Implement P0482R2, support for char8_t type.

This is not yet part of any C++ working draft, and so is controlled by the flag
-fchar8_t rather than a -std= flag. (The GCC implementation is controlled by a
flag with the same name.)

This implementation is experimental, and will be removed or revised
substantially to match the proposal as it makes its way through the C++
committee.

llvm-svn: 331244

33 files changed, 143 insertions, 19 deletions

diff --git a/clang/lib/AST/ASTContext.cpp b/clang/lib/AST/ASTContext.cpp
index d61ca589d5b..d51071440e7 100644
--- a/clang/lib/AST/ASTContext.cpp
+++ b/clang/lib/AST/ASTContext.cpp
@@ -1151,6 +1151,9 @@ void ASTContext::InitBuiltinTypes(const TargetInfo &Target,
 
   WIntTy = getFromTargetType(Target.getWIntType());
 
+  // C++20 (proposed)
+  InitBuiltinType(Char8Ty,              BuiltinType::Char8);
+
   if (LangOpts.CPlusPlus) // C++0x 3.9.1p5, extension for C++
     InitBuiltinType(Char16Ty,           BuiltinType::Char16);
   else // C99
@@ -1739,6 +1742,7 @@ TypeInfo ASTContext::getTypeInfoImpl(const Type *T) const {
     case BuiltinType::Char_U:
     case BuiltinType::UChar:
     case BuiltinType::SChar:
+    case BuiltinType::Char8:
       Width = Target->getCharWidth();
       Align = Target->getCharAlign();
       break;
@@ -5456,6 +5460,7 @@ QualType ASTContext::getPromotedIntegerType(QualType Promotable) const {
     // FIXME: Is there some better way to compute this?
     if (BT->getKind() == BuiltinType::WChar_S ||
         BT->getKind() == BuiltinType::WChar_U ||
+        BT->getKind() == BuiltinType::Char8 ||
         BT->getKind() == BuiltinType::Char16 ||
         BT->getKind() == BuiltinType::Char32) {
       bool FromIsSigned = BT->getKind() == BuiltinType::WChar_S;
@@ -6202,6 +6207,7 @@ static char getObjCEncodingForPrimitiveKind(const ASTContext *C,
     switch (kind) {
     case BuiltinType::Void:       return 'v';
     case BuiltinType::Bool:       return 'B';
+    case BuiltinType::Char8:
     case BuiltinType::Char_U:
     case BuiltinType::UChar:      return 'C';
     case BuiltinType::Char16:
diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp
index 81577248612..c540dfbbf07 100644
--- a/clang/lib/AST/ExprConstant.cpp
+++ b/clang/lib/AST/ExprConstant.cpp
@@ -7326,6 +7326,7 @@ static int EvaluateBuiltinClassifyType(const CallExpr *E,
       return pointer_type_class;
 
     case BuiltinType::WChar_U:
+    case BuiltinType::Char8:
     case BuiltinType::Char16:
     case BuiltinType::Char32:
     case BuiltinType::ObjCId:
diff --git a/clang/lib/AST/ItaniumMangle.cpp b/clang/lib/AST/ItaniumMangle.cpp
index 609d0ebc849..610400d8322 100644
--- a/clang/lib/AST/ItaniumMangle.cpp
+++ b/clang/lib/AST/ItaniumMangle.cpp
@@ -2525,6 +2525,9 @@ void CXXNameMangler::mangleType(const BuiltinType *T) {
   case BuiltinType::WChar_U:
     Out << 'w';
     break;
+  case BuiltinType::Char8:
+    Out << "Du";
+    break;
   case BuiltinType::Char16:
     Out << "Ds";
     break;
diff --git a/clang/lib/AST/MicrosoftMangle.cpp b/clang/lib/AST/MicrosoftMangle.cpp
index 1039ae87ab6..e72804b2c3b 100644
--- a/clang/lib/AST/MicrosoftMangle.cpp
+++ b/clang/lib/AST/MicrosoftMangle.cpp
@@ -1918,6 +1918,7 @@ void MicrosoftCXXNameMangler::mangleType(const BuiltinType *T, Qualifiers,
     Out << "$$T";
     break;
 
+  case BuiltinType::Char8:
   case BuiltinType::Float16:
     mangleArtificalTagType(TTK_Struct, "_Float16", {"__clang"});
     break;
diff --git a/clang/lib/AST/NSAPI.cpp b/clang/lib/AST/NSAPI.cpp
index 8adaef1fb64..eb807f0cc53 100644
--- a/clang/lib/AST/NSAPI.cpp
+++ b/clang/lib/AST/NSAPI.cpp
@@ -436,6 +436,7 @@ NSAPI::getNSNumberFactoryMethodKind(QualType T) const {
   case BuiltinType::Void:
   case BuiltinType::WChar_U:
   case BuiltinType::WChar_S:
+  case BuiltinType::Char8:
   case BuiltinType::Char16:
   case BuiltinType::Char32:
   case BuiltinType::Int128:
diff --git a/clang/lib/AST/Type.cpp b/clang/lib/AST/Type.cpp
index a2a60772b77..571dc2abfa0 100644
--- a/clang/lib/AST/Type.cpp
+++ b/clang/lib/AST/Type.cpp
@@ -1763,6 +1763,12 @@ bool Type::isWideCharType() const {
   return false;
 }
 
+bool Type::isChar8Type() const {
+  if (const BuiltinType *BT = dyn_cast<BuiltinType>(CanonicalType))
+    return BT->getKind() == BuiltinType::Char8;
+  return false;
+}
+
 bool Type::isChar16Type() const {
   if (const auto *BT = dyn_cast<BuiltinType>(CanonicalType))
     return BT->getKind() == BuiltinType::Char16;
@@ -1785,6 +1791,7 @@ bool Type::isAnyCharacterType() const {
   case BuiltinType::Char_U:
   case BuiltinType::UChar:
   case BuiltinType::WChar_U:
+  case BuiltinType::Char8:
   case BuiltinType::Char16:
   case BuiltinType::Char32:
   case BuiltinType::Char_S:
@@ -2419,6 +2426,7 @@ bool Type::isPromotableIntegerType() const {
     case BuiltinType::UShort:
     case BuiltinType::WChar_S:
     case BuiltinType::WChar_U:
+    case BuiltinType::Char8:
     case BuiltinType::Char16:
     case BuiltinType::Char32:
       return true;
@@ -2655,6 +2663,8 @@ StringRef BuiltinType::getName(const PrintingPolicy &Policy) const {
   case WChar_S:
   case WChar_U:
     return Policy.MSWChar ? "__wchar_t" : "wchar_t";
+  case Char8:
+    return "char8_t";
   case Char16:
     return "char16_t";
   case Char32:
diff --git a/clang/lib/AST/TypeLoc.cpp b/clang/lib/AST/TypeLoc.cpp
index 0ac50b31ace..57349b46f54 100644
--- a/clang/lib/AST/TypeLoc.cpp
+++ b/clang/lib/AST/TypeLoc.cpp
@@ -317,6 +317,8 @@ TypeSpecifierType BuiltinTypeLoc::getWrittenTypeSpec() const {
   case BuiltinType::Char_U:
   case BuiltinType::Char_S:
     return TST_char;
+  case BuiltinType::Char8:
+    return TST_char8;
   case BuiltinType::Char16:
     return TST_char16;
   case BuiltinType::Char32:
diff --git a/clang/lib/Analysis/PrintfFormatString.cpp b/clang/lib/Analysis/PrintfFormatString.cpp
index dfaed26564e..2043970ccd4 100644
--- a/clang/lib/Analysis/PrintfFormatString.cpp
+++ b/clang/lib/Analysis/PrintfFormatString.cpp
@@ -647,6 +647,7 @@ bool PrintfSpecifier::fixType(QualType QT, const LangOptions &LangOpt,
   case BuiltinType::Bool:
   case BuiltinType::WChar_U:
   case BuiltinType::WChar_S:
+  case BuiltinType::Char8: // FIXME: Treat like 'char'?
   case BuiltinType::Char16:
   case BuiltinType::Char32:
   case BuiltinType::UInt128:
diff --git a/clang/lib/Basic/IdentifierTable.cpp b/clang/lib/Basic/IdentifierTable.cpp
index 6b0133208ba..025104c63be 100644
--- a/clang/lib/Basic/IdentifierTable.cpp
+++ b/clang/lib/Basic/IdentifierTable.cpp
@@ -115,14 +115,15 @@ namespace {
     KEYNOOPENCL = 0x02000,
     WCHARSUPPORT = 0x04000,
     HALFSUPPORT = 0x08000,
-    KEYCONCEPTS = 0x10000,
-    KEYOBJC2    = 0x20000,
-    KEYZVECTOR  = 0x40000,
-    KEYCOROUTINES = 0x80000,
-    KEYMODULES = 0x100000,
-    KEYCXX2A = 0x200000,
+    CHAR8SUPPORT = 0x10000,
+    KEYCONCEPTS = 0x20000,
+    KEYOBJC2    = 0x40000,
+    KEYZVECTOR  = 0x80000,
+    KEYCOROUTINES = 0x100000,
+    KEYMODULES = 0x200000,
+    KEYCXX2A = 0x400000,
     KEYALLCXX = KEYCXX | KEYCXX11 | KEYCXX2A,
-    KEYALL = (0x3fffff & ~KEYNOMS18 &
+    KEYALL = (0x7fffff & ~KEYNOMS18 &
               ~KEYNOOPENCL) // KEYNOMS18 and KEYNOOPENCL are used to exclude.
   };
 
@@ -151,6 +152,7 @@ static KeywordStatus getKeywordStatus(const LangOptions &LangOpts,
   if (LangOpts.Bool && (Flags & BOOLSUPPORT)) return KS_Enabled;
   if (LangOpts.Half && (Flags & HALFSUPPORT)) return KS_Enabled;
   if (LangOpts.WChar && (Flags & WCHARSUPPORT)) return KS_Enabled;
+  if (LangOpts.Char8 && (Flags & CHAR8SUPPORT)) return KS_Enabled;
   if (LangOpts.AltiVec && (Flags & KEYALTIVEC)) return KS_Enabled;
   if (LangOpts.OpenCL && (Flags & KEYOPENCL)) return KS_Enabled;
   if (!LangOpts.CPlusPlus && (Flags & KEYNOCXX)) return KS_Enabled;
diff --git a/clang/lib/CodeGen/CGDebugInfo.cpp b/clang/lib/CodeGen/CGDebugInfo.cpp
index 474018c065e..12209723999 100644
--- a/clang/lib/CodeGen/CGDebugInfo.cpp
+++ b/clang/lib/CodeGen/CGDebugInfo.cpp
@@ -665,6 +665,7 @@ llvm::DIType *CGDebugInfo::CreateType(const BuiltinType *BT) {
   case BuiltinType::SChar:
     Encoding = llvm::dwarf::DW_ATE_signed_char;
     break;
+  case BuiltinType::Char8:
   case BuiltinType::Char16:
   case BuiltinType::Char32:
     Encoding = llvm::dwarf::DW_ATE_UTF;
diff --git a/clang/lib/CodeGen/CodeGenTypes.cpp b/clang/lib/CodeGen/CodeGenTypes.cpp
index ccb6df9a580..ce1fdf9b125 100644
--- a/clang/lib/CodeGen/CodeGenTypes.cpp
+++ b/clang/lib/CodeGen/CodeGenTypes.cpp
@@ -437,6 +437,7 @@ llvm::Type *CodeGenTypes::ConvertType(QualType T) {
     case BuiltinType::ULongLong:
     case BuiltinType::WChar_S:
     case BuiltinType::WChar_U:
+    case BuiltinType::Char8:
     case BuiltinType::Char16:
     case BuiltinType::Char32:
       ResultType = llvm::IntegerType::get(getLLVMContext(),
diff --git a/clang/lib/CodeGen/ItaniumCXXABI.cpp b/clang/lib/CodeGen/ItaniumCXXABI.cpp
index f92d7ec0632..0e35633e1d6 100644
--- a/clang/lib/CodeGen/ItaniumCXXABI.cpp
+++ b/clang/lib/CodeGen/ItaniumCXXABI.cpp
@@ -2706,6 +2706,7 @@ static bool TypeInfoIsInStandardLibrary(const BuiltinType *Ty) {
     case BuiltinType::LongDouble:
     case BuiltinType::Float16:
     case BuiltinType::Float128:
+    case BuiltinType::Char8:
     case BuiltinType::Char16:
     case BuiltinType::Char32:
     case BuiltinType::Int128:
@@ -3567,7 +3568,8 @@ void ItaniumCXXABI::EmitFundamentalRTTIDescriptors(bool DLLExport) {
       getContext().UnsignedInt128Ty,   getContext().HalfTy,
       getContext().FloatTy,            getContext().DoubleTy,
       getContext().LongDoubleTy,       getContext().Float128Ty,
-      getContext().Char16Ty,           getContext().Char32Ty
+      getContext().Char8Ty,            getContext().Char16Ty,
+      getContext().Char32Ty
   };
   for (const QualType &FundamentalType : FundamentalTypes)
     EmitFundamentalRTTIDescriptor(FundamentalType, DLLExport);
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index 331b4beaa6b..5e5dfdeecae 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -2687,6 +2687,9 @@ static void RenderCharacterOptions(const ArgList &Args, const llvm::Triple &T,
     CmdArgs.push_back("-fno-signed-char");
   }
 
+  if (Args.hasFlag(options::OPT_fchar8__t, options::OPT_fno_char8__t, false))
+    CmdArgs.push_back("-fchar8_t");
+
   if (const Arg *A = Args.getLastArg(options::OPT_fshort_wchar,
                                      options::OPT_fno_short_wchar)) {
     if (A->getOption().matches(options::OPT_fshort_wchar)) {
diff --git a/clang/lib/Format/FormatToken.cpp b/clang/lib/Format/FormatToken.cpp
index 10ac392abbf..c63f0129c48 100644
--- a/clang/lib/Format/FormatToken.cpp
+++ b/clang/lib/Format/FormatToken.cpp
@@ -57,6 +57,7 @@ bool FormatToken::isSimpleTypeSpecifier() const {
   case tok::kw_bool:
   case tok::kw___underlying_type:
   case tok::annot_typename:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw_typeof:
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index 44f29ba3b10..a6dc1678d1d 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -2362,6 +2362,7 @@ static void ParseLangArgs(LangOptions &Opts, ArgList &Args, InputKind IK,
   Opts.ImplicitModules = !Args.hasArg(OPT_fno_implicit_modules);
   Opts.CharIsSigned = Opts.OpenCL || !Args.hasArg(OPT_fno_signed_char);
   Opts.WChar = Opts.CPlusPlus && !Args.hasArg(OPT_fno_wchar);
+  Opts.Char8 = Args.hasArg(OPT_fchar8__t);
   if (const Arg *A = Args.getLastArg(OPT_fwchar_type_EQ)) {
     Opts.WCharSize = llvm::StringSwitch<unsigned>(A->getValue())
                          .Case("char", 1)
diff --git a/clang/lib/Frontend/InitPreprocessor.cpp b/clang/lib/Frontend/InitPreprocessor.cpp
index 192862db657..8a87b9f35e3 100644
--- a/clang/lib/Frontend/InitPreprocessor.cpp
+++ b/clang/lib/Frontend/InitPreprocessor.cpp
@@ -559,6 +559,10 @@ static void InitializeCPlusPlusFeatureTestMacros(const LangOptions &LangOpts,
     Builder.defineMacro("__cpp_experimental_concepts", "1");
   if (LangOpts.CoroutinesTS)
     Builder.defineMacro("__cpp_coroutines", "201703L");
+
+  // Potential future breaking changes.
+  if (LangOpts.Char8)
+    Builder.defineMacro("__cpp_char8_t", "201803");
 }
 
 static void InitializePredefinedMacros(const TargetInfo &TI,
@@ -939,6 +943,8 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
                                        InlineWidthBits));
     DEFINE_LOCK_FREE_MACRO(BOOL, Bool);
     DEFINE_LOCK_FREE_MACRO(CHAR, Char);
+    if (LangOpts.Char8)
+      DEFINE_LOCK_FREE_MACRO(CHAR8_T, Char); // Treat char8_t like char.
     DEFINE_LOCK_FREE_MACRO(CHAR16_T, Char16);
     DEFINE_LOCK_FREE_MACRO(CHAR32_T, Char32);
     DEFINE_LOCK_FREE_MACRO(WCHAR_T, WChar);
diff --git a/clang/lib/Index/USRGeneration.cpp b/clang/lib/Index/USRGeneration.cpp
index ee1c950eb69..ba536c748c3 100644
--- a/clang/lib/Index/USRGeneration.cpp
+++ b/clang/lib/Index/USRGeneration.cpp
@@ -650,6 +650,8 @@ void USRGenerator::VisitType(QualType T) {
           c = 'b'; break;
         case BuiltinType::UChar:
           c = 'c'; break;
+        case BuiltinType::Char8:
+          c = 'u'; break; // FIXME: Check this doesn't collide
         case BuiltinType::Char16:
           c = 'q'; break;
         case BuiltinType::Char32:
diff --git a/clang/lib/Lex/PPExpressions.cpp b/clang/lib/Lex/PPExpressions.cpp
index d8431827e9c..b1ed0e10c6f 100644
--- a/clang/lib/Lex/PPExpressions.cpp
+++ b/clang/lib/Lex/PPExpressions.cpp
@@ -363,7 +363,7 @@ static bool EvaluateValue(PPValue &Result, Token &PeekTok, DefinedTracker &DT,
       NumBits = TI.getChar16Width();
     else if (Literal.isUTF32())
       NumBits = TI.getChar32Width();
-    else
+    else // char or char8_t
       NumBits = TI.getCharWidth();
 
     // Set the width.
diff --git a/clang/lib/Parse/ParseDecl.cpp b/clang/lib/Parse/ParseDecl.cpp
index fc0ca613c73..e0948b37170 100644
--- a/clang/lib/Parse/ParseDecl.cpp
+++ b/clang/lib/Parse/ParseDecl.cpp
@@ -3587,6 +3587,10 @@ void Parser::ParseDeclarationSpecifiers(DeclSpec &DS,
       isInvalid = DS.SetTypeSpecType(DeclSpec::TST_wchar, Loc, PrevSpec,
                                      DiagID, Policy);
       break;
+    case tok::kw_char8_t:
+      isInvalid = DS.SetTypeSpecType(DeclSpec::TST_char8, Loc, PrevSpec,
+                                     DiagID, Policy);
+      break;
     case tok::kw_char16_t:
       isInvalid = DS.SetTypeSpecType(DeclSpec::TST_char16, Loc, PrevSpec,
                                      DiagID, Policy);
@@ -4585,6 +4589,7 @@ bool Parser::isKnownToBeTypeSpecifier(const Token &Tok) const {
   case tok::kw_void:
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw_int:
@@ -4661,6 +4666,7 @@ bool Parser::isTypeSpecifierQualifier() {
   case tok::kw_void:
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw_int:
@@ -4817,6 +4823,7 @@ bool Parser::isDeclarationSpecifier(bool DisambiguatingWithExpression) {
   case tok::kw_void:
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
 
diff --git a/clang/lib/Parse/ParseExpr.cpp b/clang/lib/Parse/ParseExpr.cpp
index ca5d3bce9cf..1cc98568ded 100644
--- a/clang/lib/Parse/ParseExpr.cpp
+++ b/clang/lib/Parse/ParseExpr.cpp
@@ -1224,6 +1224,7 @@ ExprResult Parser::ParseCastExpression(bool isUnaryExpression,
   case tok::annot_decltype:
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw_bool:
diff --git a/clang/lib/Parse/ParseExprCXX.cpp b/clang/lib/Parse/ParseExprCXX.cpp
index 8e39adf8a5d..276dea1add1 100644
--- a/clang/lib/Parse/ParseExprCXX.cpp
+++ b/clang/lib/Parse/ParseExprCXX.cpp
@@ -1962,6 +1962,9 @@ void Parser::ParseCXXSimpleTypeSpecifier(DeclSpec &DS) {
   case tok::kw_wchar_t:
     DS.SetTypeSpecType(DeclSpec::TST_wchar, Loc, PrevSpec, DiagID, Policy);
     break;
+  case tok::kw_char8_t:
+    DS.SetTypeSpecType(DeclSpec::TST_char8, Loc, PrevSpec, DiagID, Policy);
+    break;
   case tok::kw_char16_t:
     DS.SetTypeSpecType(DeclSpec::TST_char16, Loc, PrevSpec, DiagID, Policy);
     break;
diff --git a/clang/lib/Parse/ParseTentative.cpp b/clang/lib/Parse/ParseTentative.cpp
index ebd6f0f5b8e..17ff9f9be6d 100644
--- a/clang/lib/Parse/ParseTentative.cpp
+++ b/clang/lib/Parse/ParseTentative.cpp
@@ -1052,6 +1052,7 @@ Parser::isExpressionOrTypeSpecifierSimple(tok::TokenKind Kind) {
   case tok::kw_class:
   case tok::kw_typename:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw__Decimal32:
@@ -1523,6 +1524,7 @@ Parser::isCXXDeclarationSpecifier(Parser::TPResult BracedCastResult,
       
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw_bool:
@@ -1614,6 +1616,7 @@ bool Parser::isCXXDeclarationSpecifierAType() {
     // simple-type-specifier
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char8_t:
   case tok::kw_char16_t:
   case tok::kw_char32_t:
   case tok::kw_bool:
diff --git a/clang/lib/Sema/DeclSpec.cpp b/clang/lib/Sema/DeclSpec.cpp
index 2fad5a18ba6..26d62021380 100644
--- a/clang/lib/Sema/DeclSpec.cpp
+++ b/clang/lib/Sema/DeclSpec.cpp
@@ -329,6 +329,7 @@ bool Declarator::isDeclarationOfFunction() const {
     case TST_auto_type:
     case TST_bool:
     case TST_char:
+    case TST_char8:
     case TST_char16:
     case TST_char32:
     case TST_class:
@@ -499,6 +500,7 @@ const char *DeclSpec::getSpecifierName(DeclSpec::TST T,
   case DeclSpec::TST_void:        return "void";
   case DeclSpec::TST_char:        return "char";
   case DeclSpec::TST_wchar:       return Policy.MSWChar ? "__wchar_t" : "wchar_t";
+  case DeclSpec::TST_char8:       return "char8_t";
   case DeclSpec::TST_char16:      return "char16_t";
   case DeclSpec::TST_char32:      return "char32_t";
   case DeclSpec::TST_int:         return "int";
@@ -1202,7 +1204,9 @@ void DeclSpec::Finish(Sema &S, const PrintingPolicy &Policy) {
       StorageClassSpec == SCS_auto)
     S.Diag(StorageClassSpecLoc, diag::warn_auto_storage_class)
       << FixItHint::CreateRemoval(StorageClassSpecLoc);
-  if (TypeSpecType == TST_char16 || TypeSpecType == TST_char32)
+  if (TypeSpecType == TST_char8)
+    S.Diag(TSTLoc, diag::warn_cxx17_compat_unicode_type);
+  else if (TypeSpecType == TST_char16 || TypeSpecType == TST_char32)
     S.Diag(TSTLoc, diag::warn_cxx98_compat_unicode_type)
       << (TypeSpecType == TST_char16 ? "char16_t" : "char32_t");
   if (Constexpr_specified)
diff --git a/clang/lib/Sema/SemaDecl.cpp b/clang/lib/Sema/SemaDecl.cpp
index 58676101d82..732853de48c 100644
--- a/clang/lib/Sema/SemaDecl.cpp
+++ b/clang/lib/Sema/SemaDecl.cpp
@@ -148,6 +148,9 @@ bool Sema::isSimpleTypeSpecifier(tok::TokenKind Kind) const {
   case tok::kw_decltype:
     return getLangOpts().CPlusPlus;
 
+  case tok::kw_char8_t:
+    return getLangOpts().Char8;
+
   default:
     break;
   }
diff --git a/clang/lib/Sema/SemaDeclCXX.cpp b/clang/lib/Sema/SemaDeclCXX.cpp
index 2cf16221f75..4339c9a8de1 100644
--- a/clang/lib/Sema/SemaDeclCXX.cpp
+++ b/clang/lib/Sema/SemaDeclCXX.cpp
@@ -13187,6 +13187,7 @@ bool Sema::CheckLiteralOperatorDeclaration(FunctionDecl *FnDecl) {
         ParamType->isSpecificBuiltinType(BuiltinType::LongDouble) ||
         Context.hasSameType(ParamType, Context.CharTy) ||
         Context.hasSameType(ParamType, Context.WideCharTy) ||
+        Context.hasSameType(ParamType, Context.Char8Ty) ||
         Context.hasSameType(ParamType, Context.Char16Ty) ||
         Context.hasSameType(ParamType, Context.Char32Ty)) {
     } else if (const PointerType *Ptr = ParamType->getAs<PointerType>()) {
@@ -13247,10 +13248,12 @@ bool Sema::CheckLiteralOperatorDeclaration(FunctionDecl *FnDecl) {
     }
 
     QualType InnerType = PointeeType.getUnqualifiedType();
-    // Only const char *, const wchar_t*, const char16_t*, and const char32_t*
-    // are allowed as the first parameter to a two-parameter function
+    // Only const char *, const wchar_t*, const char8_t*, const char16_t*, and
+    // const char32_t* are allowed as the first parameter to a two-parameter
+    // function
     if (!(Context.hasSameType(InnerType, Context.CharTy) ||
           Context.hasSameType(InnerType, Context.WideCharTy) ||
+          Context.hasSameType(InnerType, Context.Char8Ty) ||
           Context.hasSameType(InnerType, Context.Char16Ty) ||
           Context.hasSameType(InnerType, Context.Char32Ty))) {
       Diag((*Param)->getSourceRange().getBegin(),
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
index b1ecbfad4c3..58e70a4ceaa 100644
--- a/clang/lib/Sema/SemaExpr.cpp
+++ b/clang/lib/Sema/SemaExpr.cpp
@@ -1535,6 +1535,8 @@ Sema::ActOnStringLiteral(ArrayRef<Token> StringToks, Scope *UDLScope) {
     CharTy = Context.getWideCharType();
     Kind = StringLiteral::Wide;
   } else if (Literal.isUTF8()) {
+    if (getLangOpts().Char8)
+      CharTy = Context.Char8Ty;
     Kind = StringLiteral::UTF8;
   } else if (Literal.isUTF16()) {
     CharTy = Context.Char16Ty;
@@ -3094,6 +3096,8 @@ ExprResult Sema::ActOnCharacterConstant(const Token &Tok, Scope *UDLScope) {
   QualType Ty;
   if (Literal.isWide())
     Ty = Context.WideCharTy; // L'x' -> wchar_t in C and C++.
+  else if (Literal.isUTF8() && getLangOpts().Char8)
+    Ty = Context.Char8Ty; // u8'x' -> char8_t when it exists.
   else if (Literal.isUTF16())
     Ty = Context.Char16Ty; // u'x' -> char16_t in C11 and C++11.
   else if (Literal.isUTF32())
diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp
index e44eaa5162a..be33326cd42 100644
--- a/clang/lib/Sema/SemaInit.cpp
+++ b/clang/lib/Sema/SemaInit.cpp
@@ -49,6 +49,8 @@ enum StringInitFailureKind {
   SIF_NarrowStringIntoWideChar,
   SIF_WideStringIntoChar,
   SIF_IncompatWideStringIntoWideChar,
+  SIF_UTF8StringIntoPlainChar,
+  SIF_PlainStringIntoUTF8Char,
   SIF_Other
 };
 
@@ -77,12 +79,21 @@ static StringInitFailureKind IsStringInit(Expr *Init, const ArrayType *AT,
       Context.getCanonicalType(AT->getElementType()).getUnqualifiedType();
 
   switch (SL->getKind()) {
-  case StringLiteral::Ascii:
   case StringLiteral::UTF8:
+    // char8_t array can be initialized with a UTF-8 string.
+    if (ElemTy->isChar8Type())
+      return SIF_None;
+    LLVM_FALLTHROUGH;
+  case StringLiteral::Ascii:
     // char array can be initialized with a narrow string.
     // Only allow char x[] = "foo";  not char x[] = L"foo";
     if (ElemTy->isCharType())
-      return SIF_None;
+      return (SL->getKind() == StringLiteral::UTF8 &&
+              Context.getLangOpts().Char8)
+                 ? SIF_UTF8StringIntoPlainChar
+                 : SIF_None;
+    if (ElemTy->isChar8Type())
+      return SIF_PlainStringIntoUTF8Char;
     if (IsWideCharCompatible(ElemTy, Context))
       return SIF_NarrowStringIntoWideChar;
     return SIF_Other;
@@ -94,7 +105,7 @@ static StringInitFailureKind IsStringInit(Expr *Init, const ArrayType *AT,
   case StringLiteral::UTF16:
     if (Context.typesAreCompatible(Context.Char16Ty, ElemTy))
       return SIF_None;
-    if (ElemTy->isCharType())
+    if (ElemTy->isCharType() || ElemTy->isChar8Type())
       return SIF_WideStringIntoChar;
     if (IsWideCharCompatible(ElemTy, Context))
       return SIF_IncompatWideStringIntoWideChar;
@@ -102,7 +113,7 @@ static StringInitFailureKind IsStringInit(Expr *Init, const ArrayType *AT,
   case StringLiteral::UTF32:
     if (Context.typesAreCompatible(Context.Char32Ty, ElemTy))
       return SIF_None;
-    if (ElemTy->isCharType())
+    if (ElemTy->isCharType() || ElemTy->isChar8Type())
       return SIF_WideStringIntoChar;
     if (IsWideCharCompatible(ElemTy, Context))
       return SIF_IncompatWideStringIntoWideChar;
@@ -110,7 +121,7 @@ static StringInitFailureKind IsStringInit(Expr *Init, const ArrayType *AT,
   case StringLiteral::Wide:
     if (Context.typesAreCompatible(Context.getWideCharType(), ElemTy))
       return SIF_None;
-    if (ElemTy->isCharType())
+    if (ElemTy->isCharType() || ElemTy->isChar8Type())
       return SIF_WideStringIntoChar;
     if (IsWideCharCompatible(ElemTy, Context))
       return SIF_IncompatWideStringIntoWideChar;
@@ -3185,6 +3196,8 @@ bool InitializationSequence::isAmbiguous() const {
   case FK_NarrowStringIntoWideCharArray:
   case FK_WideStringIntoCharArray:
   case FK_IncompatWideStringIntoWideChar:
+  case FK_PlainStringIntoUTF8Char:
+  case FK_UTF8StringIntoPlainChar:
   case FK_AddressOfOverloadFailed: // FIXME: Could do better
   case FK_NonConstLValueReferenceBindingToTemporary:
   case FK_NonConstLValueReferenceBindingToBitfield:
@@ -5362,6 +5375,12 @@ void InitializationSequence::InitializeFrom(Sema &S,
       case SIF_IncompatWideStringIntoWideChar:
         SetFailed(FK_IncompatWideStringIntoWideChar);
         return;
+      case SIF_PlainStringIntoUTF8Char:
+        SetFailed(FK_PlainStringIntoUTF8Char);
+        return;
+      case SIF_UTF8StringIntoPlainChar:
+        SetFailed(FK_UTF8StringIntoPlainChar);
+        return;
       case SIF_Other:
         break;
       }
@@ -7591,6 +7610,17 @@ bool InitializationSequence::Diagnose(Sema &S,
     S.Diag(Kind.getLocation(),
            diag::err_array_init_incompat_wide_string_into_wchar);
     break;
+  case FK_PlainStringIntoUTF8Char:
+    S.Diag(Kind.getLocation(),
+           diag::err_array_init_plain_string_into_char8_t);
+    S.Diag(Args.front()->getLocStart(),
+           diag::note_array_init_plain_string_into_char8_t)
+      << FixItHint::CreateInsertion(Args.front()->getLocStart(), "u8");
+    break;
+  case FK_UTF8StringIntoPlainChar:
+    S.Diag(Kind.getLocation(),
+           diag::err_array_init_utf8_string_into_char);
+    break;
   case FK_ArrayTypeMismatch:
   case FK_NonConstantArrayInit:
     S.Diag(Kind.getLocation(),
@@ -8000,6 +8030,14 @@ void InitializationSequence::dump(raw_ostream &OS) const {
       OS << "incompatible wide string into wide char array";
       break;
 
+    case FK_PlainStringIntoUTF8Char:
+      OS << "plain string literal into char8_t array";
+      break;
+
+    case FK_UTF8StringIntoPlainChar:
+      OS << "u8 string literal into char array";
+      break;
+
     case FK_ArrayTypeMismatch:
       OS << "array type mismatch";
       break;
diff --git a/clang/lib/Sema/SemaOverload.cpp b/clang/lib/Sema/SemaOverload.cpp
index 64cd52c1f65..4b5898883b3 100644
--- a/clang/lib/Sema/SemaOverload.cpp
+++ b/clang/lib/Sema/SemaOverload.cpp
@@ -7707,6 +7707,8 @@ class BuiltinOperatorOverloadBuilder {
     ArithmeticTypes.push_back(S.Context.BoolTy);
     ArithmeticTypes.push_back(S.Context.CharTy);
     ArithmeticTypes.push_back(S.Context.WCharTy);
+    if (S.Context.getLangOpts().Char8)
+      ArithmeticTypes.push_back(S.Context.Char8Ty);
     ArithmeticTypes.push_back(S.Context.Char16Ty);
     ArithmeticTypes.push_back(S.Context.Char32Ty);
     ArithmeticTypes.push_back(S.Context.SignedCharTy);
diff --git a/clang/lib/Sema/SemaTemplate.cpp b/clang/lib/Sema/SemaTemplate.cpp
index d98176ad2e5..04fdfaea75f 100644
--- a/clang/lib/Sema/SemaTemplate.cpp
+++ b/clang/lib/Sema/SemaTemplate.cpp
@@ -6771,11 +6771,11 @@ Sema::BuildExpressionFromIntegralTemplateArgument(const TemplateArgument &Arg,
 
   Expr *E;
   if (T->isAnyCharacterType()) {
-    // This does not need to handle u8 character literals because those are
-    // of type char, and so can also be covered by an ASCII character literal.
     CharacterLiteral::CharacterKind Kind;
     if (T->isWideCharType())
       Kind = CharacterLiteral::Wide;
+    else if (T->isChar8Type() && getLangOpts().Char8)
+      Kind = CharacterLiteral::UTF8;
     else if (T->isChar16Type())
       Kind = CharacterLiteral::UTF16;
     else if (T->isChar32Type())
diff --git a/clang/lib/Sema/SemaTemplateVariadic.cpp b/clang/lib/Sema/SemaTemplateVariadic.cpp
index 7aa0e317df0..37a9c260034 100644
--- a/clang/lib/Sema/SemaTemplateVariadic.cpp
+++ b/clang/lib/Sema/SemaTemplateVariadic.cpp
@@ -822,6 +822,7 @@ bool Sema::containsUnexpandedParameterPacks(Declarator &D) {
   case TST_void:
   case TST_char:
   case TST_wchar:
+  case TST_char8:
   case TST_char16:
   case TST_char32:
   case TST_int:
diff --git a/clang/lib/Sema/SemaType.cpp b/clang/lib/Sema/SemaType.cpp
index 0e71047f00e..a6491e56acd 100644
--- a/clang/lib/Sema/SemaType.cpp
+++ b/clang/lib/Sema/SemaType.cpp
@@ -1277,6 +1277,11 @@ static QualType ConvertDeclSpecToType(TypeProcessingState &state) {
       Result = Context.getUnsignedWCharType();
     }
     break;
+  case DeclSpec::TST_char8:
+      assert(DS.getTypeSpecSign() == DeclSpec::TSS_unspecified &&
+        "Unknown TSS value");
+      Result = Context.Char8Ty;
+    break;
   case DeclSpec::TST_char16:
       assert(DS.getTypeSpecSign() == DeclSpec::TSS_unspecified &&
         "Unknown TSS value");
diff --git a/clang/lib/Serialization/ASTCommon.cpp b/clang/lib/Serialization/ASTCommon.cpp
index 535aacb8c49..54cea92042a 100644
--- a/clang/lib/Serialization/ASTCommon.cpp
+++ b/clang/lib/Serialization/ASTCommon.cpp
@@ -100,6 +100,9 @@ serialization::TypeIdxFromBuiltin(const BuiltinType *BT) {
   case BuiltinType::NullPtr:
     ID = PREDEF_TYPE_NULLPTR_ID;
     break;
+  case BuiltinType::Char8:
+    ID = PREDEF_TYPE_CHAR8_ID;
+    break;
   case BuiltinType::Char16:
     ID = PREDEF_TYPE_CHAR16_ID;
     break;
diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp
index a1de22b68ec..4b0220abc30 100644
--- a/clang/lib/Serialization/ASTReader.cpp
+++ b/clang/lib/Serialization/ASTReader.cpp
@@ -6841,6 +6841,9 @@ QualType ASTReader::GetType(TypeID ID) {
     case PREDEF_TYPE_NULLPTR_ID:
       T = Context.NullPtrTy;
       break;
+    case PREDEF_TYPE_CHAR8_ID:
+      T = Context.Char8Ty;
+      break;
     case PREDEF_TYPE_CHAR16_ID:
       T = Context.Char16Ty;
       break;