diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
@@ -3663,14 +3663,14 @@ defm rwpi : BoolFOption<"rwpi",
           "Generate read-write position independent code (ARM only)">,
   NegFlag<SetFalse, [], [ClangOption, FlangOption, CC1Option]>>;
 def fplugin_EQ : Joined<["-"], "fplugin=">, Group<f_Group>,
-  Flags<[NoXarchOption]>, MetaVarName<"<dsopath>">,
+  Flags<[NoXarchOption, NoArgumentUnused]>, MetaVarName<"<dsopath>">,
   HelpText<"Load the named plugin (dynamic shared object)">;
 def fplugin_arg : Joined<["-"], "fplugin-arg-">,
-  MetaVarName<"<name>-<arg>">,
+  MetaVarName<"<name>-<arg>">, Flags<[NoArgumentUnused]>,
   HelpText<"Pass <arg> to plugin <name>">;
 def fpass_plugin_EQ : Joined<["-"], "fpass-plugin=">,
   Group<f_Group>, Visibility<[ClangOption, CC1Option, FlangOption, FC1Option]>,
-  MetaVarName<"<dsopath>">,
+  MetaVarName<"<dsopath>">, Flags<[NoArgumentUnused]>,
   HelpText<"Load pass plugin from a dynamic shared object file (only with new pass manager).">,
   MarshallingInfoStringVector<CodeGenOpts<"PassPlugins">>;
 defm tocdata : BoolOption<"m","tocdata",
@@ -4159,6 +4159,11 @@ defm unique_section_names : BoolFOption<"unique-section-names",
   NegFlag<SetFalse, [], [ClangOption, CC1Option],
           "Don't use unique names for text and data sections">,
   PosFlag<SetTrue>>;
+defm separate_named_sections : BoolFOption<"separate-named-sections",
+  CodeGenOpts<"SeparateNamedSections">, DefaultFalse,
+  PosFlag<SetTrue, [], [ClangOption, CC1Option],
+          "Use separate unique sections for named sections (ELF Only)">,
+  NegFlag<SetFalse>>;
 
 defm split_machine_functions: BoolFOption<"split-machine-functions",
   CodeGenOpts<"SplitMachineFunctions">, DefaultFalse,
@@ -4175,6 +4180,14 @@ defm strict_return : BoolFOption<"strict-return",
 
 let Flags = [TargetSpecific] in {
 defm ptrauth_intrinsics : OptInCC1FFlag<"ptrauth-intrinsics", "Enable pointer authentication intrinsics">;
+defm ptrauth_calls : OptInCC1FFlag<"ptrauth-calls", "Enable signing and authentication of all indirect calls">;
+defm ptrauth_returns : OptInCC1FFlag<"ptrauth-returns", "Enable signing and authentication of return addresses">;
+defm ptrauth_auth_traps : OptInCC1FFlag<"ptrauth-auth-traps", "Enable traps on authentication failures">;
+defm ptrauth_vtable_pointer_address_discrimination :
+  OptInCC1FFlag<"ptrauth-vtable-pointer-address-discrimination", "Enable address discrimination of vtable pointers">;
+defm ptrauth_vtable_pointer_type_discrimination :
+  OptInCC1FFlag<"ptrauth-vtable-pointer-type-discrimination", "Enable type discrimination of vtable pointers">;
+defm ptrauth_init_fini : OptInCC1FFlag<"ptrauth-init-fini", "Enable signing of function pointers in init/fini arrays">;
 }
 
 def fenable_matrix : Flag<["-"], "fenable-matrix">, Group<f_Group>,
@@ -5072,6 +5085,10 @@ def maix_small_local_dynamic_tls : Flag<["-"], "maix-small-local-dynamic-tls">,
            "where the offset from the TLS base is encoded as an "
            "immediate operand (AIX 64-bit only). "
            "This access sequence is not used for variables larger than 32KB.">;
+def maix_shared_lib_tls_model_opt : Flag<["-"], "maix-shared-lib-tls-model-opt">,
+  Group<m_ppc_Features_Group>,
+  HelpText<"For shared library loaded with the main program, change local-dynamic access(es) "
+           "to initial-exec access(es) at the function level (AIX 64-bit only).">;
 def maix_struct_return : Flag<["-"], "maix-struct-return">,
   Group<m_Group>, Visibility<[ClangOption, CC1Option]>,
   HelpText<"Return all structs in memory (PPC32 only)">,
@@ -5698,7 +5715,7 @@ def whatsloaded : Flag<["-"], "whatsloaded">;
 def why_load : Flag<["-"], "why_load">;
 def whyload : Flag<["-"], "whyload">, Alias<why_load>;
 def w : Flag<["-"], "w">, HelpText<"Suppress all warnings">,
-  Visibility<[ClangOption, CC1Option]>,
+  Visibility<[ClangOption, CC1Option, FlangOption, FC1Option]>,
   MarshallingInfoFlag<DiagnosticOpts<"IgnoreWarnings">>;
 def x : JoinedOrSeparate<["-"], "x">,
 Flags<[NoXarchOption]>,
@@ -7087,6 +7104,11 @@ def mlink_bitcode_file : Separate<["-"], "mlink-bitcode-file">,
 def mlink_builtin_bitcode : Separate<["-"], "mlink-builtin-bitcode">,
   HelpText<"Link and internalize needed symbols from the given bitcode file "
            "before performing optimizations.">;
+defm link_builtin_bitcode_postopt: BoolMOption<"link-builtin-bitcode-postopt",
+  CodeGenOpts<"LinkBitcodePostopt">, DefaultFalse,
+  PosFlag<SetTrue, [], [ClangOption], "Link builtin bitcodes after the "
+  "optimization pipeline">,
+  NegFlag<SetFalse, [], [ClangOption]>>;
 def vectorize_loops : Flag<["-"], "vectorize-loops">,
   HelpText<"Run the Loop vectorization passes">,
   MarshallingInfoFlag<CodeGenOpts<"VectorizeLoop">>;

diff --git a/clang/include/clang/Format/Format.h b/clang/include/clang/Format/Format.h
@@ -375,8 +375,25 @@ struct FormatStyle {
     ///   }
     /// \endcode
     bool AcrossComments;
-    /// Whether aligned case labels are aligned on the colon, or on the
-    /// , or on the tokens after the colon.
+    /// Whether to align the case arrows when aligning short case expressions.
+    /// \code{.java}
+    ///   true:
+    ///   i = switch (day) {
+    ///     case THURSDAY, SATURDAY -> 8;
+    ///     case WEDNESDAY          -> 9;
+    ///     default                 -> 0;
+    ///   };
+    ///
+    ///   false:
+    ///   i = switch (day) {
+    ///     case THURSDAY, SATURDAY -> 8;
+    ///     case WEDNESDAY ->          9;
+    ///     default ->                 0;
+    ///   };
+    /// \endcode
+    bool AlignCaseArrows;
+    /// Whether aligned case labels are aligned on the colon, or on the tokens
+    /// after the colon.
     /// \code
     ///   true:
     ///   switch (level) {
@@ -396,12 +413,14 @@ struct FormatStyle {
     bool operator==(const ShortCaseStatementsAlignmentStyle &R) const {
       return Enabled == R.Enabled && AcrossEmptyLines == R.AcrossEmptyLines &&
              AcrossComments == R.AcrossComments &&
+             AlignCaseArrows == R.AlignCaseArrows &&
              AlignCaseColons == R.AlignCaseColons;
     }
   };
 
   /// Style of aligning consecutive short case labels.
-  /// Only applies if ``AllowShortCaseLabelsOnASingleLine`` is ``true``.
+  /// Only applies if ``AllowShortCaseExpressionOnASingleLine`` or
+  /// ``AllowShortCaseLabelsOnASingleLine`` is ``true``.
   ///
   /// \code{.yaml}
   ///   # Example of usage:
@@ -724,6 +743,19 @@ struct FormatStyle {
   /// \version 3.5
   ShortBlockStyle AllowShortBlocksOnASingleLine;
 
+  /// Whether to merge a short switch labeled rule into a single line.
+  /// \code{.java}
+  ///   true:                               false:
+  ///   switch (a) {           vs.          switch (a) {
+  ///   case 1 -> 1;                        case 1 ->
+  ///   default -> 0;                         1;
+  ///   };                                  default ->
+  ///                                         0;
+  ///                                       };
+  /// \endcode
+  /// \version 19
+  bool AllowShortCaseExpressionOnASingleLine;
+
   /// If ``true``, short case labels will be contracted to a single line.
   /// \code
   ///   true:                                   false:
@@ -4923,6 +4955,8 @@ struct FormatStyle {
            AllowBreakBeforeNoexceptSpecifier ==
                R.AllowBreakBeforeNoexceptSpecifier &&
            AllowShortBlocksOnASingleLine == R.AllowShortBlocksOnASingleLine &&
+           AllowShortCaseExpressionOnASingleLine ==
+               R.AllowShortCaseExpressionOnASingleLine &&
            AllowShortCaseLabelsOnASingleLine ==
                R.AllowShortCaseLabelsOnASingleLine &&
            AllowShortCompoundRequirementOnASingleLine ==

diff --git a/clang/include/clang/Parse/Parser.h b/clang/include/clang/Parse/Parser.h
@@ -3632,6 +3632,13 @@ class Parser : public CodeCompletionHandler {
     // Wait constructs, we likely want to put that information in here as well.
   };
 
+  struct OpenACCWaitParseInfo {
+    bool Failed = false;
+    Expr *DevNumExpr = nullptr;
+    SourceLocation QueuesLoc;
+    SmallVector<Expr *> QueueIdExprs;
+  };
+
   /// Represents the 'error' state of parsing an OpenACC Clause, and stores
   /// whether we can continue parsing, or should give up on the directive.
   enum class OpenACCParseCanContinue { Cannot = 0, Can = 1 };
@@ -3674,7 +3681,8 @@ class Parser : public CodeCompletionHandler {
   /// Parses the clause-list for an OpenACC directive.
   SmallVector<OpenACCClause *>
   ParseOpenACCClauseList(OpenACCDirectiveKind DirKind);
-  bool ParseOpenACCWaitArgument(SourceLocation Loc, bool IsDirective);
+  OpenACCWaitParseInfo ParseOpenACCWaitArgument(SourceLocation Loc,
+                                                bool IsDirective);
   /// Parses the clause of the 'bind' argument, which can be a string literal or
   /// an ID expression.
   ExprResult ParseOpenACCBindClauseArgument();
@@ -3698,7 +3706,9 @@ class Parser : public CodeCompletionHandler {
   bool ParseOpenACCDeviceTypeList();
   /// Parses the 'async-argument', which is an integral value with two
   /// 'special' values that are likely negative (but come from Macros).
-  ExprResult ParseOpenACCAsyncArgument();
+  OpenACCIntExprParseResult ParseOpenACCAsyncArgument(OpenACCDirectiveKind DK,
+                                                      OpenACCClauseKind CK,
+                                                      SourceLocation Loc);
   /// Parses the 'size-expr', which is an integral value, or an asterisk.
   bool ParseOpenACCSizeExpr();
   /// Parses a comma delimited list of 'size-expr's.

diff --git a/clang/include/clang/Sema/Sema.h b/clang/include/clang/Sema/Sema.h
@@ -9739,6 +9739,9 @@ class Sema final : public SemaBase {
                      const PartialDiagnostic &CandidateDiag,
                      bool Complain = true, QualType TargetType = QualType());
 
+  FunctionDecl *getMoreConstrainedFunction(FunctionDecl *FD1,
+                                           FunctionDecl *FD2);
+
   ///@}
 
   //
@@ -10199,7 +10202,9 @@ class Sema final : public SemaBase {
         S.ExprEvalContexts.back().InImmediateFunctionContext =
             FD->isImmediateFunction() ||
             S.ExprEvalContexts[S.ExprEvalContexts.size() - 2]
-                .isConstantEvaluated();
+                .isConstantEvaluated() ||
+            S.ExprEvalContexts[S.ExprEvalContexts.size() - 2]
+                .isImmediateFunctionContext();
         S.ExprEvalContexts.back().InImmediateEscalatingFunctionContext =
             S.getLangOpts().CPlusPlus20 && FD->isImmediateEscalating();
       } else

diff --git a/clang/include/clang/Sema/SemaOpenACC.h b/clang/include/clang/Sema/SemaOpenACC.h
@@ -54,8 +54,14 @@ class SemaOpenACC : public SemaBase {
       bool IsZero;
     };
 
+    struct WaitDetails {
+      Expr *DevNumExpr;
+      SourceLocation QueuesLoc;
+      SmallVector<Expr *> QueueIdExprs;
+    };
+
     std::variant<std::monostate, DefaultDetails, ConditionDetails,
-                 IntExprDetails, VarListDetails>
+                 IntExprDetails, VarListDetails, WaitDetails>
         Details = std::monostate{};
 
   public:
@@ -101,16 +107,55 @@ class SemaOpenACC : public SemaBase {
     unsigned getNumIntExprs() const {
       assert((ClauseKind == OpenACCClauseKind::NumGangs ||
               ClauseKind == OpenACCClauseKind::NumWorkers ||
+              ClauseKind == OpenACCClauseKind::Async ||
               ClauseKind == OpenACCClauseKind::VectorLength) &&
              "Parsed clause kind does not have a int exprs");
+
+      // 'async' and 'wait' have an optional IntExpr, so be tolerant of that.
+      if ((ClauseKind == OpenACCClauseKind::Async ||
+           ClauseKind == OpenACCClauseKind::Wait) &&
+          std::holds_alternative<std::monostate>(Details))
+        return 0;
       return std::get<IntExprDetails>(Details).IntExprs.size();
     }
 
+    SourceLocation getQueuesLoc() const {
+      assert(ClauseKind == OpenACCClauseKind::Wait &&
+             "Parsed clause kind does not have a queues location");
+
+      if (std::holds_alternative<std::monostate>(Details))
+        return SourceLocation{};
+
+      return std::get<WaitDetails>(Details).QueuesLoc;
+    }
+
+    Expr *getDevNumExpr() const {
+      assert(ClauseKind == OpenACCClauseKind::Wait &&
+             "Parsed clause kind does not have a device number expr");
+
+      if (std::holds_alternative<std::monostate>(Details))
+        return nullptr;
+
+      return std::get<WaitDetails>(Details).DevNumExpr;
+    }
+
+    ArrayRef<Expr *> getQueueIdExprs() const {
+      assert(ClauseKind == OpenACCClauseKind::Wait &&
+             "Parsed clause kind does not have a queue id expr list");
+
+      if (std::holds_alternative<std::monostate>(Details))
+        return ArrayRef<Expr *>{std::nullopt};
+
+      return std::get<WaitDetails>(Details).QueueIdExprs;
+    }
+
     ArrayRef<Expr *> getIntExprs() {
       assert((ClauseKind == OpenACCClauseKind::NumGangs ||
               ClauseKind == OpenACCClauseKind::NumWorkers ||
+              ClauseKind == OpenACCClauseKind::Async ||
               ClauseKind == OpenACCClauseKind::VectorLength) &&
              "Parsed clause kind does not have a int exprs");
+
       return std::get<IntExprDetails>(Details).IntExprs;
     }
 
@@ -134,6 +179,8 @@ class SemaOpenACC : public SemaBase {
               ClauseKind == OpenACCClauseKind::Create ||
               ClauseKind == OpenACCClauseKind::PCreate ||
               ClauseKind == OpenACCClauseKind::PresentOrCreate ||
+              ClauseKind == OpenACCClauseKind::Attach ||
+              ClauseKind == OpenACCClauseKind::DevicePtr ||
               ClauseKind == OpenACCClauseKind::FirstPrivate) &&
              "Parsed clause kind does not have a var-list");
       return std::get<VarListDetails>(Details).VarList;
@@ -188,13 +235,15 @@ class SemaOpenACC : public SemaBase {
     void setIntExprDetails(ArrayRef<Expr *> IntExprs) {
       assert((ClauseKind == OpenACCClauseKind::NumGangs ||
               ClauseKind == OpenACCClauseKind::NumWorkers ||
+              ClauseKind == OpenACCClauseKind::Async ||
               ClauseKind == OpenACCClauseKind::VectorLength) &&
              "Parsed clause kind does not have a int exprs");
       Details = IntExprDetails{{IntExprs.begin(), IntExprs.end()}};
     }
     void setIntExprDetails(llvm::SmallVector<Expr *> &&IntExprs) {
       assert((ClauseKind == OpenACCClauseKind::NumGangs ||
               ClauseKind == OpenACCClauseKind::NumWorkers ||
+              ClauseKind == OpenACCClauseKind::Async ||
               ClauseKind == OpenACCClauseKind::VectorLength) &&
              "Parsed clause kind does not have a int exprs");
       Details = IntExprDetails{std::move(IntExprs)};
@@ -217,6 +266,8 @@ class SemaOpenACC : public SemaBase {
               ClauseKind == OpenACCClauseKind::Create ||
               ClauseKind == OpenACCClauseKind::PCreate ||
               ClauseKind == OpenACCClauseKind::PresentOrCreate ||
+              ClauseKind == OpenACCClauseKind::Attach ||
+              ClauseKind == OpenACCClauseKind::DevicePtr ||
               ClauseKind == OpenACCClauseKind::FirstPrivate) &&
              "Parsed clause kind does not have a var-list");
       assert((!IsReadOnly || ClauseKind == OpenACCClauseKind::CopyIn ||
@@ -251,6 +302,8 @@ class SemaOpenACC : public SemaBase {
               ClauseKind == OpenACCClauseKind::Create ||
               ClauseKind == OpenACCClauseKind::PCreate ||
               ClauseKind == OpenACCClauseKind::PresentOrCreate ||
+              ClauseKind == OpenACCClauseKind::Attach ||
+              ClauseKind == OpenACCClauseKind::DevicePtr ||
               ClauseKind == OpenACCClauseKind::FirstPrivate) &&
              "Parsed clause kind does not have a var-list");
       assert((!IsReadOnly || ClauseKind == OpenACCClauseKind::CopyIn ||
@@ -266,6 +319,13 @@ class SemaOpenACC : public SemaBase {
              "zero: tag only valid on copyout/create");
       Details = VarListDetails{std::move(VarList), IsReadOnly, IsZero};
     }
+
+    void setWaitDetails(Expr *DevNum, SourceLocation QueuesLoc,
+                        llvm::SmallVector<Expr *> &&IntExprs) {
+      assert(ClauseKind == OpenACCClauseKind::Wait &&
+             "Parsed clause kind does not have a wait-details");
+      Details = WaitDetails{DevNum, QueuesLoc, std::move(IntExprs)};
+    }
   };
 
   SemaOpenACC(Sema &S);
@@ -315,6 +375,10 @@ class SemaOpenACC : public SemaBase {
   /// declaration reference to a variable of the correct type.
   ExprResult ActOnVar(Expr *VarExpr);
 
+  /// Called to check the 'var' type is a variable of pointer type, necessary
+  /// for 'deviceptr' and 'attach' clauses. Returns true on success.
+  bool CheckVarIsPointerType(OpenACCClauseKind ClauseKind, Expr *VarExpr);
+
   /// Checks and creates an Array Section used in an OpenACC construct/clause.
   ExprResult ActOnArraySectionExpr(Expr *Base, SourceLocation LBLoc,
                                    Expr *LowerBound,

diff --git a/clang/include/clang/Serialization/ASTBitCodes.h b/clang/include/clang/Serialization/ASTBitCodes.h
@@ -23,6 +23,7 @@
 #include "clang/Basic/IdentifierTable.h"
 #include "clang/Basic/OperatorKinds.h"
 #include "clang/Basic/SourceLocation.h"
+#include "clang/Serialization/SourceLocationEncoding.h"
 #include "llvm/ADT/DenseMapInfo.h"
 #include "llvm/Bitstream/BitCodes.h"
 #include <cassert>
@@ -165,99 +166,95 @@ using SubmoduleID = uint32_t;
 /// The number of predefined submodule IDs.
 const unsigned int NUM_PREDEF_SUBMODULE_IDS = 1;
 
+/// 32 aligned uint64_t in the AST file. Use splitted 64-bit integer into
+/// low/high parts to keep structure alignment 32-bit (it is important
+/// because blobs in bitstream are 32-bit aligned). This structure is
+/// serialized "as is" to the AST file.
+class UnalignedUInt64 {
+  uint32_t BitLow = 0;
+  uint32_t BitHigh = 0;
+
+public:
+  UnalignedUInt64() = default;
+  UnalignedUInt64(uint64_t BitOffset) { set(BitOffset); }
+
+  void set(uint64_t Offset) {
+    BitLow = Offset;
+    BitHigh = Offset >> 32;
+  }
+
+  uint64_t get() const { return BitLow | (uint64_t(BitHigh) << 32); }
+};
+
 /// Source range/offset of a preprocessed entity.
-struct PPEntityOffset {
+class PPEntityOffset {
+  using RawLocEncoding = SourceLocationEncoding::RawLocEncoding;
+
   /// Raw source location of beginning of range.
-  SourceLocation::UIntTy Begin;
+  UnalignedUInt64 Begin;
 
   /// Raw source location of end of range.
-  SourceLocation::UIntTy End;
+  UnalignedUInt64 End;
 
   /// Offset in the AST file relative to ModuleFile::MacroOffsetsBase.
   uint32_t BitOffset;
 
-  PPEntityOffset(SourceRange R, uint32_t BitOffset)
-      : Begin(R.getBegin().getRawEncoding()), End(R.getEnd().getRawEncoding()),
-        BitOffset(BitOffset) {}
+public:
+  PPEntityOffset(RawLocEncoding Begin, RawLocEncoding End, uint32_t BitOffset)
+      : Begin(Begin), End(End), BitOffset(BitOffset) {}
 
-  SourceLocation getBegin() const {
-    return SourceLocation::getFromRawEncoding(Begin);
-  }
+  RawLocEncoding getBegin() const { return Begin.get(); }
+  RawLocEncoding getEnd() const { return End.get(); }
 
-  SourceLocation getEnd() const {
-    return SourceLocation::getFromRawEncoding(End);
-  }
+  uint32_t getOffset() const { return BitOffset; }
 };
 
 /// Source range of a skipped preprocessor region
-struct PPSkippedRange {
+class PPSkippedRange {
+  using RawLocEncoding = SourceLocationEncoding::RawLocEncoding;
+
   /// Raw source location of beginning of range.
-  SourceLocation::UIntTy Begin;
+  UnalignedUInt64 Begin;
   /// Raw source location of end of range.
-  SourceLocation::UIntTy End;
+  UnalignedUInt64 End;
 
-  PPSkippedRange(SourceRange R)
-      : Begin(R.getBegin().getRawEncoding()), End(R.getEnd().getRawEncoding()) {
-  }
+public:
+  PPSkippedRange(RawLocEncoding Begin, RawLocEncoding End)
+      : Begin(Begin), End(End) {}
 
-  SourceLocation getBegin() const {
-    return SourceLocation::getFromRawEncoding(Begin);
-  }
-  SourceLocation getEnd() const {
-    return SourceLocation::getFromRawEncoding(End);
-  }
+  RawLocEncoding getBegin() const { return Begin.get(); }
+  RawLocEncoding getEnd() const { return End.get(); }
 };
 
-/// Offset in the AST file. Use splitted 64-bit integer into low/high
-/// parts to keep structure alignment 32-bit (it is important because
-/// blobs in bitstream are 32-bit aligned). This structure is serialized
-/// "as is" to the AST file.
-struct UnderalignedInt64 {
-  uint32_t BitOffsetLow = 0;
-  uint32_t BitOffsetHigh = 0;
-
-  UnderalignedInt64() = default;
-  UnderalignedInt64(uint64_t BitOffset) { setBitOffset(BitOffset); }
-
-  void setBitOffset(uint64_t Offset) {
-    BitOffsetLow = Offset;
-    BitOffsetHigh = Offset >> 32;
-  }
-
-  uint64_t getBitOffset() const {
-    return BitOffsetLow | (uint64_t(BitOffsetHigh) << 32);
-  }
-};
+/// Source location and bit offset of a declaration. Keep
+/// structure alignment 32-bit since the blob is assumed as 32-bit aligned.
+class DeclOffset {
+  using RawLocEncoding = SourceLocationEncoding::RawLocEncoding;
 
-/// Source location and bit offset of a declaration.
-struct DeclOffset {
   /// Raw source location.
-  SourceLocation::UIntTy Loc = 0;
+  UnalignedUInt64 RawLoc;
 
-  /// Offset relative to the start of the DECLTYPES_BLOCK block. Keep
-  /// structure alignment 32-bit and avoid padding gap because undefined
-  /// value in the padding affects AST hash.
-  UnderalignedInt64 BitOffset;
+  /// Offset relative to the start of the DECLTYPES_BLOCK block.
+  UnalignedUInt64 BitOffset;
 
+public:
   DeclOffset() = default;
-  DeclOffset(SourceLocation Loc, uint64_t BitOffset,
-             uint64_t DeclTypesBlockStartOffset) {
-    setLocation(Loc);
+  DeclOffset(RawLocEncoding RawLoc, uint64_t BitOffset,
+             uint64_t DeclTypesBlockStartOffset)
+      : RawLoc(RawLoc) {
     setBitOffset(BitOffset, DeclTypesBlockStartOffset);
   }
 
-  void setLocation(SourceLocation L) { Loc = L.getRawEncoding(); }
+  void setRawLoc(RawLocEncoding Loc) { RawLoc = Loc; }
 
-  SourceLocation getLocation() const {
-    return SourceLocation::getFromRawEncoding(Loc);
-  }
+  RawLocEncoding getRawLoc() const { return RawLoc.get(); }
 
   void setBitOffset(uint64_t Offset, const uint64_t DeclTypesBlockStartOffset) {
-    BitOffset.setBitOffset(Offset - DeclTypesBlockStartOffset);
+    BitOffset.set(Offset - DeclTypesBlockStartOffset);
   }
 
   uint64_t getBitOffset(const uint64_t DeclTypesBlockStartOffset) const {
-    return BitOffset.getBitOffset() + DeclTypesBlockStartOffset;
+    return BitOffset.get() + DeclTypesBlockStartOffset;
   }
 };
 
@@ -1091,6 +1088,9 @@ enum PredefinedTypeIDs {
 // \brief WebAssembly reference types with auto numeration
 #define WASM_TYPE(Name, Id, SingletonId) PREDEF_TYPE_##Id##_ID,
 #include "clang/Basic/WebAssemblyReferenceTypes.def"
+
+  /// The placeholder type for unresolved templates.
+  PREDEF_TYPE_UNRESOLVED_TEMPLATE,
   // Sentinel value. Considered a predefined type but not useable as one.
   PREDEF_TYPE_LAST_ID
 };
@@ -1100,7 +1100,7 @@ enum PredefinedTypeIDs {
 ///
 /// Type IDs for non-predefined types will start at
 /// NUM_PREDEF_TYPE_IDs.
-const unsigned NUM_PREDEF_TYPE_IDS = 502;
+const unsigned NUM_PREDEF_TYPE_IDS = 503;
 
 // Ensure we do not overrun the predefined types we reserved
 // in the enum PredefinedTypeIDs above.

diff --git a/clang/include/clang/Serialization/ASTReader.h b/clang/include/clang/Serialization/ASTReader.h
@@ -1771,6 +1771,7 @@ class ASTReader
 
   /// Retrieve the module manager.
   ModuleManager &getModuleManager() { return ModuleMgr; }
+  const ModuleManager &getModuleManager() const { return ModuleMgr; }
 
   /// Retrieve the preprocessor.
   Preprocessor &getPreprocessor() const { return PP; }
@@ -2177,8 +2178,8 @@ class ASTReader
 
   /// Retrieve the global submodule ID given a module and its local ID
   /// number.
-  serialization::SubmoduleID
-  getGlobalSubmoduleID(ModuleFile &M, unsigned LocalID);
+  serialization::SubmoduleID getGlobalSubmoduleID(ModuleFile &M,
+                                                  unsigned LocalID) const;
 
   /// Retrieve the submodule that corresponds to a global submodule ID.
   ///
@@ -2191,7 +2192,7 @@ class ASTReader
 
   /// Retrieve the module file with a given local ID within the specified
   /// ModuleFile.
-  ModuleFile *getLocalModuleFile(ModuleFile &M, unsigned ID);
+  ModuleFile *getLocalModuleFile(ModuleFile &M, unsigned ID) const;
 
   /// Get an ID for the given module file.
   unsigned getModuleFileID(ModuleFile *M);
@@ -2227,33 +2228,46 @@ class ASTReader
     return Sema::AlignPackInfo::getFromRawEncoding(Raw);
   }
 
+  using RawLocEncoding = SourceLocationEncoding::RawLocEncoding;
+
   /// Read a source location from raw form and return it in its
   /// originating module file's source location space.
-  SourceLocation ReadUntranslatedSourceLocation(SourceLocation::UIntTy Raw,
-                                                LocSeq *Seq = nullptr) const {
+  std::pair<SourceLocation, unsigned>
+  ReadUntranslatedSourceLocation(RawLocEncoding Raw,
+                                 LocSeq *Seq = nullptr) const {
     return SourceLocationEncoding::decode(Raw, Seq);
   }
 
   /// Read a source location from raw form.
-  SourceLocation ReadSourceLocation(ModuleFile &ModuleFile,
-                                    SourceLocation::UIntTy Raw,
+  SourceLocation ReadSourceLocation(ModuleFile &MF, RawLocEncoding Raw,
                                     LocSeq *Seq = nullptr) const {
-    SourceLocation Loc = ReadUntranslatedSourceLocation(Raw, Seq);
-    return TranslateSourceLocation(ModuleFile, Loc);
+    if (!MF.ModuleOffsetMap.empty())
+      ReadModuleOffsetMap(MF);
+
+    auto [Loc, ModuleFileIndex] = ReadUntranslatedSourceLocation(Raw, Seq);
+    ModuleFile *OwningModuleFile =
+        ModuleFileIndex == 0 ? &MF : MF.DependentModules[ModuleFileIndex - 1];
+
+    assert(!SourceMgr.isLoadedSourceLocation(Loc) &&
+           "Run out source location space");
+
+    return TranslateSourceLocation(*OwningModuleFile, Loc);
   }
 
   /// Translate a source location from another module file's source
   /// location space into ours.
   SourceLocation TranslateSourceLocation(ModuleFile &ModuleFile,
                                          SourceLocation Loc) const {
-    if (!ModuleFile.ModuleOffsetMap.empty())
-      ReadModuleOffsetMap(ModuleFile);
-    assert(ModuleFile.SLocRemap.find(Loc.getOffset()) !=
-               ModuleFile.SLocRemap.end() &&
-           "Cannot find offset to remap.");
-    SourceLocation::IntTy Remap =
-        ModuleFile.SLocRemap.find(Loc.getOffset())->second;
-    return Loc.getLocWithOffset(Remap);
+    if (Loc.isInvalid())
+      return Loc;
+
+    // FIXME: TranslateSourceLocation is not re-enterable. It is problematic
+    // to call TranslateSourceLocation on a translated source location.
+    // We either need a method to know whether or not a source location is
+    // translated or refactor the code to make it clear that
+    // TranslateSourceLocation won't be called with translated source location.
+
+    return Loc.getLocWithOffset(ModuleFile.SLocEntryBaseOffset - 2);
   }
 
   /// Read a source location.

diff --git a/clang/include/clang/Serialization/ASTRecordReader.h b/clang/include/clang/Serialization/ASTRecordReader.h
@@ -272,6 +272,9 @@ class ASTRecordReader
   /// Read a list of Exprs used for a var-list.
   llvm::SmallVector<Expr *> readOpenACCVarList();
 
+  /// Read a list of Exprs used for a int-expr-list.
+  llvm::SmallVector<Expr *> readOpenACCIntExprList();
+
   /// Read an OpenACC clause, advancing Idx.
   OpenACCClause *readOpenACCClause();
 

diff --git a/clang/include/clang/Serialization/ASTRecordWriter.h b/clang/include/clang/Serialization/ASTRecordWriter.h
@@ -296,6 +296,8 @@ class ASTRecordWriter
 
   void writeOpenACCVarList(const OpenACCClauseWithVarList *C);
 
+  void writeOpenACCIntExprList(ArrayRef<Expr *> Exprs);
+
   /// Writes out a single OpenACC Clause.
   void writeOpenACCClause(const OpenACCClause *C);
 

diff --git a/clang/include/clang/Serialization/ASTWriter.h b/clang/include/clang/Serialization/ASTWriter.h
@@ -274,7 +274,7 @@ class ASTWriter : public ASTDeserializationListener,
 
   /// Offset of each type in the bitstream, indexed by
   /// the type's ID.
-  std::vector<serialization::UnderalignedInt64> TypeOffsets;
+  std::vector<serialization::UnalignedUInt64> TypeOffsets;
 
   /// The first ID number we can use for our own identifiers.
   serialization::IdentID FirstIdentID = serialization::NUM_PREDEF_IDENT_IDS;
@@ -357,6 +357,13 @@ class ASTWriter : public ASTDeserializationListener,
   /// contexts.
   llvm::DenseMap<const Decl *, unsigned> AnonymousDeclarationNumbers;
 
+  /// The external top level module during the writing process. Used to
+  /// generate signature for the module file being written.
+  ///
+  /// Only meaningful for standard C++ named modules. See the comments in
+  /// createSignatureForNamedModule() for details.
+  llvm::DenseSet<Module *> TouchedTopLevelModules;
+
   /// An update to a Decl.
   class DeclUpdate {
     /// A DeclUpdateKind.
@@ -676,6 +683,10 @@ class ASTWriter : public ASTDeserializationListener,
   void AddSourceLocation(SourceLocation Loc, RecordDataImpl &Record,
                          LocSeq *Seq = nullptr);
 
+  /// Return the raw encodings for source locations.
+  SourceLocationEncoding::RawLocEncoding
+  getRawSourceLocationEncoding(SourceLocation Loc, LocSeq *Seq = nullptr);
+
   /// Emit a source range.
   void AddSourceRange(SourceRange Range, RecordDataImpl &Record,
                       LocSeq *Seq = nullptr);

diff --git a/clang/include/clang/Serialization/ModuleFile.h b/clang/include/clang/Serialization/ModuleFile.h
@@ -295,10 +295,6 @@ class ModuleFile {
   /// AST file.
   const uint32_t *SLocEntryOffsets = nullptr;
 
-  /// Remapping table for source locations in this module.
-  ContinuousRangeMap<SourceLocation::UIntTy, SourceLocation::IntTy, 2>
-      SLocRemap;
-
   // === Identifiers ===
 
   /// The number of identifiers in this AST file.
@@ -495,7 +491,7 @@ class ModuleFile {
 
   /// Offset of each type within the bitstream, indexed by the
   /// type ID, or the representation of a Type*.
-  const UnderalignedInt64 *TypeOffsets = nullptr;
+  const UnalignedUInt64 *TypeOffsets = nullptr;
 
   /// Base type ID for types local to this module as represented in
   /// the global type ID space.
@@ -512,9 +508,17 @@ class ModuleFile {
   /// List of modules which depend on this module
   llvm::SetVector<ModuleFile *> ImportedBy;
 
-  /// List of modules which this module depends on
+  /// List of modules which this module directly imported
   llvm::SetVector<ModuleFile *> Imports;
 
+  /// List of modules which this modules dependent on. Different
+  /// from `Imports`, this includes indirectly imported modules too.
+  /// The order of DependentModules is significant. It should keep
+  /// the same order with that module file manager when we write
+  /// the current module file. The value of the member will be initialized
+  /// in `ASTReader::ReadModuleOffsetMap`.
+  llvm::SmallVector<ModuleFile *, 16> DependentModules;
+
   /// Determine whether this module was directly imported at
   /// any point during translation.
   bool isDirectlyImported() const { return DirectlyImported; }

diff --git a/clang/include/clang/Serialization/SourceLocationEncoding.h b/clang/include/clang/Serialization/SourceLocationEncoding.h
@@ -6,28 +6,33 @@
 //
 //===----------------------------------------------------------------------===//
 //
-// Source locations are stored pervasively in the AST, making up a third of
-// the size of typical serialized files. Storing them efficiently is important.
+// We wish to encode the SourceLocation from other module file not dependent
+// on the other module file. So that the source location changes from other
+// module file may not affect the contents of the current module file. Then the
+// users don't need to recompile the whole project due to a new line in a module
+// unit in the root of the dependency graph.
 //
-// We use integers optimized by VBR-encoding, because:
-//  - when abbreviations cannot be used, VBR6 encoding is our only choice
-//  - in the worst case a SourceLocation can be ~any 32-bit number, but in
-//    practice they are highly predictable
+// To achieve this, we need to encode the index of the module file into the
+// encoding of the source location. The encoding of the source location may be:
 //
-// We encode the integer so that likely values encode as small numbers that
-// turn into few VBR chunks:
-//  - the invalid sentinel location is a very common value: it encodes as 0
-//  - the "macro or not" bit is stored at the bottom of the integer
-//    (rather than at the top, as in memory), so macro locations can have
-//    small representations.
-//  - related locations (e.g. of a left and right paren pair) are usually
-//    similar, so when encoding a sequence of locations we store only
-//    differences between successive elements.
+//      |-----------------------|-----------------------|
+//      |          A            |         B         | C |
+//
+//  * A: 32 bit. The index of the module file in the module manager + 1. The +1
+//  here is necessary since we wish 0 stands for the current module file.
+//  * B: 31 bit. The offset of the source location to the module file containing
+//  it.
+//  * C: The macro bit. We rotate it to the lowest bit so that we can save some
+//  space in case the index of the module file is 0.
+//
+// Specially, if the index of the module file is 0, we allow to encode a
+// sequence of locations we store only differences between successive elements.
 //
 //===----------------------------------------------------------------------===//
 
-#include <climits>
 #include "clang/Basic/SourceLocation.h"
+#include "llvm/Support/MathExtras.h"
+#include <climits>
 
 #ifndef LLVM_CLANG_SERIALIZATION_SOURCELOCATIONENCODING_H
 #define LLVM_CLANG_SERIALIZATION_SOURCELOCATIONENCODING_H
@@ -52,9 +57,13 @@ class SourceLocationEncoding {
   friend SourceLocationSequence;
 
 public:
-  static uint64_t encode(SourceLocation Loc,
-                         SourceLocationSequence * = nullptr);
-  static SourceLocation decode(uint64_t, SourceLocationSequence * = nullptr);
+  using RawLocEncoding = uint64_t;
+
+  static RawLocEncoding encode(SourceLocation Loc, UIntTy BaseOffset,
+                               unsigned BaseModuleFileIndex,
+                               SourceLocationSequence * = nullptr);
+  static std::pair<SourceLocation, unsigned>
+  decode(RawLocEncoding, SourceLocationSequence * = nullptr);
 };
 
 /// Serialized encoding of a sequence of SourceLocations.
@@ -149,14 +158,44 @@ class SourceLocationSequence::State {
   operator SourceLocationSequence *() { return &Seq; }
 };
 
-inline uint64_t SourceLocationEncoding::encode(SourceLocation Loc,
-                                               SourceLocationSequence *Seq) {
-  return Seq ? Seq->encode(Loc) : encodeRaw(Loc.getRawEncoding());
+inline SourceLocationEncoding::RawLocEncoding
+SourceLocationEncoding::encode(SourceLocation Loc, UIntTy BaseOffset,
+                               unsigned BaseModuleFileIndex,
+                               SourceLocationSequence *Seq) {
+  // If the source location is a local source location, we can try to optimize
+  // the similar sequences to only record the differences.
+  if (!BaseOffset)
+    return Seq ? Seq->encode(Loc) : encodeRaw(Loc.getRawEncoding());
+
+  if (Loc.isInvalid())
+    return 0;
+
+  // Otherwise, the higher bits are used to store the module file index,
+  // so it is meaningless to optimize the source locations into small
+  // integers. Let's try to always use the raw encodings.
+  assert(Loc.getOffset() >= BaseOffset);
+  Loc = Loc.getLocWithOffset(-BaseOffset);
+  RawLocEncoding Encoded = encodeRaw(Loc.getRawEncoding());
+
+  // 16 bits should be sufficient to store the module file index.
+  assert(BaseModuleFileIndex < (1 << 16));
+  Encoded |= (RawLocEncoding)BaseModuleFileIndex << 32;
+  return Encoded;
 }
-inline SourceLocation
-SourceLocationEncoding::decode(uint64_t Encoded, SourceLocationSequence *Seq) {
-  return Seq ? Seq->decode(Encoded)
-             : SourceLocation::getFromRawEncoding(decodeRaw(Encoded));
+inline std::pair<SourceLocation, unsigned>
+SourceLocationEncoding::decode(RawLocEncoding Encoded,
+                               SourceLocationSequence *Seq) {
+  unsigned ModuleFileIndex = Encoded >> 32;
+
+  if (!ModuleFileIndex)
+    return {Seq ? Seq->decode(Encoded)
+                : SourceLocation::getFromRawEncoding(decodeRaw(Encoded)),
+            ModuleFileIndex};
+
+  Encoded &= llvm::maskTrailingOnes<RawLocEncoding>(32);
+  SourceLocation Loc = SourceLocation::getFromRawEncoding(decodeRaw(Encoded));
+
+  return {Loc, ModuleFileIndex};
 }
 
 } // namespace clang

diff --git a/clang/include/clang/StaticAnalyzer/Checkers/Checkers.td b/clang/include/clang/StaticAnalyzer/Checkers/Checkers.td
@@ -1397,7 +1397,7 @@ def CastValueChecker : Checker<"CastValue">,
   Documentation<NotDocumented>;
 
 def ReturnValueChecker : Checker<"ReturnValue">,
-  HelpText<"Model the guaranteed boolean return value of function calls">,
+  HelpText<"Model certain Error() methods that always return true by convention">,
   Documentation<NotDocumented>;
 
 } // end "apiModeling.llvm"

diff --git a/clang/lib/AST/ASTContext.cpp b/clang/lib/AST/ASTContext.cpp
@@ -1307,6 +1307,9 @@ void ASTContext::InitBuiltinTypes(const TargetInfo &Target,
   // Placeholder type for bound members.
   InitBuiltinType(BoundMemberTy,       BuiltinType::BoundMember);
 
+  // Placeholder type for unresolved templates.
+  InitBuiltinType(UnresolvedTemplateTy, BuiltinType::UnresolvedTemplate);
+
   // Placeholder type for pseudo-objects.
   InitBuiltinType(PseudoObjectTy,      BuiltinType::PseudoObject);
 
@@ -2255,9 +2258,8 @@ TypeInfo ASTContext::getTypeInfoImpl(const Type *T) const {
   }
   case Type::BitInt: {
     const auto *EIT = cast<BitIntType>(T);
-    Align = std::clamp<unsigned>(llvm::PowerOf2Ceil(EIT->getNumBits()),
-                                 getCharWidth(), Target->getLongLongAlign());
-    Width = llvm::alignTo(EIT->getNumBits(), Align);
+    Align = Target->getBitIntAlign(EIT->getNumBits());
+    Width = Target->getBitIntWidth(EIT->getNumBits());
     break;
   }
   case Type::Record:
@@ -3794,33 +3796,33 @@ QualType ASTContext::getDependentSizedArrayType(QualType elementType,
           numElements->isValueDependent()) &&
          "Size must be type- or value-dependent!");
 
+  SplitQualType canonElementType = getCanonicalType(elementType).split();
+
+  void *insertPos = nullptr;
+  llvm::FoldingSetNodeID ID;
+  DependentSizedArrayType::Profile(
+      ID, *this, numElements ? QualType(canonElementType.Ty, 0) : elementType,
+      ASM, elementTypeQuals, numElements);
+
+  // Look for an existing type with these properties.
+  DependentSizedArrayType *canonTy =
+    DependentSizedArrayTypes.FindNodeOrInsertPos(ID, insertPos);
+
   // Dependently-sized array types that do not have a specified number
   // of elements will have their sizes deduced from a dependent
-  // initializer.  We do no canonicalization here at all, which is okay
-  // because they can't be used in most locations.
+  // initializer.
   if (!numElements) {
+    if (canonTy)
+      return QualType(canonTy, 0);
+
     auto *newType = new (*this, alignof(DependentSizedArrayType))
         DependentSizedArrayType(elementType, QualType(), numElements, ASM,
                                 elementTypeQuals, brackets);
+    DependentSizedArrayTypes.InsertNode(newType, insertPos);
     Types.push_back(newType);
     return QualType(newType, 0);
   }
 
-  // Otherwise, we actually build a new type every time, but we
-  // also build a canonical type.
-
-  SplitQualType canonElementType = getCanonicalType(elementType).split();
-
-  void *insertPos = nullptr;
-  llvm::FoldingSetNodeID ID;
-  DependentSizedArrayType::Profile(ID, *this,
-                                   QualType(canonElementType.Ty, 0),
-                                   ASM, elementTypeQuals, numElements);
-
-  // Look for an existing type with these properties.
-  DependentSizedArrayType *canonTy =
-    DependentSizedArrayTypes.FindNodeOrInsertPos(ID, insertPos);
-
   // If we don't have one, build one.
   if (!canonTy) {
     canonTy = new (*this, alignof(DependentSizedArrayType))

diff --git a/clang/lib/AST/Decl.cpp b/clang/lib/AST/Decl.cpp
@@ -5758,3 +5758,18 @@ ExportDecl *ExportDecl::Create(ASTContext &C, DeclContext *DC,
 ExportDecl *ExportDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ExportDecl(nullptr, SourceLocation());
 }
+
+bool clang::IsArmStreamingFunction(const FunctionDecl *FD,
+                                   bool IncludeLocallyStreaming) {
+  if (IncludeLocallyStreaming)
+    if (FD->hasAttr<ArmLocallyStreamingAttr>())
+      return true;
+
+  if (const Type *Ty = FD->getType().getTypePtrOrNull())
+    if (const auto *FPT = Ty->getAs<FunctionProtoType>())
+      if (FPT->getAArch64SMEAttributes() &
+          FunctionType::SME_PStateSMEnabledMask)
+        return true;
+
+  return false;
+}
diff --git a/clang/lib/AST/Interp/ByteCodeEmitter.cpp b/clang/lib/AST/Interp/ByteCodeEmitter.cpp
@@ -82,11 +82,13 @@ Function *ByteCodeEmitter::compileFunc(const FunctionDecl *FuncDecl) {
   // InterpStack when calling the function.
   bool HasThisPointer = false;
   if (const auto *MD = dyn_cast<CXXMethodDecl>(FuncDecl)) {
-    if (MD->isImplicitObjectMemberFunction() && !IsLambdaStaticInvoker) {
-      HasThisPointer = true;
-      ParamTypes.push_back(PT_Ptr);
-      ParamOffsets.push_back(ParamOffset);
-      ParamOffset += align(primSize(PT_Ptr));
+    if (!IsLambdaStaticInvoker) {
+      HasThisPointer = MD->isInstance();
+      if (MD->isImplicitObjectMemberFunction()) {
+        ParamTypes.push_back(PT_Ptr);
+        ParamOffsets.push_back(ParamOffset);
+        ParamOffset += align(primSize(PT_Ptr));
+      }
     }
 
     // Set up lambda capture to closure record field mapping.

diff --git a/clang/lib/AST/Interp/ByteCodeExprGen.cpp b/clang/lib/AST/Interp/ByteCodeExprGen.cpp
@@ -2025,7 +2025,7 @@ bool ByteCodeExprGen<Emitter>::VisitLambdaExpr(const LambdaExpr *E) {
       if (!this->visit(Init))
         return false;
 
-      if (!this->emitSetField(*T, F.Offset, E))
+      if (!this->emitInitField(*T, F.Offset, E))
         return false;
     } else {
       if (!this->emitDupPtr(E))

diff --git a/clang/lib/AST/Interp/Pointer.cpp b/clang/lib/AST/Interp/Pointer.cpp
@@ -63,9 +63,8 @@ Pointer::~Pointer() {
 }
 
 void Pointer::operator=(const Pointer &P) {
-
   if (!this->isIntegralPointer() || !P.isBlockPointer())
-    assert(P.StorageKind == StorageKind);
+    assert(P.StorageKind == StorageKind || (this->isZero() && P.isZero()));
 
   bool WasBlockPointer = isBlockPointer();
   StorageKind = P.StorageKind;
@@ -92,7 +91,7 @@ void Pointer::operator=(const Pointer &P) {
 
 void Pointer::operator=(Pointer &&P) {
   if (!this->isIntegralPointer() || !P.isBlockPointer())
-    assert(P.StorageKind == StorageKind);
+    assert(P.StorageKind == StorageKind || (this->isZero() && P.isZero()));
 
   bool WasBlockPointer = isBlockPointer();
   StorageKind = P.StorageKind;

diff --git a/clang/lib/AST/NSAPI.cpp b/clang/lib/AST/NSAPI.cpp
@@ -454,6 +454,7 @@ NSAPI::getNSNumberFactoryMethodKind(QualType T) const {
 #define WASM_TYPE(Name, Id, SingletonId) case BuiltinType::Id:
 #include "clang/Basic/WebAssemblyReferenceTypes.def"
   case BuiltinType::BoundMember:
+  case BuiltinType::UnresolvedTemplate:
   case BuiltinType::Dependent:
   case BuiltinType::Overload:
   case BuiltinType::UnknownAny:

diff --git a/clang/lib/AST/OpenACCClause.cpp b/clang/lib/AST/OpenACCClause.cpp
@@ -127,6 +127,38 @@ OpenACCVectorLengthClause::Create(const ASTContext &C, SourceLocation BeginLoc,
       OpenACCVectorLengthClause(BeginLoc, LParenLoc, IntExpr, EndLoc);
 }
 
+OpenACCAsyncClause::OpenACCAsyncClause(SourceLocation BeginLoc,
+                                       SourceLocation LParenLoc, Expr *IntExpr,
+                                       SourceLocation EndLoc)
+    : OpenACCClauseWithSingleIntExpr(OpenACCClauseKind::Async, BeginLoc,
+                                     LParenLoc, IntExpr, EndLoc) {
+  assert((!IntExpr || IntExpr->isInstantiationDependent() ||
+          IntExpr->getType()->isIntegerType()) &&
+         "Condition expression type not scalar/dependent");
+}
+
+OpenACCAsyncClause *OpenACCAsyncClause::Create(const ASTContext &C,
+                                               SourceLocation BeginLoc,
+                                               SourceLocation LParenLoc,
+                                               Expr *IntExpr,
+                                               SourceLocation EndLoc) {
+  void *Mem =
+      C.Allocate(sizeof(OpenACCAsyncClause), alignof(OpenACCAsyncClause));
+  return new (Mem) OpenACCAsyncClause(BeginLoc, LParenLoc, IntExpr, EndLoc);
+}
+
+OpenACCWaitClause *OpenACCWaitClause::Create(
+    const ASTContext &C, SourceLocation BeginLoc, SourceLocation LParenLoc,
+    Expr *DevNumExpr, SourceLocation QueuesLoc, ArrayRef<Expr *> QueueIdExprs,
+    SourceLocation EndLoc) {
+  // Allocates enough room in trailing storage for all the int-exprs, plus a
+  // placeholder for the devnum.
+  void *Mem = C.Allocate(
+      OpenACCWaitClause::totalSizeToAlloc<Expr *>(QueueIdExprs.size() + 1));
+  return new (Mem) OpenACCWaitClause(BeginLoc, LParenLoc, DevNumExpr, QueuesLoc,
+                                     QueueIdExprs, EndLoc);
+}
+
 OpenACCNumGangsClause *OpenACCNumGangsClause::Create(const ASTContext &C,
                                                      SourceLocation BeginLoc,
                                                      SourceLocation LParenLoc,
@@ -156,6 +188,26 @@ OpenACCFirstPrivateClause *OpenACCFirstPrivateClause::Create(
       OpenACCFirstPrivateClause(BeginLoc, LParenLoc, VarList, EndLoc);
 }
 
+OpenACCAttachClause *OpenACCAttachClause::Create(const ASTContext &C,
+                                                 SourceLocation BeginLoc,
+                                                 SourceLocation LParenLoc,
+                                                 ArrayRef<Expr *> VarList,
+                                                 SourceLocation EndLoc) {
+  void *Mem =
+      C.Allocate(OpenACCAttachClause::totalSizeToAlloc<Expr *>(VarList.size()));
+  return new (Mem) OpenACCAttachClause(BeginLoc, LParenLoc, VarList, EndLoc);
+}
+
+OpenACCDevicePtrClause *OpenACCDevicePtrClause::Create(const ASTContext &C,
+                                                       SourceLocation BeginLoc,
+                                                       SourceLocation LParenLoc,
+                                                       ArrayRef<Expr *> VarList,
+                                                       SourceLocation EndLoc) {
+  void *Mem = C.Allocate(
+      OpenACCDevicePtrClause::totalSizeToAlloc<Expr *>(VarList.size()));
+  return new (Mem) OpenACCDevicePtrClause(BeginLoc, LParenLoc, VarList, EndLoc);
+}
+
 OpenACCNoCreateClause *OpenACCNoCreateClause::Create(const ASTContext &C,
                                                      SourceLocation BeginLoc,
                                                      SourceLocation LParenLoc,
@@ -267,6 +319,15 @@ void OpenACCClausePrinter::VisitVectorLengthClause(
   OS << ")";
 }
 
+void OpenACCClausePrinter::VisitAsyncClause(const OpenACCAsyncClause &C) {
+  OS << "async";
+  if (C.hasIntExpr()) {
+    OS << "(";
+    printExpr(C.getIntExpr());
+    OS << ")";
+  }
+}
+
 void OpenACCClausePrinter::VisitPrivateClause(const OpenACCPrivateClause &C) {
   OS << "private(";
   llvm::interleaveComma(C.getVarList(), OS,
@@ -282,6 +343,21 @@ void OpenACCClausePrinter::VisitFirstPrivateClause(
   OS << ")";
 }
 
+void OpenACCClausePrinter::VisitAttachClause(const OpenACCAttachClause &C) {
+  OS << "attach(";
+  llvm::interleaveComma(C.getVarList(), OS,
+                        [&](const Expr *E) { printExpr(E); });
+  OS << ")";
+}
+
+void OpenACCClausePrinter::VisitDevicePtrClause(
+    const OpenACCDevicePtrClause &C) {
+  OS << "deviceptr(";
+  llvm::interleaveComma(C.getVarList(), OS,
+                        [&](const Expr *E) { printExpr(E); });
+  OS << ")";
+}
+
 void OpenACCClausePrinter::VisitNoCreateClause(const OpenACCNoCreateClause &C) {
   OS << "no_create(";
   llvm::interleaveComma(C.getVarList(), OS,
@@ -329,3 +405,22 @@ void OpenACCClausePrinter::VisitCreateClause(const OpenACCCreateClause &C) {
                         [&](const Expr *E) { printExpr(E); });
   OS << ")";
 }
+
+void OpenACCClausePrinter::VisitWaitClause(const OpenACCWaitClause &C) {
+  OS << "wait";
+  if (!C.getLParenLoc().isInvalid()) {
+    OS << "(";
+    if (C.hasDevNumExpr()) {
+      OS << "devnum: ";
+      printExpr(C.getDevNumExpr());
+      OS << " : ";
+    }
+
+    if (C.hasQueuesTag())
+      OS << "queues: ";
+
+    llvm::interleaveComma(C.getQueueIdExprs(), OS,
+                          [&](const Expr *E) { printExpr(E); });
+    OS << ")";
+  }
+}
diff --git a/clang/lib/AST/StmtProfile.cpp b/clang/lib/AST/StmtProfile.cpp
@@ -2543,6 +2543,18 @@ void OpenACCClauseProfiler::VisitFirstPrivateClause(
     Profiler.VisitStmt(E);
 }
 
+void OpenACCClauseProfiler::VisitAttachClause(
+    const OpenACCAttachClause &Clause) {
+  for (auto *E : Clause.getVarList())
+    Profiler.VisitStmt(E);
+}
+
+void OpenACCClauseProfiler::VisitDevicePtrClause(
+    const OpenACCDevicePtrClause &Clause) {
+  for (auto *E : Clause.getVarList())
+    Profiler.VisitStmt(E);
+}
+
 void OpenACCClauseProfiler::VisitNoCreateClause(
     const OpenACCNoCreateClause &Clause) {
   for (auto *E : Clause.getVarList())
@@ -2561,6 +2573,18 @@ void OpenACCClauseProfiler::VisitVectorLengthClause(
          "vector_length clause requires a valid int expr");
   Profiler.VisitStmt(Clause.getIntExpr());
 }
+
+void OpenACCClauseProfiler::VisitAsyncClause(const OpenACCAsyncClause &Clause) {
+  if (Clause.hasIntExpr())
+    Profiler.VisitStmt(Clause.getIntExpr());
+}
+
+void OpenACCClauseProfiler::VisitWaitClause(const OpenACCWaitClause &Clause) {
+  if (Clause.hasDevNumExpr())
+    Profiler.VisitStmt(Clause.getDevNumExpr());
+  for (auto *E : Clause.getQueueIdExprs())
+    Profiler.VisitStmt(E);
+}
 } // namespace
 
 void StmtProfiler::VisitOpenACCComputeConstruct(

diff --git a/clang/lib/AST/TextNodeDumper.cpp b/clang/lib/AST/TextNodeDumper.cpp
@@ -397,10 +397,13 @@ void TextNodeDumper::Visit(const OpenACCClause *C) {
     case OpenACCClauseKind::Default:
       OS << '(' << cast<OpenACCDefaultClause>(C)->getDefaultClauseKind() << ')';
       break;
+    case OpenACCClauseKind::Async:
+    case OpenACCClauseKind::Attach:
     case OpenACCClauseKind::Copy:
     case OpenACCClauseKind::PCopy:
     case OpenACCClauseKind::PresentOrCopy:
     case OpenACCClauseKind::If:
+    case OpenACCClauseKind::DevicePtr:
     case OpenACCClauseKind::FirstPrivate:
     case OpenACCClauseKind::NoCreate:
     case OpenACCClauseKind::NumGangs:
@@ -434,6 +437,13 @@ void TextNodeDumper::Visit(const OpenACCClause *C) {
       if (cast<OpenACCCreateClause>(C)->isZero())
         OS << " : zero";
       break;
+    case OpenACCClauseKind::Wait:
+      OS << " clause";
+      if (cast<OpenACCWaitClause>(C)->hasDevNumExpr())
+        OS << " has devnum";
+      if (cast<OpenACCWaitClause>(C)->hasQueuesTag())
+        OS << " has queues tag";
+      break;
     default:
       // Nothing to do here.
       break;

diff --git a/clang/lib/AST/Type.cpp b/clang/lib/AST/Type.cpp
@@ -256,7 +256,8 @@ void DependentSizedArrayType::Profile(llvm::FoldingSetNodeID &ID,
   ID.AddPointer(ET.getAsOpaquePtr());
   ID.AddInteger(llvm::to_underlying(SizeMod));
   ID.AddInteger(TypeQuals);
-  E->Profile(ID, Context, true);
+  if (E)
+    E->Profile(ID, Context, true);
 }
 
 DependentVectorType::DependentVectorType(QualType ElementType,
@@ -3393,6 +3394,8 @@ StringRef BuiltinType::getName(const PrintingPolicy &Policy) const {
     return "<overloaded function type>";
   case BoundMember:
     return "<bound member function type>";
+  case UnresolvedTemplate:
+    return "<unresolved template type>";
   case PseudoObject:
     return "<pseudo-object type>";
   case Dependent:
@@ -4685,6 +4688,7 @@ bool Type::canHaveNullability(bool ResultIfUnknown) const {
 #include "clang/AST/BuiltinTypes.def"
       return false;
 
+    case BuiltinType::UnresolvedTemplate:
     // Dependent types that could instantiate to a pointer type.
     case BuiltinType::Dependent:
     case BuiltinType::Overload:

diff --git a/clang/lib/AST/TypeLoc.cpp b/clang/lib/AST/TypeLoc.cpp
@@ -399,6 +399,7 @@ TypeSpecifierType BuiltinTypeLoc::getWrittenTypeSpec() const {
   case BuiltinType::NullPtr:
   case BuiltinType::Overload:
   case BuiltinType::Dependent:
+  case BuiltinType::UnresolvedTemplate:
   case BuiltinType::BoundMember:
   case BuiltinType::UnknownAny:
   case BuiltinType::ARCUnbridgedCast:

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowAnalysisContext.cpp b/clang/lib/Analysis/FlowSensitive/DataflowAnalysisContext.cpp
@@ -170,7 +170,7 @@ DataflowAnalysisContext::joinFlowConditions(Atom FirstToken,
 
 Solver::Result DataflowAnalysisContext::querySolver(
     llvm::SetVector<const Formula *> Constraints) {
-  return S->solve(Constraints.getArrayRef());
+  return S.solve(Constraints.getArrayRef());
 }
 
 bool DataflowAnalysisContext::flowConditionImplies(Atom Token,
@@ -338,10 +338,10 @@ static std::unique_ptr<Logger> makeLoggerFromCommandLine() {
   return Logger::html(std::move(StreamFactory));
 }
 
-DataflowAnalysisContext::DataflowAnalysisContext(std::unique_ptr<Solver> S,
-                                                 Options Opts)
-    : S(std::move(S)), A(std::make_unique<Arena>()), Opts(Opts) {
-  assert(this->S != nullptr);
+DataflowAnalysisContext::DataflowAnalysisContext(
+    Solver &S, std::unique_ptr<Solver> &&OwnedSolver, Options Opts)
+    : S(S), OwnedSolver(std::move(OwnedSolver)), A(std::make_unique<Arena>()),
+      Opts(Opts) {
   // If the -dataflow-log command-line flag was set, synthesize a logger.
   // This is ugly but provides a uniform method for ad-hoc debugging dataflow-
   // based tools.

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -350,6 +350,17 @@ class ResultObjectVisitor : public RecursiveASTVisitor<ResultObjectVisitor> {
     return RecursiveASTVisitor<ResultObjectVisitor>::TraverseDecl(D);
   }
 
+  // Don't traverse expressions in unevaluated contexts, as we don't model
+  // fields that are only used in these.
+  // Note: The operand of the `noexcept` operator is an unevaluated operand, but
+  // nevertheless it appears in the Clang CFG, so we don't exclude it here.
+  bool TraverseDecltypeTypeLoc(DecltypeTypeLoc) { return true; }
+  bool TraverseTypeOfExprTypeLoc(TypeOfExprTypeLoc) { return true; }
+  bool TraverseCXXTypeidExpr(CXXTypeidExpr *) { return true; }
+  bool TraverseUnaryExprOrTypeTraitExpr(UnaryExprOrTypeTraitExpr *) {
+    return true;
+  }
+
   bool TraverseBindingDecl(BindingDecl *BD) {
     // `RecursiveASTVisitor` doesn't traverse holding variables for
     // `BindingDecl`s by itself, so we need to tell it to.

diff --git a/clang/lib/Analysis/FlowSensitive/Transfer.cpp b/clang/lib/Analysis/FlowSensitive/Transfer.cpp
@@ -68,6 +68,14 @@ static BoolValue &evaluateBooleanEquality(const Expr &LHS, const Expr &RHS,
     if (auto *RHSBool = dyn_cast_or_null<BoolValue>(RHSValue))
       return Env.makeIff(*LHSBool, *RHSBool);
 
+  if (auto *LHSPtr = dyn_cast_or_null<PointerValue>(LHSValue))
+    if (auto *RHSPtr = dyn_cast_or_null<PointerValue>(RHSValue))
+      // If the storage locations are the same, the pointers definitely compare
+      // the same. If the storage locations are different, they may still alias,
+      // so we fall through to the case below that returns an atom.
+      if (&LHSPtr->getPointeeLoc() == &RHSPtr->getPointeeLoc())
+        return Env.getBoolLiteralValue(true);
+
   return Env.makeAtomicBoolValue();
 }
 
@@ -556,14 +564,23 @@ class TransferVisitor : public ConstStmtVisitor<TransferVisitor> {
 
       copyRecord(*LocSrc, *LocDst, Env);
 
-      // If the expr is a glvalue, we can reasonably assume the operator is
-      // returning T& and thus we can assign it `LocDst`.
-      if (S->isGLValue()) {
+      // The assignment operator can have an arbitrary return type. We model the
+      // return value only if the return type is the same as or a base class of
+      // the destination type.
+      if (S->getType().getCanonicalType().getUnqualifiedType() !=
+          LocDst->getType().getCanonicalType().getUnqualifiedType()) {
+        auto ReturnDecl = S->getType()->getAsCXXRecordDecl();
+        auto DstDecl = LocDst->getType()->getAsCXXRecordDecl();
+        if (ReturnDecl == nullptr || DstDecl == nullptr)
+          return;
+        if (!DstDecl->isDerivedFrom(ReturnDecl))
+          return;
+      }
+
+      if (S->isGLValue())
         Env.setStorageLocation(*S, *LocDst);
-      } else if (S->getType()->isRecordType()) {
-        // Assume that the assignment returns the assigned value.
+      else
         copyRecord(*LocDst, Env.getResultObjectLocation(*S), Env);
-      }
 
       return;
     }

diff --git a/clang/lib/Basic/Cuda.cpp b/clang/lib/Basic/Cuda.cpp
@@ -14,7 +14,7 @@ struct CudaVersionMapEntry {
 };
 #define CUDA_ENTRY(major, minor)                                               \
   {                                                                            \
-#major "." #minor, CudaVersion::CUDA_##major##minor,                       \
+    #major "." #minor, CudaVersion::CUDA_##major##minor,                       \
         llvm::VersionTuple(major, minor)                                       \
   }
 
@@ -41,6 +41,7 @@ static const CudaVersionMapEntry CudaNameVersionMap[] = {
     CUDA_ENTRY(12, 1),
     CUDA_ENTRY(12, 2),
     CUDA_ENTRY(12, 3),
+    CUDA_ENTRY(12, 4),
     {"", CudaVersion::NEW, llvm::VersionTuple(std::numeric_limits<int>::max())},
     {"unknown", CudaVersion::UNKNOWN, {}} // End of list tombstone.
 };
@@ -241,7 +242,7 @@ CudaVersion MaxVersionForCudaArch(CudaArch A) {
   }
 }
 
-bool CudaFeatureEnabled(llvm::VersionTuple  Version, CudaFeature Feature) {
+bool CudaFeatureEnabled(llvm::VersionTuple Version, CudaFeature Feature) {
   return CudaFeatureEnabled(ToCudaVersion(Version), Feature);
 }
 

diff --git a/clang/lib/Basic/Module.cpp b/clang/lib/Basic/Module.cpp
@@ -140,8 +140,8 @@ bool Module::isUnimportable(const LangOptions &LangOpts,
       return true;
     }
     for (unsigned I = 0, N = Current->Requirements.size(); I != N; ++I) {
-      if (hasFeature(Current->Requirements[I].first, LangOpts, Target) !=
-              Current->Requirements[I].second) {
+      if (hasFeature(Current->Requirements[I].FeatureName, LangOpts, Target) !=
+          Current->Requirements[I].RequiredState) {
         Req = Current->Requirements[I];
         return true;
       }
@@ -319,7 +319,7 @@ bool Module::directlyUses(const Module *Requested) {
 void Module::addRequirement(StringRef Feature, bool RequiredState,
                             const LangOptions &LangOpts,
                             const TargetInfo &Target) {
-  Requirements.push_back(Requirement(std::string(Feature), RequiredState));
+  Requirements.push_back(Requirement{std::string(Feature), RequiredState});
 
   // If this feature is currently available, we're done.
   if (hasFeature(Feature, LangOpts, Target) == RequiredState)
@@ -504,9 +504,9 @@ void Module::print(raw_ostream &OS, unsigned Indent, bool Dump) const {
     for (unsigned I = 0, N = Requirements.size(); I != N; ++I) {
       if (I)
         OS << ", ";
-      if (!Requirements[I].second)
+      if (!Requirements[I].RequiredState)
         OS << "!";
-      OS << Requirements[I].first;
+      OS << Requirements[I].FeatureName;
     }
     OS << "\n";
   }

diff --git a/clang/lib/Basic/Targets.cpp b/clang/lib/Basic/Targets.cpp
@@ -760,7 +760,7 @@ using namespace clang::targets;
 TargetInfo *
 TargetInfo::CreateTargetInfo(DiagnosticsEngine &Diags,
                              const std::shared_ptr<TargetOptions> &Opts) {
-  llvm::Triple Triple(Opts->Triple);
+  llvm::Triple Triple(llvm::Triple::normalize(Opts->Triple));
 
   // Construct the target
   std::unique_ptr<TargetInfo> Target = AllocateTarget(Triple, *Opts);

diff --git a/clang/lib/Basic/Targets/AArch64.cpp b/clang/lib/Basic/Targets/AArch64.cpp
@@ -154,6 +154,7 @@ AArch64TargetInfo::AArch64TargetInfo(const llvm::Triple &Triple,
   else
     LongWidth = LongAlign = PointerWidth = PointerAlign = 32;
 
+  BitIntMaxAlign = 128;
   MaxVectorAlign = 128;
   MaxAtomicInlineWidth = 128;
   MaxAtomicPromoteWidth = 128;
@@ -224,7 +225,7 @@ bool AArch64TargetInfo::validateBranchProtection(StringRef Spec, StringRef,
                                                  BranchProtectionInfo &BPI,
                                                  StringRef &Err) const {
   llvm::ARM::ParsedBranchProtection PBP;
-  if (!llvm::ARM::parseBranchProtection(Spec, PBP, Err))
+  if (!llvm::ARM::parseBranchProtection(Spec, PBP, Err, HasPAuthLR))
     return false;
 
   BPI.SignReturnAddr =
@@ -1480,11 +1481,11 @@ AArch64leTargetInfo::AArch64leTargetInfo(const llvm::Triple &Triple,
 void AArch64leTargetInfo::setDataLayout() {
   if (getTriple().isOSBinFormatMachO()) {
     if(getTriple().isArch32Bit())
-      resetDataLayout("e-m:o-p:32:32-i64:64-i128:128-n32:64-S128", "_");
+      resetDataLayout("e-m:o-p:32:32-i64:64-i128:128-n32:64-S128-Fn32", "_");
     else
-      resetDataLayout("e-m:o-i64:64-i128:128-n32:64-S128", "_");
+      resetDataLayout("e-m:o-i64:64-i128:128-n32:64-S128-Fn32", "_");
   } else
-    resetDataLayout("e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128");
+    resetDataLayout("e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128-Fn32");
 }
 
 void AArch64leTargetInfo::getTargetDefines(const LangOptions &Opts,
@@ -1507,7 +1508,7 @@ void AArch64beTargetInfo::getTargetDefines(const LangOptions &Opts,
 
 void AArch64beTargetInfo::setDataLayout() {
   assert(!getTriple().isOSBinFormatMachO());
-  resetDataLayout("E-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128");
+  resetDataLayout("E-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128-Fn32");
 }
 
 WindowsARM64TargetInfo::WindowsARM64TargetInfo(const llvm::Triple &Triple,
@@ -1530,8 +1531,8 @@ WindowsARM64TargetInfo::WindowsARM64TargetInfo(const llvm::Triple &Triple,
 
 void WindowsARM64TargetInfo::setDataLayout() {
   resetDataLayout(Triple.isOSBinFormatMachO()
-                      ? "e-m:o-i64:64-i128:128-n32:64-S128"
-                      : "e-m:w-p:64:64-i32:32-i64:64-i128:128-n32:64-S128",
+                      ? "e-m:o-i64:64-i128:128-n32:64-S128-Fn32"
+                      : "e-m:w-p:64:64-i32:32-i64:64-i128:128-n32:64-S128-Fn32",
                   Triple.isOSBinFormatMachO() ? "_" : "");
 }
 

diff --git a/clang/lib/Basic/Targets/AMDGPU.cpp b/clang/lib/Basic/Targets/AMDGPU.cpp
@@ -232,7 +232,7 @@ AMDGPUTargetInfo::AMDGPUTargetInfo(const llvm::Triple &Triple,
 
   HasLegalHalfType = true;
   HasFloat16 = true;
-  WavefrontSize = GPUFeatures & llvm::AMDGPU::FEATURE_WAVE32 ? 32 : 64;
+  WavefrontSize = (GPUFeatures & llvm::AMDGPU::FEATURE_WAVE32) ? 32 : 64;
   AllowAMDGPUUnsafeFPAtomics = Opts.AllowAMDGPUUnsafeFPAtomics;
 
   // Set pointer width and alignment for the generic address space.

diff --git a/clang/lib/Basic/Targets/PPC.cpp b/clang/lib/Basic/Targets/PPC.cpp
@@ -91,6 +91,8 @@ bool PPCTargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
       IsISA3_1 = true;
     } else if (Feature == "+quadword-atomics") {
       HasQuadwordAtomics = true;
+    } else if (Feature == "+aix-shared-lib-tls-model-opt") {
+      HasAIXShLibTLSModelOpt = true;
     }
     // TODO: Finish this list and add an assert that we've handled them
     // all.
@@ -580,6 +582,9 @@ bool PPCTargetInfo::initFeatureMap(
   Features["aix-small-local-exec-tls"] = false;
   Features["aix-small-local-dynamic-tls"] = false;
 
+  // Turn off TLS model opt by default.
+  Features["aix-shared-lib-tls-model-opt"] = false;
+
   Features["spe"] = llvm::StringSwitch<bool>(CPU)
                         .Case("8548", true)
                         .Case("e500", true)
@@ -722,6 +727,7 @@ bool PPCTargetInfo::hasFeature(StringRef Feature) const {
       .Case("isa-v30-instructions", IsISA3_0)
       .Case("isa-v31-instructions", IsISA3_1)
       .Case("quadword-atomics", HasQuadwordAtomics)
+      .Case("aix-shared-lib-tls-model-opt", HasAIXShLibTLSModelOpt)
       .Default(false);
 }
 

diff --git a/clang/lib/Basic/Targets/PPC.h b/clang/lib/Basic/Targets/PPC.h
@@ -81,6 +81,7 @@ class LLVM_LIBRARY_VISIBILITY PPCTargetInfo : public TargetInfo {
   bool IsISA3_0 = false;
   bool IsISA3_1 = false;
   bool HasQuadwordAtomics = false;
+  bool HasAIXShLibTLSModelOpt = false;
 
 protected:
   std::string ABI;

diff --git a/clang/lib/Basic/Targets/X86.cpp b/clang/lib/Basic/Targets/X86.cpp
@@ -64,6 +64,8 @@ static const char *const GCCRegNames[] = {
     "dr0",   "dr1",   "dr2",   "dr3",   "dr6",     "dr7",
     "bnd0",  "bnd1",  "bnd2",  "bnd3",
     "tmm0",  "tmm1",  "tmm2",  "tmm3",  "tmm4",    "tmm5",  "tmm6",  "tmm7",
+    "r16",   "r17",   "r18",   "r19",   "r20",     "r21",   "r22",   "r23",
+    "r24",   "r25",   "r26",   "r27",   "r28",     "r29",   "r30",   "r31",
 };
 
 const TargetInfo::AddlRegName AddlRegNames[] = {
@@ -83,8 +85,23 @@ const TargetInfo::AddlRegName AddlRegNames[] = {
     {{"r13d", "r13w", "r13b"}, 43},
     {{"r14d", "r14w", "r14b"}, 44},
     {{"r15d", "r15w", "r15b"}, 45},
+    {{"r16d", "r16w", "r16b"}, 165},
+    {{"r17d", "r17w", "r17b"}, 166},
+    {{"r18d", "r18w", "r18b"}, 167},
+    {{"r19d", "r19w", "r19b"}, 168},
+    {{"r20d", "r20w", "r20b"}, 169},
+    {{"r21d", "r21w", "r21b"}, 170},
+    {{"r22d", "r22w", "r22b"}, 171},
+    {{"r23d", "r23w", "r23b"}, 172},
+    {{"r24d", "r24w", "r24b"}, 173},
+    {{"r25d", "r25w", "r25b"}, 174},
+    {{"r26d", "r26w", "r26b"}, 175},
+    {{"r27d", "r27w", "r27b"}, 176},
+    {{"r28d", "r28w", "r28b"}, 177},
+    {{"r29d", "r29w", "r29b"}, 178},
+    {{"r30d", "r30w", "r30b"}, 179},
+    {{"r31d", "r31w", "r31b"}, 180},
 };
-
 } // namespace targets
 } // namespace clang
 

diff --git a/clang/lib/CodeGen/BackendConsumer.h b/clang/lib/CodeGen/BackendConsumer.h
@@ -34,7 +34,6 @@ class BackendConsumer : public ASTConsumer {
   const CodeGenOptions &CodeGenOpts;
   const TargetOptions &TargetOpts;
   const LangOptions &LangOpts;
-  const FileManager &FileMgr;
   std::unique_ptr<raw_pwrite_stream> AsmOutStream;
   ASTContext *Context;
   IntrusiveRefCntPtr<llvm::vfs::FileSystem> FS;
@@ -76,7 +75,7 @@ class BackendConsumer : public ASTConsumer {
                   const PreprocessorOptions &PPOpts,
                   const CodeGenOptions &CodeGenOpts,
                   const TargetOptions &TargetOpts, const LangOptions &LangOpts,
-                  const FileManager &FileMgr, const std::string &InFile,
+                  const std::string &InFile,
                   SmallVector<LinkModule, 4> LinkModules,
                   std::unique_ptr<raw_pwrite_stream> OS, llvm::LLVMContext &C,
                   CoverageSourceInfo *CoverageInfo = nullptr);
@@ -90,8 +89,8 @@ class BackendConsumer : public ASTConsumer {
                   const PreprocessorOptions &PPOpts,
                   const CodeGenOptions &CodeGenOpts,
                   const TargetOptions &TargetOpts, const LangOptions &LangOpts,
-                  const FileManager &FileMgr, llvm::Module *Module,
-                  SmallVector<LinkModule, 4> LinkModules, llvm::LLVMContext &C,
+                  llvm::Module *Module, SmallVector<LinkModule, 4> LinkModules,
+                  llvm::LLVMContext &C,
                   CoverageSourceInfo *CoverageInfo = nullptr);
 
   llvm::Module *getModule() const;
@@ -115,10 +114,6 @@ class BackendConsumer : public ASTConsumer {
   // Links each entry in LinkModules into our module.  Returns true on error.
   bool LinkInModules(llvm::Module *M, bool ShouldLinkFiles = true);
 
-  // Load a bitcode module from -mlink-builtin-bitcode option using
-  // methods from a BackendConsumer instead of CompilerInstance
-  bool ReloadModules(llvm::Module *M);
-
   /// Get the best possible source location to represent a diagnostic that
   /// may have associated debug info.
   const FullSourceLoc getBestLocationFromDebugLoc(

diff --git a/clang/lib/CodeGen/BackendUtil.cpp b/clang/lib/CodeGen/BackendUtil.cpp
@@ -120,11 +120,6 @@ static cl::opt<PGOOptions::ColdFuncOpt> ClPGOColdFuncAttr(
                           "Mark cold functions with optnone.")));
 
 extern cl::opt<InstrProfCorrelator::ProfCorrelatorKind> ProfileCorrelate;
-
-// Re-link builtin bitcodes after optimization
-cl::opt<bool> ClRelinkBuiltinBitcodePostop(
-    "relink-builtin-bitcode-postop", cl::Optional,
-    cl::desc("Re-link builtin bitcodes after optimization."));
 } // namespace llvm
 
 namespace {
@@ -423,6 +418,7 @@ static bool initTargetOptions(DiagnosticsEngine &Diags,
   Options.UniqueSectionNames = CodeGenOpts.UniqueSectionNames;
   Options.UniqueBasicBlockSectionNames =
       CodeGenOpts.UniqueBasicBlockSectionNames;
+  Options.SeparateNamedSections = CodeGenOpts.SeparateNamedSections;
   Options.TLSSize = CodeGenOpts.TLSSize;
   Options.EnableTLSDESC = CodeGenOpts.EnableTLSDESC;
   Options.EmulatedTLS = CodeGenOpts.EmulatedTLS;
@@ -1054,11 +1050,8 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
     }
   }
 
-  // Re-link against any bitcodes supplied via the -mlink-builtin-bitcode option
-  // Some optimizations may generate new function calls that would not have
-  // been linked pre-optimization (i.e. fused sincos calls generated by
-  // AMDGPULibCalls::fold_sincos.)
-  if (ClRelinkBuiltinBitcodePostop)
+  // Link against bitcodes supplied via the -mlink-builtin-bitcode option
+  if (CodeGenOpts.LinkBitcodePostopt)
     MPM.addPass(LinkInModulesPass(BC, false));
 
   // Add a verifier pass if requested. We don't have to do this if the action

diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -513,8 +513,8 @@ static Value *emitBinaryMaybeConstrainedFPBuiltin(CodeGenFunction &CGF,
   llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
   llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
 
+  CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
   if (CGF.Builder.getIsFPConstrained()) {
-    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
     Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID, Src0->getType());
     return CGF.Builder.CreateConstrainedFPCall(F, { Src0, Src1 });
   } else {
@@ -530,8 +530,8 @@ static Value *emitBinaryExpMaybeConstrainedFPBuiltin(
   llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
   llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
 
+  CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
   if (CGF.Builder.getIsFPConstrained()) {
-    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
     Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID,
                                        {Src0->getType(), Src1->getType()});
     return CGF.Builder.CreateConstrainedFPCall(F, {Src0, Src1});
@@ -551,8 +551,8 @@ static Value *emitTernaryMaybeConstrainedFPBuiltin(CodeGenFunction &CGF,
   llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
   llvm::Value *Src2 = CGF.EmitScalarExpr(E->getArg(2));
 
+  CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
   if (CGF.Builder.getIsFPConstrained()) {
-    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
     Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID, Src0->getType());
     return CGF.Builder.CreateConstrainedFPCall(F, { Src0, Src1, Src2 });
   } else {
@@ -704,6 +704,7 @@ static Value *EmitSignBit(CodeGenFunction &CGF, Value *V) {
 
 static RValue emitLibraryCall(CodeGenFunction &CGF, const FunctionDecl *FD,
                               const CallExpr *E, llvm::Constant *calleeValue) {
+  CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
   CGCallee callee = CGCallee::forDirect(calleeValue, GlobalDecl(FD));
   return CGF.EmitCall(E->getCallee()->getType(), callee, E, ReturnValueSlot());
 }
@@ -2660,7 +2661,7 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     if (OP.hasMathErrnoOverride())
       ErrnoOverriden = OP.getMathErrnoOverride();
   }
-  // True if 'atttibute__((optnone)) is used. This attibute overrides
+  // True if 'attribute__((optnone))' is used. This attribute overrides
   // fast-math which implies math-errno.
   bool OptNone = CurFuncDecl && CurFuncDecl->hasAttr<OptimizeNoneAttr>();
 
@@ -3821,7 +3822,9 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
   case Builtin::BI__builtin_elementwise_sin:
     return RValue::get(
         emitUnaryBuiltin(*this, E, llvm::Intrinsic::sin, "elt.sin"));
-
+  case Builtin::BI__builtin_elementwise_tan:
+    return RValue::get(
+        emitUnaryBuiltin(*this, E, llvm::Intrinsic::tan, "elt.tan"));
   case Builtin::BI__builtin_elementwise_trunc:
     return RValue::get(
         emitUnaryBuiltin(*this, E, llvm::Intrinsic::trunc, "elt.trunc"));
@@ -21302,6 +21305,17 @@ Value *CodeGenFunction::EmitWebAssemblyBuiltinExpr(unsigned BuiltinID,
         CGM.getIntrinsic(Intrinsic::wasm_relaxed_dot_bf16x8_add_f32);
     return Builder.CreateCall(Callee, {LHS, RHS, Acc});
   }
+  case WebAssembly::BI__builtin_wasm_loadf16_f32: {
+    Value *Addr = EmitScalarExpr(E->getArg(0));
+    Function *Callee = CGM.getIntrinsic(Intrinsic::wasm_loadf16_f32);
+    return Builder.CreateCall(Callee, {Addr});
+  }
+  case WebAssembly::BI__builtin_wasm_storef16_f32: {
+    Value *Val = EmitScalarExpr(E->getArg(0));
+    Value *Addr = EmitScalarExpr(E->getArg(1));
+    Function *Callee = CGM.getIntrinsic(Intrinsic::wasm_storef16_f32);
+    return Builder.CreateCall(Callee, {Val, Addr});
+  }
   case WebAssembly::BI__builtin_wasm_table_get: {
     assert(E->getArg(0)->getType()->isArrayType());
     Value *Table = EmitArrayToPointerDecay(E->getArg(0)).emitRawPointer(*this);

diff --git a/clang/lib/CodeGen/CGCall.cpp b/clang/lib/CodeGen/CGCall.cpp
@@ -5050,13 +5050,14 @@ RValue CodeGenFunction::EmitCall(const CGFunctionInfo &CallInfo,
         (TargetDecl->hasAttr<TargetAttr>() ||
          (CurFuncDecl && CurFuncDecl->hasAttr<TargetAttr>())))
       checkTargetFeatures(Loc, FD);
-
-    // Some architectures (such as x86-64) have the ABI changed based on
-    // attribute-target/features. Give them a chance to diagnose.
-    CGM.getTargetCodeGenInfo().checkFunctionCallABI(
-        CGM, Loc, dyn_cast_or_null<FunctionDecl>(CurCodeDecl), FD, CallArgs);
   }
 
+  // Some architectures (such as x86-64) have the ABI changed based on
+  // attribute-target/features. Give them a chance to diagnose.
+  CGM.getTargetCodeGenInfo().checkFunctionCallABI(
+      CGM, Loc, dyn_cast_or_null<FunctionDecl>(CurCodeDecl),
+      dyn_cast_or_null<FunctionDecl>(TargetDecl), CallArgs, RetTy);
+
   // 1. Set up the arguments.
 
   // If we're using inalloca, insert the allocation after the stack save.

diff --git a/clang/lib/CodeGen/CGExprAgg.cpp b/clang/lib/CodeGen/CGExprAgg.cpp
@@ -1736,7 +1736,7 @@ void AggExprEmitter::VisitCXXParenListOrInitListExpr(
       for (const auto *Field : record->fields())
         assert(
             (Field->isUnnamedBitField() || Field->isAnonymousStructOrUnion()) &&
-            "Only unnamed bitfields or ananymous class allowed");
+            "Only unnamed bitfields or anonymous class allowed");
 #endif
       return;
     }

diff --git a/clang/lib/CodeGen/CodeGenAction.cpp b/clang/lib/CodeGen/CodeGenAction.cpp
@@ -60,10 +60,6 @@ using namespace llvm;
 
 #define DEBUG_TYPE "codegenaction"
 
-namespace llvm {
-extern cl::opt<bool> ClRelinkBuiltinBitcodePostop;
-}
-
 namespace clang {
 class BackendConsumer;
 class ClangDiagnosticHandler final : public DiagnosticHandler {
@@ -118,13 +114,12 @@ BackendConsumer::BackendConsumer(
     const HeaderSearchOptions &HeaderSearchOpts,
     const PreprocessorOptions &PPOpts, const CodeGenOptions &CodeGenOpts,
     const TargetOptions &TargetOpts, const LangOptions &LangOpts,
-    const FileManager &FileMgr, const std::string &InFile,
-    SmallVector<LinkModule, 4> LinkModules,
+    const std::string &InFile, SmallVector<LinkModule, 4> LinkModules,
     std::unique_ptr<raw_pwrite_stream> OS, LLVMContext &C,
     CoverageSourceInfo *CoverageInfo)
     : Diags(Diags), Action(Action), HeaderSearchOpts(HeaderSearchOpts),
       CodeGenOpts(CodeGenOpts), TargetOpts(TargetOpts), LangOpts(LangOpts),
-      FileMgr(FileMgr), AsmOutStream(std::move(OS)), Context(nullptr), FS(VFS),
+      AsmOutStream(std::move(OS)), Context(nullptr), FS(VFS),
       LLVMIRGeneration("irgen", "LLVM IR Generation Time"),
       LLVMIRGenerationRefCount(0),
       Gen(CreateLLVMCodeGen(Diags, InFile, std::move(VFS), HeaderSearchOpts,
@@ -144,12 +139,11 @@ BackendConsumer::BackendConsumer(
     const HeaderSearchOptions &HeaderSearchOpts,
     const PreprocessorOptions &PPOpts, const CodeGenOptions &CodeGenOpts,
     const TargetOptions &TargetOpts, const LangOptions &LangOpts,
-    const FileManager &FileMgr, llvm::Module *Module,
-    SmallVector<LinkModule, 4> LinkModules, LLVMContext &C,
-    CoverageSourceInfo *CoverageInfo)
+    llvm::Module *Module, SmallVector<LinkModule, 4> LinkModules,
+    LLVMContext &C, CoverageSourceInfo *CoverageInfo)
     : Diags(Diags), Action(Action), HeaderSearchOpts(HeaderSearchOpts),
       CodeGenOpts(CodeGenOpts), TargetOpts(TargetOpts), LangOpts(LangOpts),
-      FileMgr(FileMgr), Context(nullptr), FS(VFS),
+      Context(nullptr), FS(VFS),
       LLVMIRGeneration("irgen", "LLVM IR Generation Time"),
       LLVMIRGenerationRefCount(0),
       Gen(CreateLLVMCodeGen(Diags, "", std::move(VFS), HeaderSearchOpts, PPOpts,
@@ -232,35 +226,6 @@ void BackendConsumer::HandleInterestingDecl(DeclGroupRef D) {
     HandleTopLevelDecl(D);
 }
 
-bool BackendConsumer::ReloadModules(llvm::Module *M) {
-  for (const CodeGenOptions::BitcodeFileToLink &F :
-       CodeGenOpts.LinkBitcodeFiles) {
-    auto BCBuf = FileMgr.getBufferForFile(F.Filename);
-    if (!BCBuf) {
-      Diags.Report(diag::err_cannot_open_file)
-          << F.Filename << BCBuf.getError().message();
-      LinkModules.clear();
-      return true;
-    }
-
-    LLVMContext &Ctx = getModule()->getContext();
-    Expected<std::unique_ptr<llvm::Module>> ModuleOrErr =
-        getOwningLazyBitcodeModule(std::move(*BCBuf), Ctx);
-
-    if (!ModuleOrErr) {
-      handleAllErrors(ModuleOrErr.takeError(), [&](ErrorInfoBase &EIB) {
-        Diags.Report(diag::err_cannot_open_file) << F.Filename << EIB.message();
-      });
-      LinkModules.clear();
-      return true;
-    }
-    LinkModules.push_back({std::move(ModuleOrErr.get()), F.PropagateAttrs,
-                           F.Internalize, F.LinkFlags});
-  }
-
-  return false; // success
-}
-
 // Links each entry in LinkModules into our module.  Returns true on error.
 bool BackendConsumer::LinkInModules(llvm::Module *M, bool ShouldLinkFiles) {
   for (auto &LM : LinkModules) {
@@ -362,7 +327,7 @@ void BackendConsumer::HandleTranslationUnit(ASTContext &C) {
   }
 
   // Link each LinkModule into our module.
-  if (LinkInModules(getModule()))
+  if (!CodeGenOpts.LinkBitcodePostopt && LinkInModules(getModule()))
     return;
 
   for (auto &F : getModule()->functions()) {
@@ -1055,9 +1020,8 @@ CodeGenAction::CreateASTConsumer(CompilerInstance &CI, StringRef InFile) {
   std::unique_ptr<BackendConsumer> Result(new BackendConsumer(
       BA, CI.getDiagnostics(), &CI.getVirtualFileSystem(),
       CI.getHeaderSearchOpts(), CI.getPreprocessorOpts(), CI.getCodeGenOpts(),
-      CI.getTargetOpts(), CI.getLangOpts(), CI.getFileManager(),
-      std::string(InFile), std::move(LinkModules), std::move(OS), *VMContext,
-      CoverageInfo));
+      CI.getTargetOpts(), CI.getLangOpts(), std::string(InFile),
+      std::move(LinkModules), std::move(OS), *VMContext, CoverageInfo));
   BEConsumer = Result.get();
 
   // Enable generating macro debug info only when debug info is not disabled and
@@ -1228,11 +1192,11 @@ void CodeGenAction::ExecuteAction() {
   BackendConsumer Result(BA, CI.getDiagnostics(), &CI.getVirtualFileSystem(),
                          CI.getHeaderSearchOpts(), CI.getPreprocessorOpts(),
                          CI.getCodeGenOpts(), CI.getTargetOpts(),
-                         CI.getLangOpts(), CI.getFileManager(), TheModule.get(),
+                         CI.getLangOpts(), TheModule.get(),
                          std::move(LinkModules), *VMContext, nullptr);
 
   // Link in each pending link module.
-  if (Result.LinkInModules(&*TheModule))
+  if (!CodeGenOpts.LinkBitcodePostopt && Result.LinkInModules(&*TheModule))
     return;
 
   // PR44896: Force DiscardValueNames as false. DiscardValueNames cannot be

diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -53,6 +53,7 @@
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringSwitch.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
+#include "llvm/BinaryFormat/ELF.h"
 #include "llvm/Frontend/OpenMP/OMPIRBuilder.h"
 #include "llvm/IR/AttributeMask.h"
 #include "llvm/IR/CallingConv.h"
@@ -1190,6 +1191,37 @@ void CodeGenModule::Release() {
     if (!LangOpts.isSignReturnAddressWithAKey())
       getModule().addModuleFlag(llvm::Module::Min,
                                 "sign-return-address-with-bkey", 1);
+
+    if (getTriple().isOSLinux()) {
+      assert(getTriple().isOSBinFormatELF());
+      using namespace llvm::ELF;
+      uint64_t PAuthABIVersion =
+          (LangOpts.PointerAuthIntrinsics
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_INTRINSICS) |
+          (LangOpts.PointerAuthCalls
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_CALLS) |
+          (LangOpts.PointerAuthReturns
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_RETURNS) |
+          (LangOpts.PointerAuthAuthTraps
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_AUTHTRAPS) |
+          (LangOpts.PointerAuthVTPtrAddressDiscrimination
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_VPTRADDRDISCR) |
+          (LangOpts.PointerAuthVTPtrTypeDiscrimination
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_VPTRTYPEDISCR) |
+          (LangOpts.PointerAuthInitFini
+           << AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_INITFINI);
+      static_assert(AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_INITFINI ==
+                        AARCH64_PAUTH_PLATFORM_LLVM_LINUX_VERSION_LAST,
+                    "Update when new enum items are defined");
+      if (PAuthABIVersion != 0) {
+        getModule().addModuleFlag(llvm::Module::Error,
+                                  "aarch64-elf-pauthabi-platform",
+                                  AARCH64_PAUTH_PLATFORM_LLVM_LINUX);
+        getModule().addModuleFlag(llvm::Module::Error,
+                                  "aarch64-elf-pauthabi-version",
+                                  PAuthABIVersion);
+      }
+    }
   }
 
   if (CodeGenOpts.StackClashProtector)

diff --git a/clang/lib/CodeGen/CoverageMappingGen.cpp b/clang/lib/CodeGen/CoverageMappingGen.cpp
@@ -191,10 +191,7 @@ class SourceMappingRegion {
   bool isBranch() const { return FalseCount.has_value(); }
 
   bool isMCDCDecision() const {
-    const auto *DecisionParams =
-        std::get_if<mcdc::DecisionParameters>(&MCDCParams);
-    assert(!DecisionParams || DecisionParams->NumConditions > 0);
-    return DecisionParams;
+    return std::holds_alternative<mcdc::DecisionParameters>(MCDCParams);
   }
 
   const auto &getMCDCDecisionParams() const {

diff --git a/clang/lib/CodeGen/LinkInModulesPass.cpp b/clang/lib/CodeGen/LinkInModulesPass.cpp
@@ -28,12 +28,8 @@ PreservedAnalyses LinkInModulesPass::run(Module &M, ModuleAnalysisManager &AM) {
   if (!BC)
     return PreservedAnalyses::all();
 
-  // Re-load bitcode modules from files
-  if (BC->ReloadModules(&M))
-    report_fatal_error("Bitcode module re-loading failed, aborted!");
-
   if (BC->LinkInModules(&M, ShouldLinkFiles))
-    report_fatal_error("Bitcode module re-linking failed, aborted!");
+    report_fatal_error("Bitcode module postopt linking failed, aborted!");
 
-  return PreservedAnalyses::all();
+  return PreservedAnalyses::none();
 }
diff --git a/clang/lib/CodeGen/MicrosoftCXXABI.cpp b/clang/lib/CodeGen/MicrosoftCXXABI.cpp
@@ -1122,7 +1122,22 @@ static bool isTrivialForMSVC(const CXXRecordDecl *RD, QualType Ty,
   //   No base classes
   //   No virtual functions
   // Additionally, we need to ensure that there is a trivial copy assignment
-  // operator, a trivial destructor and no user-provided constructors.
+  // operator, a trivial destructor, no user-provided constructors and no
+  // deleted copy assignment operator.
+
+  // We need to cover two cases when checking for a deleted copy assignment
+  // operator.
+  //
+  // struct S { int& r; };
+  // The above will have an implicit copy assignment operator that is deleted
+  // and there will not be a `CXXMethodDecl` for the copy assignment operator.
+  // This is handled by the `needsImplicitCopyAssignment()` check below.
+  //
+  // struct S { S& operator=(const S&) = delete; int i; };
+  // The above will not have an implicit copy assignment operator that is
+  // deleted but there is a deleted `CXXMethodDecl` for the declared copy
+  // assignment operator. This is handled by the `isDeleted()` check below.
+
   if (RD->hasProtectedFields() || RD->hasPrivateFields())
     return false;
   if (RD->getNumBases() > 0)
@@ -1131,13 +1146,18 @@ static bool isTrivialForMSVC(const CXXRecordDecl *RD, QualType Ty,
     return false;
   if (RD->hasNonTrivialCopyAssignment())
     return false;
+  if (RD->needsImplicitCopyAssignment() && !RD->hasSimpleCopyAssignment())
+    return false;
   for (const Decl *D : RD->decls()) {
     if (auto *Ctor = dyn_cast<CXXConstructorDecl>(D)) {
       if (Ctor->isUserProvided())
         return false;
     } else if (auto *Template = dyn_cast<FunctionTemplateDecl>(D)) {
       if (isa<CXXConstructorDecl>(Template->getTemplatedDecl()))
         return false;
+    } else if (auto *MethodDecl = dyn_cast<CXXMethodDecl>(D)) {
+      if (MethodDecl->isCopyAssignmentOperator() && MethodDecl->isDeleted())
+        return false;
     }
   }
   if (RD->hasNonTrivialDestructor())

diff --git a/clang/lib/CodeGen/TargetInfo.h b/clang/lib/CodeGen/TargetInfo.h
@@ -94,7 +94,8 @@ class TargetCodeGenInfo {
   virtual void checkFunctionCallABI(CodeGenModule &CGM, SourceLocation CallLoc,
                                     const FunctionDecl *Caller,
                                     const FunctionDecl *Callee,
-                                    const CallArgList &Args) const {}
+                                    const CallArgList &Args,
+                                    QualType ReturnType) const {}
 
   /// Determines the size of struct _Unwind_Exception on this platform,
   /// in 8-bit units.  The Itanium ABI defines this as:

diff --git a/clang/lib/CodeGen/Targets/AArch64.cpp b/clang/lib/CodeGen/Targets/AArch64.cpp
@@ -8,6 +8,7 @@
 
 #include "ABIInfoImpl.h"
 #include "TargetInfo.h"
+#include "clang/AST/Decl.h"
 #include "clang/Basic/DiagnosticFrontend.h"
 #include "llvm/TargetParser/AArch64TargetParser.h"
 
@@ -170,8 +171,22 @@ class AArch64TargetCodeGenInfo : public TargetCodeGenInfo {
 
   void checkFunctionCallABI(CodeGenModule &CGM, SourceLocation CallLoc,
                             const FunctionDecl *Caller,
-                            const FunctionDecl *Callee,
-                            const CallArgList &Args) const override;
+                            const FunctionDecl *Callee, const CallArgList &Args,
+                            QualType ReturnType) const override;
+
+private:
+  // Diagnose calls between functions with incompatible Streaming SVE
+  // attributes.
+  void checkFunctionCallABIStreaming(CodeGenModule &CGM, SourceLocation CallLoc,
+                                     const FunctionDecl *Caller,
+                                     const FunctionDecl *Callee) const;
+  // Diagnose calls which must pass arguments in floating-point registers when
+  // the selected target does not have floating-point registers.
+  void checkFunctionCallABISoftFloat(CodeGenModule &CGM, SourceLocation CallLoc,
+                                     const FunctionDecl *Caller,
+                                     const FunctionDecl *Callee,
+                                     const CallArgList &Args,
+                                     QualType ReturnType) const;
 };
 
 class WindowsAArch64TargetCodeGenInfo : public AArch64TargetCodeGenInfo {
@@ -838,57 +853,56 @@ Address AArch64ABIInfo::EmitMSVAArg(CodeGenFunction &CGF, Address VAListAddr,
                           /*allowHigherAlign*/ false);
 }
 
-static bool isStreaming(const FunctionDecl *F) {
-  if (F->hasAttr<ArmLocallyStreamingAttr>())
-    return true;
-  if (const auto *T = F->getType()->getAs<FunctionProtoType>())
-    return T->getAArch64SMEAttributes() & FunctionType::SME_PStateSMEnabledMask;
-  return false;
-}
-
 static bool isStreamingCompatible(const FunctionDecl *F) {
   if (const auto *T = F->getType()->getAs<FunctionProtoType>())
     return T->getAArch64SMEAttributes() &
            FunctionType::SME_PStateSMCompatibleMask;
   return false;
 }
 
+// Report an error if an argument or return value of type Ty would need to be
+// passed in a floating-point register.
+static void diagnoseIfNeedsFPReg(DiagnosticsEngine &Diags,
+                                 const StringRef ABIName,
+                                 const AArch64ABIInfo &ABIInfo,
+                                 const QualType &Ty, const NamedDecl *D) {
+  const Type *HABase = nullptr;
+  uint64_t HAMembers = 0;
+  if (Ty->isFloatingType() || Ty->isVectorType() ||
+      ABIInfo.isHomogeneousAggregate(Ty, HABase, HAMembers)) {
+    Diags.Report(D->getLocation(), diag::err_target_unsupported_type_for_abi)
+        << D->getDeclName() << Ty << ABIName;
+  }
+}
+
+// If we are using a hard-float ABI, but do not have floating point registers,
+// then report an error for any function arguments or returns which would be
+// passed in floating-pint registers.
 void AArch64TargetCodeGenInfo::checkFunctionABI(
     CodeGenModule &CGM, const FunctionDecl *FuncDecl) const {
   const AArch64ABIInfo &ABIInfo = getABIInfo<AArch64ABIInfo>();
   const TargetInfo &TI = ABIInfo.getContext().getTargetInfo();
 
-  // If we are using a hard-float ABI, but do not have floating point
-  // registers, then report an error for any function arguments or returns
-  // which would be passed in floating-pint registers.
-  auto CheckType = [&CGM, &TI, &ABIInfo](const QualType &Ty,
-                                         const NamedDecl *D) {
-    const Type *HABase = nullptr;
-    uint64_t HAMembers = 0;
-    if (Ty->isFloatingType() || Ty->isVectorType() ||
-        ABIInfo.isHomogeneousAggregate(Ty, HABase, HAMembers)) {
-      CGM.getDiags().Report(D->getLocation(),
-                            diag::err_target_unsupported_type_for_abi)
-          << D->getDeclName() << Ty << TI.getABI();
-    }
-  };
-
   if (!TI.hasFeature("fp") && !ABIInfo.isSoftFloat()) {
-    CheckType(FuncDecl->getReturnType(), FuncDecl);
+    diagnoseIfNeedsFPReg(CGM.getDiags(), TI.getABI(), ABIInfo,
+                         FuncDecl->getReturnType(), FuncDecl);
     for (ParmVarDecl *PVD : FuncDecl->parameters()) {
-      CheckType(PVD->getType(), PVD);
+      diagnoseIfNeedsFPReg(CGM.getDiags(), TI.getABI(), ABIInfo, PVD->getType(),
+                           PVD);
     }
   }
 }
 
-void AArch64TargetCodeGenInfo::checkFunctionCallABI(
+void AArch64TargetCodeGenInfo::checkFunctionCallABIStreaming(
     CodeGenModule &CGM, SourceLocation CallLoc, const FunctionDecl *Caller,
-    const FunctionDecl *Callee, const CallArgList &Args) const {
+    const FunctionDecl *Callee) const {
   if (!Caller || !Callee || !Callee->hasAttr<AlwaysInlineAttr>())
     return;
 
-  bool CallerIsStreaming = isStreaming(Caller);
-  bool CalleeIsStreaming = isStreaming(Callee);
+  bool CallerIsStreaming =
+      IsArmStreamingFunction(Caller, /*IncludeLocallyStreaming=*/true);
+  bool CalleeIsStreaming =
+      IsArmStreamingFunction(Callee, /*IncludeLocallyStreaming=*/true);
   bool CallerIsStreamingCompatible = isStreamingCompatible(Caller);
   bool CalleeIsStreamingCompatible = isStreamingCompatible(Callee);
 
@@ -903,6 +917,37 @@ void AArch64TargetCodeGenInfo::checkFunctionCallABI(
           << Callee->getDeclName();
 }
 
+// If the target does not have floating-point registers, but we are using a
+// hard-float ABI, there is no way to pass floating-point, vector or HFA values
+// to functions, so we report an error.
+void AArch64TargetCodeGenInfo::checkFunctionCallABISoftFloat(
+    CodeGenModule &CGM, SourceLocation CallLoc, const FunctionDecl *Caller,
+    const FunctionDecl *Callee, const CallArgList &Args,
+    QualType ReturnType) const {
+  const AArch64ABIInfo &ABIInfo = getABIInfo<AArch64ABIInfo>();
+  const TargetInfo &TI = ABIInfo.getContext().getTargetInfo();
+
+  if (!Caller || TI.hasFeature("fp") || ABIInfo.isSoftFloat())
+    return;
+
+  diagnoseIfNeedsFPReg(CGM.getDiags(), TI.getABI(), ABIInfo, ReturnType,
+                       Caller);
+
+  for (const CallArg &Arg : Args)
+    diagnoseIfNeedsFPReg(CGM.getDiags(), TI.getABI(), ABIInfo, Arg.getType(),
+                         Caller);
+}
+
+void AArch64TargetCodeGenInfo::checkFunctionCallABI(CodeGenModule &CGM,
+                                                    SourceLocation CallLoc,
+                                                    const FunctionDecl *Caller,
+                                                    const FunctionDecl *Callee,
+                                                    const CallArgList &Args,
+                                                    QualType ReturnType) const {
+  checkFunctionCallABIStreaming(CGM, CallLoc, Caller, Callee);
+  checkFunctionCallABISoftFloat(CGM, CallLoc, Caller, Callee, Args, ReturnType);
+}
+
 void AArch64ABIInfo::appendAttributeMangling(TargetClonesAttr *Attr,
                                              unsigned Index,
                                              raw_ostream &Out) const {

diff --git a/clang/lib/CodeGen/Targets/X86.cpp b/clang/lib/CodeGen/Targets/X86.cpp
@@ -1482,8 +1482,8 @@ class X86_64TargetCodeGenInfo : public TargetCodeGenInfo {
 
   void checkFunctionCallABI(CodeGenModule &CGM, SourceLocation CallLoc,
                             const FunctionDecl *Caller,
-                            const FunctionDecl *Callee,
-                            const CallArgList &Args) const override;
+                            const FunctionDecl *Callee, const CallArgList &Args,
+                            QualType ReturnType) const override;
 };
 } // namespace
 
@@ -1558,9 +1558,15 @@ static bool checkAVXParam(DiagnosticsEngine &Diag, ASTContext &Ctx,
   return false;
 }
 
-void X86_64TargetCodeGenInfo::checkFunctionCallABI(
-    CodeGenModule &CGM, SourceLocation CallLoc, const FunctionDecl *Caller,
-    const FunctionDecl *Callee, const CallArgList &Args) const {
+void X86_64TargetCodeGenInfo::checkFunctionCallABI(CodeGenModule &CGM,
+                                                   SourceLocation CallLoc,
+                                                   const FunctionDecl *Caller,
+                                                   const FunctionDecl *Callee,
+                                                   const CallArgList &Args,
+                                                   QualType ReturnType) const {
+  if (!Callee)
+    return;
+
   llvm::StringMap<bool> CallerMap;
   llvm::StringMap<bool> CalleeMap;
   unsigned ArgIndex = 0;

diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
@@ -564,9 +564,9 @@ static llvm::Triple computeTargetTriple(const Driver &D,
       StringRef ObjectMode = *ObjectModeValue;
       llvm::Triple::ArchType AT = llvm::Triple::UnknownArch;
 
-      if (ObjectMode.equals("64")) {
+      if (ObjectMode == "64") {
         AT = Target.get64BitArchVariant().getArch();
-      } else if (ObjectMode.equals("32")) {
+      } else if (ObjectMode == "32") {
         AT = Target.get32BitArchVariant().getArch();
       } else {
         D.Diag(diag::err_drv_invalid_object_mode) << ObjectMode;
@@ -6694,7 +6694,7 @@ llvm::StringRef clang::driver::getDriverMode(StringRef ProgName,
   return Opt.consume_front(OptName) ? Opt : "";
 }
 
-bool driver::IsClangCL(StringRef DriverMode) { return DriverMode.equals("cl"); }
+bool driver::IsClangCL(StringRef DriverMode) { return DriverMode == "cl"; }
 
 llvm::Error driver::expandResponseFiles(SmallVectorImpl<const char *> &Args,
                                         bool ClangCLMode,

diff --git a/clang/lib/Driver/OffloadBundler.cpp b/clang/lib/Driver/OffloadBundler.cpp
@@ -113,8 +113,11 @@ bool OffloadTargetInfo::isOffloadKindValid() const {
 
 bool OffloadTargetInfo::isOffloadKindCompatible(
     const StringRef TargetOffloadKind) const {
-  if (OffloadKind == TargetOffloadKind)
+  if ((OffloadKind == TargetOffloadKind) ||
+      (OffloadKind == "hip" && TargetOffloadKind == "hipv4") ||
+      (OffloadKind == "hipv4" && TargetOffloadKind == "hip"))
     return true;
+
   if (BundlerConfig.HipOpenmpCompatible) {
     bool HIPCompatibleWithOpenMP = OffloadKind.starts_with_insensitive("hip") &&
                                    TargetOffloadKind == "openmp";

diff --git a/clang/lib/Driver/ToolChains/AIX.cpp b/clang/lib/Driver/ToolChains/AIX.cpp
@@ -481,8 +481,8 @@ static void addTocDataOptions(const llvm::opt::ArgList &Args,
 
   // Currently only supported for small code model.
   if (TOCDataGloballyinEffect &&
-      (Args.getLastArgValue(options::OPT_mcmodel_EQ).equals("large") ||
-       Args.getLastArgValue(options::OPT_mcmodel_EQ).equals("medium"))) {
+      (Args.getLastArgValue(options::OPT_mcmodel_EQ) == "large" ||
+       Args.getLastArgValue(options::OPT_mcmodel_EQ) == "medium")) {
     D.Diag(clang::diag::warn_drv_unsupported_tocdata);
     return;
   }

diff --git a/clang/lib/Driver/ToolChains/AMDGPU.cpp b/clang/lib/Driver/ToolChains/AMDGPU.cpp
@@ -732,7 +732,7 @@ AMDGPUToolChain::TranslateArgs(const DerivedArgList &Args, StringRef BoundArch,
 
   checkTargetID(*DAL);
 
-  if (!Args.getLastArgValue(options::OPT_x).equals("cl"))
+  if (Args.getLastArgValue(options::OPT_x) != "cl")
     return DAL;
 
   // Phase 1 (.cl -> .bc)