diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -7368,7 +7368,9 @@ void Sema::ProcessDeclAttributeList(
   // good to have a way to specify "these attributes must appear as a group",
   // for these. Additionally, it would be good to have a way to specify "these
   // attribute must never appear as a group" for attributes like cold and hot.
-  if (!D->hasAttr<OpenCLKernelAttr>()) {
+  if (!(D->hasAttr<OpenCLKernelAttr>() ||
+        (D->hasAttr<CUDAGlobalAttr>() &&
+         Context.getTargetInfo().getTriple().isSPIRV()))) {
     // These attributes cannot be applied to a non-kernel function.
     if (const auto *A = D->getAttr<ReqdWorkGroupSizeAttr>()) {
       // FIXME: This emits a different error message than

diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
@@ -13821,7 +13821,7 @@ QualType Sema::CheckAssignmentOperands(Expr *LHSExpr, ExprResult &RHS,
   CheckForNullPointerDereference(*this, LHSExpr);
 
   AssignedEntity AE{LHSExpr};
-  checkExprLifetime(*this, AE, RHS.get());
+  checkAssignmentLifetime(*this, AE, RHS.get());
 
   if (getLangOpts().CPlusPlus20 && LHSType.isVolatileQualified()) {
     if (CompoundType.isNull()) {

diff --git a/clang/lib/Sema/SemaFunctionEffects.cpp b/clang/lib/Sema/SemaFunctionEffects.cpp
@@ -807,7 +807,8 @@ class Analyzer {
 
     auto MaybeAddTemplateNote = [&](const Decl *D) {
       if (const FunctionDecl *FD = dyn_cast<FunctionDecl>(D)) {
-        while (FD != nullptr && FD->isTemplateInstantiation()) {
+        while (FD != nullptr && FD->isTemplateInstantiation() &&
+               FD->getPointOfInstantiation().isValid()) {
           S.Diag(FD->getPointOfInstantiation(),
                  diag::note_func_effect_from_template);
           FD = FD->getTemplateInstantiationPattern();

diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp
@@ -7401,7 +7401,7 @@ PerformConstructorInitialization(Sema &S,
 
 void Sema::checkInitializerLifetime(const InitializedEntity &Entity,
                                     Expr *Init) {
-  return sema::checkExprLifetime(*this, Entity, Init);
+  return sema::checkInitLifetime(*this, Entity, Init);
 }
 
 static void DiagnoseNarrowingInInitList(Sema &S,

diff --git a/clang/lib/Sema/SemaLambda.cpp b/clang/lib/Sema/SemaLambda.cpp
@@ -1950,8 +1950,6 @@ ExprResult Sema::ActOnLambdaExpr(SourceLocation StartLoc, Stmt *Body) {
   LambdaScopeInfo LSI = *cast<LambdaScopeInfo>(FunctionScopes.back());
   ActOnFinishFunctionBody(LSI.CallOperator, Body);
 
-  maybeAddDeclWithEffects(LSI.CallOperator);
-
   return BuildLambdaExpr(StartLoc, Body->getEndLoc(), &LSI);
 }
 
@@ -2284,6 +2282,7 @@ ExprResult Sema::BuildLambdaExpr(SourceLocation StartLoc, SourceLocation EndLoc,
     case ExpressionEvaluationContext::PotentiallyEvaluatedIfUsed:
       break;
     }
+    maybeAddDeclWithEffects(LSI->CallOperator);
   }
 
   return MaybeBindToTemporary(Lambda);

diff --git a/clang/lib/Sema/SemaOverload.cpp b/clang/lib/Sema/SemaOverload.cpp
@@ -14809,7 +14809,7 @@ ExprResult Sema::CreateOverloadedBinOp(SourceLocation OpLoc,
           // Check for a self move.
           DiagnoseSelfMove(Args[0], Args[1], OpLoc);
           // lifetime check.
-          checkExprLifetime(
+          checkAssignmentLifetime(
               *this, AssignedEntity{Args[0], dyn_cast<CXXMethodDecl>(FnDecl)},
               Args[1]);
         }

diff --git a/clang/lib/Sema/SemaStmt.cpp b/clang/lib/Sema/SemaStmt.cpp
@@ -200,23 +200,30 @@ static bool DiagnoseUnusedComparison(Sema &S, const Expr *E) {
   return true;
 }
 
-static bool DiagnoseNoDiscard(Sema &S, const WarnUnusedResultAttr *A,
-                              SourceLocation Loc, SourceRange R1,
-                              SourceRange R2, bool IsCtor) {
+static bool DiagnoseNoDiscard(Sema &S, const NamedDecl *OffendingDecl,
+                              const WarnUnusedResultAttr *A, SourceLocation Loc,
+                              SourceRange R1, SourceRange R2, bool IsCtor) {
   if (!A)
     return false;
   StringRef Msg = A->getMessage();
 
   if (Msg.empty()) {
+    if (OffendingDecl)
+      return S.Diag(Loc, diag::warn_unused_return_type)
+             << IsCtor << A << OffendingDecl << false << R1 << R2;
     if (IsCtor)
-      return S.Diag(Loc, diag::warn_unused_constructor) << A << R1 << R2;
-    return S.Diag(Loc, diag::warn_unused_result) << A << R1 << R2;
+      return S.Diag(Loc, diag::warn_unused_constructor)
+             << A << false << R1 << R2;
+    return S.Diag(Loc, diag::warn_unused_result) << A << false << R1 << R2;
   }
 
+  if (OffendingDecl)
+    return S.Diag(Loc, diag::warn_unused_return_type)
+           << IsCtor << A << OffendingDecl << true << Msg << R1 << R2;
   if (IsCtor)
-    return S.Diag(Loc, diag::warn_unused_constructor_msg) << A << Msg << R1
-                                                          << R2;
-  return S.Diag(Loc, diag::warn_unused_result_msg) << A << Msg << R1 << R2;
+    return S.Diag(Loc, diag::warn_unused_constructor)
+           << A << true << Msg << R1 << R2;
+  return S.Diag(Loc, diag::warn_unused_result) << A << true << Msg << R1 << R2;
 }
 
 void Sema::DiagnoseUnusedExprResult(const Stmt *S, unsigned DiagID) {
@@ -286,9 +293,10 @@ void Sema::DiagnoseUnusedExprResult(const Stmt *S, unsigned DiagID) {
     if (E->getType()->isVoidType())
       return;
 
-    if (DiagnoseNoDiscard(*this, cast_or_null<WarnUnusedResultAttr>(
-                                     CE->getUnusedResultAttr(Context)),
-                          Loc, R1, R2, /*isCtor=*/false))
+    auto [OffendingDecl, A] = CE->getUnusedResultAttr(Context);
+    if (DiagnoseNoDiscard(*this, OffendingDecl,
+                          cast_or_null<WarnUnusedResultAttr>(A), Loc, R1, R2,
+                          /*isCtor=*/false))
       return;
 
     // If the callee has attribute pure, const, or warn_unused_result, warn with
@@ -309,16 +317,21 @@ void Sema::DiagnoseUnusedExprResult(const Stmt *S, unsigned DiagID) {
     }
   } else if (const auto *CE = dyn_cast<CXXConstructExpr>(E)) {
     if (const CXXConstructorDecl *Ctor = CE->getConstructor()) {
+      const NamedDecl *OffendingDecl = nullptr;
       const auto *A = Ctor->getAttr<WarnUnusedResultAttr>();
-      A = A ? A : Ctor->getParent()->getAttr<WarnUnusedResultAttr>();
-      if (DiagnoseNoDiscard(*this, A, Loc, R1, R2, /*isCtor=*/true))
+      if (!A) {
+        OffendingDecl = Ctor->getParent();
+        A = OffendingDecl->getAttr<WarnUnusedResultAttr>();
+      }
+      if (DiagnoseNoDiscard(*this, OffendingDecl, A, Loc, R1, R2,
+                            /*isCtor=*/true))
         return;
     }
   } else if (const auto *ILE = dyn_cast<InitListExpr>(E)) {
     if (const TagDecl *TD = ILE->getType()->getAsTagDecl()) {
 
-      if (DiagnoseNoDiscard(*this, TD->getAttr<WarnUnusedResultAttr>(), Loc, R1,
-                            R2, /*isCtor=*/false))
+      if (DiagnoseNoDiscard(*this, TD, TD->getAttr<WarnUnusedResultAttr>(), Loc,
+                            R1, R2, /*isCtor=*/false))
         return;
     }
   } else if (ShouldSuppress)
@@ -332,8 +345,8 @@ void Sema::DiagnoseUnusedExprResult(const Stmt *S, unsigned DiagID) {
     }
     const ObjCMethodDecl *MD = ME->getMethodDecl();
     if (MD) {
-      if (DiagnoseNoDiscard(*this, MD->getAttr<WarnUnusedResultAttr>(), Loc, R1,
-                            R2, /*isCtor=*/false))
+      if (DiagnoseNoDiscard(*this, nullptr, MD->getAttr<WarnUnusedResultAttr>(),
+                            Loc, R1, R2, /*isCtor=*/false))
         return;
     }
   } else if (const PseudoObjectExpr *POE = dyn_cast<PseudoObjectExpr>(E)) {

diff --git a/clang/lib/Sema/SemaType.cpp b/clang/lib/Sema/SemaType.cpp
@@ -4882,9 +4882,17 @@ static TypeSourceInfo *GetFullTypeForDeclarator(TypeProcessingState &state,
                       cast<AutoType>(T)->getKeyword() !=
                           AutoTypeKeyword::Auto ||
                       cast<AutoType>(T)->isConstrained())) {
-            S.Diag(D.getDeclSpec().getTypeSpecTypeLoc(),
-                   diag::err_trailing_return_without_auto)
-                << T << D.getDeclSpec().getSourceRange();
+            // Attach a valid source location for diagnostics on functions with
+            // trailing return types missing 'auto'. Attempt to get the location
+            // from the declared type; if invalid, fall back to the trailing
+            // return type's location.
+            SourceLocation Loc = D.getDeclSpec().getTypeSpecTypeLoc();
+            SourceRange SR = D.getDeclSpec().getSourceRange();
+            if (Loc.isInvalid()) {
+              Loc = FTI.getTrailingReturnTypeLoc();
+              SR = D.getSourceRange();
+            }
+            S.Diag(Loc, diag::err_trailing_return_without_auto) << T << SR;
             D.setInvalidType(true);
             // FIXME: recover and fill decls in `TypeLoc`s.
             AreDeclaratorChunksValid = false;

diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp
@@ -3092,98 +3092,97 @@ ASTReader::ReadControlBlock(ModuleFile &F,
       break;
     }
 
-    case IMPORTS: {
+    case IMPORT: {
       // Validate the AST before processing any imports (otherwise, untangling
       // them can be error-prone and expensive).  A module will have a name and
       // will already have been validated, but this catches the PCH case.
       if (ASTReadResult Result = readUnhashedControlBlockOnce())
         return Result;
 
-      // Load each of the imported PCH files.
-      unsigned Idx = 0, N = Record.size();
-      while (Idx < N) {
-        // Read information about the AST file.
-        ModuleKind ImportedKind = (ModuleKind)Record[Idx++];
-        // Whether we're importing a standard c++ module.
-        bool IsImportingStdCXXModule = Record[Idx++];
-        // The import location will be the local one for now; we will adjust
-        // all import locations of module imports after the global source
-        // location info are setup, in ReadAST.
-        auto [ImportLoc, ImportModuleFileIndex] =
-            ReadUntranslatedSourceLocation(Record[Idx++]);
-        // The import location must belong to the current module file itself.
-        assert(ImportModuleFileIndex == 0);
-        off_t StoredSize = !IsImportingStdCXXModule ? (off_t)Record[Idx++] : 0;
-        time_t StoredModTime =
-            !IsImportingStdCXXModule ? (time_t)Record[Idx++] : 0;
-
-        ASTFileSignature StoredSignature;
-        if (!IsImportingStdCXXModule) {
-          auto FirstSignatureByte = Record.begin() + Idx;
-          StoredSignature = ASTFileSignature::create(
-              FirstSignatureByte, FirstSignatureByte + ASTFileSignature::size);
-          Idx += ASTFileSignature::size;
-        }
+      unsigned Idx = 0;
+      // Read information about the AST file.
+      ModuleKind ImportedKind = (ModuleKind)Record[Idx++];
+
+      // The import location will be the local one for now; we will adjust
+      // all import locations of module imports after the global source
+      // location info are setup, in ReadAST.
+      auto [ImportLoc, ImportModuleFileIndex] =
+          ReadUntranslatedSourceLocation(Record[Idx++]);
+      // The import location must belong to the current module file itself.
+      assert(ImportModuleFileIndex == 0);
+
+      StringRef ImportedName = ReadStringBlob(Record, Idx, Blob);
+
+      bool IsImportingStdCXXModule = Record[Idx++];
+
+      off_t StoredSize = 0;
+      time_t StoredModTime = 0;
+      ASTFileSignature StoredSignature;
+      std::string ImportedFile;
+
+      // For prebuilt and explicit modules first consult the file map for
+      // an override. Note that here we don't search prebuilt module
+      // directories if we're not importing standard c++ module, only the
+      // explicit name to file mappings. Also, we will still verify the
+      // size/signature making sure it is essentially the same file but
+      // perhaps in a different location.
+      if (ImportedKind == MK_PrebuiltModule || ImportedKind == MK_ExplicitModule)
+        ImportedFile = PP.getHeaderSearchInfo().getPrebuiltModuleFileName(
+            ImportedName, /*FileMapOnly*/ !IsImportingStdCXXModule);
+
+      if (IsImportingStdCXXModule && ImportedFile.empty()) {
+        Diag(diag::err_failed_to_find_module_file) << ImportedName;
+        return Missing;
+      }
 
-        std::string ImportedName = ReadString(Record, Idx);
-        std::string ImportedFile;
-
-        // For prebuilt and explicit modules first consult the file map for
-        // an override. Note that here we don't search prebuilt module
-        // directories if we're not importing standard c++ module, only the
-        // explicit name to file mappings. Also, we will still verify the
-        // size/signature making sure it is essentially the same file but
-        // perhaps in a different location.
-        if (ImportedKind == MK_PrebuiltModule || ImportedKind == MK_ExplicitModule)
-          ImportedFile = PP.getHeaderSearchInfo().getPrebuiltModuleFileName(
-              ImportedName, /*FileMapOnly*/ !IsImportingStdCXXModule);
-
-        // For C++20 Modules, we won't record the path to the imported modules
-        // in the BMI
-        if (!IsImportingStdCXXModule) {
-          if (ImportedFile.empty()) {
-            // Use BaseDirectoryAsWritten to ensure we use the same path in the
-            // ModuleCache as when writing.
-            ImportedFile = ReadPath(BaseDirectoryAsWritten, Record, Idx);
-          } else
-            SkipPath(Record, Idx);
-        } else if (ImportedFile.empty()) {
-          Diag(clang::diag::err_failed_to_find_module_file) << ImportedName;
-          return Missing;
-        }
+      if (!IsImportingStdCXXModule) {
+        StoredSize = (off_t)Record[Idx++];
+        StoredModTime = (time_t)Record[Idx++];
 
-        // If our client can't cope with us being out of date, we can't cope with
-        // our dependency being missing.
-        unsigned Capabilities = ClientLoadCapabilities;
-        if ((ClientLoadCapabilities & ARR_OutOfDate) == 0)
-          Capabilities &= ~ARR_Missing;
-
-        // Load the AST file.
-        auto Result = ReadASTCore(ImportedFile, ImportedKind, ImportLoc, &F,
-                                  Loaded, StoredSize, StoredModTime,
-                                  StoredSignature, Capabilities);
-
-        // If we diagnosed a problem, produce a backtrace.
-        bool recompilingFinalized =
-            Result == OutOfDate && (Capabilities & ARR_OutOfDate) &&
-            getModuleManager().getModuleCache().isPCMFinal(F.FileName);
-        if (isDiagnosedResult(Result, Capabilities) || recompilingFinalized)
-          Diag(diag::note_module_file_imported_by)
-              << F.FileName << !F.ModuleName.empty() << F.ModuleName;
-        if (recompilingFinalized)
-          Diag(diag::note_module_file_conflict);
-
-        switch (Result) {
-        case Failure: return Failure;
-          // If we have to ignore the dependency, we'll have to ignore this too.
-        case Missing:
-        case OutOfDate: return OutOfDate;
-        case VersionMismatch: return VersionMismatch;
-        case ConfigurationMismatch: return ConfigurationMismatch;
-        case HadErrors: return HadErrors;
-        case Success: break;
+        StringRef SignatureBytes = Blob.substr(0, ASTFileSignature::size);
+        StoredSignature = ASTFileSignature::create(SignatureBytes.begin(),
+                                                   SignatureBytes.end());
+        Blob = Blob.substr(ASTFileSignature::size);
+
+        if (ImportedFile.empty()) {
+          // Use BaseDirectoryAsWritten to ensure we use the same path in the
+          // ModuleCache as when writing.
+          ImportedFile =
+              ReadPathBlob(BaseDirectoryAsWritten, Record, Idx, Blob);
         }
       }
+
+      // If our client can't cope with us being out of date, we can't cope with
+      // our dependency being missing.
+      unsigned Capabilities = ClientLoadCapabilities;
+      if ((ClientLoadCapabilities & ARR_OutOfDate) == 0)
+        Capabilities &= ~ARR_Missing;
+
+      // Load the AST file.
+      auto Result = ReadASTCore(ImportedFile, ImportedKind, ImportLoc, &F,
+                                Loaded, StoredSize, StoredModTime,
+                                StoredSignature, Capabilities);
+
+      // If we diagnosed a problem, produce a backtrace.
+      bool recompilingFinalized =
+          Result == OutOfDate && (Capabilities & ARR_OutOfDate) &&
+          getModuleManager().getModuleCache().isPCMFinal(F.FileName);
+      if (isDiagnosedResult(Result, Capabilities) || recompilingFinalized)
+        Diag(diag::note_module_file_imported_by)
+            << F.FileName << !F.ModuleName.empty() << F.ModuleName;
+      if (recompilingFinalized)
+        Diag(diag::note_module_file_conflict);
+
+      switch (Result) {
+      case Failure: return Failure;
+        // If we have to ignore the dependency, we'll have to ignore this too.
+      case Missing:
+      case OutOfDate: return OutOfDate;
+      case VersionMismatch: return VersionMismatch;
+      case ConfigurationMismatch: return ConfigurationMismatch;
+      case HadErrors: return HadErrors;
+      case Success: break;
+      }
       break;
     }
 
@@ -5624,36 +5623,38 @@ bool ASTReader::readASTFileControlBlock(
       break;
     }
 
-    case IMPORTS: {
+    case IMPORT: {
       if (!NeedsImports)
         break;
 
-      unsigned Idx = 0, N = Record.size();
-      while (Idx < N) {
-        // Read information about the AST file.
+      unsigned Idx = 0;
+      // Read information about the AST file.
+
+      // Skip Kind
+      Idx++;
 
-        // Skip Kind
-        Idx++;
-        bool IsStandardCXXModule = Record[Idx++];
+      // Skip ImportLoc
+      Idx++;
 
-        // Skip ImportLoc
-        Idx++;
+      StringRef ModuleName = ReadStringBlob(Record, Idx, Blob);
 
-        // In C++20 Modules, we don't record the path to imported
-        // modules in the BMI files.
-        if (IsStandardCXXModule) {
-          std::string ModuleName = ReadString(Record, Idx);
-          Listener.visitImport(ModuleName, /*Filename=*/"");
-          continue;
-        }
+      bool IsStandardCXXModule = Record[Idx++];
 
-        // Skip Size, ModTime and Signature
-        Idx += 1 + 1 + ASTFileSignature::size;
-        std::string ModuleName = ReadString(Record, Idx);
-        std::string FilenameStr = ReadString(Record, Idx);
-        auto Filename = ResolveImportedPath(PathBuf, FilenameStr, ModuleDir);
-        Listener.visitImport(ModuleName, *Filename);
+      // In C++20 Modules, we don't record the path to imported
+      // modules in the BMI files.
+      if (IsStandardCXXModule) {
+        Listener.visitImport(ModuleName, /*Filename=*/"");
+        continue;
       }
+
+      // Skip Size and ModTime.
+      Idx += 1 + 1;
+      // Skip signature.
+      Blob = Blob.substr(ASTFileSignature::size);
+
+      StringRef FilenameStr = ReadStringBlob(Record, Idx, Blob);
+      auto Filename = ResolveImportedPath(PathBuf, FilenameStr, ModuleDir);
+      Listener.visitImport(ModuleName, *Filename);
       break;
     }
 
@@ -9602,6 +9603,14 @@ std::string ASTReader::ReadString(const RecordDataImpl &Record, unsigned &Idx) {
   return Result;
 }
 
+StringRef ASTReader::ReadStringBlob(const RecordDataImpl &Record, unsigned &Idx,
+                                    StringRef &Blob) {
+  unsigned Len = Record[Idx++];
+  StringRef Result = Blob.substr(0, Len);
+  Blob = Blob.substr(Len);
+  return Result;
+}
+
 std::string ASTReader::ReadPath(ModuleFile &F, const RecordData &Record,
                                 unsigned &Idx) {
   return ReadPath(F.BaseDirectory, Record, Idx);
@@ -9613,6 +9622,13 @@ std::string ASTReader::ReadPath(StringRef BaseDirectory,
   return ResolveImportedPathAndAllocate(PathBuf, Filename, BaseDirectory);
 }
 
+std::string ASTReader::ReadPathBlob(StringRef BaseDirectory,
+                                    const RecordData &Record, unsigned &Idx,
+                                    StringRef &Blob) {
+  StringRef Filename = ReadStringBlob(Record, Idx, Blob);
+  return ResolveImportedPathAndAllocate(PathBuf, Filename, BaseDirectory);
+}
+
 VersionTuple ASTReader::ReadVersionTuple(const RecordData &Record,
                                          unsigned &Idx) {
   unsigned Major = Record[Idx++];

diff --git a/clang/lib/Serialization/ASTWriter.cpp b/clang/lib/Serialization/ASTWriter.cpp
@@ -878,7 +878,7 @@ void ASTWriter::WriteBlockInfoBlock() {
   RECORD(MODULE_NAME);
   RECORD(MODULE_DIRECTORY);
   RECORD(MODULE_MAP_FILE);
-  RECORD(IMPORTS);
+  RECORD(IMPORT);
   RECORD(ORIGINAL_FILE);
   RECORD(ORIGINAL_FILE_ID);
   RECORD(INPUT_FILE_OFFSETS);
@@ -1536,34 +1536,53 @@ void ASTWriter::WriteControlBlock(Preprocessor &PP, StringRef isysroot) {
 
   // Imports
   if (Chain) {
-    serialization::ModuleManager &Mgr = Chain->getModuleManager();
-    Record.clear();
+    auto Abbrev = std::make_shared<BitCodeAbbrev>();
+    Abbrev->Add(BitCodeAbbrevOp(IMPORT));
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 3)); // Kind
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // ImportLoc
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // Module name len
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 1)); // Standard C++ mod
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // File size
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // File timestamp
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // File name len
+    Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Blob)); // Strings
+    unsigned AbbrevCode = Stream.EmitAbbrev(std::move(Abbrev));
 
-    for (ModuleFile &M : Mgr) {
+    SmallString<128> Blob;
+
+    for (ModuleFile &M : Chain->getModuleManager()) {
       // Skip modules that weren't directly imported.
       if (!M.isDirectlyImported())
         continue;
 
+      Record.clear();
+      Blob.clear();
+
+      Record.push_back(IMPORT);
       Record.push_back((unsigned)M.Kind); // FIXME: Stable encoding
-      Record.push_back(M.StandardCXXModule);
       AddSourceLocation(M.ImportLoc, Record);
+      AddStringBlob(M.ModuleName, Record, Blob);
+      Record.push_back(M.StandardCXXModule);
 
       // We don't want to hard code the information about imported modules
       // in the C++20 named modules.
-      if (!M.StandardCXXModule) {
+      if (M.StandardCXXModule) {
+        Record.push_back(0);
+        Record.push_back(0);
+        Record.push_back(0);
+      } else {
         // If we have calculated signature, there is no need to store
         // the size or timestamp.
         Record.push_back(M.Signature ? 0 : M.File.getSize());
         Record.push_back(M.Signature ? 0 : getTimestampForOutput(M.File));
-        llvm::append_range(Record, M.Signature);
-      }
 
-      AddString(M.ModuleName, Record);
+        llvm::append_range(Blob, M.Signature);
 
-      if (!M.StandardCXXModule)
-        AddPath(M.FileName, Record);
+        AddPathBlob(M.FileName, Record, Blob);
+      }
+
+      Stream.EmitRecordWithBlob(AbbrevCode, Record, Blob);
     }
-    Stream.EmitRecord(IMPORTS, Record);
   }
 
   // Write the options block.
@@ -4777,6 +4796,12 @@ void ASTWriter::AddString(StringRef Str, RecordDataImpl &Record) {
   Record.insert(Record.end(), Str.begin(), Str.end());
 }
 
+void ASTWriter::AddStringBlob(StringRef Str, RecordDataImpl &Record,
+                              SmallVectorImpl<char> &Blob) {
+  Record.push_back(Str.size());
+  Blob.insert(Blob.end(), Str.begin(), Str.end());
+}
+
 bool ASTWriter::PreparePathForOutput(SmallVectorImpl<char> &Path) {
   assert(WritingAST && "can't prepare path for output when not writing AST");
 
@@ -4805,6 +4830,13 @@ void ASTWriter::AddPath(StringRef Path, RecordDataImpl &Record) {
   AddString(FilePath, Record);
 }
 
+void ASTWriter::AddPathBlob(StringRef Path, RecordDataImpl &Record,
+                            SmallVectorImpl<char> &Blob) {
+  SmallString<128> FilePath(Path);
+  PreparePathForOutput(FilePath);
+  AddStringBlob(FilePath, Record, Blob);
+}
+
 void ASTWriter::EmitRecordWithPath(unsigned Abbrev, RecordDataRef Record,
                                    StringRef Path) {
   SmallString<128> FilePath(Path);

diff --git a/clang/lib/Serialization/GlobalModuleIndex.cpp b/clang/lib/Serialization/GlobalModuleIndex.cpp
@@ -614,62 +614,58 @@ llvm::Error GlobalModuleIndexBuilder::loadModuleFile(FileEntryRef File) {
     unsigned Code = MaybeCode.get();
 
     // Handle module dependencies.
-    if (State == ControlBlock && Code == IMPORTS) {
-      // Load each of the imported PCH files.
-      unsigned Idx = 0, N = Record.size();
-      while (Idx < N) {
-        // Read information about the AST file.
-
-        // Skip the imported kind
-        ++Idx;
-
-        // Skip if it is standard C++ module
-        ++Idx;
-
-        // Skip the import location
-        ++Idx;
-
-        // Load stored size/modification time.
-        off_t StoredSize = (off_t)Record[Idx++];
-        time_t StoredModTime = (time_t)Record[Idx++];
-
-        // Skip the stored signature.
-        // FIXME: we could read the signature out of the import and validate it.
-        auto FirstSignatureByte = Record.begin() + Idx;
-        ASTFileSignature StoredSignature = ASTFileSignature::create(
-            FirstSignatureByte, FirstSignatureByte + ASTFileSignature::size);
-        Idx += ASTFileSignature::size;
-
-        // Skip the module name (currently this is only used for prebuilt
-        // modules while here we are only dealing with cached).
-        Idx += Record[Idx] + 1;
-
-        // Retrieve the imported file name.
-        unsigned Length = Record[Idx++];
-        SmallString<128> ImportedFile(Record.begin() + Idx,
-                                      Record.begin() + Idx + Length);
-        Idx += Length;
-
-        // Find the imported module file.
-        auto DependsOnFile =
-            FileMgr.getOptionalFileRef(ImportedFile, /*OpenFile=*/false,
-                                       /*CacheFailure=*/false);
-
-        if (!DependsOnFile)
-          return llvm::createStringError(std::errc::bad_file_descriptor,
-                                         "imported file \"%s\" not found",
-                                         ImportedFile.c_str());
-
-        // Save the information in ImportedModuleFileInfo so we can verify after
-        // loading all pcms.
-        ImportedModuleFiles.insert(std::make_pair(
-            *DependsOnFile, ImportedModuleFileInfo(StoredSize, StoredModTime,
-                                                   StoredSignature)));
-
-        // Record the dependency.
-        unsigned DependsOnID = getModuleFileInfo(*DependsOnFile).ID;
-        getModuleFileInfo(File).Dependencies.push_back(DependsOnID);
-      }
+    if (State == ControlBlock && Code == IMPORT) {
+      unsigned Idx = 0;
+      // Read information about the AST file.
+
+      // Skip the imported kind
+      ++Idx;
+
+      // Skip the import location
+      ++Idx;
+
+      // Skip the module name (currently this is only used for prebuilt
+      // modules while here we are only dealing with cached).
+      Blob = Blob.substr(Record[Idx++]);
+
+      // Skip if it is standard C++ module
+      ++Idx;
+
+      // Load stored size/modification time.
+      off_t StoredSize = (off_t)Record[Idx++];
+      time_t StoredModTime = (time_t)Record[Idx++];
+
+      // Skip the stored signature.
+      // FIXME: we could read the signature out of the import and validate it.
+      StringRef SignatureBytes = Blob.substr(0, ASTFileSignature::size);
+      auto StoredSignature = ASTFileSignature::create(SignatureBytes.begin(),
+                                                      SignatureBytes.end());
+      Blob = Blob.substr(ASTFileSignature::size);
+
+      // Retrieve the imported file name.
+      unsigned Length = Record[Idx++];
+      StringRef ImportedFile = Blob.substr(0, Length);
+      Blob = Blob.substr(Length);
+
+      // Find the imported module file.
+      auto DependsOnFile =
+          FileMgr.getOptionalFileRef(ImportedFile, /*OpenFile=*/false,
+                                     /*CacheFailure=*/false);
+
+      if (!DependsOnFile)
+        return llvm::createStringError(std::errc::bad_file_descriptor,
+                                       "imported file \"%s\" not found",
+                                       std::string(ImportedFile).c_str());
+
+      // Save the information in ImportedModuleFileInfo so we can verify after
+      // loading all pcms.
+      ImportedModuleFiles.insert(std::make_pair(
+          *DependsOnFile, ImportedModuleFileInfo(StoredSize, StoredModTime,
+                                                 StoredSignature)));
+
+      // Record the dependency.
+      unsigned DependsOnID = getModuleFileInfo(*DependsOnFile).ID;
+      getModuleFileInfo(File).Dependencies.push_back(DependsOnID);
 
       continue;
     }

diff --git a/clang/test/AST/HLSL/AppendStructuredBuffer-AST.hlsl b/clang/test/AST/HLSL/AppendStructuredBuffer-AST.hlsl
@@ -30,20 +30,20 @@ AppendStructuredBuffer<int> Buffer;
 // CHECK-NEXT: CXXRecordDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit class AppendStructuredBuffer definition
 
 // CHECK: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(element_type)]]
 // CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit RawBuffer
 
-// CHECK-NOT: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &const (unsigned int) const'
+// CHECK-NOT: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'const element_type &(unsigned int) const'
 // CHECK-NOT: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &(unsigned int)'
 
 // CHECK: ClassTemplateSpecializationDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> class AppendStructuredBuffer definition
 // CHECK: TemplateArgument type 'int'
 // CHECK-NEXT: BuiltinType 0x{{[0-9A-Fa-f]+}} 'int'
 // CHECK-NEXT: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(int)]]

diff --git a/clang/test/AST/HLSL/ConsumeStructuredBuffer-AST.hlsl b/clang/test/AST/HLSL/ConsumeStructuredBuffer-AST.hlsl
@@ -30,21 +30,21 @@ ConsumeStructuredBuffer<int> Buffer;
 // CHECK-NEXT: CXXRecordDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit class ConsumeStructuredBuffer definition
 
 // CHECK: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(element_type)]]
 // CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit RawBuffer
 
-// CHECK-NOT: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &const (unsigned int) const'
+// CHECK-NOT: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'const element_type &(unsigned int) const'
 // CHECK-NOT: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &(unsigned int)'
 
 // CHECK: ClassTemplateSpecializationDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> class ConsumeStructuredBuffer definition
 
 // CHECK: TemplateArgument type 'int'
 // CHECK-NEXT: BuiltinType 0x{{[0-9A-Fa-f]+}} 'int'
 // CHECK-NEXT: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(int)]]

diff --git a/clang/test/AST/HLSL/RWBuffer-AST.hlsl b/clang/test/AST/HLSL/RWBuffer-AST.hlsl
@@ -29,12 +29,12 @@ RWBuffer<float> Buffer;
 // CHECK-NEXT: CXXRecordDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit class RWBuffer definition
 
 // CHECK: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(element_type)]]
 // CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit TypedBuffer
 
-// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &const (unsigned int) const'
+// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'const element_type &(unsigned int) const'
 // CHECK-NEXT: ParmVarDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> Idx 'unsigned int'
 // CHECK-NEXT: CompoundStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
 // CHECK-NEXT: ReturnStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
@@ -55,7 +55,7 @@ RWBuffer<float> Buffer;
 // CHECK: TemplateArgument type 'float'
 // CHECK-NEXT: BuiltinType 0x{{[0-9A-Fa-f]+}} 'float'
 // CHECK-NEXT: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
-// CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]] 
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
+// CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(float)]]
 // CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit TypedBuffer
diff --git a/clang/test/AST/HLSL/RWStructuredBuffer-AST.hlsl b/clang/test/AST/HLSL/RWStructuredBuffer-AST.hlsl
@@ -30,13 +30,13 @@ RWStructuredBuffer<int> Buffer;
 // CHECK-NEXT: CXXRecordDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit class RWStructuredBuffer definition
 
 // CHECK: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(element_type)]]
 // CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit RawBuffer
 
-// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &const (unsigned int) const'
+// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'const element_type &(unsigned int) const'
 // CHECK-NEXT: ParmVarDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> Idx 'unsigned int'
 // CHECK-NEXT: CompoundStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
 // CHECK-NEXT: ReturnStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
@@ -57,7 +57,7 @@ RWStructuredBuffer<int> Buffer;
 // CHECK: TemplateArgument type 'int'
 // CHECK-NEXT: BuiltinType 0x{{[0-9A-Fa-f]+}} 'int'
 // CHECK-NEXT: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(int)]]

diff --git a/clang/test/AST/HLSL/RasterizerOrderedStructuredBuffer-AST.hlsl b/clang/test/AST/HLSL/RasterizerOrderedStructuredBuffer-AST.hlsl
@@ -30,14 +30,14 @@ RasterizerOrderedStructuredBuffer<int> Buffer;
 // CHECK-NEXT: CXXRecordDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit class RasterizerOrderedStructuredBuffer definition
 
 // CHECK: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::is_rov]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(element_type)]]
-// CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit TypedBuffer
+// CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit RawBuffer
 
-// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &const (unsigned int) const'
+// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'const element_type &(unsigned int) const'
 // CHECK-NEXT: ParmVarDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> Idx 'unsigned int'
 // CHECK-NEXT: CompoundStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
 // CHECK-NEXT: ReturnStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
@@ -58,9 +58,9 @@ RasterizerOrderedStructuredBuffer<int> Buffer;
 // CHECK: TemplateArgument type 'int'
 // CHECK-NEXT: BuiltinType 0x{{[0-9A-Fa-f]+}} 'int'
 // CHECK-NEXT: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(UAV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::is_rov]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(int)]]
-// CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit TypedBuffer
+// CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit RawBuffer
diff --git a/clang/test/AST/HLSL/StructuredBuffer-AST.hlsl b/clang/test/AST/HLSL/StructuredBuffer-AST.hlsl
@@ -30,13 +30,13 @@ StructuredBuffer<float> Buffer;
 // CHECK-NEXT: CXXRecordDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit class StructuredBuffer definition
 
 // CHECK: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(SRV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(element_type)]]
 // CHECK-NEXT: HLSLResourceAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit RawBuffer
 
-// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'element_type &const (unsigned int) const'
+// CHECK: CXXMethodDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> operator[] 'const element_type &(unsigned int) const'
 // CHECK-NEXT: ParmVarDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> Idx 'unsigned int'
 // CHECK-NEXT: CompoundStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
 // CHECK-NEXT: ReturnStmt 0x{{[0-9A-Fa-f]+}} <<invalid sloc>>
@@ -57,7 +57,7 @@ StructuredBuffer<float> Buffer;
 // CHECK: TemplateArgument type 'float'
 // CHECK-NEXT: BuiltinType 0x{{[0-9A-Fa-f]+}} 'float'
 // CHECK-NEXT: FinalAttr 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> Implicit final
-// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit h '__hlsl_resource_t
+// CHECK-NEXT: FieldDecl 0x{{[0-9A-Fa-f]+}} <<invalid sloc>> <invalid sloc> implicit __handle '__hlsl_resource_t
 // CHECK-SAME{LITERAL}: [[hlsl::resource_class(SRV)]]
 // CHECK-SAME{LITERAL}: [[hlsl::raw_buffer]]
 // CHECK-SAME{LITERAL}: [[hlsl::contained_type(float)]]

diff --git a/clang/test/CXX/dcl.dcl/dcl.attr/dcl.attr.nodiscard/p2.cpp b/clang/test/CXX/dcl.dcl/dcl.attr/dcl.attr.nodiscard/p2.cpp
@@ -17,10 +17,10 @@ E get_e();
 // cxx11-warning@-1 {{use of the 'nodiscard' attribute is a C++17 extension}}
 
 void f() {
-  get_s(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
+  get_s(); // expected-warning {{ignoring return value of type 'S' declared with 'nodiscard' attribute}}
   get_i(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
   get_vi(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
-  get_e(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
+  get_e(); // expected-warning {{ignoring return value of type 'E' declared with 'nodiscard' attribute}}
 
   // Okay, warnings are not encouraged
   get_s_ref();
@@ -54,10 +54,10 @@ void f() {
   fp3 three;
   fp2_alias four;
 
-  one(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
-  two(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
-  three(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
-  four(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
+  one(); // expected-warning {{ignoring return value of type 'E' declared with 'nodiscard' attribute}}
+  two(); // expected-warning {{ignoring return value of type 'S' declared with 'nodiscard' attribute}}
+  three(); // expected-warning {{ignoring return value of type 'S' declared with 'nodiscard' attribute}}
+  four(); // expected-warning {{ignoring return value of type 'S' declared with 'nodiscard' attribute}}
 
   // These are all okay because of the explicit cast to void.
   (void)one();
@@ -84,8 +84,8 @@ LaterReason get_later_reason();
 // cxx11-17-warning@-1 {{use of the 'nodiscard' attribute is a C++20 extension}}
 
 void cxx20_use() {
-  get_reason(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute: reason}}
-  get_later_reason(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute: later reason}}
+  get_reason(); // expected-warning {{ignoring return value of type 'ReasonStruct' declared with 'nodiscard' attribute: reason}}
+  get_later_reason(); // expected-warning {{ignoring return value of type 'LaterReason' declared with 'nodiscard' attribute: later reason}}
   another_reason(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute: another reason}}
   conflicting_reason(); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute: special reason}}
 }
@@ -115,20 +115,20 @@ void usage() {
   S('A'); // expected-warning {{ignoring temporary created by a constructor declared with 'nodiscard' attribute: Don't let that S-Char go!}}
   S(1);
   S(2.2);
-  Y(); // expected-warning {{ignoring temporary created by a constructor declared with 'nodiscard' attribute: Don't throw me away either!}}
+  Y(); // expected-warning {{ignoring temporary of type 'Y' declared with 'nodiscard' attribute: Don't throw me away either!}}
   S s;
-  ConvertTo{}; // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute: Don't throw me away!}}
+  ConvertTo{}; // expected-warning {{ignoring return value of type 'ConvertTo' declared with 'nodiscard' attribute: Don't throw me away!}}
 
   // AST is different in C++17 mode. Before, a move ctor for ConvertTo is there
   // as well, hence the constructor warning.
 
-  // since-cxx17-warning@+2 {{ignoring return value of function declared with 'nodiscard' attribute: Don't throw me away!}}
-  // cxx11-warning@+1 {{ignoring temporary created by a constructor declared with 'nodiscard' attribute: Don't throw me away!}}
+  // since-cxx17-warning@+2 {{ignoring return value of type 'ConvertTo' declared with 'nodiscard' attribute: Don't throw me away!}}
+  // cxx11-warning@+1 {{ignoring temporary of type 'ConvertTo' declared with 'nodiscard' attribute: Don't throw me away!}}
   (ConvertTo) s;
   (int)s; // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
   (S)'c'; // expected-warning {{ignoring temporary created by a constructor declared with 'nodiscard' attribute: Don't let that S-Char go!}}
-  // since-cxx17-warning@+2 {{ignoring return value of function declared with 'nodiscard' attribute: Don't throw me away!}}
-  // cxx11-warning@+1 {{ignoring temporary created by a constructor declared with 'nodiscard' attribute: Don't throw me away!}}
+  // since-cxx17-warning@+2 {{ignoring return value of type 'ConvertTo' declared with 'nodiscard' attribute: Don't throw me away!}}
+  // cxx11-warning@+1 {{ignoring temporary of type 'ConvertTo' declared with 'nodiscard' attribute: Don't throw me away!}}
   static_cast<ConvertTo>(s);
   static_cast<int>(s); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute}}
   static_cast<double>(s); // expected-warning {{ignoring return value of function declared with 'nodiscard' attribute: Don't throw away as a double}}

diff --git a/clang/test/CXX/dcl.dcl/dcl.attr/dcl.attr.nodiscard/p3.cpp b/clang/test/CXX/dcl.dcl/dcl.attr/dcl.attr.nodiscard/p3.cpp
@@ -8,7 +8,7 @@ namespace std_example {
   error_info enable_missile_safety_mode();
   void launch_missiles();
   void test_missiles() {
-    enable_missile_safety_mode(); // expected-warning {{ignoring return value of function declared with 'nodiscard'}}
+    enable_missile_safety_mode(); // expected-warning {{ignoring return value of type 'error_info' declared with 'nodiscard'}}
     launch_missiles();
   }
 

diff --git a/clang/test/CXX/dcl.decl/dcl.meaning/dcl.fct/p2-cxx0x.cpp b/clang/test/CXX/dcl.decl/dcl.meaning/dcl.fct/p2-cxx0x.cpp
@@ -1,7 +1,18 @@
-// RUN: %clang_cc1 -fsyntax-only -verify -std=c++11 %s 
+// RUN: %clang_cc1 -fsyntax-only -std=c++11 -verify %s
+// RUN: not %clang_cc1 -fsyntax-only -std=c++11 -fno-diagnostics-show-line-numbers -fcaret-diagnostics-max-lines=1 %s 2>&1 | FileCheck %s -strict-whitespace
 
 auto a() -> int; // ok
 const auto b() -> int; // expected-error {{function with trailing return type must specify return type 'auto', not 'const auto'}}
 auto *c() -> int; // expected-error {{function with trailing return type must specify return type 'auto', not 'auto *'}}
 auto (d() -> int); // expected-error {{trailing return type may not be nested within parentheses}}
 auto e() -> auto (*)() -> auto (*)() -> void; // ok: same as void (*(*e())())();
+
+namespace GH78694 {
+
+template <typename T> struct B {
+  // CHECK:      error: function with trailing return type must specify return type 'auto', not 'void'
+  // CHECK-NEXT: {{^}}  template <class U> B(U) -> B<int>;
+  // CHECK-NEXT: {{^}}                     ~~~~~~~~^~~~~~{{$}}
+  template <class U> B(U) -> B<int>; // expected-error {{function with trailing return type must specify return type 'auto', not 'void'}}
+};
+}
diff --git a/clang/test/CodeGen/AArch64/elf-pauthabi.c b/clang/test/CodeGen/AArch64/elf-pauthabi.c
@@ -1,5 +1,3 @@
-//// TODO: also test with -fptrauth-elf-got when the driver flag is supported
-
 // RUN: %clang_cc1 -triple aarch64-linux -emit-llvm -o - \
 // RUN:   -fptrauth-intrinsics \
 // RUN:   -fptrauth-calls \
@@ -9,6 +7,7 @@
 // RUN:   -fptrauth-vtable-pointer-type-discrimination \
 // RUN:   -fptrauth-init-fini \
 // RUN:   -fptrauth-init-fini-address-discrimination \
+// RUN:   -fptrauth-elf-got \
 // RUN:   -fptrauth-indirect-gotos \
 // RUN:   -fptrauth-type-info-vtable-pointer-discrimination \
 // RUN:   -fptrauth-function-pointer-type-discrimination %s | \
@@ -42,6 +41,9 @@
 // RUN:   -fptrauth-calls -fptrauth-init-fini -fptrauth-init-fini-address-discrimination %s | \
 // RUN:   FileCheck %s --check-prefix=INITFINIADDR
 
+// RUN: %clang_cc1 -triple aarch64-linux -emit-llvm -o - \
+// RUN:   -fptrauth-elf-got %s | FileCheck %s --check-prefix=ELFGOT
+
 // RUN: %clang_cc1 -triple aarch64-linux -emit-llvm -o - \
 // RUN:   -fptrauth-indirect-gotos %s | FileCheck %s --check-prefix=GOTOS
 
@@ -54,7 +56,7 @@
 // RUN:   FileCheck %s --check-prefix=FPTRTYPE
 
 // ALL: !{i32 1, !"aarch64-elf-pauthabi-platform", i32 268435458}
-// ALL: !{i32 1, !"aarch64-elf-pauthabi-version", i32 3839}
+// ALL: !{i32 1, !"aarch64-elf-pauthabi-version", i32 4095}
 
 // INTRIN: !{i32 1, !"aarch64-elf-pauthabi-platform", i32 268435458}
 // INTRIN: !{i32 1, !"aarch64-elf-pauthabi-version", i32 1}
@@ -80,6 +82,9 @@
 // INITFINIADDR: !{i32 1, !"aarch64-elf-pauthabi-platform", i32 268435458}
 // INITFINIADDR: !{i32 1, !"aarch64-elf-pauthabi-version", i32 194}
 
+// ELFGOT: !{i32 1, !"aarch64-elf-pauthabi-platform", i32 268435458}
+// ELFGOT: !{i32 1, !"aarch64-elf-pauthabi-version", i32 256}
+
 // GOTOS: !{i32 1, !"aarch64-elf-pauthabi-platform", i32 268435458}
 // GOTOS: !{i32 1, !"aarch64-elf-pauthabi-version", i32 512}
 

diff --git a/clang/test/CodeGen/PowerPC/musttail-forward-declaration-inline.c b/clang/test/CodeGen/PowerPC/musttail-forward-declaration-inline.c
@@ -3,7 +3,7 @@
 
 inline int func2(int i);
 int external_call2(int i) {
-  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls can not be tail called on PPC}}
+  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls cannot be tail called on PPC}}
   [[clang::musttail]] return func2(i);
 }
 

diff --git a/clang/test/CodeGen/PowerPC/musttail-forward-declaration-weak.c b/clang/test/CodeGen/PowerPC/musttail-forward-declaration-weak.c
@@ -3,7 +3,7 @@
 
 int func2(int i);
 int external_call2(int i) {
-  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls can not be tail called on PPC}}
+  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls cannot be tail called on PPC}}
   [[clang::musttail]] return func2(i);
 }
 

diff --git a/clang/test/CodeGen/PowerPC/musttail-indirect.cpp b/clang/test/CodeGen/PowerPC/musttail-indirect.cpp
@@ -3,6 +3,6 @@
 
 void name(int *params) {
   auto fn = (void (*)(int *))1;
-  // expected-error@+1 {{'musttail' attribute for this call is impossible because indirect calls can not be tail called on PPC}}
+  // expected-error@+1 {{'musttail' attribute for this call is impossible because indirect calls cannot be tail called on PPC}}
   [[clang::musttail]] return fn(params);
 }
diff --git a/clang/test/CodeGen/PowerPC/musttail-inline.c b/clang/test/CodeGen/PowerPC/musttail-inline.c
@@ -7,6 +7,6 @@ inline int foo(int x) {
 
 int bar(int x)
 {
-  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls can not be tail called on PPC}}
+  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls cannot be tail called on PPC}}
   [[clang::musttail]] return foo(1);
 }
diff --git a/clang/test/CodeGen/PowerPC/musttail-undefined.c b/clang/test/CodeGen/PowerPC/musttail-undefined.c
@@ -5,6 +5,6 @@ int foo(int x);
 
 int bar(int x)
 {
-  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls can not be tail called on PPC}}
+  // expected-error@+1 {{'musttail' attribute for this call is impossible because external calls cannot be tail called on PPC}}
   [[clang::musttail]] return foo(x);
 }
diff --git a/clang/test/CodeGen/PowerPC/musttail-weak.c b/clang/test/CodeGen/PowerPC/musttail-weak.c
@@ -7,7 +7,7 @@ __attribute__((weak)) int func2(int i) {
   return 0;
 }
 int external_call2(int i) {
-  // linux-error@+2 {{'musttail' attribute for this call is impossible because external calls can not be tail called on PPC}}
+  // linux-error@+2 {{'musttail' attribute for this call is impossible because external calls cannot be tail called on PPC}}
   // aix-error@+1 {{'musttail' attribute is not supported on AIX}}
   [[clang::musttail]] return func2(i);
 }
diff --git a/clang/test/CodeGen/PowerPC/musttail.c b/clang/test/CodeGen/PowerPC/musttail.c
@@ -14,7 +14,7 @@ int foo(int x) {
 int bar(int x)
 {
   // good-no-diagnostics
-  // longcall-error@+2 {{'musttail' attribute for this call is impossible because long calls can not be tail called on PPC}}
+  // longcall-error@+2 {{'musttail' attribute for this call is impossible because long calls cannot be tail called on PPC}}
   // aix-error@+1 {{'musttail' attribute is not supported on AIX}}
  [[clang::musttail]] return foo(1);
 }
diff --git a/clang/test/CodeGen/RISCV/riscv-inline-asm.c b/clang/test/CodeGen/RISCV/riscv-inline-asm.c
@@ -33,6 +33,19 @@ void test_cf(float f, double d) {
   asm volatile("" : "=cf"(cd) : "cf"(d));
 }
 
+#if __riscv_xlen == 32
+typedef long long double_xlen_t;
+#elif __riscv_xlen == 64
+typedef __int128_t double_xlen_t;
+#endif
+double_xlen_t test_R_wide_scalar(double_xlen_t p) {
+// CHECK-LABEL: define{{.*}} {{i128|i64}} @test_R_wide_scalar(
+// CHECK: call {{i128|i64}} asm sideeffect "", "=R,R"({{i128|i64}} %{{.*}})
+  double_xlen_t ret;
+  asm volatile("" : "=R"(ret) : "R"(p));
+  return ret;
+}
+
 void test_I(void) {
 // CHECK-LABEL: define{{.*}} void @test_I()
 // CHECK: call void asm sideeffect "", "I"(i32 2047)

diff --git a/clang/test/CodeGen/X86/x86_64-PR42672.c b/clang/test/CodeGen/X86/x86_64-PR42672.c
@@ -58,7 +58,7 @@ void odd_struct(void) {
       : "=r"(str));
 #endif
 }
-// CHECK-IMPOSSIBLE_ODD: impossible constraint in asm: can't store value into a register
+// CHECK-IMPOSSIBLE_ODD: impossible constraint in asm: cannot store value into a register
 
 // Check Clang reports an error if attempting to return a big structure via a register.
 void big_struct(void) {
@@ -70,7 +70,7 @@ void big_struct(void) {
       : "=r"(str));
 #endif
 }
-// CHECK-IMPOSSIBLE_BIG: impossible constraint in asm: can't store value into a register
+// CHECK-IMPOSSIBLE_BIG: impossible constraint in asm: cannot store value into a register
 
 // Clang is able to emit LLVM IR for an 16-byte structure.
 void x_constraint_fit(void) {
@@ -103,7 +103,7 @@ void x_constraint_nofit(void) {
 
 // http://crbug.com/999160
 // Clang used to report the following message:
-//   "impossible constraint in asm: can't store struct into a register"
+//   "impossible constraint in asm: cannot store struct into a register"
 // for the assembly directive below, although there's no struct.
 void crbug_999160_regtest(void) {
 #ifdef IMPOSSIBLE_9BYTES
@@ -113,12 +113,12 @@ void crbug_999160_regtest(void) {
 #endif
 }
 
-// CHECK-IMPOSSIBLE_9BYTES: impossible constraint in asm: can't store value into a register
+// CHECK-IMPOSSIBLE_9BYTES: impossible constraint in asm: cannot store value into a register
 
 void crbug_999160_regtest_v2(void) {
 #ifdef IMPOSSIBLE_9BYTES_V2
   char buf[9];
   asm("" : "=r"(buf) : "0"(buf));
 #endif
 }
-// CHECK-IMPOSSIBLE_9BYTES_V2: impossible constraint in asm: can't store value into a register
+// CHECK-IMPOSSIBLE_9BYTES_V2: impossible constraint in asm: cannot store value into a register
diff --git a/clang/test/CodeGen/arm-mve-intrinsics/compare.c b/clang/test/CodeGen/arm-mve-intrinsics/compare.c
@@ -2376,7 +2376,7 @@ mve_pred16_t test_vcmphiq_m_n_u32(uint32x4_t a, uint32_t b, mve_pred16_t p)
 
 // CHECK-LABEL: @test_vcmpleq_f16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ole <8 x half> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ule <8 x half> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2392,7 +2392,7 @@ mve_pred16_t test_vcmpleq_f16(float16x8_t a, float16x8_t b)
 
 // CHECK-LABEL: @test_vcmpleq_f32(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ole <4 x float> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ule <4 x float> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2458,7 +2458,7 @@ mve_pred16_t test_vcmpleq_s32(int32x4_t a, int32x4_t b)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <8 x half> poison, half [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <8 x half> [[DOTSPLATINSERT]], <8 x half> poison, <8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ole <8 x half> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ule <8 x half> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2476,7 +2476,7 @@ mve_pred16_t test_vcmpleq_n_f16(float16x8_t a, float16_t b)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x float> [[DOTSPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ole <4 x float> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ule <4 x float> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2548,7 +2548,7 @@ mve_pred16_t test_vcmpleq_n_s32(int32x4_t a, int32_t b)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
 // CHECK-NEXT:    [[TMP1:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP0]])
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ole <8 x half> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ule <8 x half> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <8 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -2567,7 +2567,7 @@ mve_pred16_t test_vcmpleq_m_f16(float16x8_t a, float16x8_t b, mve_pred16_t p)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
 // CHECK-NEXT:    [[TMP1:%.*]] = call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 [[TMP0]])
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ole <4 x float> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ule <4 x float> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <4 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -2645,7 +2645,7 @@ mve_pred16_t test_vcmpleq_m_s32(int32x4_t a, int32x4_t b, mve_pred16_t p)
 // CHECK-NEXT:    [[TMP1:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP0]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <8 x half> poison, half [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <8 x half> [[DOTSPLATINSERT]], <8 x half> poison, <8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ole <8 x half> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ule <8 x half> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <8 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -2666,7 +2666,7 @@ mve_pred16_t test_vcmpleq_m_n_f16(float16x8_t a, float16_t b, mve_pred16_t p)
 // CHECK-NEXT:    [[TMP1:%.*]] = call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 [[TMP0]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x float> [[DOTSPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ole <4 x float> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ule <4 x float> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <4 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -2746,7 +2746,7 @@ mve_pred16_t test_vcmpleq_m_n_s32(int32x4_t a, int32_t b, mve_pred16_t p)
 
 // CHECK-LABEL: @test_vcmpltq_f16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp olt <8 x half> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ult <8 x half> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2762,7 +2762,7 @@ mve_pred16_t test_vcmpltq_f16(float16x8_t a, float16x8_t b)
 
 // CHECK-LABEL: @test_vcmpltq_f32(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp olt <4 x float> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ult <4 x float> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2828,7 +2828,7 @@ mve_pred16_t test_vcmpltq_s32(int32x4_t a, int32x4_t b)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <8 x half> poison, half [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <8 x half> [[DOTSPLATINSERT]], <8 x half> poison, <8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp olt <8 x half> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ult <8 x half> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2846,7 +2846,7 @@ mve_pred16_t test_vcmpltq_n_f16(float16x8_t a, float16_t b)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x float> [[DOTSPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP0:%.*]] = fcmp olt <4 x float> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP0:%.*]] = fcmp ult <4 x float> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    ret i16 [[TMP2]]
@@ -2918,7 +2918,7 @@ mve_pred16_t test_vcmpltq_n_s32(int32x4_t a, int32_t b)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
 // CHECK-NEXT:    [[TMP1:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP0]])
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp olt <8 x half> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ult <8 x half> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <8 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -2937,7 +2937,7 @@ mve_pred16_t test_vcmpltq_m_f16(float16x8_t a, float16x8_t b, mve_pred16_t p)
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = zext i16 [[P:%.*]] to i32
 // CHECK-NEXT:    [[TMP1:%.*]] = call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 [[TMP0]])
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp olt <4 x float> [[A:%.*]], [[B:%.*]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ult <4 x float> [[A:%.*]], [[B:%.*]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <4 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -3015,7 +3015,7 @@ mve_pred16_t test_vcmpltq_m_s32(int32x4_t a, int32x4_t b, mve_pred16_t p)
 // CHECK-NEXT:    [[TMP1:%.*]] = call <8 x i1> @llvm.arm.mve.pred.i2v.v8i1(i32 [[TMP0]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <8 x half> poison, half [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <8 x half> [[DOTSPLATINSERT]], <8 x half> poison, <8 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp olt <8 x half> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ult <8 x half> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <8 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v8i1(<8 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16
@@ -3036,7 +3036,7 @@ mve_pred16_t test_vcmpltq_m_n_f16(float16x8_t a, float16_t b, mve_pred16_t p)
 // CHECK-NEXT:    [[TMP1:%.*]] = call <4 x i1> @llvm.arm.mve.pred.i2v.v4i1(i32 [[TMP0]])
 // CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[B:%.*]], i64 0
 // CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <4 x float> [[DOTSPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
-// CHECK-NEXT:    [[TMP2:%.*]] = fcmp olt <4 x float> [[A:%.*]], [[DOTSPLAT]]
+// CHECK-NEXT:    [[TMP2:%.*]] = fcmp ult <4 x float> [[A:%.*]], [[DOTSPLAT]]
 // CHECK-NEXT:    [[TMP3:%.*]] = and <4 x i1> [[TMP1]], [[TMP2]]
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.arm.mve.pred.v2i.v4i1(<4 x i1> [[TMP3]])
 // CHECK-NEXT:    [[TMP5:%.*]] = trunc i32 [[TMP4]] to i16

diff --git a/clang/test/CodeGen/ptrauth-module-flags.c b/clang/test/CodeGen/ptrauth-module-flags.c
@@ -0,0 +1,8 @@
+// RUN: %clang_cc1 -triple aarch64-linux-gnu                   -emit-llvm %s  -o - | FileCheck %s --check-prefix=OFF
+// RUN: %clang_cc1 -triple aarch64-linux-gnu -fptrauth-elf-got -emit-llvm %s  -o - | FileCheck %s --check-prefix=ELFGOT
+
+// ELFGOT:      !llvm.module.flags = !{
+// ELFGOT-SAME: !1
+// ELFGOT:      !1 = !{i32 8, !"ptrauth-elf-got", i32 1}
+
+// OFF-NOT: "ptrauth-
diff --git a/clang/test/CodeGen/scoped-fence-ops.c b/clang/test/CodeGen/scoped-fence-ops.c
@@ -0,0 +1,257 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5
+// RUN: %clang_cc1 %s -emit-llvm -o - -triple=amdgcn-amd-amdhsa -ffreestanding \
+// RUN:   -fvisibility=hidden | FileCheck --check-prefix=AMDGCN %s
+// RUN: %clang_cc1 %s -emit-llvm -o - -triple=spirv64-unknown-unknown -ffreestanding \
+// RUN:   -fvisibility=hidden | FileCheck --check-prefix=SPIRV %s
+// RUN: %clang_cc1 %s -emit-llvm -o - -triple=x86_64-unknown-linux-gnu -ffreestanding \
+// RUN:   -fvisibility=hidden | FileCheck --check-prefix=X86_64 %s
+
+// AMDGCN-LABEL: define hidden void @fe1a(
+// AMDGCN-SAME: ) #[[ATTR0:[0-9]+]] {
+// AMDGCN-NEXT:  [[ENTRY:.*:]]
+// AMDGCN-NEXT:    fence syncscope("workgroup-one-as") release
+// AMDGCN-NEXT:    ret void
+//
+// SPIRV-LABEL: define hidden spir_func void @fe1a(
+// SPIRV-SAME: ) #[[ATTR0:[0-9]+]] {
+// SPIRV-NEXT:  [[ENTRY:.*:]]
+// SPIRV-NEXT:    fence syncscope("workgroup") release
+// SPIRV-NEXT:    ret void
+//
+// X86_64-LABEL: define hidden void @fe1a(
+// X86_64-SAME: ) #[[ATTR0:[0-9]+]] {
+// X86_64-NEXT:  [[ENTRY:.*:]]
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    ret void
+//
+void fe1a() {
+  __scoped_atomic_thread_fence(__ATOMIC_RELEASE, __MEMORY_SCOPE_WRKGRP);
+}
+
+// AMDGCN-LABEL: define hidden void @fe1b(
+// AMDGCN-SAME: i32 noundef [[ORD:%.*]]) #[[ATTR0]] {
+// AMDGCN-NEXT:  [[ENTRY:.*:]]
+// AMDGCN-NEXT:    [[ORD_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// AMDGCN-NEXT:    [[ORD_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[ORD_ADDR]] to ptr
+// AMDGCN-NEXT:    store i32 [[ORD]], ptr [[ORD_ADDR_ASCAST]], align 4
+// AMDGCN-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ORD_ADDR_ASCAST]], align 4
+// AMDGCN-NEXT:    switch i32 [[TMP0]], label %[[ATOMIC_SCOPE_CONTINUE:.*]] [
+// AMDGCN-NEXT:      i32 1, label %[[ACQUIRE:.*]]
+// AMDGCN-NEXT:      i32 2, label %[[ACQUIRE]]
+// AMDGCN-NEXT:      i32 3, label %[[RELEASE:.*]]
+// AMDGCN-NEXT:      i32 4, label %[[ACQREL:.*]]
+// AMDGCN-NEXT:      i32 5, label %[[SEQCST:.*]]
+// AMDGCN-NEXT:    ]
+// AMDGCN:       [[ATOMIC_SCOPE_CONTINUE]]:
+// AMDGCN-NEXT:    ret void
+// AMDGCN:       [[ACQUIRE]]:
+// AMDGCN-NEXT:    fence syncscope("workgroup-one-as") acquire
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[RELEASE]]:
+// AMDGCN-NEXT:    fence syncscope("workgroup-one-as") release
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[ACQREL]]:
+// AMDGCN-NEXT:    fence syncscope("workgroup-one-as") acq_rel
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[SEQCST]]:
+// AMDGCN-NEXT:    fence syncscope("workgroup") seq_cst
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+//
+// SPIRV-LABEL: define hidden spir_func void @fe1b(
+// SPIRV-SAME: i32 noundef [[ORD:%.*]]) #[[ATTR0]] {
+// SPIRV-NEXT:  [[ENTRY:.*:]]
+// SPIRV-NEXT:    [[ORD_ADDR:%.*]] = alloca i32, align 4
+// SPIRV-NEXT:    store i32 [[ORD]], ptr [[ORD_ADDR]], align 4
+// SPIRV-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ORD_ADDR]], align 4
+// SPIRV-NEXT:    switch i32 [[TMP0]], label %[[ATOMIC_SCOPE_CONTINUE:.*]] [
+// SPIRV-NEXT:      i32 1, label %[[ACQUIRE:.*]]
+// SPIRV-NEXT:      i32 2, label %[[ACQUIRE]]
+// SPIRV-NEXT:      i32 3, label %[[RELEASE:.*]]
+// SPIRV-NEXT:      i32 4, label %[[ACQREL:.*]]
+// SPIRV-NEXT:      i32 5, label %[[SEQCST:.*]]
+// SPIRV-NEXT:    ]
+// SPIRV:       [[ATOMIC_SCOPE_CONTINUE]]:
+// SPIRV-NEXT:    ret void
+// SPIRV:       [[ACQUIRE]]:
+// SPIRV-NEXT:    fence syncscope("workgroup") acquire
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[RELEASE]]:
+// SPIRV-NEXT:    fence syncscope("workgroup") release
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[ACQREL]]:
+// SPIRV-NEXT:    fence syncscope("workgroup") acq_rel
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[SEQCST]]:
+// SPIRV-NEXT:    fence syncscope("workgroup") seq_cst
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+//
+// X86_64-LABEL: define hidden void @fe1b(
+// X86_64-SAME: i32 noundef [[ORD:%.*]]) #[[ATTR0]] {
+// X86_64-NEXT:  [[ENTRY:.*:]]
+// X86_64-NEXT:    [[ORD_ADDR:%.*]] = alloca i32, align 4
+// X86_64-NEXT:    store i32 [[ORD]], ptr [[ORD_ADDR]], align 4
+// X86_64-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ORD_ADDR]], align 4
+// X86_64-NEXT:    switch i32 [[TMP0]], label %[[ATOMIC_SCOPE_CONTINUE:.*]] [
+// X86_64-NEXT:      i32 1, label %[[ACQUIRE:.*]]
+// X86_64-NEXT:      i32 2, label %[[ACQUIRE]]
+// X86_64-NEXT:      i32 3, label %[[RELEASE:.*]]
+// X86_64-NEXT:      i32 4, label %[[ACQREL:.*]]
+// X86_64-NEXT:      i32 5, label %[[SEQCST:.*]]
+// X86_64-NEXT:    ]
+// X86_64:       [[ATOMIC_SCOPE_CONTINUE]]:
+// X86_64-NEXT:    ret void
+// X86_64:       [[ACQUIRE]]:
+// X86_64-NEXT:    fence acquire
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[RELEASE]]:
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[ACQREL]]:
+// X86_64-NEXT:    fence acq_rel
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[SEQCST]]:
+// X86_64-NEXT:    fence seq_cst
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+//
+void fe1b(int ord) {
+  __scoped_atomic_thread_fence(ord, __MEMORY_SCOPE_WRKGRP);
+}
+
+// AMDGCN-LABEL: define hidden void @fe1c(
+// AMDGCN-SAME: i32 noundef [[SCOPE:%.*]]) #[[ATTR0]] {
+// AMDGCN-NEXT:  [[ENTRY:.*:]]
+// AMDGCN-NEXT:    [[SCOPE_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// AMDGCN-NEXT:    [[SCOPE_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SCOPE_ADDR]] to ptr
+// AMDGCN-NEXT:    store i32 [[SCOPE]], ptr [[SCOPE_ADDR_ASCAST]], align 4
+// AMDGCN-NEXT:    [[TMP0:%.*]] = load i32, ptr [[SCOPE_ADDR_ASCAST]], align 4
+// AMDGCN-NEXT:    switch i32 [[TMP0]], label %[[ATOMIC_SCOPE_CONTINUE:.*]] [
+// AMDGCN-NEXT:      i32 1, label %[[DEVICE_SCOPE:.*]]
+// AMDGCN-NEXT:      i32 0, label %[[SYSTEM_SCOPE:.*]]
+// AMDGCN-NEXT:      i32 2, label %[[WORKGROUP_SCOPE:.*]]
+// AMDGCN-NEXT:      i32 3, label %[[WAVEFRONT_SCOPE:.*]]
+// AMDGCN-NEXT:      i32 4, label %[[SINGLE_SCOPE:.*]]
+// AMDGCN-NEXT:    ]
+// AMDGCN:       [[ATOMIC_SCOPE_CONTINUE]]:
+// AMDGCN-NEXT:    ret void
+// AMDGCN:       [[DEVICE_SCOPE]]:
+// AMDGCN-NEXT:    fence syncscope("agent-one-as") release
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[SYSTEM_SCOPE]]:
+// AMDGCN-NEXT:    fence syncscope("one-as") release
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[WORKGROUP_SCOPE]]:
+// AMDGCN-NEXT:    fence syncscope("workgroup-one-as") release
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[WAVEFRONT_SCOPE]]:
+// AMDGCN-NEXT:    fence syncscope("wavefront-one-as") release
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// AMDGCN:       [[SINGLE_SCOPE]]:
+// AMDGCN-NEXT:    fence syncscope("singlethread-one-as") release
+// AMDGCN-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+//
+// SPIRV-LABEL: define hidden spir_func void @fe1c(
+// SPIRV-SAME: i32 noundef [[SCOPE:%.*]]) #[[ATTR0]] {
+// SPIRV-NEXT:  [[ENTRY:.*:]]
+// SPIRV-NEXT:    [[SCOPE_ADDR:%.*]] = alloca i32, align 4
+// SPIRV-NEXT:    store i32 [[SCOPE]], ptr [[SCOPE_ADDR]], align 4
+// SPIRV-NEXT:    [[TMP0:%.*]] = load i32, ptr [[SCOPE_ADDR]], align 4
+// SPIRV-NEXT:    switch i32 [[TMP0]], label %[[ATOMIC_SCOPE_CONTINUE:.*]] [
+// SPIRV-NEXT:      i32 1, label %[[DEVICE_SCOPE:.*]]
+// SPIRV-NEXT:      i32 0, label %[[SYSTEM_SCOPE:.*]]
+// SPIRV-NEXT:      i32 2, label %[[WORKGROUP_SCOPE:.*]]
+// SPIRV-NEXT:      i32 3, label %[[WAVEFRONT_SCOPE:.*]]
+// SPIRV-NEXT:      i32 4, label %[[SINGLE_SCOPE:.*]]
+// SPIRV-NEXT:    ]
+// SPIRV:       [[ATOMIC_SCOPE_CONTINUE]]:
+// SPIRV-NEXT:    ret void
+// SPIRV:       [[DEVICE_SCOPE]]:
+// SPIRV-NEXT:    fence syncscope("device") release
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[SYSTEM_SCOPE]]:
+// SPIRV-NEXT:    fence release
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[WORKGROUP_SCOPE]]:
+// SPIRV-NEXT:    fence syncscope("workgroup") release
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[WAVEFRONT_SCOPE]]:
+// SPIRV-NEXT:    fence syncscope("subgroup") release
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// SPIRV:       [[SINGLE_SCOPE]]:
+// SPIRV-NEXT:    fence syncscope("singlethread") release
+// SPIRV-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+//
+// X86_64-LABEL: define hidden void @fe1c(
+// X86_64-SAME: i32 noundef [[SCOPE:%.*]]) #[[ATTR0]] {
+// X86_64-NEXT:  [[ENTRY:.*:]]
+// X86_64-NEXT:    [[SCOPE_ADDR:%.*]] = alloca i32, align 4
+// X86_64-NEXT:    store i32 [[SCOPE]], ptr [[SCOPE_ADDR]], align 4
+// X86_64-NEXT:    [[TMP0:%.*]] = load i32, ptr [[SCOPE_ADDR]], align 4
+// X86_64-NEXT:    switch i32 [[TMP0]], label %[[ATOMIC_SCOPE_CONTINUE:.*]] [
+// X86_64-NEXT:      i32 1, label %[[DEVICE_SCOPE:.*]]
+// X86_64-NEXT:      i32 0, label %[[SYSTEM_SCOPE:.*]]
+// X86_64-NEXT:      i32 2, label %[[WORKGROUP_SCOPE:.*]]
+// X86_64-NEXT:      i32 3, label %[[WAVEFRONT_SCOPE:.*]]
+// X86_64-NEXT:      i32 4, label %[[SINGLE_SCOPE:.*]]
+// X86_64-NEXT:    ]
+// X86_64:       [[ATOMIC_SCOPE_CONTINUE]]:
+// X86_64-NEXT:    ret void
+// X86_64:       [[DEVICE_SCOPE]]:
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[SYSTEM_SCOPE]]:
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[WORKGROUP_SCOPE]]:
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[WAVEFRONT_SCOPE]]:
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+// X86_64:       [[SINGLE_SCOPE]]:
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    br label %[[ATOMIC_SCOPE_CONTINUE]]
+//
+void fe1c(int scope) {
+  __scoped_atomic_thread_fence(__ATOMIC_RELEASE, scope);
+}
+
+// AMDGCN-LABEL: define hidden void @fe2a(
+// AMDGCN-SAME: ) #[[ATTR0]] {
+// AMDGCN-NEXT:  [[ENTRY:.*:]]
+// AMDGCN-NEXT:    ret void
+//
+// SPIRV-LABEL: define hidden spir_func void @fe2a(
+// SPIRV-SAME: ) #[[ATTR0]] {
+// SPIRV-NEXT:  [[ENTRY:.*:]]
+// SPIRV-NEXT:    ret void
+//
+// X86_64-LABEL: define hidden void @fe2a(
+// X86_64-SAME: ) #[[ATTR0]] {
+// X86_64-NEXT:  [[ENTRY:.*:]]
+// X86_64-NEXT:    ret void
+//
+void fe2a() {
+  __scoped_atomic_thread_fence(999, __MEMORY_SCOPE_SYSTEM);
+}
+
+// AMDGCN-LABEL: define hidden void @fe2b(
+// AMDGCN-SAME: ) #[[ATTR0]] {
+// AMDGCN-NEXT:  [[ENTRY:.*:]]
+// AMDGCN-NEXT:    fence syncscope("one-as") release
+// AMDGCN-NEXT:    ret void
+//
+// SPIRV-LABEL: define hidden spir_func void @fe2b(
+// SPIRV-SAME: ) #[[ATTR0]] {
+// SPIRV-NEXT:  [[ENTRY:.*:]]
+// SPIRV-NEXT:    fence release
+// SPIRV-NEXT:    ret void
+//
+// X86_64-LABEL: define hidden void @fe2b(
+// X86_64-SAME: ) #[[ATTR0]] {
+// X86_64-NEXT:  [[ENTRY:.*:]]
+// X86_64-NEXT:    fence release
+// X86_64-NEXT:    ret void
+//
+void fe2b() {
+  __scoped_atomic_thread_fence(__ATOMIC_RELEASE, 999);
+}
diff --git a/clang/test/CodeGenCUDASPIRV/spirv-attrs.cu b/clang/test/CodeGenCUDASPIRV/spirv-attrs.cu
@@ -0,0 +1,28 @@
+// RUN: %clang_cc1 -fcuda-is-device -triple spirv64 -o - -emit-llvm -x cuda %s  | FileCheck %s
+// RUN: %clang_cc1 -fcuda-is-device -triple spirv32 -o - -emit-llvm -x cuda %s  | FileCheck %s
+
+#define __global__ __attribute__((global))
+
+__attribute__((reqd_work_group_size(128, 1, 1)))
+__global__ void reqd_work_group_size_128_1_1() {}
+
+__attribute__((work_group_size_hint(2, 2, 2)))
+__global__ void work_group_size_hint_2_2_2() {}
+
+__attribute__((vec_type_hint(int)))
+__global__ void vec_type_hint_int() {}
+
+__attribute__((intel_reqd_sub_group_size(64)))
+__global__ void intel_reqd_sub_group_size_64() {}
+
+// CHECK: define spir_kernel void @_Z28reqd_work_group_size_128_1_1v() #[[ATTR:[0-9]+]] !reqd_work_group_size ![[WG_SIZE:[0-9]+]]
+// CHECK: define spir_kernel void @_Z26work_group_size_hint_2_2_2v() #[[ATTR]] !work_group_size_hint ![[WG_HINT:[0-9]+]]
+// CHECK: define spir_kernel void @_Z17vec_type_hint_intv() #[[ATTR]] !vec_type_hint ![[VEC_HINT:[0-9]+]]
+// CHECK: define spir_kernel void @_Z28intel_reqd_sub_group_size_64v() #[[ATTR]] !intel_reqd_sub_group_size ![[SUB_GRP:[0-9]+]]
+
+// CHECK: attributes #[[ATTR]] = { {{.*}} }
+
+// CHECK: ![[WG_SIZE]] = !{i32 128, i32 1, i32 1}
+// CHECK: ![[WG_HINT]] = !{i32 2, i32 2, i32 2}
+// CHECK: ![[VEC_HINT]] = !{i32 undef, i32 1}
+// CHECK: ![[SUB_GRP]] = !{i32 64}
diff --git a/clang/test/CodeGenOpenCL/amdgpu-features.cl b/clang/test/CodeGenOpenCL/amdgpu-features.cl
@@ -32,6 +32,7 @@
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx940 -emit-llvm -o - %s | FileCheck --check-prefix=GFX940 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx941 -emit-llvm -o - %s | FileCheck --check-prefix=GFX941 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx942 -emit-llvm -o - %s | FileCheck --check-prefix=GFX942 %s
+// RUN: %clang_cc1 -triple amdgcn -target-cpu gfx950 -emit-llvm -o - %s | FileCheck --check-prefix=GFX950 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1010 -emit-llvm -o - %s | FileCheck --check-prefix=GFX1010 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1011 -emit-llvm -o - %s | FileCheck --check-prefix=GFX1011 %s
 // RUN: %clang_cc1 -triple amdgcn -target-cpu gfx1012 -emit-llvm -o - %s | FileCheck --check-prefix=GFX1012 %s
@@ -88,6 +89,7 @@
 // GFX941: "target-features"="+16-bit-insts,+atomic-buffer-global-pk-add-f16-insts,+atomic-ds-pk-add-16-insts,+atomic-fadd-rtn-insts,+atomic-flat-pk-add-16-insts,+atomic-global-pk-add-bf16-inst,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+fp8-conversion-insts,+fp8-insts,+gfx8-insts,+gfx9-insts,+gfx90a-insts,+gfx940-insts,+mai-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize64,+xf32-insts"
 // GFX942: "target-features"="+16-bit-insts,+atomic-buffer-global-pk-add-f16-insts,+atomic-ds-pk-add-16-insts,+atomic-fadd-rtn-insts,+atomic-flat-pk-add-16-insts,+atomic-global-pk-add-bf16-inst,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+fp8-conversion-insts,+fp8-insts,+gfx8-insts,+gfx9-insts,+gfx90a-insts,+gfx940-insts,+mai-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize64,+xf32-insts"
 // GFX9_4_Generic: "target-features"="+16-bit-insts,+atomic-buffer-global-pk-add-f16-insts,+atomic-ds-pk-add-16-insts,+atomic-fadd-rtn-insts,+atomic-flat-pk-add-16-insts,+atomic-global-pk-add-bf16-inst,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+gfx8-insts,+gfx9-insts,+gfx90a-insts,+gfx940-insts,+mai-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize64"
+// GFX950: "target-features"="+16-bit-insts,+atomic-buffer-global-pk-add-f16-insts,+atomic-ds-pk-add-16-insts,+atomic-fadd-rtn-insts,+atomic-flat-pk-add-16-insts,+atomic-global-pk-add-bf16-inst,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot3-insts,+dot4-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+fp8-conversion-insts,+fp8-insts,+gfx8-insts,+gfx9-insts,+gfx90a-insts,+gfx940-insts,+gfx950-insts,+mai-insts,+prng-inst,+s-memrealtime,+s-memtime-inst,+wavefrontsize64"
 // GFX1010: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dpp,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize32"
 // GFX1011: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize32"
 // GFX1012: "target-features"="+16-bit-insts,+ci-insts,+dl-insts,+dot1-insts,+dot10-insts,+dot2-insts,+dot5-insts,+dot6-insts,+dot7-insts,+dpp,+gfx10-insts,+gfx8-insts,+gfx9-insts,+s-memrealtime,+s-memtime-inst,+wavefrontsize32"

diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx950-err.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx950-err.cl
@@ -0,0 +1,16 @@
+// RUN: %clang_cc1 -O0 -cl-std=CL2.0 -triple amdgcn-amd-amdhsa -target-cpu gfx906 -emit-llvm \
+// RUN:   -verify -o - %s
+// RUN: %clang_cc1 -O0 -cl-std=CL2.0 -triple amdgcn-amd-amdhsa -target-cpu gfx90a -emit-llvm \
+// RUN:   -verify -o - %s
+// RUN: %clang_cc1 -O0 -cl-std=CL2.0 -triple amdgcn-amd-amdhsa -target-cpu gfx940 -emit-llvm \
+// RUN:   -verify -o - %s
+// RUN: %clang_cc1 -O0 -cl-std=CL2.0 -triple amdgcn-amd-amdhsa -target-cpu gfx1200 -emit-llvm \
+// RUN:   -verify -o - %s
+
+
+// REQUIRES: amdgpu-registered-target
+
+typedef unsigned int uint;
+void test_prng_b32(global uint* out, uint a) {
+  *out = __builtin_amdgcn_prng_b32(a); // expected-error{{'__builtin_amdgcn_prng_b32' needs target feature prng-inst}}
+}
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx950.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx950.cl
@@ -0,0 +1,21 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -cl-std=CL1.2 -O0 -triple amdgcn-unknown-unknown -target-cpu gfx950 -emit-llvm -o - %s | FileCheck %s
+// REQUIRES: amdgpu-registered-target
+
+typedef unsigned int uint;
+
+// CHECK-LABEL: @test_prng_b32(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// CHECK-NEXT:    store ptr addrspace(1) [[OUT:%.*]], ptr addrspace(5) [[OUT_ADDR]], align 8
+// CHECK-NEXT:    store i32 [[A:%.*]], ptr addrspace(5) [[A_ADDR]], align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr addrspace(5) [[A_ADDR]], align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.amdgcn.prng.b32(i32 [[TMP0]])
+// CHECK-NEXT:    [[TMP2:%.*]] = load ptr addrspace(1), ptr addrspace(5) [[OUT_ADDR]], align 8
+// CHECK-NEXT:    store i32 [[TMP1]], ptr addrspace(1) [[TMP2]], align 4
+// CHECK-NEXT:    ret void
+//
+void test_prng_b32(global uint* out, uint a) {
+  *out = __builtin_amdgcn_prng_b32(a);
+}
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-mfma.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-mfma.cl
@@ -2,6 +2,7 @@
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx908 -DMFMA_GFX908_TESTS -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX908
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx90a -DMFMA_GFX90A_TESTS -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX90A
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx940 -DMFMA_GFX940_TESTS -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX940
+// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx950 -DMFMA_GFX950_TESTS -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX950
 
 #pragma OPENCL EXTENSION cl_khr_fp64:enable
 
@@ -23,6 +24,7 @@ typedef short  v8s   __attribute__((ext_vector_type(8)));
 typedef short  v16s  __attribute__((ext_vector_type(16)));
 typedef short  v32s  __attribute__((ext_vector_type(32)));
 typedef double v4d   __attribute__((ext_vector_type(4)));
+typedef __bf16 v8bf16   __attribute__((ext_vector_type(8)));
 
 
 #ifdef MFMA_GFX908_TESTS
@@ -222,7 +224,7 @@ void test_mfma_f64_4x4x4f64(global double* out, double a, double b, double c)
 
 #endif // MFMA_GFX90A_TESTS
 
-#ifdef MFMA_GFX940_TESTS
+#if defined(MFMA_GFX940_TESTS) || defined(MFMA_GFX950_TESTS)
 // CHECK-GFX940-LABEL: @test_mfma_i32_16x16x32_i8
 // CHECK-GFX940: call <4 x i32> @llvm.amdgcn.mfma.i32.16x16x32.i8(i64 %a, i64 %b, <4 x i32> %c, i32 0, i32 0, i32 0)
 void test_mfma_i32_16x16x32_i8(global v4i* out, long a, long b, v4i c)
@@ -404,4 +406,29 @@ void test_smfmac_f32_32x32x32_fp8_fp8(global v16f* out, v2i a, v4i b, v16f c, in
 {
   *out = __builtin_amdgcn_smfmac_f32_32x32x32_fp8_fp8(a, b, c, idx, 0, 0);
 }
-#endif // MFMA_GFX940_TESTS
+#endif // defined(MFMA_GFX940_TESTS) || defined(MFMA_GFX950_TESTS)
+
+#ifdef MFMA_GFX950_TESTS
+
+// CHECK-GFX950-LABEL: @test_mfma_f32_16x16x32_f16(
+// CHECK-GFX950: tail call <4 x float> @llvm.amdgcn.mfma.f32.16x16x32.f16(<8 x half> %a, <8 x half> %b, <4 x float> %c, i32 1, i32 2, i32 3)
+
+v4f test_mfma_f32_16x16x32_f16(v8h a, v8h b, v4f c)
+{
+  return __builtin_amdgcn_mfma_f32_16x16x32_f16(a, b, c, 1, 2, 3);
+}
+
+// CHECK-GFX950-LABEL: @test_mfma_f32_32x32x16_f16
+// CHECK-GFX950: tail call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.f16(<8 x half> %a, <8 x half> %b, <16 x float> %c, i32 1, i32 2, i32 3)
+v16f test_mfma_f32_32x32x16_f16(v8h a, v8h b, v16f c)
+{
+  return __builtin_amdgcn_mfma_f32_32x32x16_f16(a, b, c, 1, 2, 3);
+}
+
+// CHECK-GFX950-LABEL: @test_mfma_f32_32x32x16_bf16(
+// CHECK-GFX950: tail call <16 x float> @llvm.amdgcn.mfma.f32.32x32x16.bf16(<8 x bfloat> %a, <8 x bfloat> %b, <16 x float> %c, i32 1, i32 2, i32 3)
+v16f test_mfma_f32_32x32x16_bf16(v8bf16 a, v8bf16 b, v16f c) {
+  return __builtin_amdgcn_mfma_f32_32x32x16_bf16(a, b, c, 1, 2, 3);
+}
+
+#endif
diff --git a/clang/test/Driver/amdgpu-macros.cl b/clang/test/Driver/amdgpu-macros.cl
@@ -110,6 +110,7 @@
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx940 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx940 -DFAMILY=GFX9
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx941 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx941 -DFAMILY=GFX9
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx942 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx942 -DFAMILY=GFX9
+// RUN: %clang -E -dM -target amdgcn -mcpu=gfx950 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=64 -DCPU=gfx950 -DFAMILY=GFX9
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1010 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1010 -DFAMILY=GFX10
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1011 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1011 -DFAMILY=GFX10
 // RUN: %clang -E -dM -target amdgcn -mcpu=gfx1012 %s 2>&1 | FileCheck --check-prefixes=ARCH-GCN,FAST_FMAF %s -DWAVEFRONT_SIZE=32 -DCPU=gfx1012 -DFAMILY=GFX10

diff --git a/clang/test/Driver/amdgpu-mcpu.cl b/clang/test/Driver/amdgpu-mcpu.cl
@@ -95,6 +95,7 @@
 // RUN: %clang -### -target amdgcn -mcpu=gfx940 %s 2>&1 | FileCheck --check-prefix=GFX940 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx941 %s 2>&1 | FileCheck --check-prefix=GFX941 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx942 %s 2>&1 | FileCheck --check-prefix=GFX942 %s
+// RUN: %clang -### -target amdgcn -mcpu=gfx950 %s 2>&1 | FileCheck --check-prefix=GFX950 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1010 %s 2>&1 | FileCheck --check-prefix=GFX1010 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1011 %s 2>&1 | FileCheck --check-prefix=GFX1011 %s
 // RUN: %clang -### -target amdgcn -mcpu=gfx1012 %s 2>&1 | FileCheck --check-prefix=GFX1012 %s
@@ -150,6 +151,7 @@
 // GFX940:    "-target-cpu" "gfx940"
 // GFX941:    "-target-cpu" "gfx941"
 // GFX942:    "-target-cpu" "gfx942"
+// GFX950:    "-target-cpu" "gfx950"
 // GFX1010:   "-target-cpu" "gfx1010"
 // GFX1011:   "-target-cpu" "gfx1011"
 // GFX1012:   "-target-cpu" "gfx1012"

diff --git a/clang/test/Driver/module-output.cppm b/clang/test/Driver/module-output.cppm
@@ -42,7 +42,7 @@ export module Hello;
 // CHECK: "-emit-module-interface" {{.*}}"-main-file-name" "Hello.cppm" {{.*}}"-o" "{{.*}}/output/Hello.pcm" "-x" "c++" "{{.*}}/Hello.cppm"
 // CHECK: "-emit-obj" {{.*}}"-main-file-name" "Hello.cppm" {{.*}}"-o" "{{.*}}/output/Hello.o" "-x" "pcm" "{{.*}}/output/Hello.pcm"
 
-// MULTIPLE-ARCH: option '-fmodule-output' can't be used with multiple arch options
+// MULTIPLE-ARCH: option '-fmodule-output' cannot be used with multiple arch options
 
 // CHECK-SPECIFIED: "-emit-module-interface" {{.*}}"-main-file-name" "Hello.cppm" {{.*}}"-o" "{{.*}}/pcm/Hello.pcm" "-x" "c++" "{{.*}}/Hello.cppm"
 // CHECK-SPECIFIED: "-emit-obj" {{.*}}"-main-file-name" "Hello.cppm" {{.*}}"-o" "{{.*}}/Hello.o" "-x" "pcm" "{{.*}}/pcm/Hello.pcm"

diff --git a/clang/test/Driver/relax.s b/clang/test/Driver/relax.s
@@ -8,7 +8,7 @@
 // RUN: llvm-readobj -r %t | FileCheck --check-prefix=REL %s
 
 // REL: R_X86_64_REX_GOTPCRELX foo
-// REL: R_X86_64_REX2_GOTPCRELX foo
+// REL: R_X86_64_CODE_4_GOTPCRELX foo
 
         movq	foo@GOTPCREL(%rip), %rax
         movq	foo@GOTPCREL(%rip), %r16
diff --git a/clang/test/Driver/riscv-cpus.c b/clang/test/Driver/riscv-cpus.c
@@ -104,6 +104,68 @@
 // RUN: %clang --target=riscv32 -### -c %s 2>&1 -mtune=syntacore-scr1-max | FileCheck -check-prefix=MTUNE-SYNTACORE-SCR1-MAX %s
 // MTUNE-SYNTACORE-SCR1-MAX: "-tune-cpu" "syntacore-scr1-max"
 
+// RUN: %clang --target=riscv64 -### -c %s 2>&1 -mtune=tt-ascalon-d8 | FileCheck -check-prefix=MTUNE-TT-ASCALON-D8 %s
+// MTUNE-TT-ASCALON-D8: "-tune-cpu" "tt-ascalon-d8"
+
+// RUN: %clang --target=riscv64 -### -c %s 2>&1 -mcpu=tt-ascalon-d8 | FileCheck -check-prefix=MCPU-TT-ASCALON-D8 %s
+// MCPU-TT-ASCALON-D8: "-target-cpu" "tt-ascalon-d8"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+m"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+a"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+f"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+d"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+c"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+v"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+h"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zicbom"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zicbop"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zicboz"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zicntr"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zicond"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zicsr"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zifencei"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zihintntl"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zihintpause"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zihpm"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zimop"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zmmul"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zawrs"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zfa"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zfbfmin"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zfh"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zfhmin"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zca"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zcb"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zba"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zbb"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zbs"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zkt"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvbb"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvbc"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zve32f"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zve32x"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zve64d"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zve64f"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zve64x"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvfbfmin"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvfbfwma"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvfh"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvfhmin"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvkb"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvkg"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvkn"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvknc"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvkned"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvkng"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvknhb"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvkt"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvl128b"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvl256b"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvl32b"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+zvl64b"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+svinval"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+svnapot"
+// MCPU-TT-ASCALON-D8-SAME: "-target-feature" "+svpbmt"
+
 // RUN: %clang --target=riscv64 -### -c %s 2>&1 -mcpu=veyron-v1 | FileCheck -check-prefix=MCPU-VEYRON-V1 %s
 // MCPU-VEYRON-V1: "-target-cpu" "veyron-v1"
 // MCPU-VEYRON-V1: "-target-feature" "+m"

diff --git a/clang/test/Misc/pragma-attribute-strict-subjects.c b/clang/test/Misc/pragma-attribute-strict-subjects.c
@@ -51,7 +51,7 @@ struct testRecoverStrictnessStruct { };
 #pragma clang attribute pop
 
 #pragma clang attribute push (__attribute__((abi_tag("a"))), apply_to = any(function, record(unless(is_union)), variable, enum))
-// expected-error@-1 {{attribute 'abi_tag' can't be applied to 'enum'}}
+// expected-error@-1 {{attribute 'abi_tag' cannot be applied to 'enum'}}
 
 int testRecoverExtraVar = 0;
 // CHECK-LABEL: VarDecl{{.*}} testRecoverExtraVar
@@ -188,7 +188,7 @@ struct testSubset7Struct { };
 
 
 #pragma clang attribute push (__attribute__((abi_tag("a"))), apply_to = any(record(unless(is_union)), function, variable, enum, enum_constant))
-// expected-error@-1 {{attribute 'abi_tag' can't be applied to 'enum_constant', and 'enum'}}
+// expected-error@-1 {{attribute 'abi_tag' cannot be applied to 'enum_constant', and 'enum'}}
 
 int testSubsetRecoverVar;
 // CHECK-LABEL: VarDecl{{.*}} testSubsetRecoverVar
@@ -205,7 +205,7 @@ struct testSubsetRecoverStruct { };
 #pragma clang attribute pop
 
 #pragma clang attribute push (__attribute__((abi_tag("a"))), apply_to = enum)
-// expected-error@-1 {{attribute 'abi_tag' can't be applied to 'enum'}}
+// expected-error@-1 {{attribute 'abi_tag' cannot be applied to 'enum'}}
 
 int testSubsetNoVar;
 // CHECK-LABEL: VarDecl{{.*}} testSubsetNoVar

diff --git a/clang/test/Misc/target-invalid-cpu-note/amdgcn.c b/clang/test/Misc/target-invalid-cpu-note/amdgcn.c
@@ -48,6 +48,7 @@
 // CHECK-SAME: {{^}}, gfx940
 // CHECK-SAME: {{^}}, gfx941
 // CHECK-SAME: {{^}}, gfx942
+// CHECK-SAME: {{^}}, gfx950
 // CHECK-SAME: {{^}}, gfx1010
 // CHECK-SAME: {{^}}, gfx1011
 // CHECK-SAME: {{^}}, gfx1012

diff --git a/clang/test/Misc/target-invalid-cpu-note/nvptx.c b/clang/test/Misc/target-invalid-cpu-note/nvptx.c
@@ -54,6 +54,7 @@
 // CHECK-SAME: {{^}}, gfx940
 // CHECK-SAME: {{^}}, gfx941
 // CHECK-SAME: {{^}}, gfx942
+// CHECK-SAME: {{^}}, gfx950
 // CHECK-SAME: {{^}}, gfx10-1-generic
 // CHECK-SAME: {{^}}, gfx1010
 // CHECK-SAME: {{^}}, gfx1011

diff --git a/clang/test/Misc/target-invalid-cpu-note/riscv.c b/clang/test/Misc/target-invalid-cpu-note/riscv.c
@@ -41,6 +41,7 @@
 // RISCV64-SAME: {{^}}, syntacore-scr4-rv64
 // RISCV64-SAME: {{^}}, syntacore-scr5-rv64
 // RISCV64-SAME: {{^}}, syntacore-scr7
+// RISCV64-SAME: {{^}}, tt-ascalon-d8
 // RISCV64-SAME: {{^}}, veyron-v1
 // RISCV64-SAME: {{^}}, xiangshan-nanhu
 // RISCV64-SAME: {{$}}
@@ -87,6 +88,7 @@
 // TUNE-RISCV64-SAME: {{^}}, syntacore-scr4-rv64
 // TUNE-RISCV64-SAME: {{^}}, syntacore-scr5-rv64
 // TUNE-RISCV64-SAME: {{^}}, syntacore-scr7
+// TUNE-RISCV64-SAME: {{^}}, tt-ascalon-d8
 // TUNE-RISCV64-SAME: {{^}}, veyron-v1
 // TUNE-RISCV64-SAME: {{^}}, xiangshan-nanhu
 // TUNE-RISCV64-SAME: {{^}}, generic

diff --git a/clang/test/Modules/no-eager-load.cppm b/clang/test/Modules/no-eager-load.cppm
@@ -44,7 +44,7 @@ void use() {
            // expected-note@* {{but in 'a' found a different body}}
 }
 
-// expected-error@a.cppm:* {{declaration 'foo' attached to named module 'a' can't be attached to other modules}}
+// expected-error@a.cppm:* {{declaration 'foo' attached to named module 'a' cannot be attached to other modules}}
 // expected-note@b.cppm:* {{}}
 
 //--- h.cppm
@@ -59,5 +59,5 @@ void use() {
            // expected-note@* {{but in 'a' found a different body}}
 }
 
-// expected-error@a.cppm:* {{declaration 'foo' attached to named module 'a' can't be attached to other modules}}
+// expected-error@a.cppm:* {{declaration 'foo' attached to named module 'a' cannot be attached to other modules}}
 // expected-note@b.cppm:* {{}}
diff --git a/clang/test/Modules/same-decl-in-different-modules.cppm b/clang/test/Modules/same-decl-in-different-modules.cppm
@@ -32,11 +32,11 @@ void test() {
     S<int> s;
 }
 
-// expected-error@mod1.cppm:* {{declaration 'v' attached to named module 'mod1' can't be attached to other modules}}
+// expected-error@mod1.cppm:* {{declaration 'v' attached to named module 'mod1' cannot be attached to other modules}}
 // expected-note@mod2.cppm:* {{}}
-// expected-error@mod1.cppm:* {{declaration 'func' attached to named module 'mod1' can't be attached to other modules}}
+// expected-error@mod1.cppm:* {{declaration 'func' attached to named module 'mod1' cannot be attached to other modules}}
 // expected-note@mod2.cppm:* {{}}
-// expected-error@mod1.cppm:* {{declaration 'A' attached to named module 'mod1' can't be attached to other modules}}
+// expected-error@mod1.cppm:* {{declaration 'A' attached to named module 'mod1' cannot be attached to other modules}}
 // expected-note@mod2.cppm:* {{}}
-// expected-error@mod1.cppm:* 1+{{declaration 'S' attached to named module 'mod1' can't be attached to other modules}}
+// expected-error@mod1.cppm:* 1+{{declaration 'S' attached to named module 'mod1' cannot be attached to other modules}}
 // expected-note@mod2.cppm:* 1+{{}}
diff --git a/clang/test/OpenMP/for_simd_loop_messages.cpp b/clang/test/OpenMP/for_simd_loop_messages.cpp
@@ -731,7 +731,7 @@ void test_ordered() {
   for (int i = 0; i < 16; ++i)
     ;
 #pragma omp parallel
-// expected-error@+1 {{'ordered' clause with a parameter can not be specified in '#pragma omp for simd' directive}}
+// expected-error@+1 {{'ordered' clause with a parameter cannot be specified in '#pragma omp for simd' directive}}
 #pragma omp for simd ordered(1)
   for (int i = 0; i < 16; ++i)
     ;

diff --git a/clang/test/OpenMP/masked_taskloop_simd_linear_messages.cpp b/clang/test/OpenMP/masked_taskloop_simd_linear_messages.cpp
@@ -152,7 +152,7 @@ template<class I, class C> int foomain(I argc, C **argv) {
   #pragma omp masked taskloop simd linear (S1) // expected-error {{'S1' does not refer to a value}}
   for (int k = 0; k < argc; ++k) ++k;
 #if defined(OMP52)
-  // omp52-error@+3{{step simple modifier is exclusive and can't be use with 'val', 'uval' or 'ref' modifier}}
+  // omp52-error@+3{{step simple modifier is exclusive and cannot be use with 'val', 'uval' or 'ref' modifier}}
   // expected-error@+2 {{linear variable with incomplete type 'S1'}}
   // expected-error@+1 {{argument of a linear clause should be of integral or pointer type, not 'S2'}}
   #pragma omp masked taskloop simd linear (a, b: val, B::ib)

diff --git a/clang/test/OpenMP/master_taskloop_simd_linear_messages.cpp b/clang/test/OpenMP/master_taskloop_simd_linear_messages.cpp
@@ -152,7 +152,7 @@ template<class I, class C> int foomain(I argc, C **argv) {
   #pragma omp master taskloop simd linear (S1) // expected-error {{'S1' does not refer to a value}}
   for (int k = 0; k < argc; ++k) ++k;
 #if defined(OMP52)
-  // omp52-error@+3{{step simple modifier is exclusive and can't be use with 'val', 'uval' or 'ref' modifier}}
+  // omp52-error@+3{{step simple modifier is exclusive and cannot be use with 'val', 'uval' or 'ref' modifier}}
   // expected-error@+2 {{linear variable with incomplete type 'S1'}}
   // expected-error@+1 {{argument of a linear clause should be of integral or pointer type, not 'S2'}}
   #pragma omp master taskloop simd linear (a, b: val, B::ib)

diff --git a/clang/test/OpenMP/parallel_for_simd_loop_messages.cpp b/clang/test/OpenMP/parallel_for_simd_loop_messages.cpp
@@ -638,7 +638,7 @@ void test_ordered() {
 #pragma omp parallel for simd ordered
   for (int i = 0; i < 16; ++i)
     ;
-//expected-error@+1 {{'ordered' clause with a parameter can not be specified in '#pragma omp parallel for simd' directive}}
+//expected-error@+1 {{'ordered' clause with a parameter cannot be specified in '#pragma omp parallel for simd' directive}}
 #pragma omp parallel for simd ordered(1)
   for (int i = 0; i < 16; ++i)
     ;

diff --git a/clang/test/OpenMP/parallel_for_simd_messages.cpp b/clang/test/OpenMP/parallel_for_simd_messages.cpp
@@ -94,7 +94,7 @@ void test_ordered() {
 #pragma omp parallel for simd ordered
   for (int i = 0; i < 16; ++i)
     ;
-// expected-error@+1 {{'ordered' clause with a parameter can not be specified in '#pragma omp parallel for simd' directive}}
+// expected-error@+1 {{'ordered' clause with a parameter cannot be specified in '#pragma omp parallel for simd' directive}}
 #pragma omp parallel for simd ordered(1)
   for (int i = 0; i < 16; ++i)
     ;

diff --git a/clang/test/OpenMP/parallel_masked_taskloop_simd_linear_messages.cpp b/clang/test/OpenMP/parallel_masked_taskloop_simd_linear_messages.cpp
@@ -152,7 +152,7 @@ template<class I, class C> int foomain(I argc, C **argv) {
   #pragma omp parallel masked taskloop simd linear (S1) // expected-error {{'S1' does not refer to a value}}
   for (int k = 0; k < argc; ++k) ++k;
 #if defined(OMP52)
-  // omp52-error@+3{{step simple modifier is exclusive and can't be use with 'val', 'uval' or 'ref' modifier}}
+  // omp52-error@+3{{step simple modifier is exclusive and cannot be use with 'val', 'uval' or 'ref' modifier}}
   // expected-error@+2 {{linear variable with incomplete type 'S1'}}
   // expected-error@+1 {{argument of a linear clause should be of integral or pointer type, not 'S2'}}
   #pragma omp parallel masked taskloop simd linear (a, b: val, B::ib)

diff --git a/clang/test/OpenMP/parallel_master_taskloop_simd_linear_messages.cpp b/clang/test/OpenMP/parallel_master_taskloop_simd_linear_messages.cpp
@@ -152,7 +152,7 @@ template<class I, class C> int foomain(I argc, C **argv) {
   #pragma omp parallel master taskloop simd linear (S1) // expected-error {{'S1' does not refer to a value}}
   for (int k = 0; k < argc; ++k) ++k;
 #if defined(OMP52)
-  // omp52-error@+3{{step simple modifier is exclusive and can't be use with 'val', 'uval' or 'ref' modifier}}
+  // omp52-error@+3{{step simple modifier is exclusive and cannot be use with 'val', 'uval' or 'ref' modifier}}
   // expected-error@+2 {{linear variable with incomplete type 'S1'}}
   // expected-error@+1 {{argument of a linear clause should be of integral or pointer type, not 'S2'}}
   #pragma omp parallel master taskloop simd linear (a, b: val, B::ib)

diff --git a/clang/test/OpenMP/simd_linear_messages.cpp b/clang/test/OpenMP/simd_linear_messages.cpp
@@ -142,7 +142,7 @@ template<class I, class C> int foomain(I argc, C **argv) {
   #pragma omp simd linear (S1) // expected-error {{'S1' does not refer to a value}}
   for (int k = 0; k < argc; ++k) ++k;
 #if defined(OMP52)
-  // omp52-error@+3{{step simple modifier is exclusive and can't be use with 'val', 'uval' or 'ref' modifier}}
+  // omp52-error@+3{{step simple modifier is exclusive and cannot be use with 'val', 'uval' or 'ref' modifier}}
   // expected-error@+2 {{linear variable with incomplete type 'S1'}}
   // expected-error@+1 {{argument of a linear clause should be of integral or pointer type, not 'S2'}}
   #pragma omp simd linear (a, b: val, B::ib)