // If both operands have arithmetic type, do the usual arithmetic conversions
   // to find a common type: C99 6.5.15p3,5.
   if (LHSTy->isArithmeticType() && RHSTy->isArithmeticType()) {
-    // Disallow invalid arithmetic conversions, such as those between ExtInts of
-    // different sizes, or between ExtInts and other types.
-    if (ResTy.isNull() && (LHSTy->isExtIntType() || RHSTy->isExtIntType())) {
+    // Disallow invalid arithmetic conversions, such as those between bit-
+    // precise integers types of different sizes, or between a bit-precise
+    // integer and another type.
+    if (ResTy.isNull() && (LHSTy->isBitIntType() || RHSTy->isBitIntType())) {
       Diag(QuestionLoc, diag::err_typecheck_cond_incompatible_operands)
           << LHSTy << RHSTy << LHS.get()->getSourceRange()
           << RHS.get()->getSourceRange();
@@ -10974,7 +10978,7 @@ static void DiagnoseBadShiftValues(Sema& S, ExprResult &LHS, ExprResult &RHS,
 
   QualType LHSExprType = LHS.get()->getType();
   uint64_t LeftSize = S.Context.getTypeSize(LHSExprType);
-  if (LHSExprType->isExtIntType())
+  if (LHSExprType->isBitIntType())
     LeftSize = S.Context.getIntWidth(LHSExprType);
   else if (LHSExprType->isFixedPointType()) {
     auto FXSema = S.Context.getFixedPointSemantics(LHSExprType);
@@ -12260,27 +12264,32 @@ QualType Sema::GetSignedVectorType(QualType V) {
   if (isa<ExtVectorType>(VTy)) {
     if (TypeSize == Context.getTypeSize(Context.CharTy))
       return Context.getExtVectorType(Context.CharTy, VTy->getNumElements());
-    else if (TypeSize == Context.getTypeSize(Context.ShortTy))
+    if (TypeSize == Context.getTypeSize(Context.ShortTy))
       return Context.getExtVectorType(Context.ShortTy, VTy->getNumElements());
-    else if (TypeSize == Context.getTypeSize(Context.IntTy))
+    if (TypeSize == Context.getTypeSize(Context.IntTy))
       return Context.getExtVectorType(Context.IntTy, VTy->getNumElements());
-    else if (TypeSize == Context.getTypeSize(Context.LongTy))
+    if (TypeSize == Context.getTypeSize(Context.Int128Ty))
+      return Context.getExtVectorType(Context.Int128Ty, VTy->getNumElements());
+    if (TypeSize == Context.getTypeSize(Context.LongTy))
       return Context.getExtVectorType(Context.LongTy, VTy->getNumElements());
     assert(TypeSize == Context.getTypeSize(Context.LongLongTy) &&
            "Unhandled vector element size in vector compare");
     return Context.getExtVectorType(Context.LongLongTy, VTy->getNumElements());
   }
 
+  if (TypeSize == Context.getTypeSize(Context.Int128Ty))
+    return Context.getVectorType(Context.Int128Ty, VTy->getNumElements(),
+                                 VectorType::GenericVector);
   if (TypeSize == Context.getTypeSize(Context.LongLongTy))
     return Context.getVectorType(Context.LongLongTy, VTy->getNumElements(),
                                  VectorType::GenericVector);
-  else if (TypeSize == Context.getTypeSize(Context.LongTy))
+  if (TypeSize == Context.getTypeSize(Context.LongTy))
     return Context.getVectorType(Context.LongTy, VTy->getNumElements(),
                                  VectorType::GenericVector);
-  else if (TypeSize == Context.getTypeSize(Context.IntTy))
+  if (TypeSize == Context.getTypeSize(Context.IntTy))
     return Context.getVectorType(Context.IntTy, VTy->getNumElements(),
                                  VectorType::GenericVector);
-  else if (TypeSize == Context.getTypeSize(Context.ShortTy))
+  if (TypeSize == Context.getTypeSize(Context.ShortTy))
     return Context.getVectorType(Context.ShortTy, VTy->getNumElements(),
                                  VectorType::GenericVector);
   assert(TypeSize == Context.getTypeSize(Context.CharTy) &&
diff --git a/clang/lib/Sema/SemaExprCXX.cpp b/clang/lib/Sema/SemaExprCXX.cpp
index d25f329f85e4..54f0242d2ca1 100644
--- a/clang/lib/Sema/SemaExprCXX.cpp
+++ b/clang/lib/Sema/SemaExprCXX.cpp
@@ -1346,7 +1346,7 @@ bool Sema::CheckCXXThisCapture(SourceLocation Loc, const bool Explicit,
   // implicitly capturing the *enclosing  object* by reference (see loop
   // above)).
   assert((!ByCopy ||
-          dyn_cast<LambdaScopeInfo>(FunctionScopes[MaxFunctionScopesIndex])) &&
+          isa<LambdaScopeInfo>(FunctionScopes[MaxFunctionScopesIndex])) &&
          "Only a lambda can capture the enclosing object (referred to by "
          "*this) by copy");
   QualType ThisTy = getCurrentThisType();
diff --git a/clang/lib/Sema/SemaLookup.cpp b/clang/lib/Sema/SemaLookup.cpp
index 0711e6d89383..635e93ba8460 100644
--- a/clang/lib/Sema/SemaLookup.cpp
+++ b/clang/lib/Sema/SemaLookup.cpp
@@ -2935,7 +2935,7 @@ addAssociatedClassesAndNamespaces(AssociatedLookup &Result, QualType Ty) {
     case Type::ExtVector:
     case Type::ConstantMatrix:
     case Type::Complex:
-    case Type::ExtInt:
+    case Type::BitInt:
       break;
 
     // Non-deduced auto types only get here for error cases.
diff --git a/clang/lib/Sema/SemaModule.cpp b/clang/lib/Sema/SemaModule.cpp
index af95b1a93cc4..a4b9f3c242c1 100644
--- a/clang/lib/Sema/SemaModule.cpp
+++ b/clang/lib/Sema/SemaModule.cpp
@@ -68,15 +68,8 @@ Sema::ActOnGlobalModuleFragmentDecl(SourceLocation ModuleLoc) {
 
   // We start in the global module; all those declarations are implicitly
   // module-private (though they do not have module linkage).
-  auto &Map = PP.getHeaderSearchInfo().getModuleMap();
-  auto *GlobalModule = Map.createGlobalModuleFragmentForModuleUnit(ModuleLoc);
-  assert(GlobalModule && "module creation should not fail");
-
-  // Enter the scope of the global module.
-  ModuleScopes.push_back({});
-  ModuleScopes.back().BeginLoc = ModuleLoc;
-  ModuleScopes.back().Module = GlobalModule;
-  VisibleModules.setVisible(GlobalModule, ModuleLoc);
+  Module *GlobalModule =
+      PushGlobalModuleFragment(ModuleLoc, /*IsImplicit=*/false);
 
   // All declarations created from now on are owned by the global module.
   auto *TU = Context.getTranslationUnitDecl();
@@ -390,11 +383,18 @@ DeclResult Sema::ActOnModuleImport(SourceLocation StartLoc,
   if (!ModuleScopes.empty())
     Context.addModuleInitializer(ModuleScopes.back().Module, Import);
 
-  // Re-export the module if needed.
   if (!ModuleScopes.empty() && ModuleScopes.back().ModuleInterface) {
+    // Re-export the module if the imported module is exported.
+    // Note that we don't need to add re-exported module to Imports field
+    // since `Exports` implies the module is imported already.
     if (ExportLoc.isValid() || getEnclosingExportDecl(Import))
       getCurrentModule()->Exports.emplace_back(Mod, false);
+    else
+      getCurrentModule()->Imports.insert(Mod);
   } else if (ExportLoc.isValid()) {
+    // [module.interface]p1:
+    // An export-declaration shall inhabit a namespace scope and appear in the
+    // purview of a module interface unit.
     Diag(ExportLoc, diag::err_export_not_in_module_interface);
   }
 
@@ -708,3 +708,26 @@ Decl *Sema::ActOnFinishExportDecl(Scope *S, Decl *D, SourceLocation RBraceLoc) {
 
   return D;
 }
+
+Module *Sema::PushGlobalModuleFragment(SourceLocation BeginLoc,
+                                       bool IsImplicit) {
+  ModuleMap &Map = PP.getHeaderSearchInfo().getModuleMap();
+  Module *GlobalModule =
+      Map.createGlobalModuleFragmentForModuleUnit(BeginLoc, getCurrentModule());
+  assert(GlobalModule && "module creation should not fail");
+
+  // Enter the scope of the global module.
+  ModuleScopes.push_back({BeginLoc, GlobalModule,
+                          /*ModuleInterface=*/false,
+                          /*ImplicitGlobalModuleFragment=*/IsImplicit,
+                          /*VisibleModuleSet*/{}});
+  VisibleModules.setVisible(GlobalModule, BeginLoc);
+
+  return GlobalModule;
+}
+
+void Sema::PopGlobalModuleFragment() {
+  assert(!ModuleScopes.empty() && getCurrentModule()->isGlobalModule() &&
+         "left the wrong module scope, which is not global module fragment");
+  ModuleScopes.pop_back();
+}
diff --git a/clang/lib/Sema/SemaOpenMP.cpp b/clang/lib/Sema/SemaOpenMP.cpp
index 22ae5f59d41b..ba0481874577 100644
--- a/clang/lib/Sema/SemaOpenMP.cpp
+++ b/clang/lib/Sema/SemaOpenMP.cpp
@@ -255,14 +255,14 @@ private:
     return &Stack.back().first[Size - 1];
   }
   const SharingMapTy *getTopOfStackOrNull() const {
-    return const_cast<DSAStackTy&>(*this).getTopOfStackOrNull();
+    return const_cast<DSAStackTy &>(*this).getTopOfStackOrNull();
   }
   SharingMapTy &getTopOfStack() {
     assert(!isStackEmpty() && "no current directive");
     return *getTopOfStackOrNull();
   }
   const SharingMapTy &getTopOfStack() const {
-    return const_cast<DSAStackTy&>(*this).getTopOfStack();
+    return const_cast<DSAStackTy &>(*this).getTopOfStack();
   }
 
   SharingMapTy *getSecondOnStackOrNull() {
@@ -272,7 +272,7 @@ private:
     return &Stack.back().first[Size - 2];
   }
   const SharingMapTy *getSecondOnStackOrNull() const {
-    return const_cast<DSAStackTy&>(*this).getSecondOnStackOrNull();
+    return const_cast<DSAStackTy &>(*this).getSecondOnStackOrNull();
   }
 
   /// Get the stack element at a certain level (previously returned by
@@ -286,7 +286,7 @@ private:
     return Stack.back().first[Level];
   }
   const SharingMapTy &getStackElemAtLevel(unsigned Level) const {
-    return const_cast<DSAStackTy&>(*this).getStackElemAtLevel(Level);
+    return const_cast<DSAStackTy &>(*this).getStackElemAtLevel(Level);
   }
 
   DSAVarData getDSA(const_iterator &Iter, ValueDecl *D) const;
@@ -354,9 +354,7 @@ public:
     const SharingMapTy *Top = getTopOfStackOrNull();
     return Top && Top->BodyComplete;
   }
-  void setBodyComplete() {
-    getTopOfStack().BodyComplete = true;
-  }
+  void setBodyComplete() { getTopOfStack().BodyComplete = true; }
 
   bool isForceVarCapturing() const { return ForceCapturing; }
   void setForceVarCapturing(bool V) { ForceCapturing = V; }
@@ -392,6 +390,7 @@ public:
   class ParentDirectiveScope {
     DSAStackTy &Self;
     bool Active;
+
   public:
     ParentDirectiveScope(DSAStackTy &Self, bool Activate)
         : Self(Self), Active(false) {
@@ -433,8 +432,7 @@ public:
   }
   /// Marks (or clears) declaration as possibly loop counter.
   void resetPossibleLoopCounter(const Decl *D = nullptr) {
-    getTopOfStack().PossiblyLoopCounter =
-        D ? D->getCanonicalDecl() : D;
+    getTopOfStack().PossiblyLoopCounter = D ? D->getCanonicalDecl() : D;
   }
   /// Gets the possible loop counter decl.
   const Decl *getPossiblyLoopCunter() const {
@@ -631,13 +629,10 @@ public:
   }
 
   /// Add requires decl to internal vector
-  void addRequiresDecl(OMPRequiresDecl *RD) {
-    RequiresDecls.push_back(RD);
-  }
+  void addRequiresDecl(OMPRequiresDecl *RD) { RequiresDecls.push_back(RD); }
 
   /// Checks if the defined 'requires' directive has specified type of clause.
-  template <typename ClauseType>
-  bool hasRequiresDeclWithClause() const {
+  template <typename ClauseType> bool hasRequiresDeclWithClause() const {
     return llvm::any_of(RequiresDecls, [](const OMPRequiresDecl *D) {
       return llvm::any_of(D->clauselists(), [](const OMPClause *C) {
         return isa<ClauseType>(C);
@@ -680,9 +675,7 @@ public:
 
   /// Returns the location of the first encountered atomic directive in the
   /// module.
-  SourceLocation getAtomicDirectiveLoc() const {
-    return AtomicLocation;
-  }
+  SourceLocation getAtomicDirectiveLoc() const { return AtomicLocation; }
 
   // Return previously encountered target region locations.
   ArrayRef<SourceLocation> getEncounteredTargetLocs() const {
@@ -706,8 +699,7 @@ public:
   }
   /// Set default data mapping attribute to Modifier:Kind
   void setDefaultDMAAttr(OpenMPDefaultmapClauseModifier M,
-                         OpenMPDefaultmapClauseKind Kind,
-                         SourceLocation Loc) {
+                         OpenMPDefaultmapClauseKind Kind, SourceLocation Loc) {
     DefaultmapInfo &DMI = getTopOfStack().DefaultmapMap[Kind];
     DMI.ImplicitBehavior = M;
     DMI.SLoc = Loc;
@@ -749,12 +741,10 @@ public:
                                    : getStackElemAtLevel(Level).DefaultAttr;
   }
   DefaultDataSharingAttributes getDefaultDSA() const {
-    return isStackEmpty() ? DSA_unspecified
-                          : getTopOfStack().DefaultAttr;
+    return isStackEmpty() ? DSA_unspecified : getTopOfStack().DefaultAttr;
   }
   SourceLocation getDefaultDSALocation() const {
-    return isStackEmpty() ? SourceLocation()
-                          : getTopOfStack().DefaultAttrLoc;
+    return isStackEmpty() ? SourceLocation() : getTopOfStack().DefaultAttrLoc;
   }
   OpenMPDefaultmapClauseModifier
   getDefaultmapModifier(OpenMPDefaultmapClauseKind Kind) const {
@@ -1457,8 +1447,7 @@ void DSAStackTy::addTaskgroupReductionData(const ValueDecl *D, SourceRange SR,
          "Additional reduction info may be specified only once for reduction "
          "items.");
   ReductionData.set(BOK, SR);
-  Expr *&TaskgroupReductionRef =
-      getTopOfStack().TaskgroupReductionRef;
+  Expr *&TaskgroupReductionRef = getTopOfStack().TaskgroupReductionRef;
   if (!TaskgroupReductionRef) {
     VarDecl *VD = buildVarDecl(SemaRef, SR.getBegin(),
                                SemaRef.Context.VoidPtrTy, ".task_red.");
@@ -1483,8 +1472,7 @@ void DSAStackTy::addTaskgroupReductionData(const ValueDecl *D, SourceRange SR,
          "Additional reduction info may be specified only once for reduction "
          "items.");
   ReductionData.set(ReductionRef, SR);
-  Expr *&TaskgroupReductionRef =
-      getTopOfStack().TaskgroupReductionRef;
+  Expr *&TaskgroupReductionRef = getTopOfStack().TaskgroupReductionRef;
   if (!TaskgroupReductionRef) {
     VarDecl *VD = buildVarDecl(SemaRef, SR.getBegin(),
                                SemaRef.Context.VoidPtrTy, ".task_red.");
@@ -1595,10 +1583,9 @@ static bool rejectConstNotMutableType(Sema &SemaRef, const ValueDecl *D,
   ASTContext &Context = SemaRef.getASTContext();
   bool IsClassType;
   if (isConstNotMutableType(SemaRef, Type, AcceptIfMutable, &IsClassType)) {
-    unsigned Diag = ListItemNotVar
-                        ? diag::err_omp_const_list_item
-                        : IsClassType ? diag::err_omp_const_not_mutable_variable
-                                      : diag::err_omp_const_variable;
+    unsigned Diag = ListItemNotVar ? diag::err_omp_const_list_item
+                    : IsClassType  ? diag::err_omp_const_not_mutable_variable
+                                   : diag::err_omp_const_variable;
     SemaRef.Diag(ELoc, Diag) << getOpenMPClauseName(CKind);
     if (!ListItemNotVar && D) {
       const VarDecl *VD = dyn_cast<VarDecl>(D);
@@ -1658,8 +1645,7 @@ const DSAStackTy::DSAVarData DSAStackTy::getTopDSA(ValueDecl *D,
         });
     if (IterTarget != end()) {
       const_iterator ParentIterTarget = IterTarget + 1;
-      for (const_iterator Iter = begin();
-           Iter != ParentIterTarget; ++Iter) {
+      for (const_iterator Iter = begin(); Iter != ParentIterTarget; ++Iter) {
         if (isOpenMPLocal(VD, Iter)) {
           DVar.RefExpr =
               buildDeclRefExpr(SemaRef, VD, D->getType().getNonReferenceType(),
@@ -1677,9 +1663,9 @@ const DSAStackTy::DSAVarData DSAStackTy::getTopDSA(ValueDecl *D,
           return DVar;
         }
         const_iterator End = end();
-        if (!SemaRef.isOpenMPCapturedByRef(
-                D, std::distance(ParentIterTarget, End),
-                /*OpenMPCaptureLevel=*/0)) {
+        if (!SemaRef.isOpenMPCapturedByRef(D,
+                                           std::distance(ParentIterTarget, End),
+                                           /*OpenMPCaptureLevel=*/0)) {
           DVar.RefExpr =
               buildDeclRefExpr(SemaRef, VD, D->getType().getNonReferenceType(),
                                IterTarget->ConstructLoc);
@@ -1891,9 +1877,7 @@ void Sema::InitDataSharingAttributesStack() {
 
 #define DSAStack static_cast<DSAStackTy *>(VarDataSharingAttributesStack)
 
-void Sema::pushOpenMPFunctionRegion() {
-  DSAStack->pushFunction();
-}
+void Sema::pushOpenMPFunctionRegion() { DSAStack->pushFunction(); }
 
 void Sema::popOpenMPFunctionRegion(const FunctionScopeInfo *OldFSI) {
   DSAStack->popFunction(OldFSI);
@@ -2070,8 +2054,8 @@ bool Sema::isOpenMPCapturedByRef(const ValueDecl *D, unsigned Level,
 
     DSAStack->checkMappableExprComponentListsForDeclAtLevel(
         D, Level,
-        [&IsVariableUsedInMapClause, &IsVariableAssociatedWithSection, D](
-            OMPClauseMappableExprCommon::MappableExprComponentListRef
+        [&IsVariableUsedInMapClause, &IsVariableAssociatedWithSection,
+         D](OMPClauseMappableExprCommon::MappableExprComponentListRef
                 MapExprComponents,
             OpenMPClauseKind WhereFoundClauseKind) {
           // Only the map clause information influences how a variable is
@@ -2248,7 +2232,7 @@ VarDecl *Sema::isOpenMPCapturedDecl(ValueDecl *D, bool CheckScopeInfo,
     bool OpenMPFound = false;
     for (unsigned I = StopAt + 1; I > 0; --I) {
       FunctionScopeInfo *FSI = FunctionScopes[I - 1];
-      if(!isa<CapturingScopeInfo>(FSI))
+      if (!isa<CapturingScopeInfo>(FSI))
         return nullptr;
       if (auto *RSI = dyn_cast<CapturedRegionScopeInfo>(FSI))
         if (RSI->CapRegionKind == CR_OpenMP) {
@@ -2322,9 +2306,7 @@ void Sema::startOpenMPCXXRangeFor() {
 OpenMPClauseKind Sema::isOpenMPPrivateDecl(ValueDecl *D, unsigned Level,
                                            unsigned CapLevel) const {
   assert(LangOpts.OpenMP && "OpenMP is not allowed");
-  if (DSAStack->hasExplicitDirective(
-          [](OpenMPDirectiveKind K) { return isOpenMPTaskingDirective(K); },
-          Level)) {
+  if (DSAStack->hasExplicitDirective(isOpenMPTaskingDirective, Level)) {
     bool IsTriviallyCopyable =
         D->getType().getNonReferenceType().isTriviallyCopyableType(Context) &&
         !D->getType()
@@ -2351,8 +2333,7 @@ OpenMPClauseKind Sema::isOpenMPPrivateDecl(ValueDecl *D, unsigned Level,
     }
   }
   if (isOpenMPLoopDirective(DSAStack->getCurrentDirective())) {
-    if (DSAStack->getAssociatedLoops() > 0 &&
-        !DSAStack->isLoopStarted()) {
+    if (DSAStack->getAssociatedLoops() > 0 && !DSAStack->isLoopStarted()) {
       DSAStack->resetPossibleLoopCounter(D);
       DSAStack->loopStart();
       return OMPC_private;
@@ -2519,16 +2500,16 @@ void Sema::finalizeOpenMPDelayedAnalysis(const FunctionDecl *Caller,
         << HostDevTy;
     return;
   }
-      if (!LangOpts.OpenMPIsDevice && DevTy &&
-          *DevTy == OMPDeclareTargetDeclAttr::DT_NoHost) {
-        // Diagnose nohost function called during host codegen.
-        StringRef NoHostDevTy = getOpenMPSimpleClauseTypeName(
-            OMPC_device_type, OMPC_DEVICE_TYPE_nohost);
-        Diag(Loc, diag::err_omp_wrong_device_function_call) << NoHostDevTy << 1;
-        Diag(*OMPDeclareTargetDeclAttr::getLocation(FD),
-             diag::note_omp_marked_device_type_here)
-            << NoHostDevTy;
-      }
+  if (!LangOpts.OpenMPIsDevice && DevTy &&
+      *DevTy == OMPDeclareTargetDeclAttr::DT_NoHost) {
+    // Diagnose nohost function called during host codegen.
+    StringRef NoHostDevTy = getOpenMPSimpleClauseTypeName(
+        OMPC_device_type, OMPC_DEVICE_TYPE_nohost);
+    Diag(Loc, diag::err_omp_wrong_device_function_call) << NoHostDevTy << 1;
+    Diag(*OMPDeclareTargetDeclAttr::getLocation(FD),
+         diag::note_omp_marked_device_type_here)
+        << NoHostDevTy;
+  }
 }
 
 void Sema::StartOpenMPDSABlock(OpenMPDirectiveKind DKind,
@@ -2779,7 +2760,6 @@ public:
   std::unique_ptr<CorrectionCandidateCallback> clone() override {
     return std::make_unique<VarDeclFilterCCC>(*this);
   }
-
 };
 
 class VarOrFuncDeclFilterCCC final : public CorrectionCandidateCallback {
@@ -3151,9 +3131,10 @@ applyOMPAllocateAttribute(Sema &S, VarDecl *VD,
     ML->DeclarationMarkedOpenMPAllocate(VD, A);
 }
 
-Sema::DeclGroupPtrTy Sema::ActOnOpenMPAllocateDirective(
-    SourceLocation Loc, ArrayRef<Expr *> VarList,
-    ArrayRef<OMPClause *> Clauses, DeclContext *Owner) {
+Sema::DeclGroupPtrTy
+Sema::ActOnOpenMPAllocateDirective(SourceLocation Loc, ArrayRef<Expr *> VarList,
+                                   ArrayRef<OMPClause *> Clauses,
+                                   DeclContext *Owner) {
   assert(Clauses.size() <= 2 && "Expected at most two clauses.");
   Expr *Alignment = nullptr;
   Expr *Allocator = nullptr;
@@ -3500,7 +3481,8 @@ public:
       return;
     if (auto *VD = dyn_cast<VarDecl>(E->getDecl())) {
       // Check the datasharing rules for the expressions in the clauses.
-      if (!CS) {
+      if (!CS || (isa<OMPCapturedExprDecl>(VD) && !CS->capturesVariable(VD) &&
+                  !Stack->getTopDSA(VD, /*FromParent=*/false).RefExpr)) {
         if (auto *CED = dyn_cast<OMPCapturedExprDecl>(VD))
           if (!CED->hasAttr<OMPCaptureNoInitAttr>()) {
             Visit(CED->getInit());
@@ -3819,6 +3801,10 @@ public:
   }
   void VisitOMPExecutableDirective(OMPExecutableDirective *S) {
     for (OMPClause *C : S->clauses()) {
+      // Skip analysis of arguments of private clauses for task|target
+      // directives.
+      if (isa_and_nonnull<OMPPrivateClause>(C))
+        continue;
       // Skip analysis of arguments of implicitly defined firstprivate clause
       // for task|target directives.
       // Skip analysis of arguments of implicitly defined map clause for target
@@ -3841,6 +3827,18 @@ public:
     VisitStmt(S);
   }
 
+  void VisitCallExpr(CallExpr *S) {
+    for (Stmt *C : S->arguments()) {
+      if (C) {
+        // Check implicitly captured variables in the task-based directives to
+        // check if they must be firstprivatized.
+        Visit(C);
+      }
+    }
+    if (Expr *Callee = S->getCallee())
+      if (auto *CE = dyn_cast<MemberExpr>(Callee->IgnoreParenImpCasts()))
+        Visit(CE->getBase());
+  }
   void VisitStmt(Stmt *S) {
     for (Stmt *C : S->children()) {
       if (C) {
@@ -5089,8 +5087,8 @@ static std::pair<ValueDecl *, bool> getPrivateItem(Sema &S, Expr *&RefExpr,
        !isa<CXXThisExpr>(ME->getBase()->IgnoreParenImpCasts()) ||
        !isa<FieldDecl>(ME->getMemberDecl()))) {
     if (IsArrayExpr != NoArrayExpr) {
-      S.Diag(ELoc, diag::err_omp_expected_base_var_name) << IsArrayExpr
-                                                         << ERange;
+      S.Diag(ELoc, diag::err_omp_expected_base_var_name)
+          << IsArrayExpr << ERange;
     } else {
       S.Diag(ELoc,
              AllowArraySection
@@ -5134,8 +5132,7 @@ static void checkAllocateClauses(Sema &S, DSAStackTy *Stack,
          "Expected non-dependent context.");
   auto AllocateRange =
       llvm::make_filter_range(Clauses, OMPAllocateClause::classof);
-  llvm::DenseMap<CanonicalDeclPtr<Decl>, CanonicalDeclPtr<VarDecl>>
-      DeclToCopy;
+  llvm::DenseMap<CanonicalDeclPtr<Decl>, CanonicalDeclPtr<VarDecl>> DeclToCopy;
   auto PrivateRange = llvm::make_filter_range(Clauses, [](const OMPClause *C) {
     return isOpenMPPrivate(C->getClauseKind());
   });
@@ -6018,7 +6015,7 @@ StmtResult Sema::ActOnOpenMPExecutableDirective(
     break;
   case OMPD_parallel_master:
     Res = ActOnOpenMPParallelMasterDirective(ClausesWithImplicit, AStmt,
-                                               StartLoc, EndLoc);
+                                             StartLoc, EndLoc);
     AllowedNameModifiers.push_back(OMPD_parallel);
     break;
   case OMPD_parallel_sections:
@@ -6357,6 +6354,7 @@ StmtResult Sema::ActOnOpenMPExecutableDirective(
       case OMPC_write:
       case OMPC_update:
       case OMPC_capture:
+      case OMPC_compare:
       case OMPC_seq_cst:
       case OMPC_acq_rel:
       case OMPC_acquire:
@@ -7626,10 +7624,10 @@ bool OpenMPIterationSpaceChecker::setStep(Expr *NewStep, bool Subtract) {
     // != with increment is treated as <; != with decrement is treated as >
     if (!TestIsLessOp.hasValue())
       TestIsLessOp = IsConstPos || (IsUnsigned && !Subtract);
-    if (UB && (IsConstZero ||
-               (TestIsLessOp.getValue() ?
-                  (IsConstNeg || (IsUnsigned && Subtract)) :
-                  (IsConstPos || (IsUnsigned && !Subtract))))) {
+    if (UB &&
+        (IsConstZero || (TestIsLessOp.getValue()
+                             ? (IsConstNeg || (IsUnsigned && Subtract))
+                             : (IsConstPos || (IsUnsigned && !Subtract))))) {
       SemaRef.Diag(NewStep->getExprLoc(),
                    diag::err_omp_loop_incr_not_compatible)
           << LCDecl << TestIsLessOp.getValue() << NewStep->getSourceRange();
@@ -8068,11 +8066,13 @@ calculateNumIters(Sema &SemaRef, Scope *S, SourceLocation DefaultLoc,
     return nullptr;
   llvm::APSInt LRes, SRes;
   bool IsLowerConst = false, IsStepConst = false;
-  if (Optional<llvm::APSInt> Res = Lower->getIntegerConstantExpr(SemaRef.Context)) {
+  if (Optional<llvm::APSInt> Res =
+          Lower->getIntegerConstantExpr(SemaRef.Context)) {
     LRes = *Res;
     IsLowerConst = true;
   }
-  if (Optional<llvm::APSInt> Res = Step->getIntegerConstantExpr(SemaRef.Context)) {
+  if (Optional<llvm::APSInt> Res =
+          Step->getIntegerConstantExpr(SemaRef.Context)) {
     SRes = *Res;
     IsStepConst = true;
   }
@@ -8110,7 +8110,8 @@ calculateNumIters(Sema &SemaRef, Scope *S, SourceLocation DefaultLoc,
   }
   llvm::APSInt URes;
   bool IsUpperConst = false;
-  if (Optional<llvm::APSInt> Res = Upper->getIntegerConstantExpr(SemaRef.Context)) {
+  if (Optional<llvm::APSInt> Res =
+          Upper->getIntegerConstantExpr(SemaRef.Context)) {
     URes = *Res;
     IsUpperConst = true;
   }
@@ -8566,10 +8567,12 @@ Expr *OpenMPIterationSpaceChecker::buildPreCond(
   // TODO: this can be improved by calculating min/max values but not sure that
   // it will be very effective.
   if (CondDependOnLC || InitDependOnLC)
-    return SemaRef.PerformImplicitConversion(
-        SemaRef.ActOnIntegerConstant(SourceLocation(), 1).get(),
-        SemaRef.Context.BoolTy, /*Action=*/Sema::AA_Casting,
-        /*AllowExplicit=*/true).get();
+    return SemaRef
+        .PerformImplicitConversion(
+            SemaRef.ActOnIntegerConstant(SourceLocation(), 1).get(),
+            SemaRef.Context.BoolTy, /*Action=*/Sema::AA_Casting,
+            /*AllowExplicit=*/true)
+        .get();
 
   // Try to build LB <op> UB, where <op> is <, >, <=, or >=.
   Sema::TentativeAnalysisScope Trap(SemaRef);
@@ -8579,12 +8582,11 @@ Expr *OpenMPIterationSpaceChecker::buildPreCond(
   if (!NewLB.isUsable() || !NewUB.isUsable())
     return nullptr;
 
-  ExprResult CondExpr =
-      SemaRef.BuildBinOp(S, DefaultLoc,
-                         TestIsLessOp.getValue() ?
-                           (TestIsStrictOp ? BO_LT : BO_LE) :
-                           (TestIsStrictOp ? BO_GT : BO_GE),
-                         NewLB.get(), NewUB.get());
+  ExprResult CondExpr = SemaRef.BuildBinOp(
+      S, DefaultLoc,
+      TestIsLessOp.getValue() ? (TestIsStrictOp ? BO_LT : BO_LE)
+                              : (TestIsStrictOp ? BO_GT : BO_GE),
+      NewLB.get(), NewUB.get());
   if (CondExpr.isUsable()) {
     if (!SemaRef.Context.hasSameUnqualifiedType(CondExpr.get()->getType(),
                                                 SemaRef.Context.BoolTy))
@@ -8804,6 +8806,9 @@ static bool checkOpenMPIterationSpace(
   }
   assert(((For && For->getBody()) || (CXXFor && CXXFor->getBody())) &&
          "No loop body.");
+  // Postpone analysis in dependent contexts for ranged for loops.
+  if (CXXFor && SemaRef.CurContext->isDependentContext())
+    return false;
 
   OpenMPIterationSpaceChecker ISC(SemaRef, SupportsNonRectangular, DSA,
                                   For ? For->getForLoc() : CXXFor->getForLoc());
@@ -9684,9 +9689,8 @@ checkOpenMPLoop(OpenMPDirectiveKind DKind, Expr *CollapseLoopCountExpr,
       ExprResult Iter;
 
       // Compute prod
-      ExprResult Prod =
-          SemaRef.ActOnIntegerConstant(SourceLocation(), 1).get();
-      for (unsigned int K = Cnt+1; K < NestedLoopCount; ++K)
+      ExprResult Prod = SemaRef.ActOnIntegerConstant(SourceLocation(), 1).get();
+      for (unsigned int K = Cnt + 1; K < NestedLoopCount; ++K)
         Prod = SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Mul, Prod.get(),
                                   IterSpaces[K].NumIterations);
 
@@ -9694,8 +9698,8 @@ checkOpenMPLoop(OpenMPDirectiveKind DKind, Expr *CollapseLoopCountExpr,
       // If there is at least one more inner loop to avoid
       // multiplication by 1.
       if (Cnt + 1 < NestedLoopCount)
-        Iter = SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Div,
-                                  Acc.get(), Prod.get());
+        Iter =
+            SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Div, Acc.get(), Prod.get());
       else
         Iter = Acc;
       if (!Iter.isUsable()) {
@@ -9708,12 +9712,11 @@ checkOpenMPLoop(OpenMPDirectiveKind DKind, Expr *CollapseLoopCountExpr,
       // Check if there is at least one more inner loop to avoid
       // multiplication by 1.
       if (Cnt + 1 < NestedLoopCount)
-        Prod = SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Mul,
-                                  Iter.get(), Prod.get());
+        Prod = SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Mul, Iter.get(),
+                                  Prod.get());
       else
         Prod = Iter;
-      Acc = SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Sub,
-                               Acc.get(), Prod.get());
+      Acc = SemaRef.BuildBinOp(CurScope, UpdLoc, BO_Sub, Acc.get(), Prod.get());
 
       // Build update: IS.CounterVar(Private) = IS.Start + Iter * IS.Step
       auto *VD = cast<VarDecl>(cast<DeclRefExpr>(IS.CounterVar)->getDecl());
@@ -10000,7 +10003,7 @@ StmtResult Sema::ActOnOpenMPSectionsDirective(ArrayRef<OMPClause *> Clauses,
       return StmtError();
     // All associated statements must be '#pragma omp section' except for
     // the first one.
-    for (Stmt *SectionStmt : llvm::make_range(std::next(S.begin()), S.end())) {
+    for (Stmt *SectionStmt : llvm::drop_begin(S)) {
       if (!SectionStmt || !isa<OMPSectionDirective>(SectionStmt)) {
         if (SectionStmt)
           Diag(SectionStmt->getBeginLoc(),
@@ -10382,7 +10385,7 @@ Sema::ActOnOpenMPParallelSectionsDirective(ArrayRef<OMPClause *> Clauses,
       return StmtError();
     // All associated statements must be '#pragma omp section' except for
     // the first one.
-    for (Stmt *SectionStmt : llvm::make_range(std::next(S.begin()), S.end())) {
+    for (Stmt *SectionStmt : llvm::drop_begin(S)) {
       if (!SectionStmt || !isa<OMPSectionDirective>(SectionStmt)) {
         if (SectionStmt)
           Diag(SectionStmt->getBeginLoc(),
@@ -10752,7 +10755,6 @@ private:
   bool checkBinaryOperation(BinaryOperator *AtomicBinOp, unsigned DiagId = 0,
                             unsigned NoteId = 0);
 };
-} // namespace
 
 bool OpenMPAtomicUpdateChecker::checkBinaryOperation(
     BinaryOperator *AtomicBinOp, unsigned DiagId, unsigned NoteId) {
@@ -10913,6 +10915,7 @@ bool OpenMPAtomicUpdateChecker::checkStatement(Stmt *S, unsigned DiagId,
   }
   return ErrorFound != NoError;
 }
+} // namespace
 
 StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
                                             Stmt *AStmt,
@@ -10933,9 +10936,12 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
   OpenMPClauseKind MemOrderKind = OMPC_unknown;
   SourceLocation MemOrderLoc;
   for (const OMPClause *C : Clauses) {
-    if (C->getClauseKind() == OMPC_read || C->getClauseKind() == OMPC_write ||
-        C->getClauseKind() == OMPC_update ||
-        C->getClauseKind() == OMPC_capture) {
+    switch (C->getClauseKind()) {
+    case OMPC_read:
+    case OMPC_write:
+    case OMPC_update:
+    case OMPC_capture:
+    case OMPC_compare: {
       if (AtomicKind != OMPC_unknown) {
         Diag(C->getBeginLoc(), diag::err_omp_atomic_several_clauses)
             << SourceRange(C->getBeginLoc(), C->getEndLoc());
@@ -10945,12 +10951,13 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
         AtomicKind = C->getClauseKind();
         AtomicKindLoc = C->getBeginLoc();
       }
+      break;
     }
-    if (C->getClauseKind() == OMPC_seq_cst ||
-        C->getClauseKind() == OMPC_acq_rel ||
-        C->getClauseKind() == OMPC_acquire ||
-        C->getClauseKind() == OMPC_release ||
-        C->getClauseKind() == OMPC_relaxed) {
+    case OMPC_seq_cst:
+    case OMPC_acq_rel:
+    case OMPC_acquire:
+    case OMPC_release:
+    case OMPC_relaxed: {
       if (MemOrderKind != OMPC_unknown) {
         Diag(C->getBeginLoc(), diag::err_omp_several_mem_order_clauses)
             << getOpenMPDirectiveName(OMPD_atomic) << 0
@@ -10961,6 +10968,13 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
         MemOrderKind = C->getClauseKind();
         MemOrderLoc = C->getBeginLoc();
       }
+      break;
+    }
+    // The following clauses are allowed, but we don't need to do anything here.
+    case OMPC_hint:
+      break;
+    default:
+      llvm_unreachable("unknown clause is encountered");
     }
   }
   // OpenMP 5.0, 2.17.7 atomic Construct, Restrictions
@@ -11075,8 +11089,8 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
     if (ErrorFound != NoError) {
       Diag(ErrorLoc, diag::err_omp_atomic_read_not_expression_statement)
           << ErrorRange;
-      Diag(NoteLoc, diag::note_omp_atomic_read_write) << ErrorFound
-                                                      << NoteRange;
+      Diag(NoteLoc, diag::note_omp_atomic_read_write)
+          << ErrorFound << NoteRange;
       return StmtError();
     }
     if (CurContext->isDependentContext())
@@ -11137,8 +11151,8 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
     if (ErrorFound != NoError) {
       Diag(ErrorLoc, diag::err_omp_atomic_write_not_expression_statement)
           << ErrorRange;
-      Diag(NoteLoc, diag::note_omp_atomic_read_write) << ErrorFound
-                                                      << NoteRange;
+      Diag(NoteLoc, diag::note_omp_atomic_read_write)
+          << ErrorFound << NoteRange;
       return StmtError();
     }
     if (CurContext->isDependentContext())
@@ -11154,9 +11168,10 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
     //  x = expr binop x;
     OpenMPAtomicUpdateChecker Checker(*this);
     if (Checker.checkStatement(
-            Body, (AtomicKind == OMPC_update)
-                      ? diag::err_omp_atomic_update_not_expression_statement
-                      : diag::err_omp_atomic_not_expression_statement,
+            Body,
+            (AtomicKind == OMPC_update)
+                ? diag::err_omp_atomic_update_not_expression_statement
+                : diag::err_omp_atomic_not_expression_statement,
             diag::note_omp_atomic_update))
       return StmtError();
     if (!CurContext->isDependentContext()) {
@@ -11370,15 +11385,21 @@ StmtResult Sema::ActOnOpenMPAtomicDirective(ArrayRef<OMPClause *> Clauses,
             SourceRange(Body->getBeginLoc(), Body->getBeginLoc());
         ErrorFound = NotACompoundStatement;
       }
-      if (ErrorFound != NoError) {
-        Diag(ErrorLoc, diag::err_omp_atomic_capture_not_compound_statement)
-            << ErrorRange;
-        Diag(NoteLoc, diag::note_omp_atomic_capture) << ErrorFound << NoteRange;
-        return StmtError();
-      }
-      if (CurContext->isDependentContext())
-        UE = V = E = X = nullptr;
     }
+    if (ErrorFound != NoError) {
+      Diag(ErrorLoc, diag::err_omp_atomic_capture_not_compound_statement)
+          << ErrorRange;
+      Diag(NoteLoc, diag::note_omp_atomic_capture) << ErrorFound << NoteRange;
+      return StmtError();
+    }
+    if (CurContext->isDependentContext())
+      UE = V = E = X = nullptr;
+  } else if (AtomicKind == OMPC_compare) {
+    // TODO: For now we emit an error here and in emitOMPAtomicExpr we ignore
+    // code gen.
+    unsigned DiagID = Diags.getCustomDiagID(
+        DiagnosticsEngine::Error, "atomic compare is not supported for now");
+    Diag(AtomicKindLoc, DiagID);
   }
 
   setFunctionHasBranchProtectedScope();
@@ -12283,8 +12304,8 @@ StmtResult Sema::ActOnOpenMPTargetParallelForSimdDirective(
   // define the nested loops number.
   unsigned NestedLoopCount = checkOpenMPLoop(
       OMPD_target_parallel_for_simd, getCollapseNumberExpr(Clauses),
-      getOrderedNumberExpr(Clauses), CS, *this, *DSAStack,
-      VarsWithImplicitDSA, B);
+      getOrderedNumberExpr(Clauses), CS, *this, *DSAStack, VarsWithImplicitDSA,
+      B);
   if (NestedLoopCount == 0)
     return StmtError();
 
@@ -13459,6 +13480,7 @@ OMPClause *Sema::ActOnOpenMPSingleExprClause(OpenMPClauseKind Kind, Expr *Expr,
   case OMPC_write:
   case OMPC_update:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -14290,6 +14312,7 @@ static OpenMPDirectiveKind getOpenMPCaptureRegionForClause(
   case OMPC_write:
   case OMPC_update:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -14751,6 +14774,7 @@ OMPClause *Sema::ActOnOpenMPSimpleClause(
   case OMPC_read:
   case OMPC_write:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -15056,6 +15080,7 @@ OMPClause *Sema::ActOnOpenMPSingleExprWithArgClause(
   case OMPC_write:
   case OMPC_update:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -15244,6 +15269,9 @@ OMPClause *Sema::ActOnOpenMPClause(OpenMPClauseKind Kind,
   case OMPC_capture:
     Res = ActOnOpenMPCaptureClause(StartLoc, EndLoc);
     break;
+  case OMPC_compare:
+    Res = ActOnOpenMPCompareClause(StartLoc, EndLoc);
+    break;
   case OMPC_seq_cst:
     Res = ActOnOpenMPSeqCstClause(StartLoc, EndLoc);
     break;
@@ -15390,6 +15418,11 @@ OMPClause *Sema::ActOnOpenMPCaptureClause(SourceLocation StartLoc,
   return new (Context) OMPCaptureClause(StartLoc, EndLoc);
 }
 
+OMPClause *Sema::ActOnOpenMPCompareClause(SourceLocation StartLoc,
+                                          SourceLocation EndLoc) {
+  return new (Context) OMPCompareClause(StartLoc, EndLoc);
+}
+
 OMPClause *Sema::ActOnOpenMPSeqCstClause(SourceLocation StartLoc,
                                          SourceLocation EndLoc) {
   return new (Context) OMPSeqCstClause(StartLoc, EndLoc);
@@ -15858,6 +15891,7 @@ OMPClause *Sema::ActOnOpenMPVarListClause(
   case OMPC_write:
   case OMPC_update:
   case OMPC_capture:
+  case OMPC_compare:
   case OMPC_seq_cst:
   case OMPC_acq_rel:
   case OMPC_acquire:
@@ -15981,9 +16015,8 @@ OMPClause *Sema::ActOnOpenMPPrivateClause(ArrayRef<Expr *> VarList,
       Diag(ELoc, diag::err_omp_variably_modified_type_not_supported)
           << getOpenMPClauseName(OMPC_private) << Type
           << getOpenMPDirectiveName(CurrDir);
-      bool IsDecl =
-          !VD ||
-          VD->isThisDeclarationADefinition(Context) == VarDecl::DeclarationOnly;
+      bool IsDecl = !VD || VD->isThisDeclarationADefinition(Context) ==
+                               VarDecl::DeclarationOnly;
       Diag(D->getLocation(),
            IsDecl ? diag::note_previous_decl : diag::note_defined_here)
           << D;
@@ -16247,9 +16280,8 @@ OMPClause *Sema::ActOnOpenMPFirstprivateClause(ArrayRef<Expr *> VarList,
       Diag(ELoc, diag::err_omp_variably_modified_type_not_supported)
           << getOpenMPClauseName(OMPC_firstprivate) << Type
           << getOpenMPDirectiveName(DSAStack->getCurrentDirective());
-      bool IsDecl =
-          !VD ||
-          VD->isThisDeclarationADefinition(Context) == VarDecl::DeclarationOnly;
+      bool IsDecl = !VD || VD->isThisDeclarationADefinition(Context) ==
+                               VarDecl::DeclarationOnly;
       Diag(D->getLocation(),
            IsDecl ? diag::note_previous_decl : diag::note_defined_here)
           << D;
@@ -17433,8 +17465,8 @@ static bool actOnOMPReductionKindClause(
           llvm::APInt InitValue =
               (BOK != BO_LT) ? IsSigned ? llvm::APInt::getSignedMinValue(Size)
                                         : llvm::APInt::getMinValue(Size)
-                             : IsSigned ? llvm::APInt::getSignedMaxValue(Size)
-                                        : llvm::APInt::getMaxValue(Size);
+              : IsSigned ? llvm::APInt::getSignedMaxValue(Size)
+                             : llvm::APInt::getMaxValue(Size);
           Init = IntegerLiteral::Create(Context, InitValue, IntTy, ELoc);
           if (Type->isPointerType()) {
             // Cast to pointer type.
@@ -17845,9 +17877,8 @@ bool Sema::CheckOpenMPLinearDecl(const ValueDecl *D, SourceLocation ELoc,
               !Ty->isIntegralType(Context) && !Ty->isPointerType())) {
     Diag(ELoc, diag::err_omp_linear_expected_int_or_ptr) << Type;
     if (D) {
-      bool IsDecl =
-          !VD ||
-          VD->isThisDeclarationADefinition(Context) == VarDecl::DeclarationOnly;
+      bool IsDecl = !VD || VD->isThisDeclarationADefinition(Context) ==
+                               VarDecl::DeclarationOnly;
       Diag(D->getLocation(),
            IsDecl ? diag::note_previous_decl : diag::note_defined_here)
           << D;
@@ -18059,13 +18090,12 @@ static bool FinishOpenMPLinearClause(OMPLinearClause &Clause, DeclRefExpr *IV,
     Update = SemaRef.ActOnFinishFullExpr(Update.get(), DE->getBeginLoc(),
                                          /*DiscardedValue*/ false);
 
-    // Build final: Var = InitExpr + NumIterations * Step
+    // Build final: Var = PrivCopy;
     ExprResult Final;
     if (!Info.first)
-      Final =
-          buildCounterUpdate(SemaRef, S, RefExpr->getExprLoc(), CapturedRef,
-                             InitExpr, NumIterations, Step, /*Subtract=*/false,
-                             /*IsNonRectangularLB=*/false);
+      Final = SemaRef.BuildBinOp(
+          S, RefExpr->getExprLoc(), BO_Assign, CapturedRef,
+          SemaRef.DefaultLvalueConversion(*CurPrivate).get());
     else
       Final = *CurPrivate;
     Final = SemaRef.ActOnFinishFullExpr(Final.get(), DE->getBeginLoc(),
@@ -18124,9 +18154,8 @@ OMPClause *Sema::ActOnOpenMPAlignedClause(
     if (!Ty || (!Ty->isArrayType() && !Ty->isPointerType())) {
       Diag(ELoc, diag::err_omp_aligned_expected_array_or_ptr)
           << QType << getLangOpts().CPlusPlus << ERange;
-      bool IsDecl =
-          !VD ||
-          VD->isThisDeclarationADefinition(Context) == VarDecl::DeclarationOnly;
+      bool IsDecl = !VD || VD->isThisDeclarationADefinition(Context) ==
+                               VarDecl::DeclarationOnly;
       Diag(D->getLocation(),
            IsDecl ? diag::note_previous_decl : diag::note_defined_here)
           << D;
@@ -18327,9 +18356,8 @@ OMPClause *Sema::ActOnOpenMPCopyprivateClause(ArrayRef<Expr *> VarList,
       Diag(ELoc, diag::err_omp_variably_modified_type_not_supported)
           << getOpenMPClauseName(OMPC_copyprivate) << Type
           << getOpenMPDirectiveName(DSAStack->getCurrentDirective());
-      bool IsDecl =
-          !VD ||
-          VD->isThisDeclarationADefinition(Context) == VarDecl::DeclarationOnly;
+      bool IsDecl = !VD || VD->isThisDeclarationADefinition(Context) ==
+                               VarDecl::DeclarationOnly;
       Diag(D->getLocation(),
            IsDecl ? diag::note_previous_decl : diag::note_defined_here)
           << D;
@@ -18636,22 +18664,19 @@ Sema::ActOnOpenMPDependClause(Expr *DepModifier, OpenMPDependClauseKind DepKind,
         if (!RefExpr->isValueDependent() && !RefExpr->isTypeDependent() &&
             !RefExpr->isInstantiationDependent() &&
             !RefExpr->containsUnexpandedParameterPack() &&
-            (OMPDependTFound &&
-             DSAStack->getOMPDependT().getTypePtr() == ExprTy.getTypePtr())) {
+            (!RefExpr->IgnoreParenImpCasts()->isLValue() ||
+             (OMPDependTFound &&
+              DSAStack->getOMPDependT().getTypePtr() == ExprTy.getTypePtr()))) {
           Diag(ELoc, diag::err_omp_expected_addressable_lvalue_or_array_item)
-              << (LangOpts.OpenMP >= 50 ? 1 : 0) << 1
-              << RefExpr->getSourceRange();
+              << (LangOpts.OpenMP >= 50 ? 1 : 0)
+              << (LangOpts.OpenMP >= 50 ? 1 : 0) << RefExpr->getSourceRange();
           continue;
         }
 
         auto *ASE = dyn_cast<ArraySubscriptExpr>(SimpleExpr);
-        if (!RefExpr->IgnoreParenImpCasts()->isLValue() ||
-            (ASE && !ASE->getBase()->isTypeDependent() &&
-             !ASE->getBase()
-                  ->getType()
-                  .getNonReferenceType()
-                  ->isPointerType() &&
-             !ASE->getBase()->getType().getNonReferenceType()->isArrayType())) {
+        if (ASE && !ASE->getBase()->isTypeDependent() &&
+            !ASE->getBase()->getType().getNonReferenceType()->isPointerType() &&
+            !ASE->getBase()->getType().getNonReferenceType()->isArrayType()) {
           Diag(ELoc, diag::err_omp_expected_addressable_lvalue_or_array_item)
               << (LangOpts.OpenMP >= 50 ? 1 : 0)
               << (LangOpts.OpenMP >= 50 ? 1 : 0) << RefExpr->getSourceRange();
@@ -18934,7 +18959,7 @@ public:
     if (!isa<FieldDecl>(ME->getMemberDecl())) {
       if (!NoDiagnose) {
         SemaRef.Diag(ELoc, diag::err_omp_expected_access_to_data_field)
-          << ME->getSourceRange();
+            << ME->getSourceRange();
         return false;
       }
       if (RelevantExpr)
@@ -18950,7 +18975,7 @@ public:
     if (FD->isBitField()) {
       if (!NoDiagnose) {
         SemaRef.Diag(ELoc, diag::err_omp_bit_fields_forbidden_in_clause)
-          << ME->getSourceRange() << getOpenMPClauseName(CKind);
+            << ME->getSourceRange() << getOpenMPClauseName(CKind);
         return false;
       }
       if (RelevantExpr)
@@ -18970,7 +18995,7 @@ public:
     if (CurType->isUnionType()) {
       if (!NoDiagnose) {
         SemaRef.Diag(ELoc, diag::err_omp_union_type_not_allowed)
-          << ME->getSourceRange();
+            << ME->getSourceRange();
         return false;
       }
       return RelevantExpr || Visit(E);
@@ -18997,7 +19022,7 @@ public:
     if (!E->getType()->isAnyPointerType() && !E->getType()->isArrayType()) {
       if (!NoDiagnose) {
         SemaRef.Diag(ELoc, diag::err_omp_expected_base_var_name)
-          << 0 << AE->getSourceRange();
+            << 0 << AE->getSourceRange();
         return false;
       }
       return RelevantExpr || Visit(E);
@@ -19006,8 +19031,7 @@ public:
     // If we got an array subscript that express the whole dimension we
     // can have any array expressions before. If it only expressing part of
     // the dimension, we can only have unitary-size array expressions.
-    if (checkArrayExpressionDoesNotReferToWholeSize(SemaRef, AE,
-                                                    E->getType()))
+    if (checkArrayExpressionDoesNotReferToWholeSize(SemaRef, AE, E->getType()))
       AllowWholeSizeArraySection = false;
 
     if (const auto *TE = dyn_cast<CXXThisExpr>(E->IgnoreParenCasts())) {
@@ -19037,7 +19061,7 @@ public:
            "Array sections cannot be implicitly mapped.");
     Expr *E = OASE->getBase()->IgnoreParenImpCasts();
     QualType CurType =
-      OMPArraySectionExpr::getBaseOriginalType(E).getCanonicalType();
+        OMPArraySectionExpr::getBaseOriginalType(E).getCanonicalType();
 
     // OpenMP 4.5 [2.15.5.1, map Clause, Restrictions, C++, p.1]
     //  If the type of a list item is a reference to a type T then the type
@@ -19049,14 +19073,14 @@ public:
 
     if (!IsPointer && !CurType->isArrayType()) {
       SemaRef.Diag(ELoc, diag::err_omp_expected_base_var_name)
-        << 0 << OASE->getSourceRange();
+          << 0 << OASE->getSourceRange();
       return false;
     }
 
     bool NotWhole =
-      checkArrayExpressionDoesNotReferToWholeSize(SemaRef, OASE, CurType);
+        checkArrayExpressionDoesNotReferToWholeSize(SemaRef, OASE, CurType);
     bool NotUnity =
-      checkArrayExpressionDoesNotReferToUnitySize(SemaRef, OASE, CurType);
+        checkArrayExpressionDoesNotReferToUnitySize(SemaRef, OASE, CurType);
 
     if (AllowWholeSizeArraySection) {
       // Any array section is currently allowed. Allowing a whole size array
@@ -19079,9 +19103,9 @@ public:
       // compatible with the properties of the current array section.
       if (NoDiagnose)
         return false;
-      SemaRef.Diag(
-        ELoc, diag::err_array_section_does_not_specify_contiguous_storage)
-        << OASE->getSourceRange();
+      SemaRef.Diag(ELoc,
+                   diag::err_array_section_does_not_specify_contiguous_storage)
+          << OASE->getSourceRange();
       return false;
     }
 
@@ -19180,9 +19204,7 @@ public:
     emitErrorMsg();
     return false;
   }
-  const Expr *getFoundBase() const {
-    return RelevantExpr;
-  }
+  const Expr *getFoundBase() const { return RelevantExpr; }
   explicit MapBaseChecker(
       Sema &SemaRef, OpenMPClauseKind CKind, OpenMPDirectiveKind DKind,
       OMPClauseMappableExprCommon::MappableExprComponentList &Components,
@@ -19193,9 +19215,9 @@ public:
 } // namespace
 
 /// Return the expression of the base of the mappable expression or null if it
-/// cannot be determined and do all the necessary checks to see if the expression
-/// is valid as a standalone mappable expression. In the process, record all the
-/// components of the expression.
+/// cannot be determined and do all the necessary checks to see if the
+/// expression is valid as a standalone mappable expression. In the process,
+/// record all the components of the expression.
 static const Expr *checkMapClauseExpressionBase(
     Sema &SemaRef, Expr *E,
     OMPClauseMappableExprCommon::MappableExprComponentList &CurComponents,
@@ -19385,9 +19407,9 @@ static bool checkMapConflicts(
             return true;
           }
           if (CI->getAssociatedExpression()->getStmtClass() !=
-                         SI->getAssociatedExpression()->getStmtClass() ||
-                     CI->getAssociatedDeclaration()->getCanonicalDecl() ==
-                         SI->getAssociatedDeclaration()->getCanonicalDecl()) {
+                  SI->getAssociatedExpression()->getStmtClass() ||
+              CI->getAssociatedDeclaration()->getCanonicalDecl() ==
+                  SI->getAssociatedDeclaration()->getCanonicalDecl()) {
             assert(CI != CE && SI != SE);
             SemaRef.Diag(DerivedLoc, diag::err_omp_same_pointer_dereferenced)
                 << DerivedLoc;
@@ -19608,7 +19630,7 @@ struct MappableVarListInfo {
     VarBaseDeclarations.reserve(VarList.size());
   }
 };
-}
+} // namespace
 
 // Check the validity of the provided variable list for the provided clause kind
 // \a CKind. In the check process the valid expressions, mappable expression
@@ -21341,8 +21363,7 @@ OMPClause *Sema::ActOnOpenMPInclusiveClause(ArrayRef<Expr *> VarList,
     // A list item that appears in the inclusive or exclusive clause must appear
     // in a reduction clause with the inscan modifier on the enclosing
     // worksharing-loop, worksharing-loop SIMD, or simd construct.
-    if (DVar.CKind != OMPC_reduction ||
-        DVar.Modifier != OMPC_REDUCTION_inscan)
+    if (DVar.CKind != OMPC_reduction || DVar.Modifier != OMPC_REDUCTION_inscan)
       Diag(ELoc, diag::err_omp_inclusive_exclusive_not_reduction)
           << RefExpr->getSourceRange();
 
diff --git a/clang/lib/Sema/SemaStmtAsm.cpp b/clang/lib/Sema/SemaStmtAsm.cpp
index 603611b2d86b..49f7a8d573d5 100644
--- a/clang/lib/Sema/SemaStmtAsm.cpp
+++ b/clang/lib/Sema/SemaStmtAsm.cpp
@@ -296,9 +296,9 @@ StmtResult Sema::ActOnGCCAsmStmt(SourceLocation AsmLoc, bool IsSimple,
         checkExprMemoryConstraintCompat(*this, OutputExpr, Info, false))
       return StmtError();
 
-    // Disallow _ExtInt, since the backends tend to have difficulties with
-    // non-normal sizes.
-    if (OutputExpr->getType()->isExtIntType())
+    // Disallow bit-precise integer types, since the backends tend to have
+    // difficulties with abnormal sizes.
+    if (OutputExpr->getType()->isBitIntType())
       return StmtError(
           Diag(OutputExpr->getBeginLoc(), diag::err_asm_invalid_type)
           << OutputExpr->getType() << 0 /*Input*/
@@ -429,7 +429,7 @@ StmtResult Sema::ActOnGCCAsmStmt(SourceLocation AsmLoc, bool IsSimple,
       }
     }
 
-    if (InputExpr->getType()->isExtIntType())
+    if (InputExpr->getType()->isBitIntType())
       return StmtError(
           Diag(InputExpr->getBeginLoc(), diag::err_asm_invalid_type)
           << InputExpr->getType() << 1 /*Output*/
@@ -924,7 +924,7 @@ StmtResult Sema::ActOnMSAsmStmt(SourceLocation AsmLoc, SourceLocation LBraceLoc,
   setFunctionHasBranchProtectedScope();
 
   for (uint64_t I = 0; I < NumOutputs + NumInputs; ++I) {
-    if (Exprs[I]->getType()->isExtIntType())
+    if (Exprs[I]->getType()->isBitIntType())
       return StmtError(
           Diag(Exprs[I]->getBeginLoc(), diag::err_asm_invalid_type)
           << Exprs[I]->getType() << (I < NumOutputs)
diff --git a/clang/lib/Sema/SemaTemplate.cpp b/clang/lib/Sema/SemaTemplate.cpp
index f4fd2ea5aa8e..2482f6d404ea 100644
--- a/clang/lib/Sema/SemaTemplate.cpp
+++ b/clang/lib/Sema/SemaTemplate.cpp
@@ -6142,12 +6142,12 @@ bool UnnamedLocalNoLinkageFinder::VisitPipeType(const PipeType* T) {
   return false;
 }
 
-bool UnnamedLocalNoLinkageFinder::VisitExtIntType(const ExtIntType *T) {
+bool UnnamedLocalNoLinkageFinder::VisitBitIntType(const BitIntType *T) {
   return false;
 }
 
-bool UnnamedLocalNoLinkageFinder::VisitDependentExtIntType(
-    const DependentExtIntType *T) {
+bool UnnamedLocalNoLinkageFinder::VisitDependentBitIntType(
+    const DependentBitIntType *T) {
   return false;
 }
 
@@ -7089,7 +7089,7 @@ ExprResult Sema::CheckTemplateArgument(NonTypeTemplateParmDecl *Param,
       QualType IntegerType = ParamType;
       if (const EnumType *Enum = IntegerType->getAs<EnumType>())
         IntegerType = Enum->getDecl()->getIntegerType();
-      Value = Value.extOrTrunc(IntegerType->isExtIntType()
+      Value = Value.extOrTrunc(IntegerType->isBitIntType()
                                    ? Context.getIntWidth(IntegerType)
                                    : Context.getTypeSize(IntegerType));
 
@@ -7184,7 +7184,7 @@ ExprResult Sema::CheckTemplateArgument(NonTypeTemplateParmDecl *Param,
 
       // Coerce the template argument's value to the value it will have
       // based on the template parameter's type.
-      unsigned AllowedBits = IntegerType->isExtIntType()
+      unsigned AllowedBits = IntegerType->isBitIntType()
                                  ? Context.getIntWidth(IntegerType)
                                  : Context.getTypeSize(IntegerType);
       if (Value.getBitWidth() != AllowedBits)
diff --git a/clang/lib/Sema/SemaTemplateDeduction.cpp b/clang/lib/Sema/SemaTemplateDeduction.cpp
index 81edae10335d..e9636d2b942e 100644
--- a/clang/lib/Sema/SemaTemplateDeduction.cpp
+++ b/clang/lib/Sema/SemaTemplateDeduction.cpp
@@ -1597,7 +1597,7 @@ static Sema::TemplateDeductionResult DeduceTemplateArgumentsByTypeMatch(
     case Type::ObjCObject:
     case Type::ObjCInterface:
     case Type::ObjCObjectPointer:
-    case Type::ExtInt:
+    case Type::BitInt:
       return (TDF & TDF_SkipNonDependent) ||
                      ((TDF & TDF_IgnoreQualifiers)
                           ? S.Context.hasSameUnqualifiedType(P, A)
@@ -2144,10 +2144,10 @@ static Sema::TemplateDeductionResult DeduceTemplateArgumentsByTypeMatch(
 
       return Sema::TDK_NonDeducedMismatch;
     }
-    case Type::DependentExtInt: {
-      const auto *IP = P->castAs<DependentExtIntType>();
+    case Type::DependentBitInt: {
+      const auto *IP = P->castAs<DependentBitIntType>();
 
-      if (const auto *IA = A->getAs<ExtIntType>()) {
+      if (const auto *IA = A->getAs<BitIntType>()) {
         if (IP->isUnsigned() != IA->isUnsigned())
           return Sema::TDK_NonDeducedMismatch;
 
@@ -2164,7 +2164,7 @@ static Sema::TemplateDeductionResult DeduceTemplateArgumentsByTypeMatch(
                                              Deduced);
       }
 
-      if (const auto *IA = A->getAs<DependentExtIntType>()) {
+      if (const auto *IA = A->getAs<DependentBitIntType>()) {
         if (IP->isUnsigned() != IA->isUnsigned())
           return Sema::TDK_NonDeducedMismatch;
         return Sema::TDK_Success;
@@ -5949,9 +5949,9 @@ MarkUsedTemplateParameters(ASTContext &Ctx, QualType T,
                                cast<DeducedType>(T)->getDeducedType(),
                                OnlyDeduced, Depth, Used);
     break;
-  case Type::DependentExtInt:
+  case Type::DependentBitInt:
     MarkUsedTemplateParameters(Ctx,
-                               cast<DependentExtIntType>(T)->getNumBitsExpr(),
+                               cast<DependentBitIntType>(T)->getNumBitsExpr(),
                                OnlyDeduced, Depth, Used);
     break;
 
@@ -5966,7 +5966,7 @@ MarkUsedTemplateParameters(ASTContext &Ctx, QualType T,
   case Type::ObjCObjectPointer:
   case Type::UnresolvedUsing:
   case Type::Pipe:
-  case Type::ExtInt:
+  case Type::BitInt:
 #define TYPE(Class, Base)
 #define ABSTRACT_TYPE(Class, Base)
 #define DEPENDENT_TYPE(Class, Base)
diff --git a/clang/lib/Sema/SemaTemplateVariadic.cpp b/clang/lib/Sema/SemaTemplateVariadic.cpp
index c0bb310e64fb..51c79e93ab0a 100644
--- a/clang/lib/Sema/SemaTemplateVariadic.cpp
+++ b/clang/lib/Sema/SemaTemplateVariadic.cpp
@@ -870,7 +870,7 @@ bool Sema::containsUnexpandedParameterPacks(Declarator &D) {
 
   case TST_typeofExpr:
   case TST_decltype:
-  case TST_extint:
+  case TST_bitint:
     if (DS.getRepAsExpr() &&
         DS.getRepAsExpr()->containsUnexpandedParameterPack())
       return true;
diff --git a/clang/lib/Sema/SemaType.cpp b/clang/lib/Sema/SemaType.cpp
index d2ee669debd0..7a038301a249 100644
--- a/clang/lib/Sema/SemaType.cpp
+++ b/clang/lib/Sema/SemaType.cpp
@@ -1435,12 +1435,11 @@ static QualType ConvertDeclSpecToType(TypeProcessingState &state) {
     }
     break;
   }
-  case DeclSpec::TST_extint: {
-    if (!S.Context.getTargetInfo().hasExtIntType())
-      S.Diag(DS.getTypeSpecTypeLoc(), diag::err_type_unsupported)
-        << "_ExtInt";
+  case DeclSpec::TST_bitint: {
+    if (!S.Context.getTargetInfo().hasBitIntType())
+      S.Diag(DS.getTypeSpecTypeLoc(), diag::err_type_unsupported) << "_BitInt";
     Result =
-        S.BuildExtIntType(DS.getTypeSpecSign() == TypeSpecifierSign::Unsigned,
+        S.BuildBitIntType(DS.getTypeSpecSign() == TypeSpecifierSign::Unsigned,
                           DS.getRepAsExpr(), DS.getBeginLoc());
     if (Result.isNull()) {
       Result = Context.IntTy;
@@ -2237,7 +2236,7 @@ QualType Sema::BuildWritePipeType(QualType T, SourceLocation Loc) {
   return Context.getWritePipeType(T);
 }
 
-/// Build a extended int type.
+/// Build a bit-precise integer type.
 ///
 /// \param IsUnsigned Boolean representing the signedness of the type.
 ///
@@ -2245,10 +2244,10 @@ QualType Sema::BuildWritePipeType(QualType T, SourceLocation Loc) {
 /// that.
 ///
 /// \param Loc Location of the keyword.
-QualType Sema::BuildExtIntType(bool IsUnsigned, Expr *BitWidth,
+QualType Sema::BuildBitIntType(bool IsUnsigned, Expr *BitWidth,
                                SourceLocation Loc) {
   if (BitWidth->isInstantiationDependent())
-    return Context.getDependentExtIntType(IsUnsigned, BitWidth);
+    return Context.getDependentBitIntType(IsUnsigned, BitWidth);
 
   llvm::APSInt Bits(32);
   ExprResult ICE =
@@ -2259,22 +2258,22 @@ QualType Sema::BuildExtIntType(bool IsUnsigned, Expr *BitWidth,
 
   int64_t NumBits = Bits.getSExtValue();
   if (!IsUnsigned && NumBits < 2) {
-    Diag(Loc, diag::err_ext_int_bad_size) << 0;
+    Diag(Loc, diag::err_bit_int_bad_size) << 0;
     return QualType();
   }
 
   if (IsUnsigned && NumBits < 1) {
-    Diag(Loc, diag::err_ext_int_bad_size) << 1;
+    Diag(Loc, diag::err_bit_int_bad_size) << 1;
     return QualType();
   }
 
   if (NumBits > llvm::IntegerType::MAX_INT_BITS) {
-    Diag(Loc, diag::err_ext_int_max_size) << IsUnsigned
-                                          << llvm::IntegerType::MAX_INT_BITS;
+    Diag(Loc, diag::err_bit_int_max_size)
+        << IsUnsigned << llvm::IntegerType::MAX_INT_BITS;
     return QualType();
   }
 
-  return Context.getExtIntType(IsUnsigned, NumBits);
+  return Context.getBitIntType(IsUnsigned, NumBits);
 }
 
 /// Check whether the specified array bound can be evaluated using the relevant
@@ -3941,6 +3940,20 @@ static CallingConv getCCForDeclaratorChunk(
         break;
       }
     }
+  } else if (S.getLangOpts().CUDA) {
+    // If we're compiling CUDA/HIP code and targeting SPIR-V we need to make
+    // sure the kernels will be marked with the right calling convention so that
+    // they will be visible by the APIs that ingest SPIR-V.
+    llvm::Triple Triple = S.Context.getTargetInfo().getTriple();
+    if (Triple.getArch() == llvm::Triple::spirv32 ||
+        Triple.getArch() == llvm::Triple::spirv64) {
+      for (const ParsedAttr &AL : D.getDeclSpec().getAttributes()) {
+        if (AL.getKind() == ParsedAttr::AT_CUDAGlobal) {
+          CC = CC_OpenCLKernel;
+          break;
+        }
+      }
+    }
   }
 
   return CC;
@@ -6077,11 +6090,11 @@ namespace {
       TL.getValueLoc().initializeFullCopy(TInfo->getTypeLoc());
     }
 
-    void VisitExtIntTypeLoc(ExtIntTypeLoc TL) {
+    void VisitExtIntTypeLoc(BitIntTypeLoc TL) {
       TL.setNameLoc(DS.getTypeSpecTypeLoc());
     }
 
-    void VisitDependentExtIntTypeLoc(DependentExtIntTypeLoc TL) {
+    void VisitDependentExtIntTypeLoc(DependentBitIntTypeLoc TL) {
       TL.setNameLoc(DS.getTypeSpecTypeLoc());
     }
 
@@ -6211,7 +6224,7 @@ namespace {
       assert(Chunk.Kind == DeclaratorChunk::Pipe);
       TL.setKWLoc(Chunk.Loc);
     }
-    void VisitExtIntTypeLoc(ExtIntTypeLoc TL) {
+    void VisitBitIntTypeLoc(BitIntTypeLoc TL) {
       TL.setNameLoc(Chunk.Loc);
     }
     void VisitMacroQualifiedTypeLoc(MacroQualifiedTypeLoc TL) {
@@ -6528,7 +6541,6 @@ static void HandleBTFTypeTagAttribute(QualType &Type, const ParsedAttr &Attr,
   StringRef BTFTypeTag = StrLiteral->getString();
   Type = State.getAttributedType(
       ::new (Ctx) BTFTypeTagAttr(Ctx, Attr, BTFTypeTag), Type, Type);
-  return;
 }
 
 /// HandleAddressSpaceTypeAttribute - Process an address_space attribute on the
@@ -9079,9 +9091,8 @@ QualType Sema::BuildAtomicType(QualType T, SourceLocation Loc) {
     else if (!T.isTriviallyCopyableType(Context))
       // Some other non-trivially-copyable type (probably a C++ class)
       DisallowedKind = 7;
-    else if (T->isExtIntType()) {
-        DisallowedKind = 8;
-    }
+    else if (T->isBitIntType())
+      DisallowedKind = 8;
 
     if (DisallowedKind != -1) {
       Diag(Loc, diag::err_atomic_specifier_bad_type) << DisallowedKind << T;
diff --git a/clang/lib/Sema/TreeTransform.h b/clang/lib/Sema/TreeTransform.h
index 7f3326c13263..298a3f7a83d8 100644
--- a/clang/lib/Sema/TreeTransform.h
+++ b/clang/lib/Sema/TreeTransform.h
@@ -933,6 +933,11 @@ public:
   /// the UnresolvedUsingTypenameDecl was transformed to.
   QualType RebuildUnresolvedUsingType(SourceLocation NameLoc, Decl *D);
 
+  /// Build a new type found via an alias.
+  QualType RebuildUsingType(UsingShadowDecl *Found, QualType Underlying) {
+    return SemaRef.Context.getUsingType(Found, Underlying);
+  }
+
   /// Build a new typedef type.
   QualType RebuildTypedefType(TypedefNameDecl *Typedef) {
     return SemaRef.Context.getTypeDeclType(Typedef);
@@ -1195,12 +1200,12 @@ public:
   QualType RebuildPipeType(QualType ValueType, SourceLocation KWLoc,
                            bool isReadPipe);
 
-   /// Build an extended int given its value type.
-  QualType RebuildExtIntType(bool IsUnsigned, unsigned NumBits,
+  /// Build a bit-precise int given its value type.
+  QualType RebuildBitIntType(bool IsUnsigned, unsigned NumBits,
                              SourceLocation Loc);
 
-  /// Build a dependent extended int given its value type.
-  QualType RebuildDependentExtIntType(bool IsUnsigned, Expr *NumBitsExpr,
+  /// Build a dependent bit-precise int given its value type.
+  QualType RebuildDependentBitIntType(bool IsUnsigned, Expr *NumBitsExpr,
                                       SourceLocation Loc);
 
   /// Build a new template name given a nested name specifier, a flag
@@ -6072,9 +6077,9 @@ QualType TreeTransform<Derived>::TransformFunctionNoProtoType(
   return Result;
 }
 
-template<typename Derived> QualType
-TreeTransform<Derived>::TransformUnresolvedUsingType(TypeLocBuilder &TLB,
-                                                 UnresolvedUsingTypeLoc TL) {
+template <typename Derived>
+QualType TreeTransform<Derived>::TransformUnresolvedUsingType(
+    TypeLocBuilder &TLB, UnresolvedUsingTypeLoc TL) {
   const UnresolvedUsingType *T = TL.getTypePtr();
   Decl *D = getDerived().TransformDecl(TL.getNameLoc(), T->getDecl());
   if (!D)
@@ -6095,6 +6100,32 @@ TreeTransform<Derived>::TransformUnresolvedUsingType(TypeLocBuilder &TLB,
   return Result;
 }
 
+template <typename Derived>
+QualType TreeTransform<Derived>::TransformUsingType(TypeLocBuilder &TLB,
+                                                    UsingTypeLoc TL) {
+  const UsingType *T = TL.getTypePtr();
+
+  auto *Found = cast_or_null<UsingShadowDecl>(getDerived().TransformDecl(
+      TL.getLocalSourceRange().getBegin(), T->getFoundDecl()));
+  if (!Found)
+    return QualType();
+
+  QualType Underlying = getDerived().TransformType(T->desugar());
+  if (Underlying.isNull())
+    return QualType();
+
+  QualType Result = TL.getType();
+  if (getDerived().AlwaysRebuild() || Found != T->getFoundDecl() ||
+      Underlying != T->getUnderlyingType()) {
+    Result = getDerived().RebuildUsingType(Found, Underlying);
+    if (Result.isNull())
+      return QualType();
+  }
+
+  TLB.pushTypeSpec(Result).setNameLoc(TL.getNameLoc());
+  return Result;
+}
+
 template<typename Derived>
 QualType TreeTransform<Derived>::TransformTypedefType(TypeLocBuilder &TLB,
                                                       TypedefTypeLoc TL) {
@@ -6430,27 +6461,27 @@ QualType TreeTransform<Derived>::TransformPipeType(TypeLocBuilder &TLB,
 }
 
 template <typename Derived>
-QualType TreeTransform<Derived>::TransformExtIntType(TypeLocBuilder &TLB,
-                                                     ExtIntTypeLoc TL) {
-  const ExtIntType *EIT = TL.getTypePtr();
+QualType TreeTransform<Derived>::TransformBitIntType(TypeLocBuilder &TLB,
+                                                     BitIntTypeLoc TL) {
+  const BitIntType *EIT = TL.getTypePtr();
   QualType Result = TL.getType();
 
   if (getDerived().AlwaysRebuild()) {
-    Result = getDerived().RebuildExtIntType(EIT->isUnsigned(),
+    Result = getDerived().RebuildBitIntType(EIT->isUnsigned(),
                                             EIT->getNumBits(), TL.getNameLoc());
     if (Result.isNull())
       return QualType();
   }
 
-  ExtIntTypeLoc NewTL = TLB.push<ExtIntTypeLoc>(Result);
+  BitIntTypeLoc NewTL = TLB.push<BitIntTypeLoc>(Result);
   NewTL.setNameLoc(TL.getNameLoc());
   return Result;
 }
 
 template <typename Derived>
-QualType TreeTransform<Derived>::TransformDependentExtIntType(
-    TypeLocBuilder &TLB, DependentExtIntTypeLoc TL) {
-  const DependentExtIntType *EIT = TL.getTypePtr();
+QualType TreeTransform<Derived>::TransformDependentBitIntType(
+    TypeLocBuilder &TLB, DependentBitIntTypeLoc TL) {
+  const DependentBitIntType *EIT = TL.getTypePtr();
 
   EnterExpressionEvaluationContext Unevaluated(
       SemaRef, Sema::ExpressionEvaluationContext::ConstantEvaluated);
@@ -6463,18 +6494,18 @@ QualType TreeTransform<Derived>::TransformDependentExtIntType(
   QualType Result = TL.getType();
 
   if (getDerived().AlwaysRebuild() || BitsExpr.get() != EIT->getNumBitsExpr()) {
-    Result = getDerived().RebuildDependentExtIntType(
+    Result = getDerived().RebuildDependentBitIntType(
         EIT->isUnsigned(), BitsExpr.get(), TL.getNameLoc());
 
     if (Result.isNull())
       return QualType();
   }
 
-  if (isa<DependentExtIntType>(Result)) {
-    DependentExtIntTypeLoc NewTL = TLB.push<DependentExtIntTypeLoc>(Result);
+  if (isa<DependentBitIntType>(Result)) {
+    DependentBitIntTypeLoc NewTL = TLB.push<DependentBitIntTypeLoc>(Result);
     NewTL.setNameLoc(TL.getNameLoc());
   } else {
-    ExtIntTypeLoc NewTL = TLB.push<ExtIntTypeLoc>(Result);
+    BitIntTypeLoc NewTL = TLB.push<BitIntTypeLoc>(Result);
     NewTL.setNameLoc(TL.getNameLoc());
   }
   return Result;
@@ -9431,6 +9462,13 @@ TreeTransform<Derived>::TransformOMPCaptureClause(OMPCaptureClause *C) {
 
 template <typename Derived>
 OMPClause *
+TreeTransform<Derived>::TransformOMPCompareClause(OMPCompareClause *C) {
+  // No need to rebuild this clause, no template-dependent parameters.
+  return C;
+}
+
+template <typename Derived>
+OMPClause *
 TreeTransform<Derived>::TransformOMPSeqCstClause(OMPSeqCstClause *C) {
   // No need to rebuild this clause, no template-dependent parameters.
   return C;
@@ -14462,7 +14500,6 @@ QualType TreeTransform<Derived>::RebuildUnresolvedUsingType(SourceLocation Loc,
   if (D->isInvalidDecl()) return QualType();
 
   // FIXME: Doesn't account for ObjCInterfaceDecl!
-  TypeDecl *Ty;
   if (auto *UPD = dyn_cast<UsingPackDecl>(D)) {
     // A valid resolved using typename pack expansion decl can have multiple
     // UsingDecls, but they must each have exactly one type, and it must be
@@ -14498,17 +14535,18 @@ QualType TreeTransform<Derived>::RebuildUnresolvedUsingType(SourceLocation Loc,
     // A valid resolved using typename decl points to exactly one type decl.
     assert(++Using->shadow_begin() == Using->shadow_end());
 
-    NamedDecl *Target = Using->shadow_begin()->getTargetDecl();
-    if (SemaRef.DiagnoseUseOfDecl(Target, Loc))
+    UsingShadowDecl *Shadow = *Using->shadow_begin();
+    if (SemaRef.DiagnoseUseOfDecl(Shadow->getTargetDecl(), Loc))
       return QualType();
-    Ty = cast<TypeDecl>(Target);
+    return SemaRef.Context.getUsingType(
+        Shadow, SemaRef.Context.getTypeDeclType(
+                    cast<TypeDecl>(Shadow->getTargetDecl())));
   } else {
     assert(isa<UnresolvedUsingTypenameDecl>(D) &&
            "UnresolvedUsingTypenameDecl transformed to non-using decl");
-    Ty = cast<UnresolvedUsingTypenameDecl>(D);
+    return SemaRef.Context.getTypeDeclType(
+        cast<UnresolvedUsingTypenameDecl>(D));
   }
-
-  return SemaRef.Context.getTypeDeclType(Ty);
 }
 
 template <typename Derived>
@@ -14557,20 +14595,20 @@ QualType TreeTransform<Derived>::RebuildPipeType(QualType ValueType,
 }
 
 template <typename Derived>
-QualType TreeTransform<Derived>::RebuildExtIntType(bool IsUnsigned,
+QualType TreeTransform<Derived>::RebuildBitIntType(bool IsUnsigned,
                                                    unsigned NumBits,
                                                    SourceLocation Loc) {
   llvm::APInt NumBitsAP(SemaRef.Context.getIntWidth(SemaRef.Context.IntTy),
                         NumBits, true);
   IntegerLiteral *Bits = IntegerLiteral::Create(SemaRef.Context, NumBitsAP,
                                                 SemaRef.Context.IntTy, Loc);
-  return SemaRef.BuildExtIntType(IsUnsigned, Bits, Loc);
+  return SemaRef.BuildBitIntType(IsUnsigned, Bits, Loc);
 }
 
 template <typename Derived>
-QualType TreeTransform<Derived>::RebuildDependentExtIntType(
+QualType TreeTransform<Derived>::RebuildDependentBitIntType(
     bool IsUnsigned, Expr *NumBitsExpr, SourceLocation Loc) {
-  return SemaRef.BuildExtIntType(IsUnsigned, NumBitsExpr, Loc);
+  return SemaRef.BuildBitIntType(IsUnsigned, NumBitsExpr, Loc);
 }
 
 template<typename Derived>
diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp
index a033bccbe506..f93e0d2ed1c4 100644
--- a/clang/lib/Serialization/ASTReader.cpp
+++ b/clang/lib/Serialization/ASTReader.cpp
@@ -6607,6 +6607,10 @@ void TypeLocReader::VisitUnresolvedUsingTypeLoc(UnresolvedUsingTypeLoc TL) {
   TL.setNameLoc(readSourceLocation());
 }
 
+void TypeLocReader::VisitUsingTypeLoc(UsingTypeLoc TL) {
+  TL.setNameLoc(readSourceLocation());
+}
+
 void TypeLocReader::VisitTypedefTypeLoc(TypedefTypeLoc TL) {
   TL.setNameLoc(readSourceLocation());
 }
@@ -6772,11 +6776,11 @@ void TypeLocReader::VisitPipeTypeLoc(PipeTypeLoc TL) {
   TL.setKWLoc(readSourceLocation());
 }
 
-void TypeLocReader::VisitExtIntTypeLoc(clang::ExtIntTypeLoc TL) {
+void TypeLocReader::VisitBitIntTypeLoc(clang::BitIntTypeLoc TL) {
   TL.setNameLoc(readSourceLocation());
 }
-void TypeLocReader::VisitDependentExtIntTypeLoc(
-    clang::DependentExtIntTypeLoc TL) {
+void TypeLocReader::VisitDependentBitIntTypeLoc(
+    clang::DependentBitIntTypeLoc TL) {
   TL.setNameLoc(readSourceLocation());
 }
 
@@ -11761,6 +11765,9 @@ OMPClause *OMPClauseReader::readClause() {
   case llvm::omp::OMPC_capture:
     C = new (Context) OMPCaptureClause();
     break;
+  case llvm::omp::OMPC_compare:
+    C = new (Context) OMPCompareClause();
+    break;
   case llvm::omp::OMPC_seq_cst:
     C = new (Context) OMPSeqCstClause();
     break;
@@ -12119,6 +12126,8 @@ void OMPClauseReader::VisitOMPUpdateClause(OMPUpdateClause *C) {
 
 void OMPClauseReader::VisitOMPCaptureClause(OMPCaptureClause *) {}
 
+void OMPClauseReader::VisitOMPCompareClause(OMPCompareClause *) {}
+
 void OMPClauseReader::VisitOMPSeqCstClause(OMPSeqCstClause *) {}
 
 void OMPClauseReader::VisitOMPAcqRelClause(OMPAcqRelClause *) {}
diff --git a/clang/lib/Serialization/ASTReaderDecl.cpp b/clang/lib/Serialization/ASTReaderDecl.cpp
index 62a31f299d6b..2144befcdb14 100644
--- a/clang/lib/Serialization/ASTReaderDecl.cpp
+++ b/clang/lib/Serialization/ASTReaderDecl.cpp
@@ -2948,6 +2948,7 @@ uint64_t ASTReader::getGlobalBitOffset(ModuleFile &M, uint64_t LocalOffset) {
 static bool isSameTemplateParameterList(const ASTContext &C,
                                         const TemplateParameterList *X,
                                         const TemplateParameterList *Y);
+static bool isSameEntity(NamedDecl *X, NamedDecl *Y);
 
 /// Determine whether two template parameters are similar enough
 /// that they may be used in declarations of the same template.
@@ -2967,7 +2968,9 @@ static bool isSameTemplateParameter(const NamedDecl *X,
     if (!TXTC != !TYTC)
       return false;
     if (TXTC && TYTC) {
-      if (TXTC->getNamedConcept() != TYTC->getNamedConcept())
+      auto *NCX = TXTC->getNamedConcept();
+      auto *NCY = TYTC->getNamedConcept();
+      if (!NCX || !NCY || !isSameEntity(NCX, NCY))
         return false;
       if (TXTC->hasExplicitTemplateArgs() != TYTC->hasExplicitTemplateArgs())
         return false;
@@ -3111,11 +3114,12 @@ static bool hasSameOverloadableAttrs(const FunctionDecl *A,
 
 /// Determine whether the two declarations refer to the same entity.
 static bool isSameEntity(NamedDecl *X, NamedDecl *Y) {
-  assert(X->getDeclName() == Y->getDeclName() && "Declaration name mismatch!");
-
   if (X == Y)
     return true;
 
+  if (X->getDeclName() != Y->getDeclName())
+    return false;
+
   // Must be in the same context.
   //
   // Note that we can't use DeclContext::Equals here, because the DeclContexts
diff --git a/clang/lib/Serialization/ASTWriter.cpp b/clang/lib/Serialization/ASTWriter.cpp
index a1972f5c6496..65a780e67510 100644
--- a/clang/lib/Serialization/ASTWriter.cpp
+++ b/clang/lib/Serialization/ASTWriter.cpp
@@ -396,6 +396,10 @@ void TypeLocWriter::VisitUnresolvedUsingTypeLoc(UnresolvedUsingTypeLoc TL) {
   Record.AddSourceLocation(TL.getNameLoc());
 }
 
+void TypeLocWriter::VisitUsingTypeLoc(UsingTypeLoc TL) {
+  Record.AddSourceLocation(TL.getNameLoc());
+}
+
 void TypeLocWriter::VisitTypedefTypeLoc(TypedefTypeLoc TL) {
   Record.AddSourceLocation(TL.getNameLoc());
 }
@@ -562,11 +566,11 @@ void TypeLocWriter::VisitPipeTypeLoc(PipeTypeLoc TL) {
   Record.AddSourceLocation(TL.getKWLoc());
 }
 
-void TypeLocWriter::VisitExtIntTypeLoc(clang::ExtIntTypeLoc TL) {
+void TypeLocWriter::VisitBitIntTypeLoc(clang::BitIntTypeLoc TL) {
   Record.AddSourceLocation(TL.getNameLoc());
 }
-void TypeLocWriter::VisitDependentExtIntTypeLoc(
-    clang::DependentExtIntTypeLoc TL) {
+void TypeLocWriter::VisitDependentBitIntTypeLoc(
+    clang::DependentBitIntTypeLoc TL) {
   Record.AddSourceLocation(TL.getNameLoc());
 }
 
@@ -6248,6 +6252,8 @@ void OMPClauseWriter::VisitOMPUpdateClause(OMPUpdateClause *C) {
 
 void OMPClauseWriter::VisitOMPCaptureClause(OMPCaptureClause *) {}
 
+void OMPClauseWriter::VisitOMPCompareClause(OMPCompareClause *) {}
+
 void OMPClauseWriter::VisitOMPSeqCstClause(OMPSeqCstClause *) {}
 
 void OMPClauseWriter::VisitOMPAcqRelClause(OMPAcqRelClause *) {}
diff --git a/clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
index 13781b336426..4a56156de4b2 100644
--- a/clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
@@ -66,7 +66,8 @@ bool BuiltinFunctionChecker::evalCall(const CallEvent &Call,
   case Builtin::BI__builtin_expect:
   case Builtin::BI__builtin_expect_with_probability:
   case Builtin::BI__builtin_assume_aligned:
-  case Builtin::BI__builtin_addressof: {
+  case Builtin::BI__builtin_addressof:
+  case Builtin::BI__builtin_function_start: {
     // For __builtin_unpredictable, __builtin_expect,
     // __builtin_expect_with_probability and __builtin_assume_aligned,
     // just return the value of the subexpression.
diff --git a/clang/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
index 8da482a2aec9..6ea39fb95e9a 100644
--- a/clang/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
@@ -56,9 +56,8 @@ private:
 
 void ConversionChecker::checkPreStmt(const ImplicitCastExpr *Cast,
                                      CheckerContext &C) const {
-  // TODO: For now we only warn about DeclRefExpr, to avoid noise. Warn for
-  // calculations also.
-  if (!isa<DeclRefExpr>(Cast->IgnoreParenImpCasts()))
+  // Don't warn for implicit conversions to bool
+  if (Cast->getType()->isBooleanType())
     return;
 
   // Don't warn for loss of sign/precision in macros.
@@ -70,6 +69,9 @@ void ConversionChecker::checkPreStmt(const ImplicitCastExpr *Cast,
   const Stmt *Parent = PM.getParent(Cast);
   if (!Parent)
     return;
+  // Dont warn if this is part of an explicit cast
+  if (isa<ExplicitCastExpr>(Parent))
+    return;
 
   bool LossOfSign = false;
   bool LossOfPrecision = false;
@@ -78,8 +80,10 @@ void ConversionChecker::checkPreStmt(const ImplicitCastExpr *Cast,
   if (const auto *B = dyn_cast<BinaryOperator>(Parent)) {
     BinaryOperator::Opcode Opc = B->getOpcode();
     if (Opc == BO_Assign) {
-      LossOfSign = isLossOfSign(Cast, C);
-      LossOfPrecision = isLossOfPrecision(Cast, Cast->getType(), C);
+      if (!Cast->IgnoreParenImpCasts()->isEvaluatable(C.getASTContext())) {
+        LossOfSign = isLossOfSign(Cast, C);
+        LossOfPrecision = isLossOfPrecision(Cast, Cast->getType(), C);
+      }
     } else if (Opc == BO_AddAssign || Opc == BO_SubAssign) {
       // No loss of sign.
       LossOfPrecision = isLossOfPrecision(Cast, B->getLHS()->getType(), C);
@@ -98,7 +102,12 @@ void ConversionChecker::checkPreStmt(const ImplicitCastExpr *Cast,
     } else if (B->isRelationalOp() || B->isMultiplicativeOp()) {
       LossOfSign = isLossOfSign(Cast, C);
     }
-  } else if (isa<DeclStmt>(Parent)) {
+  } else if (isa<DeclStmt, ReturnStmt>(Parent)) {
+    if (!Cast->IgnoreParenImpCasts()->isEvaluatable(C.getASTContext())) {
+      LossOfSign = isLossOfSign(Cast, C);
+      LossOfPrecision = isLossOfPrecision(Cast, Cast->getType(), C);
+    }
+  } else {
     LossOfSign = isLossOfSign(Cast, C);
     LossOfPrecision = isLossOfPrecision(Cast, Cast->getType(), C);
   }
diff --git a/clang/lib/StaticAnalyzer/Checkers/FuchsiaHandleChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/FuchsiaHandleChecker.cpp
index e3f4be0726c8..6e4801aa8e91 100644
--- a/clang/lib/StaticAnalyzer/Checkers/FuchsiaHandleChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/FuchsiaHandleChecker.cpp
@@ -389,7 +389,7 @@ void FuchsiaHandleChecker::checkPostCall(const CallEvent &Call,
         llvm::raw_string_ostream OS(SBuf);
         OS << "Function '" << FuncDecl->getDeclName()
            << "' returns an open handle";
-        return OS.str();
+        return SBuf;
       } else
         return "";
     });
@@ -405,7 +405,7 @@ void FuchsiaHandleChecker::checkPostCall(const CallEvent &Call,
         llvm::raw_string_ostream OS(SBuf);
         OS << "Function '" << FuncDecl->getDeclName()
            << "' returns an unowned handle";
-        return OS.str();
+        return SBuf;
       } else
         return "";
     });
@@ -439,7 +439,7 @@ void FuchsiaHandleChecker::checkPostCall(const CallEvent &Call,
               llvm::raw_string_ostream OS(SBuf);
               OS << "Handle released through " << ParamDiagIdx
                  << llvm::getOrdinalSuffix(ParamDiagIdx) << " parameter";
-              return OS.str();
+              return SBuf;
             } else
               return "";
           });
@@ -453,7 +453,7 @@ void FuchsiaHandleChecker::checkPostCall(const CallEvent &Call,
             llvm::raw_string_ostream OS(SBuf);
             OS << "Handle allocated through " << ParamDiagIdx
                << llvm::getOrdinalSuffix(ParamDiagIdx) << " parameter";
-            return OS.str();
+            return SBuf;
           } else
             return "";
         });
@@ -467,7 +467,7 @@ void FuchsiaHandleChecker::checkPostCall(const CallEvent &Call,
             llvm::raw_string_ostream OS(SBuf);
             OS << "Unowned handle allocated through " << ParamDiagIdx
                << llvm::getOrdinalSuffix(ParamDiagIdx) << " parameter";
-            return OS.str();
+            return SBuf;
           } else
             return "";
         });
diff --git a/clang/lib/StaticAnalyzer/Checkers/SmartPtr.h b/clang/lib/StaticAnalyzer/Checkers/SmartPtr.h
index 6a40f8eda5fa..b4352b450c7f 100644
--- a/clang/lib/StaticAnalyzer/Checkers/SmartPtr.h
+++ b/clang/lib/StaticAnalyzer/Checkers/SmartPtr.h
@@ -25,8 +25,6 @@ bool isStdSmartPtrCall(const CallEvent &Call);
 bool isStdSmartPtr(const CXXRecordDecl *RD);
 bool isStdSmartPtr(const Expr *E);
 
-bool isStdSmartPtr(const CXXRecordDecl *RD);
-
 /// Returns whether the smart pointer is null or not.
 bool isNullSmartPtr(const ProgramStateRef State, const MemRegion *ThisRegion);
 
diff --git a/clang/lib/StaticAnalyzer/Core/CallEvent.cpp b/clang/lib/StaticAnalyzer/Core/CallEvent.cpp
index 764dad3e7ab4..ae46709340d3 100644
--- a/clang/lib/StaticAnalyzer/Core/CallEvent.cpp
+++ b/clang/lib/StaticAnalyzer/Core/CallEvent.cpp
@@ -762,9 +762,9 @@ void CXXInstanceCall::getInitialStackFrameContents(
       QualType Ty = Ctx.getPointerType(Ctx.getRecordType(Class));
 
       // FIXME: CallEvent maybe shouldn't be directly accessing StoreManager.
-      bool Failed;
-      ThisVal = StateMgr.getStoreManager().attemptDownCast(ThisVal, Ty, Failed);
-      if (Failed) {
+      Optional<SVal> V =
+          StateMgr.getStoreManager().evalBaseToDerived(ThisVal, Ty);
+      if (!V.hasValue()) {
         // We might have suffered some sort of placement new earlier, so
         // we're constructing in a completely unexpected storage.
         // Fall back to a generic pointer cast for this-value.
@@ -772,7 +772,8 @@ void CXXInstanceCall::getInitialStackFrameContents(
         const CXXRecordDecl *StaticClass = StaticMD->getParent();
         QualType StaticTy = Ctx.getPointerType(Ctx.getRecordType(StaticClass));
         ThisVal = SVB.evalCast(ThisVal, Ty, StaticTy);
-      }
+      } else
+        ThisVal = *V;
     }
 
     if (!ThisVal.isUnknown())
diff --git a/clang/lib/StaticAnalyzer/Core/ExprEngineC.cpp b/clang/lib/StaticAnalyzer/Core/ExprEngineC.cpp
index 69d67cf9b465..637e4edfd778 100644
--- a/clang/lib/StaticAnalyzer/Core/ExprEngineC.cpp
+++ b/clang/lib/StaticAnalyzer/Core/ExprEngineC.cpp
@@ -439,14 +439,15 @@ void ExprEngine::VisitCast(const CastExpr *CastE, const Expr *Ex,
         if (CastE->isGLValue())
           resultType = getContext().getPointerType(resultType);
 
-        bool Failed = false;
-
-        // Check if the value being cast evaluates to 0.
-        if (val.isZeroConstant())
-          Failed = true;
-        // Else, evaluate the cast.
-        else
-          val = getStoreManager().attemptDownCast(val, T, Failed);
+        bool Failed = true;
+
+        // Check if the value being cast does not evaluates to 0.
+        if (!val.isZeroConstant())
+          if (Optional<SVal> V =
+                  StateMgr.getStoreManager().evalBaseToDerived(val, T)) {
+            val = *V;
+            Failed = false;
+          }
 
         if (Failed) {
           if (T->isReferenceType()) {
@@ -478,14 +479,13 @@ void ExprEngine::VisitCast(const CastExpr *CastE, const Expr *Ex,
         if (CastE->isGLValue())
           resultType = getContext().getPointerType(resultType);
 
-        bool Failed = false;
-
         if (!val.isConstant()) {
-          val = getStoreManager().attemptDownCast(val, T, Failed);
+          Optional<SVal> V = getStoreManager().evalBaseToDerived(val, T);
+          val = V ? *V : UnknownVal();
         }
 
         // Failed to cast or the result is unknown, fall back to conservative.
-        if (Failed || val.isUnknown()) {
+        if (val.isUnknown()) {
           val =
             svalBuilder.conjureSymbolVal(nullptr, CastE, LCtx, resultType,
                                          currBldrCtx->blockCount());
diff --git a/clang/lib/StaticAnalyzer/Core/HTMLDiagnostics.cpp b/clang/lib/StaticAnalyzer/Core/HTMLDiagnostics.cpp
index 3b847d6f0d87..b4578385a147 100644
--- a/clang/lib/StaticAnalyzer/Core/HTMLDiagnostics.cpp
+++ b/clang/lib/StaticAnalyzer/Core/HTMLDiagnostics.cpp
@@ -410,7 +410,7 @@ std::string HTMLDiagnostics::GenerateHTML(const PathDiagnostic& D, Rewriter &R,
     }
 
     // Append files to the main report file in the order they appear in the path
-    for (auto I : llvm::make_range(FileIDs.begin() + 1, FileIDs.end())) {
+    for (auto I : llvm::drop_begin(FileIDs)) {
       std::string s;
       llvm::raw_string_ostream os(s);
 
@@ -437,7 +437,7 @@ std::string HTMLDiagnostics::GenerateHTML(const PathDiagnostic& D, Rewriter &R,
   for (auto BI : *Buf)
     os << BI;
 
-  return os.str();
+  return file;
 }
 
 void HTMLDiagnostics::dumpCoverageData(
@@ -534,7 +534,7 @@ document.addEventListener("DOMContentLoaded", function() {
 </form>
 )<<<";
 
-  return os.str();
+  return s;
 }
 
 void HTMLDiagnostics::FinalizeHTML(const PathDiagnostic& D, Rewriter &R,
@@ -1202,7 +1202,7 @@ std::string getSpanBeginForControl(const char *ClassName, unsigned Index) {
   std::string Result;
   llvm::raw_string_ostream OS(Result);
   OS << "<span id=\"" << ClassName << Index << "\">";
-  return OS.str();
+  return Result;
 }
 
 std::string getSpanBeginForControlStart(unsigned Index) {
diff --git a/clang/lib/StaticAnalyzer/Core/MemRegion.cpp b/clang/lib/StaticAnalyzer/Core/MemRegion.cpp
index f77fcb030a15..58bea12f8783 100644
--- a/clang/lib/StaticAnalyzer/Core/MemRegion.cpp
+++ b/clang/lib/StaticAnalyzer/Core/MemRegion.cpp
@@ -444,7 +444,7 @@ std::string MemRegion::getString() const {
   std::string s;
   llvm::raw_string_ostream os(s);
   dumpToStream(os);
-  return os.str();
+  return s;
 }
 
 void MemRegion::dumpToStream(raw_ostream &os) const {
diff --git a/clang/lib/StaticAnalyzer/Core/RangeConstraintManager.cpp b/clang/lib/StaticAnalyzer/Core/RangeConstraintManager.cpp
index 23c67c64f975..4b0d4942e528 100644
--- a/clang/lib/StaticAnalyzer/Core/RangeConstraintManager.cpp
+++ b/clang/lib/StaticAnalyzer/Core/RangeConstraintManager.cpp
@@ -110,6 +110,14 @@ public:
 
 RangeSet::ContainerType RangeSet::Factory::EmptySet{};
 
+RangeSet RangeSet::Factory::add(RangeSet LHS, RangeSet RHS) {
+  ContainerType Result;
+  Result.reserve(LHS.size() + RHS.size());
+  std::merge(LHS.begin(), LHS.end(), RHS.begin(), RHS.end(),
+             std::back_inserter(Result));
+  return makePersistent(std::move(Result));
+}
+
 RangeSet RangeSet::Factory::add(RangeSet Original, Range Element) {
   ContainerType Result;
   Result.reserve(Original.size() + 1);
@@ -126,6 +134,186 @@ RangeSet RangeSet::Factory::add(RangeSet Original, const llvm::APSInt &Point) {
   return add(Original, Range(Point));
 }
 
+RangeSet RangeSet::Factory::unite(RangeSet LHS, RangeSet RHS) {
+  ContainerType Result = unite(*LHS.Impl, *RHS.Impl);
+  return makePersistent(std::move(Result));
+}
+
+RangeSet RangeSet::Factory::unite(RangeSet Original, Range R) {
+  ContainerType Result;
+  Result.push_back(R);
+  Result = unite(*Original.Impl, Result);
+  return makePersistent(std::move(Result));
+}
+
+RangeSet RangeSet::Factory::unite(RangeSet Original, llvm::APSInt Point) {
+  return unite(Original, Range(ValueFactory.getValue(Point)));
+}
+
+RangeSet RangeSet::Factory::unite(RangeSet Original, llvm::APSInt From,
+                                  llvm::APSInt To) {
+  return unite(Original,
+               Range(ValueFactory.getValue(From), ValueFactory.getValue(To)));
+}
+
+template <typename T>
+void swapIterators(T &First, T &FirstEnd, T &Second, T &SecondEnd) {
+  std::swap(First, Second);
+  std::swap(FirstEnd, SecondEnd);
+}
+
+RangeSet::ContainerType RangeSet::Factory::unite(const ContainerType &LHS,
+                                                 const ContainerType &RHS) {
+  if (LHS.empty())
+    return RHS;
+  if (RHS.empty())
+    return LHS;
+
+  using llvm::APSInt;
+  using iterator = ContainerType::const_iterator;
+
+  iterator First = LHS.begin();
+  iterator FirstEnd = LHS.end();
+  iterator Second = RHS.begin();
+  iterator SecondEnd = RHS.end();
+  APSIntType Ty = APSIntType(First->From());
+  const APSInt Min = Ty.getMinValue();
+
+  // Handle a corner case first when both range sets start from MIN.
+  // This helps to avoid complicated conditions below. Specifically, this
+  // particular check for `MIN` is not needed in the loop below every time
+  // when we do `Second->From() - One` operation.
+  if (Min == First->From() && Min == Second->From()) {
+    if (First->To() > Second->To()) {
+      //    [ First    ]--->
+      //    [ Second ]----->
+      // MIN^
+      // The Second range is entirely inside the First one.
+
+      // Check if Second is the last in its RangeSet.
+      if (++Second == SecondEnd)
+        //    [ First     ]--[ First + 1 ]--->
+        //    [ Second ]--------------------->
+        // MIN^
+        // The Union is equal to First's RangeSet.
+        return LHS;
+    } else {
+      // case 1: [ First ]----->
+      // case 2: [ First   ]--->
+      //         [ Second  ]--->
+      //      MIN^
+      // The First range is entirely inside or equal to the Second one.
+
+      // Check if First is the last in its RangeSet.
+      if (++First == FirstEnd)
+        //    [ First ]----------------------->
+        //    [ Second  ]--[ Second + 1 ]---->
+        // MIN^
+        // The Union is equal to Second's RangeSet.
+        return RHS;
+    }
+  }
+
+  const APSInt One = Ty.getValue(1);
+  ContainerType Result;
+
+  // This is called when there are no ranges left in one of the ranges.
+  // Append the rest of the ranges from another range set to the Result
+  // and return with that.
+  const auto AppendTheRest = [&Result](iterator I, iterator E) {
+    Result.append(I, E);
+    return Result;
+  };
+
+  while (true) {
+    // We want to keep the following invariant at all times:
+    // ---[ First ------>
+    // -----[ Second --->
+    if (First->From() > Second->From())
+      swapIterators(First, FirstEnd, Second, SecondEnd);
+
+    // The Union definitely starts with First->From().
+    // ----------[ First ------>
+    // ------------[ Second --->
+    // ----------[ Union ------>
+    // UnionStart^
+    const llvm::APSInt &UnionStart = First->From();
+
+    // Loop where the invariant holds.
+    while (true) {
+      // Skip all enclosed ranges.
+      // ---[                  First                     ]--->
+      // -----[ Second ]--[ Second + 1 ]--[ Second + N ]----->
+      while (First->To() >= Second->To()) {
+        // Check if Second is the last in its RangeSet.
+        if (++Second == SecondEnd) {
+          // Append the Union.
+          // ---[ Union      ]--->
+          // -----[ Second ]----->
+          // --------[ First ]--->
+          //         UnionEnd^
+          Result.emplace_back(UnionStart, First->To());
+          // ---[ Union ]----------------->
+          // --------------[ First + 1]--->
+          // Append all remaining ranges from the First's RangeSet.
+          return AppendTheRest(++First, FirstEnd);
+        }
+      }
+
+      // Check if First and Second are disjoint. It means that we find
+      // the end of the Union. Exit the loop and append the Union.
+      // ---[ First ]=------------->
+      // ------------=[ Second ]--->
+      // ----MinusOne^
+      if (First->To() < Second->From() - One)
+        break;
+
+      // First is entirely inside the Union. Go next.
+      // ---[ Union ----------->
+      // ---- [ First ]-------->
+      // -------[ Second ]----->
+      // Check if First is the last in its RangeSet.
+      if (++First == FirstEnd) {
+        // Append the Union.
+        // ---[ Union       ]--->
+        // -----[ First ]------->
+        // --------[ Second ]--->
+        //          UnionEnd^
+        Result.emplace_back(UnionStart, Second->To());
+        // ---[ Union ]------------------>
+        // --------------[ Second + 1]--->
+        // Append all remaining ranges from the Second's RangeSet.
+        return AppendTheRest(++Second, SecondEnd);
+      }
+
+      // We know that we are at one of the two cases:
+      // case 1: --[ First ]--------->
+      // case 2: ----[ First ]------->
+      // --------[ Second ]---------->
+      // In both cases First starts after Second->From().
+      // Make sure that the loop invariant holds.
+      swapIterators(First, FirstEnd, Second, SecondEnd);
+    }
+
+    // Here First and Second are disjoint.
+    // Append the Union.
+    // ---[ Union    ]--------------->
+    // -----------------[ Second ]--->
+    // ------[ First ]--------------->
+    //       UnionEnd^
+    Result.emplace_back(UnionStart, First->To());
+
+    // Check if First is the last in its RangeSet.
+    if (++First == FirstEnd)
+      // ---[ Union ]--------------->
+      // --------------[ Second ]--->
+      // Append all remaining ranges from the Second's RangeSet.
+      return AppendTheRest(Second, SecondEnd);
+  }
+
+  llvm_unreachable("Normally, we should not reach here");
+}
+
 RangeSet RangeSet::Factory::getRangeSet(Range From) {
   ContainerType Result;
   Result.push_back(From);
@@ -155,13 +343,6 @@ RangeSet::ContainerType *RangeSet::Factory::construct(ContainerType &&From) {
   return new (Buffer) ContainerType(std::move(From));
 }
 
-RangeSet RangeSet::Factory::add(RangeSet LHS, RangeSet RHS) {
-  ContainerType Result;
-  std::merge(LHS.begin(), LHS.end(), RHS.begin(), RHS.end(),
-             std::back_inserter(Result));
-  return makePersistent(std::move(Result));
-}
-
 const llvm::APSInt &RangeSet::getMinValue() const {
   assert(!isEmpty());
   return begin()->From();
@@ -325,11 +506,6 @@ RangeSet RangeSet::Factory::intersect(const RangeSet::ContainerType &LHS,
   const_iterator First = LHS.begin(), Second = RHS.begin(),
                  FirstEnd = LHS.end(), SecondEnd = RHS.end();
 
-  const auto SwapIterators = [&First, &FirstEnd, &Second, &SecondEnd]() {
-    std::swap(First, Second);
-    std::swap(FirstEnd, SecondEnd);
-  };
-
   // If we ran out of ranges in one set, but not in the other,
   // it means that those elements are definitely not in the
   // intersection.
@@ -339,7 +515,7 @@ RangeSet RangeSet::Factory::intersect(const RangeSet::ContainerType &LHS,
     //    ----[ First ---------------------->
     //    --------[ Second ----------------->
     if (Second->From() < First->From())
-      SwapIterators();
+      swapIterators(First, FirstEnd, Second, SecondEnd);
 
     // Loop where the invariant holds:
     do {
@@ -373,7 +549,7 @@ RangeSet RangeSet::Factory::intersect(const RangeSet::ContainerType &LHS,
       if (Second->To() > First->To()) {
         // Here we make a decision to keep First as the "longer"
         // range.
-        SwapIterators();
+        swapIterators(First, FirstEnd, Second, SecondEnd);
       }
 
       // At this point, we have the following situation:
@@ -2191,42 +2367,6 @@ LLVM_NODISCARD ProgramStateRef reAssume(ProgramStateRef State,
                                      Constraint->getMaxValue(), true);
 }
 
-// Simplify the given symbol with the help of the SValBuilder. In
-// SValBuilder::symplifySval, we traverse the symbol tree and query the
-// constraint values for the sub-trees and if a value is a constant we do the
-// constant folding. Compound symbols might collapse to simpler symbol tree
-// that is still possible to further simplify. Thus, we do the simplification on
-// a new symbol tree until we reach the simplest form, i.e. the fixpoint.
-//
-// Consider the following symbol `(b * b) * b * b` which has this tree:
-//       *
-//      / \
-//     *   b
-//    /  \
-//   /    b
-// (b * b)
-// Now, if the `b * b == 1` new constraint is added then during the first
-// iteration we have the following transformations:
-//       *                  *
-//      / \                / \
-//     *   b     -->      b   b
-//    /  \
-//   /    b
-//  1
-// We need another iteration to reach the final result `1`.
-LLVM_NODISCARD
-static SVal simplifyUntilFixpoint(SValBuilder &SVB, ProgramStateRef State,
-                                  const SymbolRef Sym) {
-  SVal Val = SVB.makeSymbolVal(Sym);
-  SVal SimplifiedVal = SVB.simplifySVal(State, Val);
-  // Do the simplification until we can.
-  while (SimplifiedVal != Val) {
-    Val = SimplifiedVal;
-    SimplifiedVal = SVB.simplifySVal(State, Val);
-  }
-  return SimplifiedVal;
-}
-
 // Iterate over all symbols and try to simplify them. Once a symbol is
 // simplified then we check if we can merge the simplified symbol's equivalence
 // class to this class. This way, we simplify not just the symbols but the
@@ -2238,8 +2378,7 @@ EquivalenceClass::simplify(SValBuilder &SVB, RangeSet::Factory &F,
   SymbolSet ClassMembers = Class.getClassMembers(State);
   for (const SymbolRef &MemberSym : ClassMembers) {
 
-    const SVal SimplifiedMemberVal =
-        simplifyUntilFixpoint(SVB, State, MemberSym);
+    const SVal SimplifiedMemberVal = simplifyToSVal(State, MemberSym);
     const SymbolRef SimplifiedMemberSym = SimplifiedMemberVal.getAsSymbol();
 
     // The symbol is collapsed to a constant, check if the current State is
diff --git a/clang/lib/StaticAnalyzer/Core/SValBuilder.cpp b/clang/lib/StaticAnalyzer/Core/SValBuilder.cpp
index 71bfc86ab8f7..8edcef319088 100644
--- a/clang/lib/StaticAnalyzer/Core/SValBuilder.cpp
+++ b/clang/lib/StaticAnalyzer/Core/SValBuilder.cpp
@@ -459,13 +459,23 @@ SVal SValBuilder::evalBinOp(ProgramStateRef state, BinaryOperator::Opcode op,
     return evalBinOpLN(state, op, *LV, rhs.castAs<NonLoc>(), type);
   }
 
-  if (Optional<Loc> RV = rhs.getAs<Loc>()) {
-    // Support pointer arithmetic where the addend is on the left
-    // and the pointer on the right.
-    assert(op == BO_Add);
+  if (const Optional<Loc> RV = rhs.getAs<Loc>()) {
+    const auto IsCommutative = [](BinaryOperatorKind Op) {
+      return Op == BO_Mul || Op == BO_Add || Op == BO_And || Op == BO_Xor ||
+             Op == BO_Or;
+    };
+
+    if (IsCommutative(op)) {
+      // Swap operands.
+      return evalBinOpLN(state, op, *RV, lhs.castAs<NonLoc>(), type);
+    }
 
-    // Commute the operands.
-    return evalBinOpLN(state, op, *RV, lhs.castAs<NonLoc>(), type);
+    // If the right operand is a concrete int location then we have nothing
+    // better but to treat it as a simple nonloc.
+    if (auto RV = rhs.getAs<loc::ConcreteInt>()) {
+      const nonloc::ConcreteInt RhsAsLoc = makeIntVal(RV->getValue());
+      return evalBinOpNN(state, op, lhs.castAs<NonLoc>(), RhsAsLoc, type);
+    }
   }
 
   return evalBinOpNN(state, op, lhs.castAs<NonLoc>(), rhs.castAs<NonLoc>(),
diff --git a/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp b/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
index 4ca35dd06ae5..dad8a7b3caae 100644
--- a/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ b/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -21,6 +21,35 @@ using namespace ento;
 
 namespace {
 class SimpleSValBuilder : public SValBuilder {
+
+  // With one `simplifySValOnce` call, a compound symbols might collapse to
+  // simpler symbol tree that is still possible to further simplify. Thus, we
+  // do the simplification on a new symbol tree until we reach the simplest
+  // form, i.e. the fixpoint.
+  // Consider the following symbol `(b * b) * b * b` which has this tree:
+  //       *
+  //      / \
+  //     *   b
+  //    /  \
+  //   /    b
+  // (b * b)
+  // Now, if the `b * b == 1` new constraint is added then during the first
+  // iteration we have the following transformations:
+  //       *                  *
+  //      / \                / \
+  //     *   b     -->      b   b
+  //    /  \
+  //   /    b
+  //  1
+  // We need another iteration to reach the final result `1`.
+  SVal simplifyUntilFixpoint(ProgramStateRef State, SVal Val);
+
+  // Recursively descends into symbolic expressions and replaces symbols
+  // with their known values (in the sense of the getKnownValue() method).
+  // We traverse the symbol tree and query the constraint values for the
+  // sub-trees and if a value is a constant we do the constant folding.
+  SVal simplifySValOnce(ProgramStateRef State, SVal V);
+
 public:
   SimpleSValBuilder(llvm::BumpPtrAllocator &alloc, ASTContext &context,
                     ProgramStateManager &stateMgr)
@@ -40,8 +69,6 @@ public:
   ///  (integer) value, that value is returned. Otherwise, returns NULL.
   const llvm::APSInt *getKnownValue(ProgramStateRef state, SVal V) override;
 
-  /// Recursively descends into symbolic expressions and replaces symbols
-  /// with their known values (in the sense of the getKnownValue() method).
   SVal simplifySVal(ProgramStateRef State, SVal V) override;
 
   SVal MakeSymIntVal(const SymExpr *LHS, BinaryOperator::Opcode op,
@@ -1105,7 +1132,20 @@ const llvm::APSInt *SimpleSValBuilder::getKnownValue(ProgramStateRef state,
   return nullptr;
 }
 
+SVal SimpleSValBuilder::simplifyUntilFixpoint(ProgramStateRef State, SVal Val) {
+  SVal SimplifiedVal = simplifySValOnce(State, Val);
+  while (SimplifiedVal != Val) {
+    Val = SimplifiedVal;
+    SimplifiedVal = simplifySValOnce(State, Val);
+  }
+  return SimplifiedVal;
+}
+
 SVal SimpleSValBuilder::simplifySVal(ProgramStateRef State, SVal V) {
+  return simplifyUntilFixpoint(State, V);
+}
+
+SVal SimpleSValBuilder::simplifySValOnce(ProgramStateRef State, SVal V) {
   // For now, this function tries to constant-fold symbols inside a
   // nonloc::SymbolVal, and does nothing else. More simplifications should
   // be possible, such as constant-folding an index in an ElementRegion.
diff --git a/clang/lib/StaticAnalyzer/Core/Store.cpp b/clang/lib/StaticAnalyzer/Core/Store.cpp
index 05feb1325c93..2bcdb0faf5da 100644
--- a/clang/lib/StaticAnalyzer/Core/Store.cpp
+++ b/clang/lib/StaticAnalyzer/Core/Store.cpp
@@ -314,10 +314,7 @@ static const CXXRecordDecl *getCXXRecordType(const MemRegion *MR) {
   return nullptr;
 }
 
-SVal StoreManager::attemptDownCast(SVal Base, QualType TargetType,
-                                   bool &Failed) {
-  Failed = false;
-
+Optional<SVal> StoreManager::evalBaseToDerived(SVal Base, QualType TargetType) {
   const MemRegion *MR = Base.getAsRegion();
   if (!MR)
     return UnknownVal();
@@ -392,7 +389,9 @@ SVal StoreManager::attemptDownCast(SVal Base, QualType TargetType,
   }
 
   // We failed if the region we ended up with has perfect type info.
-  Failed = isa<TypedValueRegion>(MR);
+  if (isa<TypedValueRegion>(MR))
+    return None;
+
   return UnknownVal();
 }
 
diff --git a/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp b/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
index f692c68045ee..f6ddcb763f9d 100644
--- a/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
+++ b/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp
@@ -35,6 +35,7 @@
 #include "clang/StaticAnalyzer/Core/PathSensitive/AnalysisManager.h"
 #include "clang/StaticAnalyzer/Core/PathSensitive/ExprEngine.h"
 #include "llvm/ADT/PostOrderIterator.h"
+#include "llvm/ADT/ScopeExit.h"
 #include "llvm/ADT/Statistic.h"
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Path.h"
@@ -493,13 +494,11 @@ void AnalysisConsumer::HandleDeclsCallGraph(const unsigned LocalTUDeclsSize) {
   }
 }
 
-static bool isBisonFile(ASTContext &C) {
+static bool fileContainsString(StringRef Substring, ASTContext &C) {
   const SourceManager &SM = C.getSourceManager();
   FileID FID = SM.getMainFileID();
   StringRef Buffer = SM.getBufferOrFake(FID).getBuffer();
-  if (Buffer.startswith("/* A Bison parser, made by"))
-    return true;
-  return false;
+  return Buffer.contains(Substring);
 }
 
 void AnalysisConsumer::runAnalysisOnTranslationUnit(ASTContext &C) {
@@ -546,38 +545,48 @@ void AnalysisConsumer::reportAnalyzerProgress(StringRef S) {
 }
 
 void AnalysisConsumer::HandleTranslationUnit(ASTContext &C) {
-
   // Don't run the actions if an error has occurred with parsing the file.
   DiagnosticsEngine &Diags = PP.getDiagnostics();
   if (Diags.hasErrorOccurred() || Diags.hasFatalErrorOccurred())
     return;
 
-  if (isBisonFile(C)) {
+  // Explicitly destroy the PathDiagnosticConsumer.  This will flush its output.
+  // FIXME: This should be replaced with something that doesn't rely on
+  // side-effects in PathDiagnosticConsumer's destructor. This is required when
+  // used with option -disable-free.
+  const auto DiagFlusherScopeExit =
+      llvm::make_scope_exit([this] { Mgr.reset(); });
+
+  if (Opts->ShouldIgnoreBisonGeneratedFiles &&
+      fileContainsString("/* A Bison parser, made by", C)) {
     reportAnalyzerProgress("Skipping bison-generated file\n");
-  } else if (Opts->DisableAllCheckers) {
+    return;
+  }
 
-    // Don't analyze if the user explicitly asked for no checks to be performed
-    // on this file.
+  if (Opts->ShouldIgnoreFlexGeneratedFiles &&
+      fileContainsString("/* A lexical scanner generated by flex", C)) {
+    reportAnalyzerProgress("Skipping flex-generated file\n");
+    return;
+  }
+
+  // Don't analyze if the user explicitly asked for no checks to be performed
+  // on this file.
+  if (Opts->DisableAllCheckers) {
     reportAnalyzerProgress("All checks are disabled using a supplied option\n");
-  } else {
-    // Otherwise, just run the analysis.
-    runAnalysisOnTranslationUnit(C);
+    return;
   }
 
+  // Otherwise, just run the analysis.
+  runAnalysisOnTranslationUnit(C);
+
   // Count how many basic blocks we have not covered.
   NumBlocksInAnalyzedFunctions = FunctionSummaries.getTotalNumBasicBlocks();
   NumVisitedBlocksInAnalyzedFunctions =
       FunctionSummaries.getTotalNumVisitedBasicBlocks();
   if (NumBlocksInAnalyzedFunctions > 0)
     PercentReachableBlocks =
-      (FunctionSummaries.getTotalNumVisitedBasicBlocks() * 100) /
+        (FunctionSummaries.getTotalNumVisitedBasicBlocks() * 100) /
         NumBlocksInAnalyzedFunctions;
-
-  // Explicitly destroy the PathDiagnosticConsumer.  This will flush its output.
-  // FIXME: This should be replaced with something that doesn't rely on
-  // side-effects in PathDiagnosticConsumer's destructor. This is required when
-  // used with option -disable-free.
-  Mgr.reset();
 }
 
 AnalysisConsumer::AnalysisMode
diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp b/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp
index f7c711690d7e..acceec690c11 100644
--- a/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp
+++ b/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp
@@ -9,66 +9,54 @@
 #include "clang/Tooling/DependencyScanning/DependencyScanningFilesystem.h"
 #include "clang/Lex/DependencyDirectivesSourceMinimizer.h"
 #include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/SmallVectorMemoryBuffer.h"
 #include "llvm/Support/Threading.h"
 
 using namespace clang;
 using namespace tooling;
 using namespace dependencies;
 
-CachedFileSystemEntry CachedFileSystemEntry::createFileEntry(
-    StringRef Filename, llvm::vfs::FileSystem &FS, bool Minimize) {
+llvm::ErrorOr<llvm::vfs::Status>
+CachedFileSystemEntry::initFile(StringRef Filename, llvm::vfs::FileSystem &FS) {
   // Load the file and its content from the file system.
-  llvm::ErrorOr<std::unique_ptr<llvm::vfs::File>> MaybeFile =
-      FS.openFileForRead(Filename);
+  auto MaybeFile = FS.openFileForRead(Filename);
   if (!MaybeFile)
     return MaybeFile.getError();
-  llvm::ErrorOr<llvm::vfs::Status> Stat = (*MaybeFile)->status();
-  if (!Stat)
-    return Stat.getError();
+  auto File = std::move(*MaybeFile);
+
+  auto MaybeStat = File->status();
+  if (!MaybeStat)
+    return MaybeStat.getError();
+  auto Stat = std::move(*MaybeStat);
 
-  llvm::vfs::File &F = **MaybeFile;
-  llvm::ErrorOr<std::unique_ptr<llvm::MemoryBuffer>> MaybeBuffer =
-      F.getBuffer(Stat->getName());
+  auto MaybeBuffer = File->getBuffer(Stat.getName());
   if (!MaybeBuffer)
     return MaybeBuffer.getError();
+  auto Buffer = std::move(*MaybeBuffer);
+
+  OriginalContents = std::move(Buffer);
+  return Stat;
+}
+
+void CachedFileSystemEntry::minimizeFile() {
+  assert(OriginalContents && "minimizing missing contents");
 
   llvm::SmallString<1024> MinimizedFileContents;
   // Minimize the file down to directives that might affect the dependencies.
-  const auto &Buffer = *MaybeBuffer;
   SmallVector<minimize_source_to_dependency_directives::Token, 64> Tokens;
-  if (!Minimize || minimizeSourceToDependencyDirectives(
-                       Buffer->getBuffer(), MinimizedFileContents, Tokens)) {
-    // Use the original file unless requested otherwise, or
-    // if the minimization failed.
-    // FIXME: Propage the diagnostic if desired by the client.
-    CachedFileSystemEntry Result;
-    Result.MaybeStat = std::move(*Stat);
-    Result.Contents.reserve(Buffer->getBufferSize() + 1);
-    Result.Contents.append(Buffer->getBufferStart(), Buffer->getBufferEnd());
-    // Implicitly null terminate the contents for Clang's lexer.
-    Result.Contents.push_back('\0');
-    Result.Contents.pop_back();
-    return Result;
+  if (minimizeSourceToDependencyDirectives(OriginalContents->getBuffer(),
+                                           MinimizedFileContents, Tokens)) {
+    // FIXME: Propagate the diagnostic if desired by the client.
+    // Use the original file if the minimization failed.
+    MinimizedContentsStorage =
+        llvm::MemoryBuffer::getMemBuffer(*OriginalContents);
+    MinimizedContentsAccess.store(MinimizedContentsStorage.get());
+    return;
   }
 
-  CachedFileSystemEntry Result;
-  size_t Size = MinimizedFileContents.size();
-  Result.MaybeStat = llvm::vfs::Status(Stat->getName(), Stat->getUniqueID(),
-                                       Stat->getLastModificationTime(),
-                                       Stat->getUser(), Stat->getGroup(), Size,
-                                       Stat->getType(), Stat->getPermissions());
   // The contents produced by the minimizer must be null terminated.
   assert(MinimizedFileContents.data()[MinimizedFileContents.size()] == '\0' &&
          "not null terminated contents");
-  // Even though there's an implicit null terminator in the minimized contents,
-  // we want to temporarily make it explicit. This will ensure that the
-  // std::move will preserve it even if it needs to do a copy if the
-  // SmallString still has the small capacity.
-  MinimizedFileContents.push_back('\0');
-  Result.Contents = std::move(MinimizedFileContents);
-  // Now make the null terminator implicit again, so that Clang's lexer can find
-  // it right where the buffer ends.
-  Result.Contents.pop_back();
 
   // Compute the skipped PP ranges that speedup skipping over inactive
   // preprocessor blocks.
@@ -86,20 +74,20 @@ CachedFileSystemEntry CachedFileSystemEntry::createFileEntry(
     }
     Mapping[Range.Offset] = Range.Length;
   }
-  Result.PPSkippedRangeMapping = std::move(Mapping);
-
-  return Result;
-}
-
-CachedFileSystemEntry
-CachedFileSystemEntry::createDirectoryEntry(llvm::vfs::Status &&Stat) {
-  assert(Stat.isDirectory() && "not a directory!");
-  auto Result = CachedFileSystemEntry();
-  Result.MaybeStat = std::move(Stat);
-  return Result;
+  PPSkippedRangeMapping = std::move(Mapping);
+
+  MinimizedContentsStorage = std::make_unique<llvm::SmallVectorMemoryBuffer>(
+      std::move(MinimizedFileContents));
+  // The algorithm in `getOrCreateFileSystemEntry` uses the presence of
+  // minimized contents to decide whether an entry is up-to-date or not.
+  // If it is up-to-date, the skipped range mappings must be already computed.
+  // This is why we need to store the minimized contents **after** storing the
+  // skipped range mappings. Failing to do so would lead to a data race.
+  MinimizedContentsAccess.store(MinimizedContentsStorage.get());
 }
 
-DependencyScanningFilesystemSharedCache::SingleCache::SingleCache() {
+DependencyScanningFilesystemSharedCache::
+    DependencyScanningFilesystemSharedCache() {
   // This heuristic was chosen using a empirical testing on a
   // reasonably high core machine (iMacPro 18 cores / 36 threads). The cache
   // sharding gives a performance edge by reducing the lock contention.
@@ -111,19 +99,13 @@ DependencyScanningFilesystemSharedCache::SingleCache::SingleCache() {
 }
 
 DependencyScanningFilesystemSharedCache::SharedFileSystemEntry &
-DependencyScanningFilesystemSharedCache::SingleCache::get(StringRef Key) {
+DependencyScanningFilesystemSharedCache::get(StringRef Key) {
   CacheShard &Shard = CacheShards[llvm::hash_value(Key) % NumShards];
-  std::unique_lock<std::mutex> LockGuard(Shard.CacheLock);
+  std::lock_guard<std::mutex> LockGuard(Shard.CacheLock);
   auto It = Shard.Cache.try_emplace(Key);
   return It.first->getValue();
 }
 
-DependencyScanningFilesystemSharedCache::SharedFileSystemEntry &
-DependencyScanningFilesystemSharedCache::get(StringRef Key, bool Minimized) {
-  SingleCache &Cache = Minimized ? CacheMinimized : CacheOriginal;
-  return Cache.get(Key);
-}
-
 /// Whitelist file extensions that should be minimized, treating no extension as
 /// a source file that should be minimized.
 ///
@@ -134,15 +116,14 @@ static bool shouldMinimizeBasedOnExtension(StringRef Filename) {
   if (Ext.empty())
     return true; // C++ standard library
   return llvm::StringSwitch<bool>(Ext)
-    .CasesLower(".c", ".cc", ".cpp", ".c++", ".cxx", true)
-    .CasesLower(".h", ".hh", ".hpp", ".h++", ".hxx", true)
-    .CasesLower(".m", ".mm", true)
-    .CasesLower(".i", ".ii", ".mi", ".mmi", true)
-    .CasesLower(".def", ".inc", true)
-    .Default(false);
+      .CasesLower(".c", ".cc", ".cpp", ".c++", ".cxx", true)
+      .CasesLower(".h", ".hh", ".hpp", ".h++", ".hxx", true)
+      .CasesLower(".m", ".mm", true)
+      .CasesLower(".i", ".ii", ".mi", ".mmi", true)
+      .CasesLower(".def", ".inc", true)
+      .Default(false);
 }
 
-
 static bool shouldCacheStatFailures(StringRef Filename) {
   StringRef Ext = llvm::sys::path::extension(Filename);
   if (Ext.empty())
@@ -167,38 +148,33 @@ bool DependencyScanningWorkerFilesystem::shouldMinimize(StringRef RawFilename) {
   return !NotToBeMinimized.contains(Filename);
 }
 
-CachedFileSystemEntry DependencyScanningWorkerFilesystem::createFileSystemEntry(
-    llvm::ErrorOr<llvm::vfs::Status> &&MaybeStatus, StringRef Filename,
-    bool ShouldMinimize) {
-  if (!MaybeStatus)
-    return CachedFileSystemEntry(MaybeStatus.getError());
-
-  if (MaybeStatus->isDirectory())
-    return CachedFileSystemEntry::createDirectoryEntry(std::move(*MaybeStatus));
-
-  return CachedFileSystemEntry::createFileEntry(Filename, getUnderlyingFS(),
-                                                ShouldMinimize);
+void CachedFileSystemEntry::init(llvm::ErrorOr<llvm::vfs::Status> &&MaybeStatus,
+                                 StringRef Filename,
+                                 llvm::vfs::FileSystem &FS) {
+  if (!MaybeStatus || MaybeStatus->isDirectory())
+    MaybeStat = std::move(MaybeStatus);
+  else
+    MaybeStat = initFile(Filename, FS);
 }
 
-llvm::ErrorOr<const CachedFileSystemEntry *>
+llvm::ErrorOr<EntryRef>
 DependencyScanningWorkerFilesystem::getOrCreateFileSystemEntry(
-    const StringRef Filename) {
-  bool ShouldMinimize = shouldMinimize(Filename);
+    StringRef Filename) {
+  bool ShouldBeMinimized = shouldMinimize(Filename);
 
-  if (const auto *Entry = Cache.getCachedEntry(Filename, ShouldMinimize))
-    return Entry;
+  const auto *Entry = LocalCache.getCachedEntry(Filename);
+  if (Entry && !Entry->needsUpdate(ShouldBeMinimized))
+    return EntryRef(ShouldBeMinimized, *Entry);
 
   // FIXME: Handle PCM/PCH files.
   // FIXME: Handle module map files.
 
-  DependencyScanningFilesystemSharedCache::SharedFileSystemEntry
-      &SharedCacheEntry = SharedCache.get(Filename, ShouldMinimize);
-  const CachedFileSystemEntry *Result;
+  auto &SharedCacheEntry = SharedCache.get(Filename);
   {
-    std::unique_lock<std::mutex> LockGuard(SharedCacheEntry.ValueLock);
+    std::lock_guard<std::mutex> LockGuard(SharedCacheEntry.ValueLock);
     CachedFileSystemEntry &CacheEntry = SharedCacheEntry.Value;
 
-    if (!CacheEntry.isValid()) {
+    if (!CacheEntry.isInitialized()) {
       auto MaybeStatus = getUnderlyingFS().status(Filename);
       if (!MaybeStatus && !shouldCacheStatFailures(Filename))
         // HACK: We need to always restat non source files if the stat fails.
@@ -206,27 +182,30 @@ DependencyScanningWorkerFilesystem::getOrCreateFileSystemEntry(
         //   files before building them, and then looks for them again. If we
         //   cache the stat failure, it won't see them the second time.
         return MaybeStatus.getError();
-      CacheEntry = createFileSystemEntry(std::move(MaybeStatus), Filename,
-                                         ShouldMinimize);
+      CacheEntry.init(std::move(MaybeStatus), Filename, getUnderlyingFS());
     }
 
-    Result = &CacheEntry;
+    // Checking `needsUpdate` verifies the entry represents an opened file.
+    // Only checking `needsMinimization` could lead to minimization of files
+    // that we failed to load (such files don't have `OriginalContents`).
+    if (CacheEntry.needsUpdate(ShouldBeMinimized))
+      CacheEntry.minimizeFile();
   }
 
   // Store the result in the local cache.
-  Cache.setCachedEntry(Filename, ShouldMinimize, Result);
-  return Result;
+  Entry = &SharedCacheEntry.Value;
+  return EntryRef(ShouldBeMinimized, *Entry);
 }
 
 llvm::ErrorOr<llvm::vfs::Status>
 DependencyScanningWorkerFilesystem::status(const Twine &Path) {
   SmallString<256> OwnedFilename;
   StringRef Filename = Path.toStringRef(OwnedFilename);
-  const llvm::ErrorOr<const CachedFileSystemEntry *> Result =
-      getOrCreateFileSystemEntry(Filename);
+
+  llvm::ErrorOr<EntryRef> Result = getOrCreateFileSystemEntry(Filename);
   if (!Result)
     return Result.getError();
-  return (*Result)->getStatus();
+  return Result->getStatus();
 }
 
 namespace {
@@ -240,7 +219,7 @@ public:
       : Buffer(std::move(Buffer)), Stat(std::move(Stat)) {}
 
   static llvm::ErrorOr<std::unique_ptr<llvm::vfs::File>>
-  create(const CachedFileSystemEntry *Entry,
+  create(EntryRef Entry,
          ExcludedPreprocessorDirectiveSkipMapping *PPSkipMappings);
 
   llvm::ErrorOr<llvm::vfs::Status> status() override { return Stat; }
@@ -261,21 +240,22 @@ private:
 } // end anonymous namespace
 
 llvm::ErrorOr<std::unique_ptr<llvm::vfs::File>> MinimizedVFSFile::create(
-    const CachedFileSystemEntry *Entry,
-    ExcludedPreprocessorDirectiveSkipMapping *PPSkipMappings) {
-  if (Entry->isDirectory())
-    return llvm::ErrorOr<std::unique_ptr<llvm::vfs::File>>(
-        std::make_error_code(std::errc::is_a_directory));
-  llvm::ErrorOr<StringRef> Contents = Entry->getContents();
+    EntryRef Entry, ExcludedPreprocessorDirectiveSkipMapping *PPSkipMappings) {
+  if (Entry.isDirectory())
+    return std::make_error_code(std::errc::is_a_directory);
+
+  llvm::ErrorOr<StringRef> Contents = Entry.getContents();
   if (!Contents)
     return Contents.getError();
   auto Result = std::make_unique<MinimizedVFSFile>(
-      llvm::MemoryBuffer::getMemBuffer(*Contents, Entry->getName(),
+      llvm::MemoryBuffer::getMemBuffer(*Contents, Entry.getName(),
                                        /*RequiresNullTerminator=*/false),
-      *Entry->getStatus());
-  if (!Entry->getPPSkippedRangeMapping().empty() && PPSkipMappings)
-    (*PPSkipMappings)[Result->Buffer->getBufferStart()] =
-        &Entry->getPPSkippedRangeMapping();
+      *Entry.getStatus());
+
+  const auto *EntrySkipMappings = Entry.getPPSkippedRangeMapping();
+  if (EntrySkipMappings && !EntrySkipMappings->empty() && PPSkipMappings)
+    (*PPSkipMappings)[Result->Buffer->getBufferStart()] = EntrySkipMappings;
+
   return llvm::ErrorOr<std::unique_ptr<llvm::vfs::File>>(
       std::unique_ptr<llvm::vfs::File>(std::move(Result)));
 }
@@ -285,8 +265,7 @@ DependencyScanningWorkerFilesystem::openFileForRead(const Twine &Path) {
   SmallString<256> OwnedFilename;
   StringRef Filename = Path.toStringRef(OwnedFilename);
 
-  const llvm::ErrorOr<const CachedFileSystemEntry *> Result =
-      getOrCreateFileSystemEntry(Filename);
+  llvm::ErrorOr<EntryRef> Result = getOrCreateFileSystemEntry(Filename);
   if (!Result)
     return Result.getError();
   return MinimizedVFSFile::create(Result.get(), PPSkipMappings);
diff --git a/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp b/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp
index 383a850301a1..086215e7a573 100644
--- a/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp
+++ b/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp
@@ -37,9 +37,13 @@ CompilerInvocation ModuleDepCollector::makeInvocationForModuleBuildWithoutPaths(
   CI.getLangOpts()->resetNonModularOptions();
   CI.getPreprocessorOpts().resetNonModularOptions();
 
-  // Remove options incompatible with explicit module build.
+  // Remove options incompatible with explicit module build or are likely to
+  // differ between identical modules discovered from different translation
+  // units.
   CI.getFrontendOpts().Inputs.clear();
   CI.getFrontendOpts().OutputFile.clear();
+  CI.getCodeGenOpts().MainFileName.clear();
+  CI.getCodeGenOpts().DwarfDebugFlags.clear();
 
   CI.getFrontendOpts().ProgramAction = frontend::GenerateModule;
   CI.getLangOpts()->ModuleName = Deps.ID.ModuleName;
@@ -233,7 +237,13 @@ ModuleID ModuleDepCollectorPP::handleTopLevelModule(const Module *M) {
                                    .getHeaderSearchInfo()
                                    .getModuleMap()
                                    .getModuleMapFileForUniquing(M);
-  MD.ClangModuleMapFile = std::string(ModuleMap ? ModuleMap->getName() : "");
+
+  if (ModuleMap) {
+    StringRef Path = ModuleMap->tryGetRealPathName();
+    if (Path.empty())
+      Path = ModuleMap->getName();
+    MD.ClangModuleMapFile = std::string(Path);
+  }
 
   serialization::ModuleFile *MF =
       MDC.ScanInstance.getASTReader()->getModuleManager().lookup(
diff --git a/clang/lib/Tooling/Syntax/Tokens.cpp b/clang/lib/Tooling/Syntax/Tokens.cpp
index 8a31e776d030..e2014f965c90 100644
--- a/clang/lib/Tooling/Syntax/Tokens.cpp
+++ b/clang/lib/Tooling/Syntax/Tokens.cpp
@@ -927,5 +927,5 @@ std::string TokenBuffer::dumpForTests() const {
           M.EndExpanded);
     }
   }
-  return OS.str();
+  return Dump;
 }
diff --git a/clang/lib/Tooling/Syntax/Tree.cpp b/clang/lib/Tooling/Syntax/Tree.cpp
index 1e3a90f3a316..c813865e95cd 100644
--- a/clang/lib/Tooling/Syntax/Tree.cpp
+++ b/clang/lib/Tooling/Syntax/Tree.cpp
@@ -263,7 +263,7 @@ std::string syntax::Node::dumpTokens(const SourceManager &SM) const {
       OS << " ";
     }
   });
-  return OS.str();
+  return Storage;
 }
 
 void syntax::Node::assertInvariants() const {