diff --git a/clang/lib/Sema/SemaTemplate.cpp b/clang/lib/Sema/SemaTemplate.cpp
@@ -881,20 +881,23 @@ bool Sema::DiagnoseUninstantiableTemplate(SourceLocation PointOfInstantiation,
   return true;
 }
 
-/// DiagnoseTemplateParameterShadow - Produce a diagnostic complaining
-/// that the template parameter 'PrevDecl' is being shadowed by a new
-/// declaration at location Loc. Returns true to indicate that this is
-/// an error, and false otherwise.
-void Sema::DiagnoseTemplateParameterShadow(SourceLocation Loc, Decl *PrevDecl) {
+void Sema::DiagnoseTemplateParameterShadow(SourceLocation Loc, Decl *PrevDecl,
+                                           bool SupportedForCompatibility) {
   assert(PrevDecl->isTemplateParameter() && "Not a template parameter");
 
-  // C++ [temp.local]p4:
-  //   A template-parameter shall not be redeclared within its
-  //   scope (including nested scopes).
+  // C++23 [temp.local]p6:
+  //   The name of a template-parameter shall not be bound to any following.
+  //   declaration whose locus is contained by the scope to which the
+  //   template-parameter belongs.
   //
-  // Make this a warning when MSVC compatibility is requested.
-  unsigned DiagId = getLangOpts().MSVCCompat ? diag::ext_template_param_shadow
-                                             : diag::err_template_param_shadow;
+  // When MSVC compatibility is enabled, the diagnostic is always a warning
+  // by default. Otherwise, it an error unless SupportedForCompatibility is
+  // true, in which case it is a default-to-error warning.
+  unsigned DiagId =
+      getLangOpts().MSVCCompat
+          ? diag::ext_template_param_shadow
+          : (SupportedForCompatibility ? diag::ext_compat_template_param_shadow
+                                       : diag::err_template_param_shadow);
   const auto *ND = cast<NamedDecl>(PrevDecl);
   Diag(Loc, DiagId) << ND->getDeclName();
   NoteTemplateParameterLocation(*ND);
@@ -8502,9 +8505,7 @@ Sema::CheckTemplateDeclScope(Scope *S, TemplateParameterList *TemplateParams) {
     return false;
 
   // Find the nearest enclosing declaration scope.
-  while ((S->getFlags() & Scope::DeclScope) == 0 ||
-         (S->getFlags() & Scope::TemplateParamScope) != 0)
-    S = S->getParent();
+  S = S->getDeclParent();
 
   // C++ [temp.pre]p6: [P2096]
   //   A template, explicit specialization, or partial specialization shall not
@@ -10619,11 +10620,8 @@ DeclResult Sema::ActOnExplicitInstantiation(Scope *S,
     return true;
   }
 
-  // The scope passed in may not be a decl scope.  Zip up the scope tree until
-  // we find one that is.
-  while ((S->getFlags() & Scope::DeclScope) == 0 ||
-         (S->getFlags() & Scope::TemplateParamScope) != 0)
-    S = S->getParent();
+  // Get the innermost enclosing declaration scope.
+  S = S->getDeclParent();
 
   // Determine the type of the declaration.
   TypeSourceInfo *T = GetTypeForDeclarator(D);

diff --git a/clang/lib/StaticAnalyzer/Checkers/ArrayBoundCheckerV2.cpp b/clang/lib/StaticAnalyzer/Checkers/ArrayBoundCheckerV2.cpp
@@ -301,21 +301,27 @@ compareValueToThreshold(ProgramStateRef State, NonLoc Value, NonLoc Threshold,
   // calling `evalBinOpNN`:
   if (isNegative(SVB, State, Value) && isUnsigned(SVB, Threshold)) {
     if (CheckEquality) {
-      // negative_value == unsigned_value is always false
+      // negative_value == unsigned_threshold is always false
       return {nullptr, State};
     }
-    // negative_value < unsigned_value is always false
+    // negative_value < unsigned_threshold is always true
     return {State, nullptr};
   }
   if (isUnsigned(SVB, Value) && isNegative(SVB, State, Threshold)) {
-    // unsigned_value == negative_value and unsigned_value < negative_value are
-    // both always false
+    // unsigned_value == negative_threshold and
+    // unsigned_value < negative_threshold are both always false
     return {nullptr, State};
   }
-  // FIXME: these special cases are sufficient for handling real-world
+  // FIXME: These special cases are sufficient for handling real-world
   // comparisons, but in theory there could be contrived situations where
   // automatic conversion of a symbolic value (which can be negative and can be
   // positive) leads to incorrect results.
+  // NOTE: We NEED to use the `evalBinOpNN` call in the "common" case, because
+  // we want to ensure that assumptions coming from this precondition and
+  // assumptions coming from regular C/C++ operator calls are represented by
+  // constraints on the same symbolic expression. A solution that would
+  // evaluate these "mathematical" compariosns through a separate pathway would
+  // be a step backwards in this sense.
 
   const BinaryOperatorKind OpKind = CheckEquality ? BO_EQ : BO_LT;
   auto BelowThreshold =

diff --git a/clang/lib/StaticAnalyzer/Checkers/CStringChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/CStringChecker.cpp
@@ -124,48 +124,47 @@ class CStringChecker : public Checker< eval::Call,
                                      const CallEvent &)>;
 
   CallDescriptionMap<FnCheck> Callbacks = {
-      {{CDF_MaybeBuiltin, {"memcpy"}, 3},
+      {{CDM::CLibrary, {"memcpy"}, 3},
        std::bind(&CStringChecker::evalMemcpy, _1, _2, _3, CK_Regular)},
-      {{CDF_MaybeBuiltin, {"wmemcpy"}, 3},
+      {{CDM::CLibrary, {"wmemcpy"}, 3},
        std::bind(&CStringChecker::evalMemcpy, _1, _2, _3, CK_Wide)},
-      {{CDF_MaybeBuiltin, {"mempcpy"}, 3},
+      {{CDM::CLibrary, {"mempcpy"}, 3},
        std::bind(&CStringChecker::evalMempcpy, _1, _2, _3, CK_Regular)},
-      {{CDF_None, {"wmempcpy"}, 3},
+      {{CDM::Unspecified, {"wmempcpy"}, 3},
        std::bind(&CStringChecker::evalMempcpy, _1, _2, _3, CK_Wide)},
-      {{CDF_MaybeBuiltin, {"memcmp"}, 3},
+      {{CDM::CLibrary, {"memcmp"}, 3},
        std::bind(&CStringChecker::evalMemcmp, _1, _2, _3, CK_Regular)},
-      {{CDF_MaybeBuiltin, {"wmemcmp"}, 3},
+      {{CDM::CLibrary, {"wmemcmp"}, 3},
        std::bind(&CStringChecker::evalMemcmp, _1, _2, _3, CK_Wide)},
-      {{CDF_MaybeBuiltin, {"memmove"}, 3},
+      {{CDM::CLibrary, {"memmove"}, 3},
        std::bind(&CStringChecker::evalMemmove, _1, _2, _3, CK_Regular)},
-      {{CDF_MaybeBuiltin, {"wmemmove"}, 3},
+      {{CDM::CLibrary, {"wmemmove"}, 3},
        std::bind(&CStringChecker::evalMemmove, _1, _2, _3, CK_Wide)},
-      {{CDF_MaybeBuiltin, {"memset"}, 3}, &CStringChecker::evalMemset},
-      {{CDF_MaybeBuiltin, {"explicit_memset"}, 3}, &CStringChecker::evalMemset},
-      {{CDF_MaybeBuiltin, {"strcpy"}, 2}, &CStringChecker::evalStrcpy},
-      {{CDF_MaybeBuiltin, {"strncpy"}, 3}, &CStringChecker::evalStrncpy},
-      {{CDF_MaybeBuiltin, {"stpcpy"}, 2}, &CStringChecker::evalStpcpy},
-      {{CDF_MaybeBuiltin, {"strlcpy"}, 3}, &CStringChecker::evalStrlcpy},
-      {{CDF_MaybeBuiltin, {"strcat"}, 2}, &CStringChecker::evalStrcat},
-      {{CDF_MaybeBuiltin, {"strncat"}, 3}, &CStringChecker::evalStrncat},
-      {{CDF_MaybeBuiltin, {"strlcat"}, 3}, &CStringChecker::evalStrlcat},
-      {{CDF_MaybeBuiltin, {"strlen"}, 1}, &CStringChecker::evalstrLength},
-      {{CDF_MaybeBuiltin, {"wcslen"}, 1}, &CStringChecker::evalstrLength},
-      {{CDF_MaybeBuiltin, {"strnlen"}, 2}, &CStringChecker::evalstrnLength},
-      {{CDF_MaybeBuiltin, {"wcsnlen"}, 2}, &CStringChecker::evalstrnLength},
-      {{CDF_MaybeBuiltin, {"strcmp"}, 2}, &CStringChecker::evalStrcmp},
-      {{CDF_MaybeBuiltin, {"strncmp"}, 3}, &CStringChecker::evalStrncmp},
-      {{CDF_MaybeBuiltin, {"strcasecmp"}, 2}, &CStringChecker::evalStrcasecmp},
-      {{CDF_MaybeBuiltin, {"strncasecmp"}, 3},
-       &CStringChecker::evalStrncasecmp},
-      {{CDF_MaybeBuiltin, {"strsep"}, 2}, &CStringChecker::evalStrsep},
-      {{CDF_MaybeBuiltin, {"bcopy"}, 3}, &CStringChecker::evalBcopy},
-      {{CDF_MaybeBuiltin, {"bcmp"}, 3},
+      {{CDM::CLibrary, {"memset"}, 3}, &CStringChecker::evalMemset},
+      {{CDM::CLibrary, {"explicit_memset"}, 3}, &CStringChecker::evalMemset},
+      {{CDM::CLibrary, {"strcpy"}, 2}, &CStringChecker::evalStrcpy},
+      {{CDM::CLibrary, {"strncpy"}, 3}, &CStringChecker::evalStrncpy},
+      {{CDM::CLibrary, {"stpcpy"}, 2}, &CStringChecker::evalStpcpy},
+      {{CDM::CLibrary, {"strlcpy"}, 3}, &CStringChecker::evalStrlcpy},
+      {{CDM::CLibrary, {"strcat"}, 2}, &CStringChecker::evalStrcat},
+      {{CDM::CLibrary, {"strncat"}, 3}, &CStringChecker::evalStrncat},
+      {{CDM::CLibrary, {"strlcat"}, 3}, &CStringChecker::evalStrlcat},
+      {{CDM::CLibrary, {"strlen"}, 1}, &CStringChecker::evalstrLength},
+      {{CDM::CLibrary, {"wcslen"}, 1}, &CStringChecker::evalstrLength},
+      {{CDM::CLibrary, {"strnlen"}, 2}, &CStringChecker::evalstrnLength},
+      {{CDM::CLibrary, {"wcsnlen"}, 2}, &CStringChecker::evalstrnLength},
+      {{CDM::CLibrary, {"strcmp"}, 2}, &CStringChecker::evalStrcmp},
+      {{CDM::CLibrary, {"strncmp"}, 3}, &CStringChecker::evalStrncmp},
+      {{CDM::CLibrary, {"strcasecmp"}, 2}, &CStringChecker::evalStrcasecmp},
+      {{CDM::CLibrary, {"strncasecmp"}, 3}, &CStringChecker::evalStrncasecmp},
+      {{CDM::CLibrary, {"strsep"}, 2}, &CStringChecker::evalStrsep},
+      {{CDM::CLibrary, {"bcopy"}, 3}, &CStringChecker::evalBcopy},
+      {{CDM::CLibrary, {"bcmp"}, 3},
        std::bind(&CStringChecker::evalMemcmp, _1, _2, _3, CK_Regular)},
-      {{CDF_MaybeBuiltin, {"bzero"}, 2}, &CStringChecker::evalBzero},
-      {{CDF_MaybeBuiltin, {"explicit_bzero"}, 2}, &CStringChecker::evalBzero},
-      {{CDF_MaybeBuiltin, {"sprintf"}, 2}, &CStringChecker::evalSprintf},
-      {{CDF_MaybeBuiltin, {"snprintf"}, 2}, &CStringChecker::evalSnprintf},
+      {{CDM::CLibrary, {"bzero"}, 2}, &CStringChecker::evalBzero},
+      {{CDM::CLibrary, {"explicit_bzero"}, 2}, &CStringChecker::evalBzero},
+      {{CDM::CLibrary, {"sprintf"}, 2}, &CStringChecker::evalSprintf},
+      {{CDM::CLibrary, {"snprintf"}, 2}, &CStringChecker::evalSnprintf},
   };
 
   // These require a bit of special handling.

diff --git a/clang/lib/StaticAnalyzer/Checkers/CheckerDocumentation.cpp b/clang/lib/StaticAnalyzer/Checkers/CheckerDocumentation.cpp
@@ -226,7 +226,7 @@ class CheckerDocumentation : public Checker< check::PreStmt<ReturnStmt>,
   /// first one wins.
   ///
   /// eval::Call
-  bool evalCall(const CallExpr *CE, CheckerContext &C) const { return true; }
+  bool evalCall(const CallEvent &Call, CheckerContext &C) const { return true; }
 
   /// Handles assumptions on symbolic values.
   ///

diff --git a/clang/lib/StaticAnalyzer/Checkers/GenericTaintChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/GenericTaintChecker.cpp
@@ -718,28 +718,24 @@ void GenericTaintChecker::initTaintRules(CheckerContext &C) const {
       {{{"isupper"}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
       {{{"isxdigit"}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
 
-      {{CDF_MaybeBuiltin, {BI.getName(Builtin::BIstrncat)}},
+      {{CDM::CLibrary, {BI.getName(Builtin::BIstrncat)}},
        TR::Prop({{1, 2}}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {BI.getName(Builtin::BIstrlcpy)}},
+      {{CDM::CLibrary, {BI.getName(Builtin::BIstrlcpy)}},
        TR::Prop({{1, 2}}, {{0}})},
-      {{CDF_MaybeBuiltin, {BI.getName(Builtin::BIstrlcat)}},
+      {{CDM::CLibrary, {BI.getName(Builtin::BIstrlcat)}},
        TR::Prop({{1, 2}}, {{0}})},
-      {{CDF_MaybeBuiltin, {{"snprintf"}}},
+      {{CDM::CLibrary, {{"snprintf"}}},
        TR::Prop({{1}, 3}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"sprintf"}}},
+      {{CDM::CLibrary, {{"sprintf"}}},
        TR::Prop({{1}, 2}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"strcpy"}}},
+      {{CDM::CLibrary, {{"strcpy"}}}, TR::Prop({{1}}, {{0, ReturnValueIndex}})},
+      {{CDM::CLibrary, {{"stpcpy"}}}, TR::Prop({{1}}, {{0, ReturnValueIndex}})},
+      {{CDM::CLibrary, {{"strcat"}}}, TR::Prop({{1}}, {{0, ReturnValueIndex}})},
+      {{CDM::CLibrary, {{"wcsncat"}}},
        TR::Prop({{1}}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"stpcpy"}}},
-       TR::Prop({{1}}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"strcat"}}},
-       TR::Prop({{1}}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"wcsncat"}}},
-       TR::Prop({{1}}, {{0, ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"strdup"}}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"strdupa"}}},
-       TR::Prop({{0}}, {{ReturnValueIndex}})},
-      {{CDF_MaybeBuiltin, {{"wcsdup"}}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
+      {{CDM::CLibrary, {{"strdup"}}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
+      {{CDM::CLibrary, {{"strdupa"}}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
+      {{CDM::CLibrary, {{"wcsdup"}}}, TR::Prop({{0}}, {{ReturnValueIndex}})},
 
       // Sinks
       {{{"system"}}, TR::Sink({{0}}, MsgSanitizeSystemArgs)},
@@ -753,31 +749,29 @@ void GenericTaintChecker::initTaintRules(CheckerContext &C) const {
       {{{"execvp"}}, TR::Sink({{0, 1}}, MsgSanitizeSystemArgs)},
       {{{"execvpe"}}, TR::Sink({{0, 1, 2}}, MsgSanitizeSystemArgs)},
       {{{"dlopen"}}, TR::Sink({{0}}, MsgSanitizeSystemArgs)},
-      {{CDF_MaybeBuiltin, {{"malloc"}}}, TR::Sink({{0}}, MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {{"calloc"}}}, TR::Sink({{0}}, MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {{"alloca"}}}, TR::Sink({{0}}, MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {{"memccpy"}}},
-       TR::Sink({{3}}, MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {{"realloc"}}},
-       TR::Sink({{1}}, MsgTaintedBufferSize)},
+      {{CDM::CLibrary, {{"malloc"}}}, TR::Sink({{0}}, MsgTaintedBufferSize)},
+      {{CDM::CLibrary, {{"calloc"}}}, TR::Sink({{0}}, MsgTaintedBufferSize)},
+      {{CDM::CLibrary, {{"alloca"}}}, TR::Sink({{0}}, MsgTaintedBufferSize)},
+      {{CDM::CLibrary, {{"memccpy"}}}, TR::Sink({{3}}, MsgTaintedBufferSize)},
+      {{CDM::CLibrary, {{"realloc"}}}, TR::Sink({{1}}, MsgTaintedBufferSize)},
       {{{{"setproctitle"}}}, TR::Sink({{0}, 1}, MsgUncontrolledFormatString)},
       {{{{"setproctitle_fast"}}},
        TR::Sink({{0}, 1}, MsgUncontrolledFormatString)},
 
       // SinkProps
-      {{CDF_MaybeBuiltin, BI.getName(Builtin::BImemcpy)},
+      {{CDM::CLibrary, BI.getName(Builtin::BImemcpy)},
        TR::SinkProp({{2}}, {{1, 2}}, {{0, ReturnValueIndex}},
                     MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {BI.getName(Builtin::BImemmove)}},
+      {{CDM::CLibrary, {BI.getName(Builtin::BImemmove)}},
        TR::SinkProp({{2}}, {{1, 2}}, {{0, ReturnValueIndex}},
                     MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {BI.getName(Builtin::BIstrncpy)}},
+      {{CDM::CLibrary, {BI.getName(Builtin::BIstrncpy)}},
        TR::SinkProp({{2}}, {{1, 2}}, {{0, ReturnValueIndex}},
                     MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {BI.getName(Builtin::BIstrndup)}},
+      {{CDM::CLibrary, {BI.getName(Builtin::BIstrndup)}},
        TR::SinkProp({{1}}, {{0, 1}}, {{ReturnValueIndex}},
                     MsgTaintedBufferSize)},
-      {{CDF_MaybeBuiltin, {{"bcopy"}}},
+      {{CDM::CLibrary, {{"bcopy"}}},
        TR::SinkProp({{2}}, {{0, 2}}, {{1}}, MsgTaintedBufferSize)}};
 
   // `getenv` returns taint only in untrusted environments.

diff --git a/clang/lib/StaticAnalyzer/Checkers/MallocChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/MallocChecker.cpp
@@ -410,13 +410,13 @@ class MallocChecker
       {{{"malloc"}, 3}, &MallocChecker::checkKernelMalloc},
       {{{"calloc"}, 2}, &MallocChecker::checkCalloc},
       {{{"valloc"}, 1}, &MallocChecker::checkBasicAlloc},
-      {{CDF_MaybeBuiltin, {"strndup"}, 2}, &MallocChecker::checkStrdup},
-      {{CDF_MaybeBuiltin, {"strdup"}, 1}, &MallocChecker::checkStrdup},
+      {{CDM::CLibrary, {"strndup"}, 2}, &MallocChecker::checkStrdup},
+      {{CDM::CLibrary, {"strdup"}, 1}, &MallocChecker::checkStrdup},
       {{{"_strdup"}, 1}, &MallocChecker::checkStrdup},
       {{{"kmalloc"}, 2}, &MallocChecker::checkKernelMalloc},
       {{{"if_nameindex"}, 1}, &MallocChecker::checkIfNameIndex},
-      {{CDF_MaybeBuiltin, {"wcsdup"}, 1}, &MallocChecker::checkStrdup},
-      {{CDF_MaybeBuiltin, {"_wcsdup"}, 1}, &MallocChecker::checkStrdup},
+      {{CDM::CLibrary, {"wcsdup"}, 1}, &MallocChecker::checkStrdup},
+      {{CDM::CLibrary, {"_wcsdup"}, 1}, &MallocChecker::checkStrdup},
       {{{"g_malloc"}, 1}, &MallocChecker::checkBasicAlloc},
       {{{"g_malloc0"}, 1}, &MallocChecker::checkGMalloc0},
       {{{"g_try_malloc"}, 1}, &MallocChecker::checkBasicAlloc},

diff --git a/clang/lib/StaticAnalyzer/Core/CallDescription.cpp b/clang/lib/StaticAnalyzer/Core/CallDescription.cpp
@@ -35,13 +35,13 @@ static MaybeCount readRequiredParams(MaybeCount RequiredArgs,
   return std::nullopt;
 }
 
-ento::CallDescription::CallDescription(CallDescriptionFlags Flags,
+ento::CallDescription::CallDescription(Mode MatchAs,
                                        ArrayRef<StringRef> QualifiedName,
                                        MaybeCount RequiredArgs /*= None*/,
                                        MaybeCount RequiredParams /*= None*/)
     : RequiredArgs(RequiredArgs),
       RequiredParams(readRequiredParams(RequiredArgs, RequiredParams)),
-      Flags(Flags) {
+      MatchAs(MatchAs) {
   assert(!QualifiedName.empty());
   this->QualifiedName.reserve(QualifiedName.size());
   llvm::transform(QualifiedName, std::back_inserter(this->QualifiedName),
@@ -52,7 +52,8 @@ ento::CallDescription::CallDescription(CallDescriptionFlags Flags,
 ento::CallDescription::CallDescription(ArrayRef<StringRef> QualifiedName,
                                        MaybeCount RequiredArgs /*= None*/,
                                        MaybeCount RequiredParams /*= None*/)
-    : CallDescription(CDF_None, QualifiedName, RequiredArgs, RequiredParams) {}
+    : CallDescription(Mode::Unspecified, QualifiedName, RequiredArgs,
+                      RequiredParams) {}
 
 bool ento::CallDescription::matches(const CallEvent &Call) const {
   // FIXME: Add ObjC Message support.
@@ -74,14 +75,20 @@ bool ento::CallDescription::matchesAsWritten(const CallExpr &CE) const {
   return matchesImpl(FD, CE.getNumArgs(), FD->param_size());
 }
 
-bool ento::CallDescription::matchesImpl(const FunctionDecl *Callee,
-                                        size_t ArgCount,
+bool ento::CallDescription::matchesImpl(const FunctionDecl *FD, size_t ArgCount,
                                         size_t ParamCount) const {
-  const auto *FD = Callee;
   if (!FD)
     return false;
 
-  if (Flags & CDF_MaybeBuiltin) {
+  const bool isMethod = isa<CXXMethodDecl>(FD);
+
+  if (MatchAs == Mode::SimpleFunc && isMethod)
+    return false;
+
+  if (MatchAs == Mode::CXXMethod && !isMethod)
+    return false;
+
+  if (MatchAs == Mode::CLibrary) {
     return CheckerContext::isCLibraryFunction(FD, getFunctionName()) &&
            (!RequiredArgs || *RequiredArgs <= ArgCount) &&
            (!RequiredParams || *RequiredParams <= ParamCount);

diff --git a/clang/lib/Tooling/CMakeLists.txt b/clang/lib/Tooling/CMakeLists.txt
@@ -135,4 +135,5 @@ add_clang_library(clangTooling
   clangRewrite
   clangSerialization
   clangToolingCore
+  ${LLVM_PTHREAD_LIB}
   )
diff --git a/clang/test/AST/Interp/arrays.cpp b/clang/test/AST/Interp/arrays.cpp
@@ -138,6 +138,8 @@ constexpr int dependent[4] = {
 static_assert(dependent[2] == dependent[0], "");
 static_assert(dependent[3] == dependent[1], "");
 
+union { char x[]; } r = {0};
+
 #pragma clang diagnostic push
 #pragma clang diagnostic ignored "-Wc99-extensions"
 #pragma clang diagnostic ignored "-Winitializer-overrides"

diff --git a/clang/test/AST/Interp/builtin-functions.cpp b/clang/test/AST/Interp/builtin-functions.cpp
@@ -34,6 +34,14 @@ namespace strcmp {
   static_assert(__builtin_strcmp(kFoobar, kFoobazfoobar + 6) == 0, ""); // both-error {{not an integral constant}} \
                                                                         // both-note {{dereferenced one-past-the-end}} \
                                                                         // expected-note {{in call to}}
+
+  /// Used to assert because we're passing a dummy pointer to
+  /// __builtin_strcmp() when evaluating the return statement.
+  constexpr bool char_memchr_mutable() {
+    char buffer[] = "mutable";
+    return __builtin_strcmp(buffer, "mutable") == 0;
+  }
+  static_assert(char_memchr_mutable(), "");
 }
 
 /// Copied from constant-expression-cxx11.cpp

diff --git a/clang/test/AST/Interp/complex.cpp b/clang/test/AST/Interp/complex.cpp
@@ -1,8 +1,5 @@
-// RUN: %clang_cc1 -fexperimental-new-constant-interpreter -verify -Wno-unused-value %s
-// RUN: %clang_cc1 -verify=ref -Wno-unused-value %s
-
-// expected-no-diagnostics
-// ref-no-diagnostics
+// RUN: %clang_cc1 -fexperimental-new-constant-interpreter -verify=both,expected -Wno-unused-value %s
+// RUN: %clang_cc1 -verify=both,ref -Wno-unused-value %s
 
 constexpr _Complex double z1 = {1.0, 2.0};
 static_assert(__real(z1) == 1.0, "");
@@ -256,3 +253,16 @@ namespace DeclRefCopy {
   }
   static_assert(localComplexArray() == (24 + 42), "");
 }
+
+namespace Builtin {
+  constexpr _Complex float A = __builtin_complex(10.0f, 20.0f);
+  static_assert(__real(A) == 10, "");
+  static_assert(__imag(A) == 20, "");
+
+  constexpr _Complex double B = __builtin_complex(10.0, 20.0);
+  static_assert(__real(B) == 10, "");
+  static_assert(__imag(B) == 20, "");
+
+
+  constexpr _Complex float C = __builtin_complex(10.0f, 20.0); // both-error {{arguments are of different types}}
+}
diff --git a/clang/test/AST/Interp/literals.cpp b/clang/test/AST/Interp/literals.cpp
@@ -1168,3 +1168,10 @@ namespace incdecbool {
 
 
 }
+
+#if __cplusplus >= 201402L
+constexpr int externvar1() { // both-error {{never produces a constant expression}}
+  extern char arr[]; // both-note {{declared here}}
+  return arr[0]; // both-note {{read of non-constexpr variable 'arr'}}
+}
+#endif
diff --git a/clang/test/Analysis/analyzer-config.c b/clang/test/Analysis/analyzer-config.c
@@ -129,7 +129,7 @@
 // CHECK-NEXT: unix.DynamicMemoryModeling:Optimistic = false
 // CHECK-NEXT: unix.Errno:AllowErrnoReadOutsideConditionExpressions = true
 // CHECK-NEXT: unix.StdCLibraryFunctions:DisplayLoadedSummaries = false
-// CHECK-NEXT: unix.StdCLibraryFunctions:ModelPOSIX = false
+// CHECK-NEXT: unix.StdCLibraryFunctions:ModelPOSIX = true
 // CHECK-NEXT: unroll-loops = false
 // CHECK-NEXT: verbose-report-filename = false
 // CHECK-NEXT: widen-loops = false
diff --git a/clang/test/Analysis/std-c-library-functions.c b/clang/test/Analysis/std-c-library-functions.c
@@ -3,6 +3,7 @@
 // RUN:   -analyzer-checker=unix.StdCLibraryFunctions \
 // RUN:   -analyzer-checker=debug.ExprInspection \
 // RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -analyzer-config unix.StdCLibraryFunctions:ModelPOSIX=false \
 // RUN:   -triple i686-unknown-linux \
 // RUN:   -verify
 
@@ -11,6 +12,7 @@
 // RUN:   -analyzer-checker=unix.StdCLibraryFunctions \
 // RUN:   -analyzer-checker=debug.ExprInspection \
 // RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -analyzer-config unix.StdCLibraryFunctions:ModelPOSIX=false \
 // RUN:   -triple x86_64-unknown-linux \
 // RUN:   -verify
 
@@ -19,6 +21,7 @@
 // RUN:   -analyzer-checker=unix.StdCLibraryFunctions \
 // RUN:   -analyzer-checker=debug.ExprInspection \
 // RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -analyzer-config unix.StdCLibraryFunctions:ModelPOSIX=false \
 // RUN:   -triple armv7-a15-linux \
 // RUN:   -verify
 
@@ -27,6 +30,7 @@
 // RUN:   -analyzer-checker=unix.StdCLibraryFunctions \
 // RUN:   -analyzer-checker=debug.ExprInspection \
 // RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -analyzer-config unix.StdCLibraryFunctions:ModelPOSIX=false \
 // RUN:   -triple thumbv7-a15-linux \
 // RUN:   -verify
 
@@ -36,6 +40,7 @@
 // RUN:   -analyzer-config unix.StdCLibraryFunctions:DisplayLoadedSummaries=true \
 // RUN:   -analyzer-checker=debug.ExprInspection \
 // RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -analyzer-config unix.StdCLibraryFunctions:ModelPOSIX=false \
 // RUN:   -triple i686-unknown-linux 2>&1 | FileCheck %s
 
 //      CHECK: Loaded summary for: int isalnum(int)

diff --git a/clang/test/CXX/temp/temp.res/temp.local/p6.cpp b/clang/test/CXX/temp/temp.res/temp.local/p6.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -verify %s -fcxx-exceptions -std=c++1y
+// RUN: %clang_cc1 -verify %s -fcxx-exceptions -std=c++1y -Wno-error=strict-primary-template-shadow
 
 namespace N {}
 
@@ -127,16 +127,30 @@ template<int T> struct Z { // expected-note 16{{declared here}}
 template<typename T> // expected-note {{declared here}}
 void f(int T) {} // expected-error {{declaration of 'T' shadows template parameter}}
 
-// FIXME: These are ill-formed: a template-parameter shall not have the same name as the template name.
 namespace A {
   template<typename T> struct T {};  // expected-error{{declaration of 'T' shadows template parameter}}
                                      // expected-note@-1{{template parameter is declared here}}
+  template<typename T> struct U {
+    template<typename V> struct V {}; // expected-error{{declaration of 'V' shadows template parameter}}
+                                      // expected-note@-1{{template parameter is declared here}}
+  };
 }
 namespace B {
-  template<typename T> void T() {}
+  template<typename T> void T() {} // expected-warning{{declaration of 'T' shadows template parameter}}
+                                   // expected-note@-1{{template parameter is declared here}}
+
+  template<typename T> struct U {
+    template<typename V> void V(); // expected-warning{{declaration of 'V' shadows template parameter}}
+                                   // expected-note@-1{{template parameter is declared here}}
+  };
 }
 namespace C {
-  template<typename T> int T;
+  template<typename T> int T; // expected-warning{{declaration of 'T' shadows template parameter}}
+                              // expected-note@-1{{template parameter is declared here}}
+  template<typename T> struct U {
+    template<typename V> static int V; // expected-warning{{declaration of 'V' shadows template parameter}}
+                                       // expected-note@-1{{template parameter is declared here}}
+  };
 }
 
 namespace PR28023 {

diff --git a/clang/test/CodeGen/PowerPC/builtins-ppc-xlcompat-rotate.c b/clang/test/CodeGen/PowerPC/builtins-ppc-xlcompat-rotate.c
@@ -16,11 +16,8 @@ void test_builtin_ppc_rldimi() {
   // CHECK:       %res = alloca i64, align 8
   // CHECK-NEXT:  [[RA:%[0-9]+]] = load i64, ptr @ull, align 8
   // CHECK-NEXT:  [[RB:%[0-9]+]] = load i64, ptr @ull, align 8
-  // CHECK-NEXT:  [[RC:%[0-9]+]] = call i64 @llvm.fshl.i64(i64 [[RA]], i64 [[RA]], i64 63)
-  // CHECK-NEXT:  [[RD:%[0-9]+]] = and i64 [[RC]], 72057593769492480
-  // CHECK-NEXT:  [[RE:%[0-9]+]] = and i64 [[RB]], -72057593769492481
-  // CHECK-NEXT:  [[RF:%[0-9]+]] = or i64 [[RD]], [[RE]]
-  // CHECK-NEXT:  store i64 [[RF]], ptr %res, align 8
+  // CHECK-NEXT:  [[RC:%[0-9]+]] = call i64 @llvm.ppc.rldimi(i64 [[RA]], i64 [[RB]], i32 63, i64 72057593769492480)
+  // CHECK-NEXT:  store i64 [[RC]], ptr %res, align 8
   // CHECK-NEXT:  ret void
 
   /*shift = 63, mask = 0x00FFFFFFF0000000 = 72057593769492480, ~mask = 0xFF0000000FFFFFFF = -72057593769492481*/
@@ -32,11 +29,8 @@ void test_builtin_ppc_rlwimi() {
   // CHECK:       %res = alloca i32, align 4
   // CHECK-NEXT:  [[RA:%[0-9]+]] = load i32, ptr @ui, align 4
   // CHECK-NEXT:  [[RB:%[0-9]+]] = load i32, ptr @ui, align 4
-  // CHECK-NEXT:  [[RC:%[0-9]+]] = call i32 @llvm.fshl.i32(i32 [[RA]], i32 [[RA]], i32 31)
-  // CHECK-NEXT:  [[RD:%[0-9]+]] = and i32 [[RC]], 16776960
-  // CHECK-NEXT:  [[RE:%[0-9]+]] = and i32 [[RB]], -16776961
-  // CHECK-NEXT:  [[RF:%[0-9]+]] = or i32 [[RD]], [[RE]]
-  // CHECK-NEXT:  store i32 [[RF]], ptr %res, align 4
+  // CHECK-NEXT:  [[RC:%[0-9]+]] = call i32 @llvm.ppc.rlwimi(i32 [[RA]], i32 [[RB]], i32 31, i32 16776960)
+  // CHECK-NEXT:  store i32 [[RC]], ptr %res, align 4
   // CHECK-NEXT:  ret void
 
   /*shift = 31, mask = 0xFFFF00 = 16776960, ~mask = 0xFFFFFFFFFF0000FF = -16776961*/
@@ -47,9 +41,8 @@ void test_builtin_ppc_rlwnm() {
   // CHECK-LABEL: test_builtin_ppc_rlwnm
   // CHECK:       %res = alloca i32, align 4
   // CHECK-NEXT:  [[RA:%[0-9]+]] = load i32, ptr @ui, align 4
-  // CHECK-NEXT:  [[RB:%[0-9]+]] = call i32 @llvm.fshl.i32(i32 [[RA]], i32 [[RA]], i32 31)
-  // CHECK-NEXT:  [[RC:%[0-9]+]] = and i32 [[RB]], 511
-  // CHECK-NEXT:  store i32 [[RC]], ptr %res, align 4
+  // CHECK-NEXT:  [[RB:%[0-9]+]] = call i32 @llvm.ppc.rlwnm(i32 [[RA]], i32 31, i32 511)
+  // CHECK-NEXT:  store i32 [[RB]], ptr %res, align 4
   // CHECK-NEXT:  ret void
 
   /*shift = 31, mask = 0x1FF = 511*/
@@ -63,9 +56,8 @@ void test_builtin_ppc_rlwnm2(unsigned int shift) {
   // CHECK-NEXT:  store i32 %shift, ptr %shift.addr, align 4
   // CHECK-NEXT:  [[RA:%[0-9]+]] = load i32, ptr @ui, align 4
   // CHECK-NEXT:  [[RB:%[0-9]+]] = load i32, ptr %shift.addr, align 4
-  // CHECK-NEXT:  [[RC:%[0-9]+]] = call i32 @llvm.fshl.i32(i32 [[RA]], i32 [[RA]], i32 [[RB]])
-  // CHECK-NEXT:  [[RD:%[0-9]+]] = and i32 [[RC]], 511
-  // CHECK-NEXT:  store i32 [[RD]], ptr %res, align 4
+  // CHECK-NEXT:  [[RC:%[0-9]+]] = call i32 @llvm.ppc.rlwnm(i32 [[RA]], i32 [[RB]], i32 511)
+  // CHECK-NEXT:  store i32 [[RC]], ptr %res, align 4
   // CHECK-NEXT:  ret void
 
   /*mask = 0x1FF = 511*/

diff --git a/clang/test/Layout/dump-complete-invalid.cpp b/clang/test/Layout/dump-complete-invalid.cpp
@@ -0,0 +1,6 @@
+// RUN: %clang_cc1 -verify -fsyntax-only -fdump-record-layouts-complete %s
+
+struct Incomplete; // expected-note {{forward declaration}}
+
+// Check we don't crash on trying to print out an invalid declaration.
+struct Invalid : Incomplete {}; // expected-error {{base class has incomplete type}}
diff --git a/clang/test/Layout/dump-complete.cpp b/clang/test/Layout/dump-complete.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -emit-llvm-only -fdump-record-layouts-complete %s | FileCheck %s
+// RUN: %clang_cc1 -fsyntax-only -fdump-record-layouts-complete %s | FileCheck %s
 
 struct a {
   int x;
@@ -12,7 +12,62 @@ class c {};
 
 class d;
 
+template <typename>
+struct s {
+  int x;
+};
+
+template <typename T>
+struct ts {
+  T x;
+};
+
+template <>
+struct ts<void> {
+  float f;
+};
+
+void f() {
+  ts<int> a;
+  ts<double> b;
+  ts<void> c;
+}
+
+namespace gh83671 {
+template <class _Tp, _Tp __v>
+struct integral_constant {
+  static constexpr const _Tp value = __v;
+  typedef integral_constant type;
+};
+
+template <bool _Val>
+using _BoolConstant = integral_constant<bool, _Val>;
+
+template <class _Tp, class _Up>
+struct is_same : _BoolConstant<__is_same(_Tp, _Up)> {};
+
+template < class _Tp >
+class numeric_limits {};
+
+template < class _Tp >
+class numeric_limits< const _Tp > : public numeric_limits< _Tp > {};
+}
+
+namespace gh83684 {
+template <class Pointer>
+struct AllocationResult {
+  Pointer ptr = nullptr;
+  int count = 0;
+};
+}
+
 // CHECK:          0 | struct a
 // CHECK:          0 | struct b
 // CHECK:          0 | class c
+// CHECK:          0 | struct ts<void>
+// CHECK-NEXT:     0 |   float
+// CHECK:          0 | struct ts<int>
+// CHECK:          0 | struct ts<double>
 // CHECK-NOT:      0 | class d
+// CHECK-NOT:      0 | struct s
+// CHECK-NOT:      0 | struct AllocationResult
diff --git a/clang/test/Parser/cxx-declarator-attribute-crash.cpp b/clang/test/Parser/cxx-declarator-attribute-crash.cpp
@@ -0,0 +1,8 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+
+// expected-error@+5{{brackets are not allowed here}}
+// expected-error@+4{{a type specifier is required for all declarations}}
+// expected-warning@+3{{unknown attribute 'h' ignored}}
+// expected-error@+2{{definition of variable with array type}}
+// expected-error@+1{{expected ';'}}
+[][[h]]l
diff --git a/clang/unittests/StaticAnalyzer/CallDescriptionTest.cpp b/clang/unittests/StaticAnalyzer/CallDescriptionTest.cpp
@@ -488,11 +488,10 @@ TEST(CallDescription, NegativeMatchQualifiedNames) {
 }
 
 TEST(CallDescription, MatchBuiltins) {
-  // Test CDF_MaybeBuiltin - a flag that allows matching weird builtins.
+  // Test CDM::CLibrary - a flag that allows matching weird builtins.
   EXPECT_TRUE(tooling::runToolOnCode(
       std::unique_ptr<FrontendAction>(new CallDescriptionAction<>(
-          {{{{"memset"}, 3}, false},
-           {{CDF_MaybeBuiltin, {"memset"}, 3}, true}})),
+          {{{{"memset"}, 3}, false}, {{CDM::CLibrary, {"memset"}, 3}, true}})),
       "void foo() {"
       "  int x;"
       "  __builtin___memset_chk(&x, 0, sizeof(x),"
@@ -503,8 +502,8 @@ TEST(CallDescription, MatchBuiltins) {
     SCOPED_TRACE("multiple similar builtins");
     EXPECT_TRUE(tooling::runToolOnCode(
         std::unique_ptr<FrontendAction>(new CallDescriptionAction<>(
-            {{{CDF_MaybeBuiltin, {"memcpy"}, 3}, false},
-             {{CDF_MaybeBuiltin, {"wmemcpy"}, 3}, true}})),
+            {{{CDM::CLibrary, {"memcpy"}, 3}, false},
+             {{CDM::CLibrary, {"wmemcpy"}, 3}, true}})),
         R"(void foo(wchar_t *x, wchar_t *y) {
             __builtin_wmemcpy(x, y, sizeof(wchar_t));
           })"));
@@ -513,17 +512,17 @@ TEST(CallDescription, MatchBuiltins) {
     SCOPED_TRACE("multiple similar builtins reversed order");
     EXPECT_TRUE(tooling::runToolOnCode(
         std::unique_ptr<FrontendAction>(new CallDescriptionAction<>(
-            {{{CDF_MaybeBuiltin, {"wmemcpy"}, 3}, true},
-             {{CDF_MaybeBuiltin, {"memcpy"}, 3}, false}})),
+            {{{CDM::CLibrary, {"wmemcpy"}, 3}, true},
+             {{CDM::CLibrary, {"memcpy"}, 3}, false}})),
         R"(void foo(wchar_t *x, wchar_t *y) {
             __builtin_wmemcpy(x, y, sizeof(wchar_t));
           })"));
   }
   {
     SCOPED_TRACE("lookbehind and lookahead mismatches");
     EXPECT_TRUE(tooling::runToolOnCode(
-        std::unique_ptr<FrontendAction>(new CallDescriptionAction<>(
-            {{{CDF_MaybeBuiltin, {"func"}}, false}})),
+        std::unique_ptr<FrontendAction>(
+            new CallDescriptionAction<>({{{CDM::CLibrary, {"func"}}, false}})),
         R"(
           void funcXXX();
           void XXXfunc();
@@ -537,8 +536,8 @@ TEST(CallDescription, MatchBuiltins) {
   {
     SCOPED_TRACE("lookbehind and lookahead matches");
     EXPECT_TRUE(tooling::runToolOnCode(
-        std::unique_ptr<FrontendAction>(new CallDescriptionAction<>(
-            {{{CDF_MaybeBuiltin, {"func"}}, true}})),
+        std::unique_ptr<FrontendAction>(
+            new CallDescriptionAction<>({{{CDM::CLibrary, {"func"}}, true}})),
         R"(
           void func();
           void func_XXX();

diff --git a/compiler-rt/lib/builtins/CMakeLists.txt b/compiler-rt/lib/builtins/CMakeLists.txt
@@ -190,6 +190,7 @@ set(GENERIC_SOURCES
 
 # We only build BF16 files when "__bf16" is available.
 set(BF16_SOURCES
+    extendbfsf2.c
     truncdfbf2.c
     truncsfbf2.c
 )

diff --git a/compiler-rt/lib/builtins/extendbfsf2.c b/compiler-rt/lib/builtins/extendbfsf2.c
@@ -0,0 +1,13 @@
+//===-- lib/extendbfsf2.c - bfloat -> single conversion -----------*- C -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#define SRC_BFLOAT16
+#define DST_SINGLE
+#include "fp_extend_impl.inc"
+
+COMPILER_RT_ABI float __extendbfsf2(src_t a) { return __extendXfYf2__(a); }
diff --git a/compiler-rt/lib/builtins/fp_extend.h b/compiler-rt/lib/builtins/fp_extend.h
@@ -81,6 +81,21 @@ static inline int src_rep_t_clz_impl(src_rep_t a) {
 
 #define src_rep_t_clz src_rep_t_clz_impl
 
+#elif defined SRC_BFLOAT16
+#ifdef COMPILER_RT_HAS_BFLOAT16
+typedef __bf16 src_t;
+#else
+typedef uint16_t src_t;
+#endif
+typedef uint16_t src_rep_t;
+#define SRC_REP_C UINT16_C
+static const int srcBits = sizeof(src_t) * CHAR_BIT;
+static const int srcSigFracBits = 7;
+// -1 accounts for the sign bit.
+// srcBits - srcSigFracBits - 1
+static const int srcExpBits = 8;
+#define src_rep_t_clz __builtin_clz
+
 #else
 #error Source should be half, single, or double precision!
 #endif // end source precision

diff --git a/compiler-rt/lib/sanitizer_common/sanitizer_posix_libcdep.cpp b/compiler-rt/lib/sanitizer_common/sanitizer_posix_libcdep.cpp
@@ -91,12 +91,12 @@ static rlim_t getlim(int res) {
 
 static void setlim(int res, rlim_t lim) {
   struct rlimit rlim;
-  if (getrlimit(res, const_cast<struct rlimit *>(&rlim))) {
+  if (getrlimit(res, &rlim)) {
     Report("ERROR: %s getrlimit() failed %d\n", SanitizerToolName, errno);
     Die();
   }
   rlim.rlim_cur = lim;
-  if (setrlimit(res, const_cast<struct rlimit *>(&rlim))) {
+  if (setrlimit(res, &rlim)) {
     Report("ERROR: %s setrlimit() failed %d\n", SanitizerToolName, errno);
     Die();
   }

diff --git a/compiler-rt/lib/scudo/standalone/checksum.cpp b/compiler-rt/lib/scudo/standalone/checksum.cpp
@@ -19,6 +19,8 @@
 #else
 #include <sys/auxv.h>
 #endif
+#elif defined(__loongarch__)
+#include <sys/auxv.h>
 #endif
 
 namespace scudo {
@@ -75,6 +77,20 @@ bool hasHardwareCRC32() {
   return !!(getauxval(AT_HWCAP) & HWCAP_CRC32);
 #endif // SCUDO_FUCHSIA
 }
+#elif defined(__loongarch__)
+// The definition is only pulled in by <sys/auxv.h> since glibc 2.38, so
+// supply it if missing.
+#ifndef HWCAP_LOONGARCH_CRC32
+#define HWCAP_LOONGARCH_CRC32 (1 << 6)
+#endif
+// Query HWCAP for platform capability, according to *Software Development and
+// Build Convention for LoongArch Architectures* v0.1, Section 9.1.
+//
+// Link:
+// https://github.com/loongson/la-softdev-convention/blob/v0.1/la-softdev-convention.adoc#kernel-development
+bool hasHardwareCRC32() {
+  return !!(getauxval(AT_HWCAP) & HWCAP_LOONGARCH_CRC32);
+}
 #else
 // No hardware CRC32 implemented in Scudo for other architectures.
 bool hasHardwareCRC32() { return false; }

diff --git a/compiler-rt/lib/scudo/standalone/checksum.h b/compiler-rt/lib/scudo/standalone/checksum.h
@@ -30,6 +30,10 @@
 #include <arm_acle.h>
 #define CRC32_INTRINSIC FIRST_32_SECOND_64(__crc32cw, __crc32cd)
 #endif
+#ifdef __loongarch__
+#include <larchintrin.h>
+#define CRC32_INTRINSIC FIRST_32_SECOND_64(__crcc_w_w_w, __crcc_w_d_w)
+#endif
 
 namespace scudo {
 

diff --git a/compiler-rt/lib/scudo/standalone/crc32_hw.cpp b/compiler-rt/lib/scudo/standalone/crc32_hw.cpp
@@ -17,4 +17,13 @@ u32 computeHardwareCRC32(u32 Crc, uptr Data) {
 #endif // defined(__CRC32__) || defined(__SSE4_2__) ||
        // defined(__ARM_FEATURE_CRC32)
 
+#if defined(__loongarch__)
+u32 computeHardwareCRC32(u32 Crc, uptr Data) {
+  // The LoongArch CRC intrinsics have the two input arguments swapped, and
+  // expect them to be signed.
+  return static_cast<u32>(
+      CRC32_INTRINSIC(static_cast<long>(Data), static_cast<int>(Crc)));
+}
+#endif // defined(__loongarch__)
+
 } // namespace scudo
diff --git a/flang/docs/conf.py b/flang/docs/conf.py
@@ -22,13 +22,24 @@
 # Add any Sphinx extension module names here, as strings. They can be extensions
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
 extensions = [
-    "myst_parser",
     "sphinx.ext.todo",
     "sphinx.ext.mathjax",
     "sphinx.ext.intersphinx",
     "sphinx.ext.autodoc",
 ]
 
+# When building man pages, we do not use the markdown pages,
+# So, we can continue without the myst_parser dependencies.
+# Doing so reduces dependencies of some packaged llvm distributions.
+try:
+    import myst_parser
+
+    extensions.append("myst_parser")
+except ImportError:
+    if not tags.has("builder-man"):
+        raise
+
+
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ["_templates"]
 myst_heading_anchors = 6

diff --git a/flang/include/flang/Optimizer/CodeGen/CodeGen.h b/flang/include/flang/Optimizer/CodeGen/CodeGen.h
@@ -18,6 +18,8 @@
 
 namespace fir {
 
+class LLVMTypeConverter;
+
 struct NameUniquer;
 
 #define GEN_PASS_DECL_FIRTOLLVMLOWERING
@@ -80,6 +82,11 @@ std::unique_ptr<mlir::Pass> createLLVMDialectToLLVMPass(
 std::unique_ptr<mlir::Pass> createBoxedProcedurePass();
 std::unique_ptr<mlir::Pass> createBoxedProcedurePass(bool useThunks);
 
+/// Populate the given list with patterns that convert from FIR to LLVM.
+void populateFIRToLLVMConversionPatterns(fir::LLVMTypeConverter &converter,
+                                         mlir::RewritePatternSet &patterns,
+                                         fir::FIRToLLVMPassOptions &options);
+
 // declarative passes
 #define GEN_PASS_REGISTRATION
 #include "flang/Optimizer/CodeGen/CGPasses.h.inc"

diff --git a/flang/lib/Optimizer/CodeGen/CodeGen.cpp b/flang/lib/Optimizer/CodeGen/CodeGen.cpp
@@ -3937,30 +3937,7 @@ class FIRToLLVMLowering
                                          options.applyTBAA || applyTBAA,
                                          options.forceUnifiedTBAATree, *dl};
     mlir::RewritePatternSet pattern(context);
-    pattern.insert<
-        AbsentOpConversion, AddcOpConversion, AddrOfOpConversion,
-        AllocaOpConversion, AllocMemOpConversion, BoxAddrOpConversion,
-        BoxCharLenOpConversion, BoxDimsOpConversion, BoxEleSizeOpConversion,
-        BoxIsAllocOpConversion, BoxIsArrayOpConversion, BoxIsPtrOpConversion,
-        BoxOffsetOpConversion, BoxProcHostOpConversion, BoxRankOpConversion,
-        BoxTypeCodeOpConversion, BoxTypeDescOpConversion, CallOpConversion,
-        CmpcOpConversion, ConstcOpConversion, ConvertOpConversion,
-        CoordinateOpConversion, DTEntryOpConversion, DivcOpConversion,
-        EmboxOpConversion, EmboxCharOpConversion, EmboxProcOpConversion,
-        ExtractValueOpConversion, FieldIndexOpConversion, FirEndOpConversion,
-        FreeMemOpConversion, GlobalLenOpConversion, GlobalOpConversion,
-        HasValueOpConversion, InsertOnRangeOpConversion,
-        InsertValueOpConversion, IsPresentOpConversion,
-        LenParamIndexOpConversion, LoadOpConversion, MulcOpConversion,
-        NegcOpConversion, NoReassocOpConversion, SelectCaseOpConversion,
-        SelectOpConversion, SelectRankOpConversion, SelectTypeOpConversion,
-        ShapeOpConversion, ShapeShiftOpConversion, ShiftOpConversion,
-        SliceOpConversion, StoreOpConversion, StringLitOpConversion,
-        SubcOpConversion, TypeDescOpConversion, TypeInfoOpConversion,
-        UnboxCharOpConversion, UnboxProcOpConversion, UndefOpConversion,
-        UnreachableOpConversion, UnrealizedConversionCastOpConversion,
-        XArrayCoorOpConversion, XEmboxOpConversion, XReboxOpConversion,
-        ZeroOpConversion>(typeConverter, options);
+    fir::populateFIRToLLVMConversionPatterns(typeConverter, pattern, options);
     mlir::populateFuncToLLVMConversionPatterns(typeConverter, pattern);
     mlir::populateOpenMPToLLVMConversionPatterns(typeConverter, pattern);
     mlir::arith::populateArithToLLVMConversionPatterns(typeConverter, pattern);
@@ -4072,3 +4049,32 @@ fir::createLLVMDialectToLLVMPass(llvm::raw_ostream &output,
                                  fir::LLVMIRLoweringPrinter printer) {
   return std::make_unique<LLVMIRLoweringPass>(output, printer);
 }
+
+void fir::populateFIRToLLVMConversionPatterns(
+    fir::LLVMTypeConverter &converter, mlir::RewritePatternSet &patterns,
+    fir::FIRToLLVMPassOptions &options) {
+  patterns.insert<
+      AbsentOpConversion, AddcOpConversion, AddrOfOpConversion,
+      AllocaOpConversion, AllocMemOpConversion, BoxAddrOpConversion,
+      BoxCharLenOpConversion, BoxDimsOpConversion, BoxEleSizeOpConversion,
+      BoxIsAllocOpConversion, BoxIsArrayOpConversion, BoxIsPtrOpConversion,
+      BoxOffsetOpConversion, BoxProcHostOpConversion, BoxRankOpConversion,
+      BoxTypeCodeOpConversion, BoxTypeDescOpConversion, CallOpConversion,
+      CmpcOpConversion, ConstcOpConversion, ConvertOpConversion,
+      CoordinateOpConversion, DTEntryOpConversion, DivcOpConversion,
+      EmboxOpConversion, EmboxCharOpConversion, EmboxProcOpConversion,
+      ExtractValueOpConversion, FieldIndexOpConversion, FirEndOpConversion,
+      FreeMemOpConversion, GlobalLenOpConversion, GlobalOpConversion,
+      HasValueOpConversion, InsertOnRangeOpConversion, InsertValueOpConversion,
+      IsPresentOpConversion, LenParamIndexOpConversion, LoadOpConversion,
+      MulcOpConversion, NegcOpConversion, NoReassocOpConversion,
+      SelectCaseOpConversion, SelectOpConversion, SelectRankOpConversion,
+      SelectTypeOpConversion, ShapeOpConversion, ShapeShiftOpConversion,
+      ShiftOpConversion, SliceOpConversion, StoreOpConversion,
+      StringLitOpConversion, SubcOpConversion, TypeDescOpConversion,
+      TypeInfoOpConversion, UnboxCharOpConversion, UnboxProcOpConversion,
+      UndefOpConversion, UnreachableOpConversion,
+      UnrealizedConversionCastOpConversion, XArrayCoorOpConversion,
+      XEmboxOpConversion, XReboxOpConversion, ZeroOpConversion>(converter,
+                                                                options);
+}
diff --git a/flang/lib/Optimizer/Transforms/MemoryAllocation.cpp b/flang/lib/Optimizer/Transforms/MemoryAllocation.cpp
@@ -200,7 +200,8 @@ class MemoryAllocationOpt
       return keepStackAllocation(alloca, &func.front(), options);
     });
 
-    patterns.insert<AllocaOpConversion>(context, analysis.getReturns(func));
+    llvm::SmallVector<mlir::Operation *> returnOps = analysis.getReturns(func);
+    patterns.insert<AllocaOpConversion>(context, returnOps);
     if (mlir::failed(
             mlir::applyPartialConversion(func, target, std::move(patterns)))) {
       mlir::emitError(func.getLoc(),

diff --git a/flang/lib/Optimizer/Transforms/SimplifyIntrinsics.cpp b/flang/lib/Optimizer/Transforms/SimplifyIntrinsics.cpp
@@ -656,7 +656,7 @@ static void genRuntimeMinMaxlocBody(fir::FirOpBuilder &builder,
                                     unsigned rank, int maskRank,
                                     mlir::Type elementType,
                                     mlir::Type maskElemType,
-                                    mlir::Type resultElemTy) {
+                                    mlir::Type resultElemTy, bool isDim) {
   auto init = [isMax](fir::FirOpBuilder builder, mlir::Location loc,
                       mlir::Type elementType) {
     if (auto ty = elementType.dyn_cast<mlir::FloatType>()) {
@@ -858,16 +858,27 @@ static void genRuntimeMinMaxlocBody(fir::FirOpBuilder &builder,
                             maskElemType, resultArr, maskRank == 0);
 
   // Store newly created output array to the reference passed in
-  fir::SequenceType::Shape resultShape(1, rank);
-  mlir::Type outputArrTy = fir::SequenceType::get(resultShape, resultElemTy);
-  mlir::Type outputHeapTy = fir::HeapType::get(outputArrTy);
-  mlir::Type outputBoxTy = fir::BoxType::get(outputHeapTy);
-  mlir::Type outputRefTy = builder.getRefType(outputBoxTy);
-  mlir::Value outputArr = builder.create<fir::ConvertOp>(
-      loc, outputRefTy, funcOp.front().getArgument(0));
-
-  // Store nearly created array to output array
-  builder.create<fir::StoreOp>(loc, resultArr, outputArr);
+  if (isDim) {
+    mlir::Type resultBoxTy =
+        fir::BoxType::get(fir::HeapType::get(resultElemTy));
+    mlir::Value outputArr = builder.create<fir::ConvertOp>(
+        loc, builder.getRefType(resultBoxTy), funcOp.front().getArgument(0));
+    mlir::Value resultArrScalar = builder.create<fir::ConvertOp>(
+        loc, fir::HeapType::get(resultElemTy), resultArrInit);
+    mlir::Value resultBox =
+        builder.create<fir::EmboxOp>(loc, resultBoxTy, resultArrScalar);
+    builder.create<fir::StoreOp>(loc, resultBox, outputArr);
+  } else {
+    fir::SequenceType::Shape resultShape(1, rank);
+    mlir::Type outputArrTy = fir::SequenceType::get(resultShape, resultElemTy);
+    mlir::Type outputHeapTy = fir::HeapType::get(outputArrTy);
+    mlir::Type outputBoxTy = fir::BoxType::get(outputHeapTy);
+    mlir::Type outputRefTy = builder.getRefType(outputBoxTy);
+    mlir::Value outputArr = builder.create<fir::ConvertOp>(
+        loc, outputRefTy, funcOp.front().getArgument(0));
+    builder.create<fir::StoreOp>(loc, resultArr, outputArr);
+  }
+
   builder.create<mlir::func::ReturnOp>(loc);
 }
 
@@ -1146,11 +1157,14 @@ void SimplifyIntrinsicsPass::simplifyMinMaxlocReduction(
 
   mlir::Operation::operand_range args = call.getArgs();
 
-  mlir::Value back = args[6];
+  mlir::SymbolRefAttr callee = call.getCalleeAttr();
+  mlir::StringRef funcNameBase = callee.getLeafReference().getValue();
+  bool isDim = funcNameBase.ends_with("Dim");
+  mlir::Value back = args[isDim ? 7 : 6];
   if (isTrueOrNotConstant(back))
     return;
 
-  mlir::Value mask = args[5];
+  mlir::Value mask = args[isDim ? 6 : 5];
   mlir::Value maskDef = findMaskDef(mask);
 
   // maskDef is set to NULL when the defining op is not one we accept.
@@ -1159,10 +1173,8 @@ void SimplifyIntrinsicsPass::simplifyMinMaxlocReduction(
   if (maskDef == NULL)
     return;
 
-  mlir::SymbolRefAttr callee = call.getCalleeAttr();
-  mlir::StringRef funcNameBase = callee.getLeafReference().getValue();
   unsigned rank = getDimCount(args[1]);
-  if (funcNameBase.ends_with("Dim") || !(rank > 0))
+  if ((isDim && rank != 1) || !(rank > 0))
     return;
 
   fir::FirOpBuilder builder{getSimplificationBuilder(call, kindMap)};
@@ -1203,22 +1215,24 @@ void SimplifyIntrinsicsPass::simplifyMinMaxlocReduction(
 
   llvm::raw_string_ostream nameOS(funcName);
   outType.print(nameOS);
+  if (isDim)
+    nameOS << '_' << inputType;
   nameOS << '_' << fmfString;
 
   auto typeGenerator = [rank](fir::FirOpBuilder &builder) {
     return genRuntimeMinlocType(builder, rank);
   };
   auto bodyGenerator = [rank, maskRank, inputType, logicalElemType, outType,
-                        isMax](fir::FirOpBuilder &builder,
-                               mlir::func::FuncOp &funcOp) {
+                        isMax, isDim](fir::FirOpBuilder &builder,
+                                      mlir::func::FuncOp &funcOp) {
     genRuntimeMinMaxlocBody(builder, funcOp, isMax, rank, maskRank, inputType,
-                            logicalElemType, outType);
+                            logicalElemType, outType, isDim);
   };
 
   mlir::func::FuncOp newFunc =
       getOrCreateFunction(builder, funcName, typeGenerator, bodyGenerator);
   builder.create<fir::CallOp>(loc, newFunc,
-                              mlir::ValueRange{args[0], args[1], args[5]});
+                              mlir::ValueRange{args[0], args[1], mask});
   call->dropAllReferences();
   call->erase();
 }

diff --git a/flang/test/Fir/memory-allocation-opt.fir b/flang/test/Fir/memory-allocation-opt.fir
@@ -1,4 +1,6 @@
 // RUN: fir-opt --memory-allocation-opt="dynamic-array-on-heap=true maximum-array-alloc-size=1024" %s | FileCheck %s
+// FIXME: started crashing on windows https://github.com/llvm/llvm-project/issues/83534
+// UNSUPPORTED: system-windows
 
 // Test for size of array being too big.
 

diff --git a/flang/test/Transforms/simplifyintrinsics.fir b/flang/test/Transforms/simplifyintrinsics.fir
@@ -2098,13 +2098,13 @@ func.func @_QPtestminloc_doesntwork1d_back(%arg0: !fir.ref<!fir.array<10xi32>> {
 // CHECK-NOT:         fir.call @_FortranAMinlocInteger4x1_i32_contract_simplified({{.*}}) fastmath<contract> : (!fir.ref<!fir.box<none>>, !fir.box<none>, !fir.box<none>) -> ()
 
 // -----
-// Check Minloc is not simplified when DIM arg is set
+// Check Minloc is simplified when DIM arg is set so long as the result is scalar
 
-func.func @_QPtestminloc_doesntwork1d_dim(%arg0: !fir.ref<!fir.array<10xi32>> {fir.bindc_name = "a"}) -> !fir.array<1xi32> {
+func.func @_QPtestminloc_1d_dim(%arg0: !fir.ref<!fir.array<10xi32>> {fir.bindc_name = "a"}) -> !fir.array<1xi32> {
   %0 = fir.alloca !fir.box<!fir.heap<i32>>
   %c10 = arith.constant 10 : index
   %c1 = arith.constant 1 : index
-  %1 = fir.alloca !fir.array<1xi32> {bindc_name = "testminloc_doesntwork1d_dim", uniq_name = "_QFtestminloc_doesntwork1d_dimEtestminloc_doesntwork1d_dim"}
+  %1 = fir.alloca !fir.array<1xi32> {bindc_name = "testminloc_1d_dim", uniq_name = "_QFtestminloc_1d_dimEtestminloc_1d_dim"}
   %2 = fir.shape %c1 : (index) -> !fir.shape<1>
   %3 = fir.array_load %1(%2) : (!fir.ref<!fir.array<1xi32>>, !fir.shape<1>) -> !fir.array<1xi32>
   %4 = fir.shape %c10 : (index) -> !fir.shape<1>
@@ -2139,11 +2139,62 @@ func.func @_QPtestminloc_doesntwork1d_dim(%arg0: !fir.ref<!fir.array<10xi32>> {f
   %21 = fir.load %1 : !fir.ref<!fir.array<1xi32>>
   return %21 : !fir.array<1xi32>
 }
-// CHECK-LABEL:   func.func @_QPtestminloc_doesntwork1d_dim(
+// CHECK-LABEL:   func.func @_QPtestminloc_1d_dim(
 // CHECK-SAME:                                             %[[ARR:.*]]: !fir.ref<!fir.array<10xi32>> {fir.bindc_name = "a"}) -> !fir.array<1xi32> {
-// CHECK-NOT:         fir.call @_FortranAMinlocDimx1_i32_contract_simplified({{.*}}) fastmath<contract> : (!fir.ref<!fir.box<none>>, !fir.box<none>, !fir.box<none>) -> ()
-// CHECK:             fir.call @_FortranAMinlocDim({{.*}}) fastmath<contract> : (!fir.ref<!fir.box<none>>, !fir.box<none>, i32, i32, !fir.ref<i8>, i32, !fir.box<none>, i1) -> none
-// CHECK-NOT:         fir.call @_FortranAMinlocDimx1_i32_contract_simplified({{.*}}) fastmath<contract> : (!fir.ref<!fir.box<none>>, !fir.box<none>, !fir.box<none>) -> ()
+// CHECK:             fir.call @_FortranAMinlocDimx1_i32_i32_contract_simplified({{.*}}) fastmath<contract> : (!fir.ref<!fir.box<none>>, !fir.box<none>, !fir.box<none>) -> ()
+
+// CHECK-LABEL:  func.func private @_FortranAMinlocDimx1_i32_i32_contract_simplified(%arg0: !fir.ref<!fir.box<none>>, %arg1: !fir.box<none>, %arg2: !fir.box<none>) attributes {llvm.linkage = #llvm.linkage<linkonce_odr>} {
+// CHECK-NEXT:    %[[V0:.*]] = fir.alloca i32
+// CHECK-NEXT:    %c0_i32 = arith.constant 0 : i32
+// CHECK-NEXT:    %c1 = arith.constant 1 : index
+// CHECK-NEXT:    %[[V1:.*]] = fir.allocmem !fir.array<1xi32>
+// CHECK-NEXT:    %[[V2:.*]] = fir.shape %c1 : (index) -> !fir.shape<1>
+// CHECK-NEXT:    %[[V3:.*]] = fir.embox %[[V1]](%[[V2]]) : (!fir.heap<!fir.array<1xi32>>, !fir.shape<1>) -> !fir.box<!fir.heap<!fir.array<1xi32>>>
+// CHECK-NEXT:    %c0 = arith.constant 0 : index
+// CHECK-NEXT:    %[[V4:.*]] = fir.coordinate_of %[[V3]], %c0 : (!fir.box<!fir.heap<!fir.array<1xi32>>>, index) -> !fir.ref<i32>
+// CHECK-NEXT:    fir.store %c0_i32 to %[[V4]] : !fir.ref<i32>
+// CHECK-NEXT:    %c0_0 = arith.constant 0 : index
+// CHECK-NEXT:    %[[V5:.*]] = fir.convert %arg1 : (!fir.box<none>) -> !fir.box<!fir.array<?xi32>>
+// CHECK-NEXT:    %c1_i32 = arith.constant 1 : i32
+// CHECK-NEXT:    %c0_i32_1 = arith.constant 0 : i32
+// CHECK-NEXT:    fir.store %c0_i32_1 to %[[V0]] : !fir.ref<i32>
+// CHECK-NEXT:    %c2147483647_i32 = arith.constant 2147483647 : i32
+// CHECK-NEXT:    %c1_2 = arith.constant 1 : index
+// CHECK-NEXT:    %c0_3 = arith.constant 0 : index
+// CHECK-NEXT:    %[[V6:.*]]:3 = fir.box_dims %[[V5]], %c0_3 : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+// CHECK-NEXT:    %[[V7:.*]] = arith.subi %[[V6]]#1, %c1_2 : index
+// CHECK-NEXT:    %[[V8:.*]] = fir.do_loop %arg3 = %c0_0 to %[[V7]] step %c1_2 iter_args(%arg4 = %c2147483647_i32) -> (i32) {
+// CHECK-NEXT:      %c1_i32_4 = arith.constant 1 : i32
+// CHECK-NEXT:      %[[ISFIRST:.*]] = fir.load %[[FLAG_ALLOC]] : !fir.ref<i32>
+// CHECK-NEXT:      %[[V12:.*]] = fir.coordinate_of %[[V5]], %arg3 : (!fir.box<!fir.array<?xi32>>, index) -> !fir.ref<i32>
+// CHECK-NEXT:      %[[V13:.*]] = fir.load %[[V12]] : !fir.ref<i32>
+// CHECK-NEXT:      %[[V14:.*]] = arith.cmpi slt, %[[V13]], %arg4 : i32
+// CHECK-NEXT:      %[[ISFIRSTL:.*]] = fir.convert %[[ISFIRST]] : (i32) -> i1
+// CHECK-NEXT:      %true = arith.constant true
+// CHECK-NEXT:      %[[ISFIRSTNOT:.*]] = arith.xori %[[ISFIRSTL]], %true : i1
+// CHECK-NEXT:      %[[ORCOND:.*]] = arith.ori %[[V14]], %[[ISFIRSTNOT]] : i1
+// CHECK-NEXT:      %[[V15:.*]] = fir.if %[[ORCOND]] -> (i32) {
+// CHECK-NEXT:        fir.store %c1_i32_4 to %[[V0]] : !fir.ref<i32>
+// CHECK-NEXT:        %c1_i32_5 = arith.constant 1 : i32
+// CHECK-NEXT:        %c0_6 = arith.constant 0 : index
+// CHECK-NEXT:        %[[V16:.*]] = fir.coordinate_of %[[V3]], %c0_6 : (!fir.box<!fir.heap<!fir.array<1xi32>>>, index) -> !fir.ref<i32>
+// CHECK-NEXT:        %[[V17:.*]] = fir.convert %arg3 : (index) -> i32
+// CHECK-NEXT:        %[[V18:.*]] = arith.addi %[[V17]], %c1_i32_5 : i32
+// CHECK-NEXT:        fir.store %[[V18]] to %[[V16]] : !fir.ref<i32>
+// CHECK-NEXT:        fir.result %[[V13]] : i32
+// CHECK-NEXT:      } else {
+// CHECK-NEXT:        fir.result %arg4 : i32
+// CHECK-NEXT:      }
+// CHECK-NEXT:      fir.result %[[V15]] : i32
+// CHECK-NEXT:    }
+// CHECK-NEXT:    %[[V11:.*]] = fir.convert %arg0 : (!fir.ref<!fir.box<none>>) -> !fir.ref<!fir.box<!fir.heap<i32>>>
+// CHECK-NEXT:    %[[V12:.*]] = fir.convert %[[V1]] : (!fir.heap<!fir.array<1xi32>>) -> !fir.heap<i32>
+// CHECK-NEXT:    %[[V13:.*]] = fir.embox %[[V12]] : (!fir.heap<i32>) -> !fir.box<!fir.heap<i32>>
+// CHECK-NEXT:    fir.store %[[V13]] to %[[V11]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+// CHECK-NEXT:    return
+// CHECK-NEXT:  }
+
+
 
 // -----
 // Check Minloc is not simplified when dimension of inputArr is unknown

diff --git a/libc/cmake/modules/LLVMLibCHeaderRules.cmake b/libc/cmake/modules/LLVMLibCHeaderRules.cmake
@@ -141,6 +141,7 @@ function(add_gen_header target_name)
 
   if(LIBC_TARGET_OS_IS_GPU)
     file(MAKE_DIRECTORY ${LIBC_INCLUDE_DIR}/llvm-libc-decls)
+    file(MAKE_DIRECTORY ${LIBC_INCLUDE_DIR}/llvm-libc-decls/gpu)
     set(decl_out_file ${LIBC_INCLUDE_DIR}/llvm-libc-decls/${relative_path})
     add_custom_command(
       OUTPUT ${decl_out_file}

diff --git a/libc/config/baremetal/arm/headers.txt b/libc/config/baremetal/arm/headers.txt
@@ -4,6 +4,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.fenv
     libc.include.errno
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.math
     libc.include.stdio

diff --git a/libc/config/baremetal/riscv/headers.txt b/libc/config/baremetal/riscv/headers.txt
@@ -4,6 +4,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.fenv
     libc.include.errno
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.math
     libc.include.stdio

diff --git a/libc/config/darwin/arm/headers.txt b/libc/config/darwin/arm/headers.txt
@@ -3,6 +3,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.errno
     libc.include.fenv
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.limits
     libc.include.math

diff --git a/libc/config/darwin/x86_64/headers.txt b/libc/config/darwin/x86_64/headers.txt
@@ -4,6 +4,7 @@ set(TARGET_PUBLIC_HEADERS
     # Fenv is currently disabled.
     #libc.include.fenv
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.limits
     libc.include.math

diff --git a/libc/config/gpu/headers.txt b/libc/config/gpu/headers.txt
@@ -3,6 +3,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.ctype
     libc.include.string
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.limits
     libc.include.math

diff --git a/libc/config/linux/aarch64/headers.txt b/libc/config/linux/aarch64/headers.txt
@@ -5,6 +5,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.features
     libc.include.fenv
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.limits
     libc.include.math

diff --git a/libc/config/linux/arm/headers.txt b/libc/config/linux/arm/headers.txt
@@ -3,6 +3,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.fenv
     libc.include.errno
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.math
     libc.include.stdckdint

diff --git a/libc/config/linux/riscv/headers.txt b/libc/config/linux/riscv/headers.txt
@@ -7,6 +7,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.features
     libc.include.fenv
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.limits
     libc.include.math

diff --git a/libc/config/linux/x86_64/headers.txt b/libc/config/linux/x86_64/headers.txt
@@ -7,6 +7,7 @@ set(TARGET_PUBLIC_HEADERS
     libc.include.features
     libc.include.fenv
     libc.include.float
+    libc.include.stdint
     libc.include.inttypes
     libc.include.limits
     libc.include.math

diff --git a/libc/include/CMakeLists.txt b/libc/include/CMakeLists.txt
@@ -84,6 +84,14 @@ add_gen_header(
     .llvm-libc-macros.float_macros
 )
 
+add_gen_header(
+  stdint
+  DEF_FILE stdint.h.def
+  GEN_HDR stdint.h
+  DEPENDS
+    .llvm-libc-macros.stdint_macros
+)
+
 add_gen_header(
   limits
   DEF_FILE limits.h.def

diff --git a/libc/include/llvm-libc-macros/CMakeLists.txt b/libc/include/llvm-libc-macros/CMakeLists.txt
@@ -67,6 +67,12 @@ add_macro_header(
     file-seek-macros.h
 )
 
+add_macro_header(
+  stdint_macros
+  HDR
+    stdint-macros.h
+)
+
 add_macro_header(
   float_macros
   HDR

diff --git a/libc/include/llvm-libc-macros/stdint-macros.h b/libc/include/llvm-libc-macros/stdint-macros.h
diff --git a/libc/include/llvm-libc-types/float128.h b/libc/include/llvm-libc-types/float128.h
@@ -19,18 +19,17 @@
 //   https://github.com/llvm/llvm-project/issues/80195
 #if defined(__STDC_IEC_60559_BFP__) && !defined(__clang__) &&                  \
     !defined(__cplusplus)
-// Use _Float128 C23 type.
-#define LIBC_COMPILER_HAS_C23_FLOAT128
+#define LIBC_TYPES_HAS_FLOAT128
 typedef _Float128 float128;
 #elif defined(__FLOAT128__) || defined(__SIZEOF_FLOAT128__)
 // Use __float128 type.  gcc and clang sometime use __SIZEOF_FLOAT128__ to
 // notify the availability of __float128.
 // clang also uses __FLOAT128__ macro to notify the availability of __float128
 // type: https://reviews.llvm.org/D15120
-#define LIBC_COMPILER_HAS_FLOAT128_EXTENSION
+#define LIBC_TYPES_HAS_FLOAT128
 typedef __float128 float128;
 #elif (LDBL_MANT_DIG == 113)
-// Use long double.
+#define LIBC_TYPES_HAS_FLOAT128
 typedef long double float128;
 #endif
 

diff --git a/clang/lib/AST/Interp/ByteCodeGenError.cpp → libc/include/stdint.h.def b/clang/lib/AST/Interp/ByteCodeGenError.cpp → libc/include/stdint.h.def
@@ -1,14 +1,14 @@
-//===--- ByteCodeGenError.h - Byte code generation error --------*- C++ -*-===//
+//===-- C standard library header stdint.h --------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
-#include "ByteCodeGenError.h"
+#ifndef LLVM_LIBC_STDINT_H
+#define LLVM_LIBC_STDINT_H
 
-using namespace clang;
-using namespace clang::interp;
+#include <llvm-libc-macros/stdint-macros.h>
 
-char ByteCodeGenError::ID;
+#endif // LLVM_LIBC_STDINT_H
diff --git a/libc/spec/stdc.td b/libc/spec/stdc.td
@@ -963,6 +963,8 @@ def StdC : StandardSpec<"stdc"> {
       ]
   >;
 
+  HeaderSpec StdInt = HeaderSpec<"StdInt.h">;
+
   HeaderSpec Limits = HeaderSpec<"limits.h">;
 
   NamedType SigAtomicT = NamedType<"sig_atomic_t">;
@@ -1268,6 +1270,7 @@ def StdC : StandardSpec<"stdc"> {
     Errno,
     Fenv,
     Float,
+    StdInt,
     Limits,
     Math,
     String,

diff --git a/libc/src/__support/FPUtil/ManipulationFunctions.h b/libc/src/__support/FPUtil/ManipulationFunctions.h
@@ -233,8 +233,8 @@ LIBC_INLINE T nextafter(T from, U to) {
 } // namespace fputil
 } // namespace LIBC_NAMESPACE
 
-#ifdef LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#ifdef LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 #include "x86_64/NextAfterLongDouble.h"
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
 #endif // LLVM_LIBC_SRC___SUPPORT_FPUTIL_MANIPULATIONFUNCTIONS_H
diff --git a/libc/src/__support/FPUtil/NormalFloat.h b/libc/src/__support/FPUtil/NormalFloat.h
@@ -172,7 +172,7 @@ template <typename T> struct NormalFloat {
   }
 };
 
-#ifdef LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#ifdef LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 template <>
 LIBC_INLINE void
 NormalFloat<long double>::init_from_bits(FPBits<long double> bits) {
@@ -261,7 +261,7 @@ template <> LIBC_INLINE NormalFloat<long double>::operator long double() const {
   result.set_implicit_bit(1);
   return result.get_val();
 }
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
 } // namespace fputil
 } // namespace LIBC_NAMESPACE

diff --git a/libc/src/__support/FPUtil/dyadic_float.h b/libc/src/__support/FPUtil/dyadic_float.h
@@ -19,7 +19,7 @@
 
 namespace LIBC_NAMESPACE::fputil {
 
-// A generic class to perform comuptations of high precision floating points.
+// A generic class to perform computations of high precision floating points.
 // We store the value in dyadic format, including 3 fields:
 //   sign    : boolean value - false means positive, true means negative
 //   exponent: the exponent value of the least significant bit of the mantissa.

diff --git a/libc/src/__support/FPUtil/generic/sqrt.h b/libc/src/__support/FPUtil/generic/sqrt.h
@@ -27,11 +27,11 @@ template <typename T> struct SpecialLongDouble {
   static constexpr bool VALUE = false;
 };
 
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 template <> struct SpecialLongDouble<long double> {
   static constexpr bool VALUE = true;
 };
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
 template <typename T>
 LIBC_INLINE void normalize(int &exponent,
@@ -43,12 +43,12 @@ LIBC_INLINE void normalize(int &exponent,
   mantissa <<= shift;
 }
 
-#ifdef LIBC_LONG_DOUBLE_IS_FLOAT64
+#ifdef LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64
 template <>
 LIBC_INLINE void normalize<long double>(int &exponent, uint64_t &mantissa) {
   normalize<double>(exponent, mantissa);
 }
-#elif !defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#elif !defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 template <>
 LIBC_INLINE void normalize<long double>(int &exponent, UInt128 &mantissa) {
   const uint64_t hi_bits = static_cast<uint64_t>(mantissa >> 64);

diff --git a/libc/src/__support/FPUtil/generic/sqrt_80_bit_long_double.h b/libc/src/__support/FPUtil/generic/sqrt_80_bit_long_double.h
@@ -34,7 +34,7 @@ LIBC_INLINE long double sqrt(long double x);
 
 // Correctly rounded SQRT for all rounding modes.
 // Shift-and-add algorithm.
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 LIBC_INLINE long double sqrt(long double x) {
   using LDBits = FPBits<long double>;
   using StorageType = typename LDBits::StorageType;
@@ -130,7 +130,7 @@ LIBC_INLINE long double sqrt(long double x) {
     return out.get_val();
   }
 }
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
 } // namespace x86
 } // namespace fputil

diff --git a/libc/src/__support/FPUtil/x86_64/sqrt.h b/libc/src/__support/FPUtil/x86_64/sqrt.h
@@ -33,7 +33,7 @@ template <> LIBC_INLINE double sqrt<double>(double x) {
   return result;
 }
 
-#ifdef LIBC_LONG_DOUBLE_IS_FLOAT64
+#ifdef LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64
 template <> LIBC_INLINE long double sqrt<long double>(long double x) {
   long double result;
   __asm__ __volatile__("sqrtsd %x1, %x0" : "=x"(result) : "x"(x));

diff --git a/libc/src/__support/GPU/CMakeLists.txt b/libc/src/__support/GPU/CMakeLists.txt
@@ -1,11 +1,9 @@
-if(NOT LIBC_TARGET_OS_IS_GPU)
+if(NOT EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/${LIBC_TARGET_ARCHITECTURE})
   return()
 endif()
 
-foreach(target nvptx amdgpu generic)
-  add_subdirectory(${target})
-  list(APPEND target_gpu_utils libc.src.__support.GPU.${target}.${target}_utils)
-endforeach()
+add_subdirectory(${LIBC_TARGET_ARCHITECTURE})
+set(target_gpu_utils libc.src.__support.GPU.${LIBC_TARGET_ARCHITECTURE}.${LIBC_TARGET_ARCHITECTURE}_utils)
 
 add_header_library(
   utils

diff --git a/libc/src/__support/float_to_string.h b/libc/src/__support/float_to_string.h
@@ -615,7 +615,7 @@ class FloatToString {
   }
 };
 
-#if !defined(LIBC_LONG_DOUBLE_IS_FLOAT64) &&                                   \
+#if !defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64) &&                             \
     !defined(LIBC_COPT_FLOAT_TO_STR_NO_SPECIALIZE_LD)
 // --------------------------- LONG DOUBLE FUNCTIONS ---------------------------
 
@@ -837,7 +837,7 @@ template <> class FloatToString<long double> {
   }
 };
 
-#endif // !LIBC_LONG_DOUBLE_IS_FLOAT64 &&
+#endif // !LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64 &&
        // !LIBC_COPT_FLOAT_TO_STR_NO_SPECIALIZE_LD
 
 } // namespace LIBC_NAMESPACE

diff --git a/libc/src/__support/macros/properties/types.h b/libc/src/__support/macros/properties/types.h
@@ -19,14 +19,14 @@
 
 // 'long double' properties.
 #if (LDBL_MANT_DIG == 53)
-#define LIBC_LONG_DOUBLE_IS_FLOAT64
+#define LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64
 #elif (LDBL_MANT_DIG == 64)
-#define LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#define LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 #elif (LDBL_MANT_DIG == 113)
-#define LIBC_LONG_DOUBLE_IS_FLOAT128
+#define LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128
 #endif
 
-// float16 support.
+// -- float16 support ---------------------------------------------------------
 // TODO: move this logic to "llvm-libc-types/float16.h"
 #if defined(LIBC_TARGET_ARCH_IS_X86_64) && defined(LIBC_TARGET_CPU_HAS_SSE2)
 #if (defined(LIBC_COMPILER_CLANG_VER) && (LIBC_COMPILER_CLANG_VER >= 1500)) || \
@@ -50,11 +50,8 @@ using float16 = _Float16;
 #endif
 #endif
 
-// float128 support.
-#if defined(LIBC_COMPILER_HAS_C23_FLOAT128) ||                                 \
-    defined(LIBC_COMPILER_HAS_FLOAT128_EXTENSION) ||                           \
-    defined(LIBC_LONG_DOUBLE_IS_FLOAT128)
-#define LIBC_TYPES_HAS_FLOAT128
-#endif
+// -- float128 support --------------------------------------------------------
+// LIBC_TYPES_HAS_FLOAT128 and 'float128' type are provided by
+// "include/llvm-libc-types/float128.h"
 
 #endif // LLVM_LIBC_SRC___SUPPORT_MACROS_PROPERTIES_TYPES_H
diff --git a/libc/src/__support/str_to_float.h b/libc/src/__support/str_to_float.h
@@ -51,13 +51,13 @@ template <class T> LIBC_INLINE void set_implicit_bit(fputil::FPBits<T> &) {
   return;
 }
 
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 template <>
 LIBC_INLINE void
 set_implicit_bit<long double>(fputil::FPBits<long double> &result) {
   result.set_implicit_bit(result.get_biased_exponent() != 0);
 }
-#endif
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
 // This Eisel-Lemire implementation is based on the algorithm described in the
 // paper Number Parsing at a Gigabyte per Second, Software: Practice and
@@ -176,7 +176,7 @@ eisel_lemire(ExpandedFloat<T> init_num,
   return output;
 }
 
-#if !defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if !defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
 template <>
 LIBC_INLINE cpp::optional<ExpandedFloat<long double>>
 eisel_lemire<long double>(ExpandedFloat<long double> init_num,
@@ -297,7 +297,7 @@ eisel_lemire<long double>(ExpandedFloat<long double> init_num,
   output.exponent = exp2;
   return output;
 }
-#endif
+#endif // !defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
 
 // The nth item in POWERS_OF_TWO represents the greatest power of two less than
 // 10^n. This tells us how much we can safely shift without overshooting.
@@ -460,7 +460,7 @@ template <> class ClingerConsts<double> {
   static constexpr double MAX_EXACT_INT = 9007199254740991.0;
 };
 
-#if defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
 template <> class ClingerConsts<long double> {
 public:
   static constexpr long double POWERS_OF_TEN_ARRAY[] = {
@@ -473,7 +473,7 @@ template <> class ClingerConsts<long double> {
   static constexpr long double MAX_EXACT_INT =
       ClingerConsts<double>::MAX_EXACT_INT;
 };
-#elif defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 template <> class ClingerConsts<long double> {
 public:
   static constexpr long double POWERS_OF_TEN_ARRAY[] = {
@@ -484,7 +484,7 @@ template <> class ClingerConsts<long double> {
   static constexpr int32_t DIGITS_IN_MANTISSA = 21;
   static constexpr long double MAX_EXACT_INT = 18446744073709551615.0L;
 };
-#else
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
 template <> class ClingerConsts<long double> {
 public:
   static constexpr long double POWERS_OF_TEN_ARRAY[] = {
@@ -498,6 +498,8 @@ template <> class ClingerConsts<long double> {
   static constexpr long double MAX_EXACT_INT =
       10384593717069655257060992658440191.0L;
 };
+#else
+#error "Unknown long double type"
 #endif
 
 // Take an exact mantissa and exponent and attempt to convert it using only

diff --git a/libc/test/src/__support/FPUtil/fpbits_test.cpp b/libc/test/src/__support/FPUtil/fpbits_test.cpp
@@ -503,7 +503,7 @@ TEST(LlvmLibcFPBitsTest, X86LongDoubleType) {
 }
 #else
 TEST(LlvmLibcFPBitsTest, LongDoubleType) {
-#if defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   return; // The tests for the "double" type cover for this case.
 #else
   using LongDoubleBits = FPBits<long double>;

diff --git a/libc/test/src/__support/str_to_long_double_test.cpp b/libc/test/src/__support/str_to_long_double_test.cpp
@@ -7,13 +7,13 @@ namespace LIBC_NAMESPACE {
 using LlvmLibcStrToLongDblTest = LlvmLibcStrToFloatTest<long double>;
 using LIBC_NAMESPACE::operator""_u128;
 
-#if defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
 
 TEST_F(LlvmLibcStrToLongDblTest, EiselLemireFloat64AsLongDouble) {
   eisel_lemire_test(123, 0, 0x1EC00000000000, 1029);
 }
 
-#elif defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 
 TEST_F(LlvmLibcStrToLongDblTest, EiselLemireFloat80Simple) {
   eisel_lemire_test(123, 0, 0xf600000000000000, 16389);
@@ -54,7 +54,7 @@ TEST_F(LlvmLibcStrToLongDblTest, EiselLemireFloat80Fallback) {
   ASSERT_FALSE(internal::eisel_lemire<long double>({1, -1000}).has_value());
 }
 
-#else // Quad precision long double
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
 
 TEST_F(LlvmLibcStrToLongDblTest, EiselLemireFloat128Simple) {
   eisel_lemire_test(123, 0, 0x1ec00'00000000'00000000'00000000_u128, 16389);
@@ -77,6 +77,8 @@ TEST_F(LlvmLibcStrToLongDblTest, EiselLemireFloat128Fallback) {
                    .has_value());
 }
 
+#else
+#error "Unknown long double type"
 #endif
 
 } // namespace LIBC_NAMESPACE
diff --git a/libc/test/src/math/smoke/nanl_test.cpp b/libc/test/src/math/smoke/nanl_test.cpp
@@ -12,12 +12,14 @@
 #include "test/UnitTest/Test.h"
 #include <signal.h>
 
-#if defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
 #define SELECT_LONG_DOUBLE(val, _, __) val
-#elif defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 #define SELECT_LONG_DOUBLE(_, val, __) val
-#else
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
 #define SELECT_LONG_DOUBLE(_, __, val) val
+#else
+#error "Unknown long double type"
 #endif
 
 class LlvmLibcNanlTest : public LIBC_NAMESPACE::testing::Test {

diff --git a/libc/test/src/stdio/sprintf_test.cpp b/libc/test/src/stdio/sprintf_test.cpp
@@ -775,29 +775,29 @@ TEST_F(LlvmLibcSPrintfTest, FloatHexExpConv) {
   // Length Modifier Tests.
 
   written = LIBC_NAMESPACE::sprintf(buff, "%La", 0.1L);
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
   ASSERT_STREQ_LEN(written, buff, "0xc.ccccccccccccccdp-7");
-#elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   ASSERT_STREQ_LEN(written, buff, "0x1.999999999999ap-4");
-#else // 128 bit long double
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
   ASSERT_STREQ_LEN(written, buff, "0x1.999999999999999999999999999ap-4");
 #endif
 
   written = LIBC_NAMESPACE::sprintf(buff, "%La", 1.0e1000L);
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
   ASSERT_STREQ_LEN(written, buff, "0xf.38db1f9dd3dac05p+3318");
-#elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   ASSERT_STREQ_LEN(written, buff, "inf");
-#else // 128 bit long double
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
   ASSERT_STREQ_LEN(written, buff, "0x1.e71b63f3ba7b580af1a52d2a7379p+3321");
 #endif
 
   written = LIBC_NAMESPACE::sprintf(buff, "%La", 1.0e-1000L);
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
   ASSERT_STREQ_LEN(written, buff, "0x8.68a9188a89e1467p-3325");
-#elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   ASSERT_STREQ_LEN(written, buff, "0x0p+0");
-#else // 128 bit long double
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
   ASSERT_STREQ_LEN(written, buff, "0x1.0d152311513c28ce202627c06ec2p-3322");
 #endif
 
@@ -899,20 +899,20 @@ TEST_F(LlvmLibcSPrintfTest, FloatHexExpConv) {
   ASSERT_STREQ_LEN(written, buff, "0x0p+0");
 
   written = LIBC_NAMESPACE::sprintf(buff, "%.1La", 0.1L);
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
   ASSERT_STREQ_LEN(written, buff, "0xc.dp-7");
-#elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
-#else // 128 bit long double
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
   ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
 #endif
 
   written = LIBC_NAMESPACE::sprintf(buff, "%.1La", 0xf.fffffffffffffffp16380L);
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
   ASSERT_STREQ_LEN(written, buff, "0x1.0p+16384");
-#elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   ASSERT_STREQ_LEN(written, buff, "inf");
-#else // 128 bit long double
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
   ASSERT_STREQ_LEN(written, buff, "0x2.0p+16383");
 #endif
 
@@ -1158,8 +1158,8 @@ TEST_F(LlvmLibcSPrintfTest, FloatDecimalConv) {
 
 // Some float128 systems (specifically the ones used for aarch64 buildbots)
 // don't respect signs for long double NaNs.
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80) ||                                \
-    defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80) ||                          \
+    defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   written = LIBC_NAMESPACE::sprintf(buff, "%LF", -ld_nan);
   ASSERT_STREQ_LEN(written, buff, "-NAN");
 #endif
@@ -1354,20 +1354,20 @@ TEST_F(LlvmLibcSPrintfTest, FloatDecimalConv) {
 
   /*
     written = LIBC_NAMESPACE::sprintf(buff, "%.1La", 0.1L);
-  #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+  #if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
     ASSERT_STREQ_LEN(written, buff, "0xc.dp-7");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
   #endif
 
     written = LIBC_NAMESPACE::sprintf(buff, "%.1La",
-  0xf.fffffffffffffffp16380L); #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
-    ASSERT_STREQ_LEN(written, buff, "0x1.0p+16384");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  0xf.fffffffffffffffp16380L); #if
+  defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80) ASSERT_STREQ_LEN(written, buff,
+  "0x1.0p+16384"); #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "inf");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x2.0p+16383");
   #endif
   */
@@ -1603,8 +1603,8 @@ TEST_F(LlvmLibcSPrintfTest, FloatDecimalLongDoubleConv) {
 
   // Length Modifier Tests.
 
-  // TODO(michaelrj): Add tests for LIBC_LONG_DOUBLE_IS_FLOAT64 and 128 bit long
-  // double systems.
+  // TODO(michaelrj): Add tests for LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64 and 128
+  // bit long double systems.
   // TODO(michaelrj): Fix the tests to only depend on the digits the long double
   // is accurate for.
 
@@ -1614,7 +1614,7 @@ TEST_F(LlvmLibcSPrintfTest, FloatDecimalLongDoubleConv) {
   written = LIBC_NAMESPACE::sprintf(buff, "%.Lf", -2.5L);
   ASSERT_STREQ_LEN(written, buff, "-2");
 
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 
   written = LIBC_NAMESPACE::sprintf(buff, "%Lf", 1e100L);
   ASSERT_STREQ_LEN(written, buff,
@@ -1930,7 +1930,7 @@ TEST_F(LlvmLibcSPrintfTest, FloatDecimalLongDoubleConv) {
       "570449525088342437216896462077260223998756027453411520977536701491759878"
       "422771447006016890777855573925295187921971811871399320142563330377888532"
       "179817332113");
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 }
 
 TEST_F(LlvmLibcSPrintfTest, FloatExponentConv) {
@@ -2171,20 +2171,20 @@ TEST_F(LlvmLibcSPrintfTest, FloatExponentConv) {
 
   /*
     written = LIBC_NAMESPACE::sprintf(buff, "%.1La", 0.1L);
-  #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+  #if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
     ASSERT_STREQ_LEN(written, buff, "0xc.dp-7");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
   #endif
 
     written = LIBC_NAMESPACE::sprintf(buff, "%.1La",
-  0xf.fffffffffffffffp16380L); #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
-    ASSERT_STREQ_LEN(written, buff, "0x1.0p+16384");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  0xf.fffffffffffffffp16380L); #if
+  defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80) ASSERT_STREQ_LEN(written, buff,
+  "0x1.0p+16384"); #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "inf");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x2.0p+16383");
   #endif
   */
@@ -2423,7 +2423,7 @@ TEST_F(LlvmLibcSPrintfTest, FloatExponentLongDoubleConv) {
   ForceRoundingMode r(RoundingMode::Nearest);
   // Length Modifier Tests.
 
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
   written = LIBC_NAMESPACE::sprintf(buff, "%.9Le", 1000000000500000000.1L);
   ASSERT_STREQ_LEN(written, buff, "1.000000001e+18");
 
@@ -2783,34 +2783,34 @@ TEST_F(LlvmLibcSPrintfTest, FloatAutoConv) {
   written = LIBC_NAMESPACE::sprintf(buff, "%.10g", 0x1.0p-1074);
   ASSERT_STREQ_LEN(written, buff, "4.940656458e-324");
 
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 
   written = LIBC_NAMESPACE::sprintf(buff, "%.60Lg", 0xa.aaaaaaaaaaaaaabp-7L);
   ASSERT_STREQ_LEN(
       written, buff,
       "0.0833333333333333333355920878593448009041821933351457118988037");
 
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
   // Long double precision tests.
   // These are currently commented out because they require long double support
   // that isn't ready yet.
   /*
     written = LIBC_NAMESPACE::sprintf(buff, "%.1La", 0.1L);
-  #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+  #if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
     ASSERT_STREQ_LEN(written, buff, "0xc.dp-7");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x1.ap-4");
   #endif
 
     written = LIBC_NAMESPACE::sprintf(buff, "%.1La",
-  0xf.fffffffffffffffp16380L); #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
-    ASSERT_STREQ_LEN(written, buff, "0x1.0p+16384");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  0xf.fffffffffffffffp16380L); #if
+  defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80) ASSERT_STREQ_LEN(written, buff,
+  "0x1.0p+16384"); #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "inf");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x2.0p+16383");
   #endif
   */
@@ -3053,7 +3053,7 @@ TEST_F(LlvmLibcSPrintfTest, FloatAutoLongDoubleConv) {
 
   // Length Modifier Tests.
 
-#if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 
   written = LIBC_NAMESPACE::sprintf(buff, "%Lg", 0xf.fffffffffffffffp+16380L);
   ASSERT_STREQ_LEN(written, buff, "1.18973e+4932");
@@ -3064,7 +3064,7 @@ TEST_F(LlvmLibcSPrintfTest, FloatAutoLongDoubleConv) {
   written = LIBC_NAMESPACE::sprintf(buff, "%Lg", 9.99999999999e-100L);
   ASSERT_STREQ_LEN(written, buff, "1e-99");
 
-#endif // LIBC_LONG_DOUBLE_IS_X86_FLOAT80
+#endif // LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80
 
   // TODO: Uncomment the below tests after long double support is added
   /*
@@ -3171,29 +3171,29 @@ TEST_F(LlvmLibcSPrintfTest, FloatAutoLongDoubleConv) {
 */
   /*
     written = LIBC_NAMESPACE::sprintf(buff, "%La", 0.1L);
-  #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+  #if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
     ASSERT_STREQ_LEN(written, buff, "0xc.ccccccccccccccdp-7");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "0x1.999999999999ap-4");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x1.999999999999999999999999999ap-4");
   #endif
 
     written = LIBC_NAMESPACE::sprintf(buff, "%La", 1.0e1000L);
-  #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+  #if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
     ASSERT_STREQ_LEN(written, buff, "0xf.38db1f9dd3dac05p+3318");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "inf");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x1.e71b63f3ba7b580af1a52d2a7379p+3321");
   #endif
 
     written = LIBC_NAMESPACE::sprintf(buff, "%La", 1.0e-1000L);
-  #if defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+  #if defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
     ASSERT_STREQ_LEN(written, buff, "0x8.68a9188a89e1467p-3325");
-  #elif defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
     ASSERT_STREQ_LEN(written, buff, "0x0p+0");
-  #else // 128 bit long double
+  #elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
     ASSERT_STREQ_LEN(written, buff, "0x1.0d152311513c28ce202627c06ec2p-3322");
   #endif
   */

diff --git a/libc/test/src/stdio/sscanf_test.cpp b/libc/test/src/stdio/sscanf_test.cpp
@@ -322,7 +322,7 @@ TEST(LlvmLibcSScanfTest, FloatConvLengthModifier) {
   EXPECT_EQ(ret_val, 1);
 // 1e600 may be larger than the maximum long double (if long double is double).
 // In that case both of these should be evaluated as inf.
-#ifdef LIBC_LONG_DOUBLE_IS_FLOAT64
+#ifdef LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64
   EXPECT_FP_EQ(ld_result, d_inf);
 #else
   EXPECT_FP_EQ(ld_result, 1.0e600L);

diff --git a/libc/test/src/stdlib/strtold_test.cpp b/libc/test/src/stdlib/strtold_test.cpp
@@ -15,17 +15,19 @@
 
 #include <stddef.h>
 
-#if defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
 #define SELECT_CONST(val, _, __) val
-#elif defined(LIBC_LONG_DOUBLE_IS_X86_FLOAT80)
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_X86_FLOAT80)
 #define SELECT_CONST(_, val, __) val
-#else
+#elif defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT128)
 #define SELECT_CONST(_, __, val) val
+#else
+#error "Unknown long double type"
 #endif
 
 class LlvmLibcStrToLDTest : public LIBC_NAMESPACE::testing::Test {
 public:
-#if defined(LIBC_LONG_DOUBLE_IS_FLOAT64)
+#if defined(LIBC_TYPES_LONG_DOUBLE_IS_FLOAT64)
   void run_test(const char *inputString, const ptrdiff_t expectedStrLen,
                 const uint64_t expectedRawData, const int expectedErrno = 0)
 #else

diff --git a/libcxx/include/CMakeLists.txt b/libcxx/include/CMakeLists.txt
@@ -431,7 +431,7 @@ set(files
   __fwd/bit_reference.h
   __fwd/complex.h
   __fwd/fstream.h
-  __fwd/hash.h
+  __fwd/functional.h
   __fwd/ios.h
   __fwd/istream.h
   __fwd/mdspan.h

diff --git a/libcxx/include/__atomic/atomic_base.h b/libcxx/include/__atomic/atomic_base.h
@@ -104,24 +104,20 @@ struct __atomic_base // false
 
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void wait(_Tp __v, memory_order __m = memory_order_seq_cst) const
       volatile _NOEXCEPT {
-    std::__cxx_atomic_wait(std::addressof(__a_), __v, __m);
+    std::__atomic_wait(*this, __v, __m);
   }
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void
   wait(_Tp __v, memory_order __m = memory_order_seq_cst) const _NOEXCEPT {
-    std::__cxx_atomic_wait(std::addressof(__a_), __v, __m);
+    std::__atomic_wait(*this, __v, __m);
   }
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_one() volatile _NOEXCEPT {
-    std::__cxx_atomic_notify_one(std::addressof(__a_));
-  }
-  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_one() _NOEXCEPT {
-    std::__cxx_atomic_notify_one(std::addressof(__a_));
+    std::__atomic_notify_one(*this);
   }
+  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_one() _NOEXCEPT { std::__atomic_notify_one(*this); }
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_all() volatile _NOEXCEPT {
-    std::__cxx_atomic_notify_all(std::addressof(__a_));
-  }
-  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_all() _NOEXCEPT {
-    std::__cxx_atomic_notify_all(std::addressof(__a_));
+    std::__atomic_notify_all(*this);
   }
+  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_all() _NOEXCEPT { std::__atomic_notify_all(*this); }
 
 #if _LIBCPP_STD_VER >= 20
   _LIBCPP_HIDE_FROM_ABI constexpr __atomic_base() noexcept(is_nothrow_default_constructible_v<_Tp>) : __a_(_Tp()) {}
@@ -200,6 +196,32 @@ struct __atomic_base<_Tp, true> : public __atomic_base<_Tp, false> {
   _LIBCPP_HIDE_FROM_ABI _Tp operator^=(_Tp __op) _NOEXCEPT { return fetch_xor(__op) ^ __op; }
 };
 
+// Here we need _IsIntegral because the default template argument is not enough
+// e.g  __atomic_base<int> is __atomic_base<int, true>, which inherits from
+// __atomic_base<int, false> and the caller of the wait function is
+// __atomic_base<int, false>. So specializing __atomic_base<_Tp> does not work
+template <class _Tp, bool _IsIntegral>
+struct __atomic_waitable_traits<__atomic_base<_Tp, _IsIntegral> > {
+  static _LIBCPP_HIDE_FROM_ABI _Tp __atomic_load(const __atomic_base<_Tp, _IsIntegral>& __a, memory_order __order) {
+    return __a.load(__order);
+  }
+
+  static _LIBCPP_HIDE_FROM_ABI _Tp
+  __atomic_load(const volatile __atomic_base<_Tp, _IsIntegral>& __this, memory_order __order) {
+    return __this.load(__order);
+  }
+
+  static _LIBCPP_HIDE_FROM_ABI const __cxx_atomic_impl<_Tp>*
+  __atomic_contention_address(const __atomic_base<_Tp, _IsIntegral>& __a) {
+    return std::addressof(__a.__a_);
+  }
+
+  static _LIBCPP_HIDE_FROM_ABI const volatile __cxx_atomic_impl<_Tp>*
+  __atomic_contention_address(const volatile __atomic_base<_Tp, _IsIntegral>& __this) {
+    return std::addressof(__this.__a_);
+  }
+};
+
 _LIBCPP_END_NAMESPACE_STD
 
 #endif // _LIBCPP___ATOMIC_ATOMIC_BASE_H
diff --git a/libcxx/include/__atomic/atomic_flag.h b/libcxx/include/__atomic/atomic_flag.h
@@ -16,6 +16,7 @@
 #include <__availability>
 #include <__chrono/duration.h>
 #include <__config>
+#include <__memory/addressof.h>
 #include <__thread/support.h>
 #include <cstdint>
 
@@ -50,20 +51,20 @@ struct atomic_flag {
 
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void wait(bool __v, memory_order __m = memory_order_seq_cst) const
       volatile _NOEXCEPT {
-    __cxx_atomic_wait(&__a_, _LIBCPP_ATOMIC_FLAG_TYPE(__v), __m);
+    std::__atomic_wait(*this, _LIBCPP_ATOMIC_FLAG_TYPE(__v), __m);
   }
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void
   wait(bool __v, memory_order __m = memory_order_seq_cst) const _NOEXCEPT {
-    __cxx_atomic_wait(&__a_, _LIBCPP_ATOMIC_FLAG_TYPE(__v), __m);
+    std::__atomic_wait(*this, _LIBCPP_ATOMIC_FLAG_TYPE(__v), __m);
   }
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_one() volatile _NOEXCEPT {
-    __cxx_atomic_notify_one(&__a_);
+    std::__atomic_notify_one(*this);
   }
-  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_one() _NOEXCEPT { __cxx_atomic_notify_one(&__a_); }
+  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_one() _NOEXCEPT { std::__atomic_notify_one(*this); }
   _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_all() volatile _NOEXCEPT {
-    __cxx_atomic_notify_all(&__a_);
+    std::__atomic_notify_all(*this);
   }
-  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_all() _NOEXCEPT { __cxx_atomic_notify_all(&__a_); }
+  _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void notify_all() _NOEXCEPT { std::__atomic_notify_all(*this); }
 
 #if _LIBCPP_STD_VER >= 20
   _LIBCPP_HIDE_FROM_ABI constexpr atomic_flag() _NOEXCEPT : __a_(false) {}
@@ -78,6 +79,28 @@ struct atomic_flag {
   atomic_flag& operator=(const atomic_flag&) volatile = delete;
 };
 
+template <>
+struct __atomic_waitable_traits<atomic_flag> {
+  static _LIBCPP_HIDE_FROM_ABI _LIBCPP_ATOMIC_FLAG_TYPE __atomic_load(const atomic_flag& __a, memory_order __order) {
+    return std::__cxx_atomic_load(&__a.__a_, __order);
+  }
+
+  static _LIBCPP_HIDE_FROM_ABI _LIBCPP_ATOMIC_FLAG_TYPE
+  __atomic_load(const volatile atomic_flag& __a, memory_order __order) {
+    return std::__cxx_atomic_load(&__a.__a_, __order);
+  }
+
+  static _LIBCPP_HIDE_FROM_ABI const __cxx_atomic_impl<_LIBCPP_ATOMIC_FLAG_TYPE>*
+  __atomic_contention_address(const atomic_flag& __a) {
+    return std::addressof(__a.__a_);
+  }
+
+  static _LIBCPP_HIDE_FROM_ABI const volatile __cxx_atomic_impl<_LIBCPP_ATOMIC_FLAG_TYPE>*
+  __atomic_contention_address(const volatile atomic_flag& __a) {
+    return std::addressof(__a.__a_);
+  }
+};
+
 inline _LIBCPP_HIDE_FROM_ABI bool atomic_flag_test(const volatile atomic_flag* __o) _NOEXCEPT { return __o->test(); }
 
 inline _LIBCPP_HIDE_FROM_ABI bool atomic_flag_test(const atomic_flag* __o) _NOEXCEPT { return __o->test(); }

diff --git a/libcxx/include/__atomic/atomic_sync.h b/libcxx/include/__atomic/atomic_sync.h
@@ -18,7 +18,11 @@
 #include <__memory/addressof.h>
 #include <__thread/poll_with_backoff.h>
 #include <__thread/support.h>
+#include <__type_traits/conjunction.h>
 #include <__type_traits/decay.h>
+#include <__type_traits/invoke.h>
+#include <__type_traits/void_t.h>
+#include <__utility/declval.h>
 #include <cstring>
 
 #if !defined(_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER)
@@ -27,15 +31,40 @@
 
 _LIBCPP_BEGIN_NAMESPACE_STD
 
-template <class _Atp, class _Poll>
-struct __libcpp_atomic_wait_poll_impl {
-  _Atp* __a_;
+// The customisation points to enable the following functions:
+// - __atomic_wait
+// - __atomic_wait_unless
+// - __atomic_notify_one
+// - __atomic_notify_all
+// Note that std::atomic<T>::wait was back-ported to C++03
+// The below implementations look ugly to support C++03
+template <class _Tp, class = void>
+struct __atomic_waitable_traits {
+  template <class _AtomicWaitable>
+  static void __atomic_load(_AtomicWaitable&&, memory_order) = delete;
+
+  template <class _AtomicWaitable>
+  static void __atomic_contention_address(_AtomicWaitable&&) = delete;
+};
+
+template <class _Tp, class = void>
+struct __atomic_waitable : false_type {};
+
+template <class _Tp>
+struct __atomic_waitable< _Tp,
+                          __void_t<decltype(__atomic_waitable_traits<__decay_t<_Tp> >::__atomic_load(
+                                       std::declval<const _Tp&>(), std::declval<memory_order>())),
+                                   decltype(__atomic_waitable_traits<__decay_t<_Tp> >::__atomic_contention_address(
+                                       std::declval<const _Tp&>()))> > : true_type {};
+
+template <class _AtomicWaitable, class _Poll>
+struct __atomic_wait_poll_impl {
+  const _AtomicWaitable& __a_;
   _Poll __poll_;
   memory_order __order_;
 
-  _LIBCPP_AVAILABILITY_SYNC
   _LIBCPP_HIDE_FROM_ABI bool operator()() const {
-    auto __current_val = std::__cxx_atomic_load(__a_, __order_);
+    auto __current_val = __atomic_waitable_traits<__decay_t<_AtomicWaitable> >::__atomic_load(__a_, __order_);
     return __poll_(__current_val);
   }
 };
@@ -56,42 +85,45 @@ __libcpp_atomic_monitor(__cxx_atomic_contention_t const volatile*);
 _LIBCPP_AVAILABILITY_SYNC _LIBCPP_EXPORTED_FROM_ABI void
 __libcpp_atomic_wait(__cxx_atomic_contention_t const volatile*, __cxx_contention_t);
 
-template <class _Atp, class _Poll>
-struct __libcpp_atomic_wait_backoff_impl {
-  _Atp* __a_;
+template <class _AtomicWaitable, class _Poll>
+struct __atomic_wait_backoff_impl {
+  const _AtomicWaitable& __a_;
   _Poll __poll_;
   memory_order __order_;
 
+  using __waitable_traits = __atomic_waitable_traits<__decay_t<_AtomicWaitable> >;
+
   _LIBCPP_AVAILABILITY_SYNC
   _LIBCPP_HIDE_FROM_ABI bool
-  __poll_or_get_monitor(__cxx_atomic_contention_t const volatile*, __cxx_contention_t& __monitor) const {
-    // In case the atomic can be waited on directly, the monitor value is just
-    // the value of the atomic.
+  __update_monitor_val_and_poll(__cxx_atomic_contention_t const volatile*, __cxx_contention_t& __monitor_val) const {
+    // In case the contention type happens to be __cxx_atomic_contention_t, i.e. __cxx_atomic_impl<int64_t>,
+    // the platform wait is directly monitoring the atomic value itself.
     // `__poll_` takes the current value of the atomic as an in-out argument
     // to potentially modify it. After it returns, `__monitor` has a value
     // which can be safely waited on by `std::__libcpp_atomic_wait` without any
     // ABA style issues.
-    __monitor = std::__cxx_atomic_load(__a_, __order_);
-    return __poll_(__monitor);
+    __monitor_val = __waitable_traits::__atomic_load(__a_, __order_);
+    return __poll_(__monitor_val);
   }
 
   _LIBCPP_AVAILABILITY_SYNC
-  _LIBCPP_HIDE_FROM_ABI bool __poll_or_get_monitor(void const volatile*, __cxx_contention_t& __monitor) const {
-    // In case we must wait on an atomic from the pool, the monitor comes from
-    // `std::__libcpp_atomic_monitor`.
-    // Only then we may read from `__a_`. This is the "event count" pattern.
-    __monitor          = std::__libcpp_atomic_monitor(__a_);
-    auto __current_val = std::__cxx_atomic_load(__a_, __order_);
+  _LIBCPP_HIDE_FROM_ABI bool
+  __update_monitor_val_and_poll(void const volatile* __contention_address, __cxx_contention_t& __monitor_val) const {
+    // In case the contention type is anything else, platform wait is monitoring a __cxx_atomic_contention_t
+    // from the global pool, the monitor comes from __libcpp_atomic_monitor
+    __monitor_val      = std::__libcpp_atomic_monitor(__contention_address);
+    auto __current_val = __waitable_traits::__atomic_load(__a_, __order_);
     return __poll_(__current_val);
   }
 
   _LIBCPP_AVAILABILITY_SYNC
   _LIBCPP_HIDE_FROM_ABI bool operator()(chrono::nanoseconds __elapsed) const {
     if (__elapsed > chrono::microseconds(64)) {
-      __cxx_contention_t __monitor;
-      if (__poll_or_get_monitor(__a_, __monitor))
+      auto __contention_address = __waitable_traits::__atomic_contention_address(__a_);
+      __cxx_contention_t __monitor_val;
+      if (__update_monitor_val_and_poll(__contention_address, __monitor_val))
         return true;
-      std::__libcpp_atomic_wait(__a_, __monitor);
+      std::__libcpp_atomic_wait(__contention_address, __monitor_val);
     } else if (__elapsed > chrono::microseconds(4))
       __libcpp_thread_yield();
     else {
@@ -105,29 +137,44 @@ struct __libcpp_atomic_wait_backoff_impl {
 // predicate (is the loaded value unequal to `old`?), the predicate function is
 // specified as an argument. The loaded value is given as an in-out argument to
 // the predicate. If the predicate function returns `true`,
-// `_cxx_atomic_wait_unless` will return. If the predicate function returns
+// `__atomic_wait_unless` will return. If the predicate function returns
 // `false`, it must set the argument to its current understanding of the atomic
 // value. The predicate function must not return `false` spuriously.
-template <class _Atp, class _Poll>
+template <class _AtomicWaitable, class _Poll>
 _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void
-__cxx_atomic_wait_unless(_Atp* __a, _Poll&& __poll, memory_order __order) {
-  __libcpp_atomic_wait_poll_impl<_Atp, __decay_t<_Poll> > __poll_fn       = {__a, __poll, __order};
-  __libcpp_atomic_wait_backoff_impl<_Atp, __decay_t<_Poll> > __backoff_fn = {__a, __poll, __order};
-  (void)std::__libcpp_thread_poll_with_backoff(__poll_fn, __backoff_fn);
+__atomic_wait_unless(const _AtomicWaitable& __a, _Poll&& __poll, memory_order __order) {
+  static_assert(__atomic_waitable<_AtomicWaitable>::value, "");
+  __atomic_wait_poll_impl<_AtomicWaitable, __decay_t<_Poll> > __poll_impl     = {__a, __poll, __order};
+  __atomic_wait_backoff_impl<_AtomicWaitable, __decay_t<_Poll> > __backoff_fn = {__a, __poll, __order};
+  std::__libcpp_thread_poll_with_backoff(__poll_impl, __backoff_fn);
+}
+
+template <class _AtomicWaitable>
+_LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void __atomic_notify_one(const _AtomicWaitable& __a) {
+  static_assert(__atomic_waitable<_AtomicWaitable>::value, "");
+  std::__cxx_atomic_notify_one(__atomic_waitable_traits<__decay_t<_AtomicWaitable> >::__atomic_contention_address(__a));
+}
+
+template <class _AtomicWaitable>
+_LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void __atomic_notify_all(const _AtomicWaitable& __a) {
+  static_assert(__atomic_waitable<_AtomicWaitable>::value, "");
+  std::__cxx_atomic_notify_all(__atomic_waitable_traits<__decay_t<_AtomicWaitable> >::__atomic_contention_address(__a));
 }
 
 #else // _LIBCPP_HAS_NO_THREADS
 
-template <class _Tp>
-_LIBCPP_HIDE_FROM_ABI void __cxx_atomic_notify_all(__cxx_atomic_impl<_Tp> const volatile*) {}
-template <class _Tp>
-_LIBCPP_HIDE_FROM_ABI void __cxx_atomic_notify_one(__cxx_atomic_impl<_Tp> const volatile*) {}
-template <class _Atp, class _Poll>
-_LIBCPP_HIDE_FROM_ABI void __cxx_atomic_wait_unless(_Atp* __a, _Poll&& __poll, memory_order __order) {
-  __libcpp_atomic_wait_poll_impl<_Atp, __decay_t<_Poll> > __poll_fn = {__a, __poll, __order};
-  (void)std::__libcpp_thread_poll_with_backoff(__poll_fn, __spinning_backoff_policy());
+template <class _AtomicWaitable, class _Poll>
+_LIBCPP_HIDE_FROM_ABI void __atomic_wait_unless(const _AtomicWaitable& __a, _Poll&& __poll, memory_order __order) {
+  __atomic_wait_poll_impl<_AtomicWaitable, __decay_t<_Poll> > __poll_fn = {__a, __poll, __order};
+  std::__libcpp_thread_poll_with_backoff(__poll_fn, __spinning_backoff_policy());
 }
 
+template <class _AtomicWaitable>
+_LIBCPP_HIDE_FROM_ABI void __atomic_notify_one(const _AtomicWaitable&) {}
+
+template <class _AtomicWaitable>
+_LIBCPP_HIDE_FROM_ABI void __atomic_notify_all(const _AtomicWaitable&) {}
+
 #endif // _LIBCPP_HAS_NO_THREADS
 
 template <typename _Tp>
@@ -143,11 +190,12 @@ struct __atomic_compare_unequal_to {
   }
 };
 
-template <class _Atp, class _Tp>
+template <class _AtomicWaitable, class _Up>
 _LIBCPP_AVAILABILITY_SYNC _LIBCPP_HIDE_FROM_ABI void
-__cxx_atomic_wait(_Atp* __a, _Tp const __val, memory_order __order) {
-  __atomic_compare_unequal_to<_Tp> __poll_fn = {__val};
-  std::__cxx_atomic_wait_unless(__a, __poll_fn, __order);
+__atomic_wait(_AtomicWaitable& __a, _Up __val, memory_order __order) {
+  static_assert(__atomic_waitable<_AtomicWaitable>::value, "");
+  __atomic_compare_unequal_to<_Up> __nonatomic_equal = {__val};
+  std::__atomic_wait_unless(__a, __nonatomic_equal, __order);
 }
 
 _LIBCPP_END_NAMESPACE_STD

diff --git a/libcxx/include/__filesystem/path.h b/libcxx/include/__filesystem/path.h
@@ -14,9 +14,8 @@
 #include <__algorithm/replace_copy.h>
 #include <__availability>
 #include <__config>
-#include <__functional/hash.h>
 #include <__functional/unary_function.h>
-#include <__fwd/hash.h>
+#include <__fwd/functional.h>
 #include <__iterator/back_insert_iterator.h>
 #include <__iterator/iterator_traits.h>
 #include <__type_traits/decay.h>

diff --git a/libcxx/include/__format/parser_std_format_spec.h b/libcxx/include/__format/parser_std_format_spec.h
@@ -355,10 +355,10 @@ class _LIBCPP_TEMPLATE_VIS __parser {
   _LIBCPP_HIDE_FROM_ABI constexpr typename _ParseContext::iterator __parse(_ParseContext& __ctx, __fields __fields) {
     auto __begin = __ctx.begin();
     auto __end   = __ctx.end();
-    if (__begin == __end || *__begin == _CharT('}'))
+    if (__begin == __end || *__begin == _CharT('}') || (__fields.__use_range_fill_ && *__begin == _CharT(':')))
       return __begin;
 
-    if (__parse_fill_align(__begin, __end, __fields.__use_range_fill_) && __begin == __end)
+    if (__parse_fill_align(__begin, __end) && __begin == __end)
       return __begin;
 
     if (__fields.__sign_) {
@@ -574,12 +574,10 @@ class _LIBCPP_TEMPLATE_VIS __parser {
     return false;
   }
 
-  _LIBCPP_HIDE_FROM_ABI constexpr void __validate_fill_character(_CharT __fill, bool __use_range_fill) {
+  _LIBCPP_HIDE_FROM_ABI constexpr void __validate_fill_character(_CharT __fill) {
     // The forbidden fill characters all code points formed from a single code unit, thus the
     // check can be omitted when more code units are used.
-    if (__use_range_fill && (__fill == _CharT('{') || __fill == _CharT(':')))
-      std::__throw_format_error("The fill option contains an invalid value");
-    else if (__fill == _CharT('{'))
+    if (__fill == _CharT('{'))
       std::__throw_format_error("The fill option contains an invalid value");
   }
 
@@ -590,7 +588,7 @@ class _LIBCPP_TEMPLATE_VIS __parser {
 #    ifndef _LIBCPP_HAS_NO_WIDE_CHARACTERS
           || (same_as<_CharT, wchar_t> && sizeof(wchar_t) == 2)
 #    endif
-  _LIBCPP_HIDE_FROM_ABI constexpr bool __parse_fill_align(_Iterator& __begin, _Iterator __end, bool __use_range_fill) {
+  _LIBCPP_HIDE_FROM_ABI constexpr bool __parse_fill_align(_Iterator& __begin, _Iterator __end) {
     _LIBCPP_ASSERT_VALID_ELEMENT_ACCESS(
         __begin != __end,
         "when called with an empty input the function will cause "
@@ -606,7 +604,7 @@ class _LIBCPP_TEMPLATE_VIS __parser {
         // The forbidden fill characters all are code points encoded
         // in one code unit, thus the check can be omitted when more
         // code units are used.
-        __validate_fill_character(*__begin, __use_range_fill);
+        __validate_fill_character(*__begin);
 
       std::copy_n(__begin, __code_units, std::addressof(__fill_.__data[0]));
       __begin += __code_units + 1;
@@ -623,7 +621,7 @@ class _LIBCPP_TEMPLATE_VIS __parser {
 #    ifndef _LIBCPP_HAS_NO_WIDE_CHARACTERS
   template <contiguous_iterator _Iterator>
     requires(same_as<_CharT, wchar_t> && sizeof(wchar_t) == 4)
-  _LIBCPP_HIDE_FROM_ABI constexpr bool __parse_fill_align(_Iterator& __begin, _Iterator __end, bool __use_range_fill) {
+  _LIBCPP_HIDE_FROM_ABI constexpr bool __parse_fill_align(_Iterator& __begin, _Iterator __end) {
     _LIBCPP_ASSERT_VALID_ELEMENT_ACCESS(
         __begin != __end,
         "when called with an empty input the function will cause "
@@ -632,7 +630,7 @@ class _LIBCPP_TEMPLATE_VIS __parser {
       if (!__unicode::__is_scalar_value(*__begin))
         std::__throw_format_error("The fill option contains an invalid value");
 
-      __validate_fill_character(*__begin, __use_range_fill);
+      __validate_fill_character(*__begin);
 
       __fill_.__data[0] = *__begin;
       __begin += 2;
@@ -651,14 +649,14 @@ class _LIBCPP_TEMPLATE_VIS __parser {
 #  else // _LIBCPP_HAS_NO_UNICODE
   // range-fill and tuple-fill are identical
   template <contiguous_iterator _Iterator>
-  _LIBCPP_HIDE_FROM_ABI constexpr bool __parse_fill_align(_Iterator& __begin, _Iterator __end, bool __use_range_fill) {
+  _LIBCPP_HIDE_FROM_ABI constexpr bool __parse_fill_align(_Iterator& __begin, _Iterator __end) {
     _LIBCPP_ASSERT_VALID_ELEMENT_ACCESS(
         __begin != __end,
         "when called with an empty input the function will cause "
         "undefined behavior by evaluating data not in the input");
     if (__begin + 1 != __end) {
       if (__parse_alignment(*(__begin + 1))) {
-        __validate_fill_character(*__begin, __use_range_fill);
+        __validate_fill_character(*__begin);
 
         __fill_.__data[0] = *__begin;
         __begin += 2;

diff --git a/libcxx/include/__functional/bind.h b/libcxx/include/__functional/bind.h
@@ -13,6 +13,7 @@
 #include <__config>
 #include <__functional/invoke.h>
 #include <__functional/weak_result_type.h>
+#include <__fwd/functional.h>
 #include <__type_traits/decay.h>
 #include <__type_traits/is_reference_wrapper.h>
 #include <__type_traits/is_void.h>
-Original file line number
+Diff line change
@@ Expand Up / @@ -963,6 +963,8 @@ def StdC : StandardSpec<"stdc"> { @@
           ]
       >;
+      HeaderSpec StdInt = HeaderSpec<"StdInt.h">;
       HeaderSpec Limits = HeaderSpec<"limits.h">;
       NamedType SigAtomicT = NamedType<"sig_atomic_t">;
@@ Expand Down Expand Up / @@ -1268,6 +1270,7 @@ def StdC : StandardSpec<"stdc"> { @@
         Errno,
         Fenv,
         Float,
+        StdInt,
         Limits,
         Math,
         String,
@@ Expand Down @@