Vendor import of llvm-project branch release/17.x llvmorg-17.0.0-rc4-10-g0176e8729ea4.vendor/llvm-project/llvmorg-17.0.0-rc4-10-g0176e8729ea4

21 files changed, 297 insertions, 54 deletions

diff --git a/clang/include/clang/AST/ExprConcepts.h b/clang/include/clang/AST/ExprConcepts.h
index d900e980852b..13d4568119eb 100644
--- a/clang/include/clang/AST/ExprConcepts.h
+++ b/clang/include/clang/AST/ExprConcepts.h
@@ -14,20 +14,21 @@
 #ifndef LLVM_CLANG_AST_EXPRCONCEPTS_H
 #define LLVM_CLANG_AST_EXPRCONCEPTS_H
 
-#include "clang/AST/ASTContext.h"
 #include "clang/AST/ASTConcept.h"
+#include "clang/AST/ASTContext.h"
 #include "clang/AST/Decl.h"
-#include "clang/AST/DeclarationName.h"
 #include "clang/AST/DeclTemplate.h"
+#include "clang/AST/DeclarationName.h"
 #include "clang/AST/Expr.h"
 #include "clang/AST/NestedNameSpecifier.h"
 #include "clang/AST/TemplateBase.h"
 #include "clang/AST/Type.h"
 #include "clang/Basic/SourceLocation.h"
+#include "llvm/ADT/STLFunctionalExtras.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/TrailingObjects.h"
-#include <utility>
 #include <string>
+#include <utility>
 
 namespace clang {
 class ASTStmtReader;
@@ -467,6 +468,13 @@ public:
   }
 };
 
+using EntityPrinter = llvm::function_ref<void(llvm::raw_ostream &)>;
+
+/// \brief create a Requirement::SubstitutionDiagnostic with only a
+/// SubstitutedEntity and DiagLoc using Sema's allocator.
+Requirement::SubstitutionDiagnostic *
+createSubstDiagAt(Sema &S, SourceLocation Location, EntityPrinter Printer);
+
 } // namespace concepts
 
 /// C++2a [expr.prim.req]:
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
index a3506df7d4e5..f09d1129b128 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -2386,7 +2386,7 @@ void CodeGenModule::SetLLVMFunctionAttributesForDefinition(const Decl *D,
   // functions. If the current target's C++ ABI requires this and this is a
   // member function, set its alignment accordingly.
   if (getTarget().getCXXABI().areMemberFunctionsAligned()) {
-    if (F->getPointerAlignment(getDataLayout()) < 2 && isa<CXXMethodDecl>(D))
+    if (isa<CXXMethodDecl>(D) && F->getPointerAlignment(getDataLayout()) < 2)
       F->setAlignment(std::max(llvm::Align(2), F->getAlign().valueOrOne()));
   }
 
diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
index f6ea4d0b4366..bdbdad9362e1 100644
--- a/clang/lib/Driver/Driver.cpp
+++ b/clang/lib/Driver/Driver.cpp
@@ -4936,6 +4936,12 @@ void Driver::BuildJobs(Compilation &C) const {
   (void)C.getArgs().hasArg(options::OPT_driver_mode);
   (void)C.getArgs().hasArg(options::OPT_rsp_quoting);
 
+  bool HasAssembleJob = llvm::any_of(C.getJobs(), [](auto &J) {
+    // Match ClangAs and other derived assemblers of Tool. ClangAs uses a
+    // longer ShortName "clang integrated assembler" while other assemblers just
+    // use "assembler".
+    return strstr(J.getCreator().getShortName(), "assembler");
+  });
   for (Arg *A : C.getArgs()) {
     // FIXME: It would be nice to be able to send the argument to the
     // DiagnosticsEngine, so that extra values, position, and so on could be
@@ -4965,7 +4971,7 @@ void Driver::BuildJobs(Compilation &C) const {
       // already been warned about.
       if (!IsCLMode() || !A->getOption().matches(options::OPT_UNKNOWN)) {
         if (A->getOption().hasFlag(options::TargetSpecific) &&
-            !A->isIgnoredTargetSpecific()) {
+            !A->isIgnoredTargetSpecific() && !HasAssembleJob) {
           Diag(diag::err_drv_unsupported_opt_for_target)
               << A->getSpelling() << getTargetTriple();
         } else {
diff --git a/clang/lib/Driver/ToolChains/AIX.cpp b/clang/lib/Driver/ToolChains/AIX.cpp
index 97217eba9ca0..bfc86d9f3471 100644
--- a/clang/lib/Driver/ToolChains/AIX.cpp
+++ b/clang/lib/Driver/ToolChains/AIX.cpp
@@ -30,6 +30,7 @@ void aix::Assembler::ConstructJob(Compilation &C, const JobAction &JA,
                                   const InputInfoList &Inputs,
                                   const ArgList &Args,
                                   const char *LinkingOutput) const {
+  const Driver &D = getToolChain().getDriver();
   ArgStringList CmdArgs;
 
   const bool IsArch32Bit = getToolChain().getTriple().isArch32Bit();
@@ -38,6 +39,11 @@ void aix::Assembler::ConstructJob(Compilation &C, const JobAction &JA,
   if (!IsArch32Bit && !IsArch64Bit)
     llvm_unreachable("Unsupported bit width value.");
 
+  if (Arg *A = C.getArgs().getLastArg(options::OPT_G)) {
+    D.Diag(diag::err_drv_unsupported_opt_for_target)
+        << A->getSpelling() << D.getTargetTriple();
+  }
+
   // Specify the mode in which the as(1) command operates.
   if (IsArch32Bit) {
     CmdArgs.push_back("-a32");
diff --git a/clang/lib/Driver/ToolChains/Arch/X86.cpp b/clang/lib/Driver/ToolChains/Arch/X86.cpp
index 4383b8004143..cf2bc63d74ad 100644
--- a/clang/lib/Driver/ToolChains/Arch/X86.cpp
+++ b/clang/lib/Driver/ToolChains/Arch/X86.cpp
@@ -118,13 +118,7 @@ std::string x86::getX86TargetCPU(const Driver &D, const ArgList &Args,
 
 void x86::getX86TargetFeatures(const Driver &D, const llvm::Triple &Triple,
                                const ArgList &Args,
-                               std::vector<StringRef> &Features, bool ForAS) {
-  if (ForAS) {
-    // Some target-specific options are only handled in AddX86TargetArgs, which
-    // is not called by ClangAs::ConstructJob. Claim them here.
-    Args.claimAllArgs(options::OPT_mfpmath_EQ);
-  }
-
+                               std::vector<StringRef> &Features) {
   // Claim and report unsupported -mabi=. Note: we don't support "sysv_abi" or
   // "ms_abi" as default function attributes.
   if (const Arg *A = Args.getLastArg(clang::driver::options::OPT_mabi_EQ)) {
diff --git a/clang/lib/Driver/ToolChains/Arch/X86.h b/clang/lib/Driver/ToolChains/Arch/X86.h
index 762a1fa6f4d5..e07387f3ece3 100644
--- a/clang/lib/Driver/ToolChains/Arch/X86.h
+++ b/clang/lib/Driver/ToolChains/Arch/X86.h
@@ -26,7 +26,7 @@ std::string getX86TargetCPU(const Driver &D, const llvm::opt::ArgList &Args,
 
 void getX86TargetFeatures(const Driver &D, const llvm::Triple &Triple,
                           const llvm::opt::ArgList &Args,
-                          std::vector<llvm::StringRef> &Features, bool ForAS);
+                          std::vector<llvm::StringRef> &Features);
 
 } // end namespace x86
 } // end namespace target
diff --git a/clang/lib/Driver/ToolChains/CommonArgs.cpp b/clang/lib/Driver/ToolChains/CommonArgs.cpp
index 8766d34eec53..0d6907b8e5c7 100644
--- a/clang/lib/Driver/ToolChains/CommonArgs.cpp
+++ b/clang/lib/Driver/ToolChains/CommonArgs.cpp
@@ -528,7 +528,7 @@ void tools::getTargetFeatures(const Driver &D, const llvm::Triple &Triple,
     break;
   case llvm::Triple::x86:
   case llvm::Triple::x86_64:
-    x86::getX86TargetFeatures(D, Triple, Args, Features, ForAS);
+    x86::getX86TargetFeatures(D, Triple, Args, Features);
     break;
   case llvm::Triple::hexagon:
     hexagon::getHexagonTargetFeatures(D, Triple, Args, Features);
diff --git a/clang/lib/Frontend/FrontendAction.cpp b/clang/lib/Frontend/FrontendAction.cpp
index c6f958a6077b..0bd4b01ff79d 100644
--- a/clang/lib/Frontend/FrontendAction.cpp
+++ b/clang/lib/Frontend/FrontendAction.cpp
@@ -15,6 +15,7 @@
 #include "clang/Basic/FileEntry.h"
 #include "clang/Basic/LangStandard.h"
 #include "clang/Basic/Sarif.h"
+#include "clang/Basic/Stack.h"
 #include "clang/Frontend/ASTUnit.h"
 #include "clang/Frontend/CompilerInstance.h"
 #include "clang/Frontend/FrontendDiagnostic.h"
@@ -1150,6 +1151,10 @@ void ASTFrontendAction::ExecuteAction() {
   CompilerInstance &CI = getCompilerInstance();
   if (!CI.hasPreprocessor())
     return;
+  // This is a fallback: If the client forgets to invoke this, we mark the
+  // current stack as the bottom. Though not optimal, this could help prevent
+  // stack overflow during deep recursion.
+  clang::noteBottomOfStack();
 
   // FIXME: Move the truncation aspect of this into Sema, we delayed this till
   // here so the source manager would be initialized.
diff --git a/clang/lib/Sema/SemaExprCXX.cpp b/clang/lib/Sema/SemaExprCXX.cpp
index 423d5372a6f6..1cff4a75790e 100644
--- a/clang/lib/Sema/SemaExprCXX.cpp
+++ b/clang/lib/Sema/SemaExprCXX.cpp
@@ -19,6 +19,7 @@
 #include "clang/AST/CharUnits.h"
 #include "clang/AST/DeclObjC.h"
 #include "clang/AST/ExprCXX.h"
+#include "clang/AST/ExprConcepts.h"
 #include "clang/AST/ExprObjC.h"
 #include "clang/AST/RecursiveASTVisitor.h"
 #include "clang/AST/Type.h"
@@ -9072,16 +9073,24 @@ Sema::BuildExprRequirement(
     MultiLevelTemplateArgumentList MLTAL(Param, TAL.asArray(),
                                          /*Final=*/false);
     MLTAL.addOuterRetainedLevels(TPL->getDepth());
-    Expr *IDC = Param->getTypeConstraint()->getImmediatelyDeclaredConstraint();
+    const TypeConstraint *TC = Param->getTypeConstraint();
+    assert(TC && "Type Constraint cannot be null here");
+    auto *IDC = TC->getImmediatelyDeclaredConstraint();
+    assert(IDC && "ImmediatelyDeclaredConstraint can't be null here.");
     ExprResult Constraint = SubstExpr(IDC, MLTAL);
     if (Constraint.isInvalid()) {
-      Status = concepts::ExprRequirement::SS_ExprSubstitutionFailure;
-    } else {
-      SubstitutedConstraintExpr =
-          cast<ConceptSpecializationExpr>(Constraint.get());
-      if (!SubstitutedConstraintExpr->isSatisfied())
-        Status = concepts::ExprRequirement::SS_ConstraintsNotSatisfied;
-    }
+      return new (Context) concepts::ExprRequirement(
+          concepts::createSubstDiagAt(*this, IDC->getExprLoc(),
+                                      [&](llvm::raw_ostream &OS) {
+                                        IDC->printPretty(OS, /*Helper=*/nullptr,
+                                                         getPrintingPolicy());
+                                      }),
+          IsSimple, NoexceptLoc, ReturnTypeRequirement);
+    }
+    SubstitutedConstraintExpr =
+        cast<ConceptSpecializationExpr>(Constraint.get());
+    if (!SubstitutedConstraintExpr->isSatisfied())
+      Status = concepts::ExprRequirement::SS_ConstraintsNotSatisfied;
   }
   return new (Context) concepts::ExprRequirement(E, IsSimple, NoexceptLoc,
                                                  ReturnTypeRequirement, Status,
diff --git a/clang/lib/Sema/SemaTemplateInstantiate.cpp b/clang/lib/Sema/SemaTemplateInstantiate.cpp
index 8702e2ca3a1b..394006a57747 100644
--- a/clang/lib/Sema/SemaTemplateInstantiate.cpp
+++ b/clang/lib/Sema/SemaTemplateInstantiate.cpp
@@ -2276,9 +2276,9 @@ QualType TemplateInstantiator::TransformSubstTemplateTypeParmPackType(
       getPackIndex(Pack), Arg, TL.getNameLoc());
 }
 
-template<typename EntityPrinter>
 static concepts::Requirement::SubstitutionDiagnostic *
-createSubstDiag(Sema &S, TemplateDeductionInfo &Info, EntityPrinter Printer) {
+createSubstDiag(Sema &S, TemplateDeductionInfo &Info,
+                concepts::EntityPrinter Printer) {
   SmallString<128> Message;
   SourceLocation ErrorLoc;
   if (Info.hasSFINAEDiagnostic()) {
@@ -2302,6 +2302,19 @@ createSubstDiag(Sema &S, TemplateDeductionInfo &Info, EntityPrinter Printer) {
       StringRef(MessageBuf, Message.size())};
 }
 
+concepts::Requirement::SubstitutionDiagnostic *
+concepts::createSubstDiagAt(Sema &S, SourceLocation Location,
+                            EntityPrinter Printer) {
+  SmallString<128> Entity;
+  llvm::raw_svector_ostream OS(Entity);
+  Printer(OS);
+  char *EntityBuf = new (S.Context) char[Entity.size()];
+  llvm::copy(Entity, EntityBuf);
+  return new (S.Context) concepts::Requirement::SubstitutionDiagnostic{
+      /*SubstitutedEntity=*/StringRef(EntityBuf, Entity.size()),
+      /*DiagLoc=*/Location, /*DiagMessage=*/StringRef()};
+}
+
 ExprResult TemplateInstantiator::TransformRequiresTypeParams(
     SourceLocation KWLoc, SourceLocation RBraceLoc, const RequiresExpr *RE,
     RequiresExprBodyDecl *Body, ArrayRef<ParmVarDecl *> Params,
diff --git a/clang/lib/Tooling/Inclusions/Stdlib/StdSymbolMap.inc b/clang/lib/Tooling/Inclusions/Stdlib/StdSymbolMap.inc
index a08ec11e77a4..b46bd2e4d7a4 100644
--- a/clang/lib/Tooling/Inclusions/Stdlib/StdSymbolMap.inc
+++ b/clang/lib/Tooling/Inclusions/Stdlib/StdSymbolMap.inc
@@ -3773,6 +3773,33 @@ SYMBOL(viewable_range, std::ranges::, <ranges>)
 SYMBOL(wistream_view, std::ranges::, <ranges>)
 SYMBOL(zip_transform_view, std::ranges::, <ranges>)
 SYMBOL(zip_view, std::ranges::, <ranges>)
+SYMBOL(all, std::ranges::views::, <ranges>)
+SYMBOL(all_t, std::ranges::views::, <ranges>)
+SYMBOL(as_const, std::ranges::views::, <ranges>)
+SYMBOL(as_rvalue, std::ranges::views::, <ranges>)
+SYMBOL(common, std::ranges::views::, <ranges>)
+SYMBOL(counted, std::ranges::views::, <ranges>)
+SYMBOL(drop, std::ranges::views::, <ranges>)
+SYMBOL(drop_while, std::ranges::views::, <ranges>)
+SYMBOL(elements, std::ranges::views::, <ranges>)
+SYMBOL(empty, std::ranges::views::, <ranges>)
+SYMBOL(filter, std::ranges::views::, <ranges>)
+SYMBOL(iota, std::ranges::views::, <ranges>)
+SYMBOL(istream, std::ranges::views::, <ranges>)
+SYMBOL(istream, std::ranges::views::, <iosfwd>)
+SYMBOL(join, std::ranges::views::, <ranges>)
+SYMBOL(join_with, std::ranges::views::, <ranges>)
+SYMBOL(keys, std::ranges::views::, <ranges>)
+SYMBOL(lazy_split, std::ranges::views::, <ranges>)
+SYMBOL(reverse, std::ranges::views::, <ranges>)
+SYMBOL(single, std::ranges::views::, <ranges>)
+SYMBOL(split, std::ranges::views::, <ranges>)
+SYMBOL(take, std::ranges::views::, <ranges>)
+SYMBOL(take_while, std::ranges::views::, <ranges>)
+SYMBOL(transform, std::ranges::views::, <ranges>)
+SYMBOL(values, std::ranges::views::, <ranges>)
+SYMBOL(zip, std::ranges::views::, <ranges>)
+SYMBOL(zip_transform, std::ranges::views::, <ranges>)
 SYMBOL(ECMAScript, std::regex_constants::, <regex>)
 SYMBOL(awk, std::regex_constants::, <regex>)
 SYMBOL(basic, std::regex_constants::, <regex>)
@@ -3817,3 +3844,30 @@ SYMBOL(get_id, std::this_thread::, <thread>)
 SYMBOL(sleep_for, std::this_thread::, <thread>)
 SYMBOL(sleep_until, std::this_thread::, <thread>)
 SYMBOL(yield, std::this_thread::, <thread>)
+SYMBOL(all, std::views::, <ranges>)
+SYMBOL(all_t, std::views::, <ranges>)
+SYMBOL(as_const, std::views::, <ranges>)
+SYMBOL(as_rvalue, std::views::, <ranges>)
+SYMBOL(common, std::views::, <ranges>)
+SYMBOL(counted, std::views::, <ranges>)
+SYMBOL(drop, std::views::, <ranges>)
+SYMBOL(drop_while, std::views::, <ranges>)
+SYMBOL(elements, std::views::, <ranges>)
+SYMBOL(empty, std::views::, <ranges>)
+SYMBOL(filter, std::views::, <ranges>)
+SYMBOL(iota, std::views::, <ranges>)
+SYMBOL(istream, std::views::, <ranges>)
+SYMBOL(istream, std::views::, <iosfwd>)
+SYMBOL(join, std::views::, <ranges>)
+SYMBOL(join_with, std::views::, <ranges>)
+SYMBOL(keys, std::views::, <ranges>)
+SYMBOL(lazy_split, std::views::, <ranges>)
+SYMBOL(reverse, std::views::, <ranges>)
+SYMBOL(single, std::views::, <ranges>)
+SYMBOL(split, std::views::, <ranges>)
+SYMBOL(take, std::views::, <ranges>)
+SYMBOL(take_while, std::views::, <ranges>)
+SYMBOL(transform, std::views::, <ranges>)
+SYMBOL(values, std::views::, <ranges>)
+SYMBOL(zip, std::views::, <ranges>)
+SYMBOL(zip_transform, std::views::, <ranges>)
diff --git a/compiler-rt/lib/builtins/aarch64/lse.S b/compiler-rt/lib/builtins/aarch64/lse.S
index 5dc0d5320b5a..1fe18f4a4681 100644
--- a/compiler-rt/lib/builtins/aarch64/lse.S
+++ b/compiler-rt/lib/builtins/aarch64/lse.S
@@ -7,7 +7,7 @@
 // Out-of-line LSE atomics helpers. Ported from libgcc library.
 // N = {1, 2, 4, 8}
 // M = {1, 2, 4, 8, 16}
-// ORDER = {'relax', 'acq', 'rel', 'acq_rel'}
+// ORDER = {'relax', 'acq', 'rel', 'acq_rel', 'sync'}
 // Routines implemented:
 //
 //  iM __aarch64_casM_ORDER(iM expected, iM desired, iM *ptr)
@@ -35,8 +35,8 @@ HIDDEN(___aarch64_have_lse_atomics)
 #endif
 
 // Generate mnemonics for
-// L_cas:                                 SIZE: 1,2,4,8,16 MODEL: 1,2,3,4
-// L_swp L_ldadd L_ldclr L_ldeor L_ldset: SIZE: 1,2,4,8    MODEL: 1,2,3,4
+// L_cas:                                 SIZE: 1,2,4,8,16 MODEL: 1,2,3,4,5
+// L_swp L_ldadd L_ldclr L_ldeor L_ldset: SIZE: 1,2,4,8    MODEL: 1,2,3,4,5
 
 #if SIZE == 1
 #define S b
@@ -64,24 +64,44 @@ HIDDEN(___aarch64_have_lse_atomics)
 #define L
 #define M 0x000000
 #define N 0x000000
+#define BARRIER
 #elif MODEL == 2
 #define SUFF _acq
 #define A a
 #define L
 #define M 0x400000
 #define N 0x800000
+#define BARRIER
 #elif MODEL == 3
 #define SUFF _rel
 #define A
 #define L l
 #define M 0x008000
 #define N 0x400000
+#define BARRIER
 #elif MODEL == 4
 #define SUFF _acq_rel
 #define A a
 #define L l
 #define M 0x408000
 #define N 0xc00000
+#define BARRIER
+#elif MODEL == 5
+#define SUFF _sync
+#ifdef L_swp
+// swp has _acq semantics.
+#define A a
+#define L
+#define M 0x400000
+#define N 0x800000
+#else
+// All other _sync functions have _seq semantics.
+#define A a
+#define L l
+#define M 0x408000
+#define N 0xc00000
+#endif
+#define BARRIER dmb ish
 #else
 #error
 #endif // MODEL
@@ -96,7 +116,12 @@ HIDDEN(___aarch64_have_lse_atomics)
 #endif
 
 #define NAME(BASE) GLUE4(__aarch64_, BASE, SIZE, SUFF)
+#if MODEL == 5
+// Drop A for _sync functions.
+#define LDXR GLUE3(ld, xr, S)
+#else
 #define LDXR GLUE4(ld, A, xr, S)
+#endif
 #define STXR GLUE4(st, L, xr, S)
 
 // Define temporary registers.
@@ -136,9 +161,15 @@ DEFINE_COMPILERRT_OUTLINE_FUNCTION_UNMANGLED(NAME(cas))
         STXR   w(tmp1), s(1), [x2]
         cbnz   w(tmp1), 0b
 1:
+        BARRIER
         ret
 #else
+#if MODEL == 5
+// Drop A for _sync functions.
+#define LDXP GLUE2(ld, xp)
+#else
 #define LDXP GLUE3(ld, A, xp)
+#endif
 #define STXP GLUE3(st, L, xp)
 #ifdef HAS_ASM_LSE
 #define CASP GLUE3(casp, A, L)  x0, x1, x2, x3, [x4]
@@ -159,6 +190,7 @@ DEFINE_COMPILERRT_OUTLINE_FUNCTION_UNMANGLED(NAME(cas))
         STXP   w(tmp2), x2, x3, [x4]
         cbnz   w(tmp2), 0b
 1:
+        BARRIER
         ret
 #endif
 END_COMPILERRT_OUTLINE_FUNCTION(NAME(cas))
@@ -180,6 +212,7 @@ DEFINE_COMPILERRT_OUTLINE_FUNCTION_UNMANGLED(NAME(swp))
         LDXR   s(0), [x1]
         STXR   w(tmp1), s(tmp0), [x1]
         cbnz   w(tmp1), 0b
+        BARRIER
         ret
 END_COMPILERRT_OUTLINE_FUNCTION(NAME(swp))
 #endif // L_swp
@@ -224,6 +257,7 @@ DEFINE_COMPILERRT_OUTLINE_FUNCTION_UNMANGLED(NAME(LDNM))
         OP     s(tmp1), s(0), s(tmp0)
         STXR   w(tmp2), s(tmp1), [x1]
         cbnz   w(tmp2), 0b
+        BARRIER
         ret
 END_COMPILERRT_OUTLINE_FUNCTION(NAME(LDNM))
 #endif // L_ldadd L_ldclr L_ldeor L_ldset
diff --git a/compiler-rt/lib/sanitizer_common/sanitizer_common_interceptors_format.inc b/compiler-rt/lib/sanitizer_common/sanitizer_common_interceptors_format.inc
index 220abb89c3be..24485900644b 100644
--- a/compiler-rt/lib/sanitizer_common/sanitizer_common_interceptors_format.inc
+++ b/compiler-rt/lib/sanitizer_common/sanitizer_common_interceptors_format.inc
@@ -340,11 +340,19 @@ static void scanf_common(void *ctx, int n_inputs, bool allowGnuMalloc,
       size = 0;
     }
     COMMON_INTERCEPTOR_WRITE_RANGE(ctx, argp, size);
-    // For %ms/%mc, write the allocated output buffer as well.
+    // For %mc/%mC/%ms/%m[/%mS, write the allocated output buffer as well.
     if (dir.allocate) {
-      char *buf = *(char **)argp;
-      if (buf)
-        COMMON_INTERCEPTOR_WRITE_RANGE(ctx, buf, internal_strlen(buf) + 1);
+      if (char *buf = *(char **)argp) {
+        if (dir.convSpecifier == 'c')
+          size = 1;
+        else if (dir.convSpecifier == 'C')
+          size = sizeof(wchar_t);
+        else if (dir.convSpecifier == 'S')
+          size = (internal_wcslen((wchar_t *)buf) + 1) * sizeof(wchar_t);
+        else  // 's' or '['
+          size = internal_strlen(buf) + 1;
+        COMMON_INTERCEPTOR_WRITE_RANGE(ctx, buf, size);
+      }
     }
   }
 }
diff --git a/libcxx/include/__config b/libcxx/include/__config
index 9759d3b9e8e0..43f8a20031ff 100644
--- a/libcxx/include/__config
+++ b/libcxx/include/__config
@@ -208,19 +208,16 @@
 
 // HARDENING {
 
-// TODO(hardening): remove this in LLVM 18.
-// This is for backward compatibility -- make enabling `_LIBCPP_ENABLE_ASSERTIONS` (which predates hardening modes)
-// equivalent to setting the hardened mode.
-#  ifdef _LIBCPP_ENABLE_ASSERTIONS
-#    warning "_LIBCPP_ENABLE_ASSERTIONS is deprecated, please use _LIBCPP_ENABLE_HARDENED_MODE instead."
-#    if _LIBCPP_ENABLE_ASSERTIONS != 0 && _LIBCPP_ENABLE_ASSERTIONS != 1
-#      error "_LIBCPP_ENABLE_ASSERTIONS must be set to 0 or 1"
-#    endif
-#    if _LIBCPP_ENABLE_ASSERTIONS
-#      define _LIBCPP_ENABLE_HARDENED_MODE 1
-#    endif
+#  ifndef _LIBCPP_ENABLE_ASSERTIONS
+#    define _LIBCPP_ENABLE_ASSERTIONS _LIBCPP_ENABLE_ASSERTIONS_DEFAULT
+#  endif
+#  if _LIBCPP_ENABLE_ASSERTIONS != 0 && _LIBCPP_ENABLE_ASSERTIONS != 1
+#    error "_LIBCPP_ENABLE_ASSERTIONS must be set to 0 or 1"
 #  endif
 
+// NOTE: These modes are experimental and are not stable yet in LLVM 17. Please refrain from using them and use the
+// documented libc++ "safe" mode instead.
+//
 // Enables the hardened mode which consists of all checks intended to be used in production. Hardened mode prioritizes
 // security-critical checks that can be done with relatively little overhead in constant time. Mutually exclusive with
 // `_LIBCPP_ENABLE_DEBUG_MODE`.
@@ -275,6 +272,11 @@
 #    error "Only one of _LIBCPP_ENABLE_HARDENED_MODE and _LIBCPP_ENABLE_DEBUG_MODE can be enabled."
 #  endif
 
+#  if _LIBCPP_ENABLE_ASSERTIONS && (_LIBCPP_ENABLE_HARDENED_MODE || _LIBCPP_ENABLE_DEBUG_MODE)
+#    error                                                                                                             \
+        "_LIBCPP_ENABLE_ASSERTIONS is mutually exclusive with _LIBCPP_ENABLE_HARDENED_MODE and _LIBCPP_ENABLE_DEBUG_MODE."
+#  endif
+
 // Hardened mode checks.
 
 // clang-format off
@@ -303,6 +305,18 @@
 #    define _LIBCPP_ASSERT_INTERNAL(expression, message)                  _LIBCPP_ASSERT(expression, message)
 #    define _LIBCPP_ASSERT_UNCATEGORIZED(expression, message)             _LIBCPP_ASSERT(expression, message)
 
+// Safe mode checks.
+
+#  elif _LIBCPP_ENABLE_ASSERTIONS
+
+// All checks enabled.
+#    define _LIBCPP_ASSERT_VALID_INPUT_RANGE(expression, message)         _LIBCPP_ASSERT(expression, message)
+#    define _LIBCPP_ASSERT_VALID_ELEMENT_ACCESS(expression, message)      _LIBCPP_ASSERT(expression, message)
+#    define _LIBCPP_ASSERT_NON_OVERLAPPING_RANGES(expression, message)    _LIBCPP_ASSERT(expression, message)
+#    define _LIBCPP_ASSERT_COMPATIBLE_ALLOCATOR(expression, message)      _LIBCPP_ASSERT(expression, message)
+#    define _LIBCPP_ASSERT_INTERNAL(expression, message)                  _LIBCPP_ASSERT(expression, message)
+#    define _LIBCPP_ASSERT_UNCATEGORIZED(expression, message)             _LIBCPP_ASSERT(expression, message)
+
 // Disable all checks if hardening is not enabled.
 
 #  else
diff --git a/llvm/include/llvm/Analysis/LazyValueInfo.h b/llvm/include/llvm/Analysis/LazyValueInfo.h
index b109b7f7e65a..7b2bfdac75a8 100644
--- a/llvm/include/llvm/Analysis/LazyValueInfo.h
+++ b/llvm/include/llvm/Analysis/LazyValueInfo.h
@@ -115,6 +115,9 @@ public:
   /// PredBB to OldSucc to be from PredBB to NewSucc instead.
   void threadEdge(BasicBlock *PredBB, BasicBlock *OldSucc, BasicBlock *NewSucc);
 
+  /// Remove information related to this value from the cache.
+  void forgetValue(Value *V);
+
   /// Inform the analysis cache that we have erased a block.
   void eraseBlock(BasicBlock *BB);
 
diff --git a/llvm/lib/Analysis/LazyValueInfo.cpp b/llvm/lib/Analysis/LazyValueInfo.cpp
index 33651783cb17..2ba6036056d9 100644
--- a/llvm/lib/Analysis/LazyValueInfo.cpp
+++ b/llvm/lib/Analysis/LazyValueInfo.cpp
@@ -465,6 +465,10 @@ public:
     F.print(OS, &Writer);
   }
 
+  /// This is part of the update interface to remove information related to this
+  /// value from the cache.
+  void forgetValue(Value *V) { TheCache.eraseValue(V); }
+
   /// This is part of the update interface to inform the cache
   /// that a block has been deleted.
   void eraseBlock(BasicBlock *BB) {
@@ -1969,6 +1973,11 @@ void LazyValueInfo::threadEdge(BasicBlock *PredBB, BasicBlock *OldSucc,
   }
 }
 
+void LazyValueInfo::forgetValue(Value *V) {
+  if (PImpl)
+    getImpl(PImpl, AC, nullptr).forgetValue(V);
+}
+
 void LazyValueInfo::eraseBlock(BasicBlock *BB) {
   if (PImpl) {
     getImpl(PImpl, AC, BB->getModule()).eraseBlock(BB);
diff --git a/llvm/lib/Analysis/ScalarEvolution.cpp b/llvm/lib/Analysis/ScalarEvolution.cpp
index 111d4d30aab9..39ab48b4a48e 100644
--- a/llvm/lib/Analysis/ScalarEvolution.cpp
+++ b/llvm/lib/Analysis/ScalarEvolution.cpp
@@ -6833,7 +6833,7 @@ const ConstantRange &ScalarEvolution::getRangeRef(
         if (llvm::isKnownNonZero(V, DL))
           MinVal = Align;
         ConservativeResult = ConservativeResult.intersectWith(
-            {MinVal, MaxVal + 1}, RangeType);
+            ConstantRange::getNonEmpty(MinVal, MaxVal + 1), RangeType);
       }
     }
 
diff --git a/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp b/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp
index 952f454f8f6a..7979ac9a5fb7 100644
--- a/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp
+++ b/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp
@@ -1424,7 +1424,17 @@ bool ComplexDeinterleavingGraph::identifyNodes(Instruction *RootI) {
   // CompositeNode we should choose only one either Real or Imag instruction to
   // use as an anchor for generating complex instruction.
   auto It = RootToNode.find(RootI);
-  if (It != RootToNode.end() && It->second->Real == RootI) {
+  if (It != RootToNode.end()) {
+    auto RootNode = It->second;
+    assert(RootNode->Operation ==
+           ComplexDeinterleavingOperation::ReductionOperation);
+    // Find out which part, Real or Imag, comes later, and only if we come to
+    // the latest part, add it to OrderedRoots.
+    auto *R = cast<Instruction>(RootNode->Real);
+    auto *I = cast<Instruction>(RootNode->Imag);
+    auto *ReplacementAnchor = R->comesBefore(I) ? I : R;
+    if (ReplacementAnchor != RootI)
+      return false;
     OrderedRoots.push_back(RootI);
     return true;
   }
diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
index 0605dfa63793..c7a6dd7deb45 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -13840,7 +13840,17 @@ bool AArch64TargetLowering::getTgtMemIntrinsic(IntrinsicInfo &Info,
   case Intrinsic::aarch64_neon_ld4:
   case Intrinsic::aarch64_neon_ld1x2:
   case Intrinsic::aarch64_neon_ld1x3:
-  case Intrinsic::aarch64_neon_ld1x4:
+  case Intrinsic::aarch64_neon_ld1x4: {
+    Info.opc = ISD::INTRINSIC_W_CHAIN;
+    uint64_t NumElts = DL.getTypeSizeInBits(I.getType()) / 64;
+    Info.memVT = EVT::getVectorVT(I.getType()->getContext(), MVT::i64, NumElts);
+    Info.ptrVal = I.getArgOperand(I.arg_size() - 1);
+    Info.offset = 0;
+    Info.align.reset();
+    // volatile loads with NEON intrinsics not supported
+    Info.flags = MachineMemOperand::MOLoad;
+    return true;
+  }
   case Intrinsic::aarch64_neon_ld2lane:
   case Intrinsic::aarch64_neon_ld3lane:
   case Intrinsic::aarch64_neon_ld4lane:
@@ -13848,9 +13858,13 @@ bool AArch64TargetLowering::getTgtMemIntrinsic(IntrinsicInfo &Info,
   case Intrinsic::aarch64_neon_ld3r:
   case Intrinsic::aarch64_neon_ld4r: {
     Info.opc = ISD::INTRINSIC_W_CHAIN;
-    // Conservatively set memVT to the entire set of vectors loaded.
-    uint64_t NumElts = DL.getTypeSizeInBits(I.getType()) / 64;
-    Info.memVT = EVT::getVectorVT(I.getType()->getContext(), MVT::i64, NumElts);
+    // ldx return struct with the same vec type
+    Type *RetTy = I.getType();
+    auto *StructTy = cast<StructType>(RetTy);
+    unsigned NumElts = StructTy->getNumElements();
+    Type *VecTy = StructTy->getElementType(0);
+    MVT EleVT = MVT::getVT(VecTy).getVectorElementType();
+    Info.memVT = EVT::getVectorVT(I.getType()->getContext(), EleVT, NumElts);
     Info.ptrVal = I.getArgOperand(I.arg_size() - 1);
     Info.offset = 0;
     Info.align.reset();
@@ -13863,20 +13877,40 @@ bool AArch64TargetLowering::getTgtMemIntrinsic(IntrinsicInfo &Info,
   case Intrinsic::aarch64_neon_st4:
   case Intrinsic::aarch64_neon_st1x2:
   case Intrinsic::aarch64_neon_st1x3:
-  case Intrinsic::aarch64_neon_st1x4:
+  case Intrinsic::aarch64_neon_st1x4: {
+    Info.opc = ISD::INTRINSIC_VOID;
+    unsigned NumElts = 0;
+    for (const Value *Arg : I.args()) {
+      Type *ArgTy = Arg->getType();
+      if (!ArgTy->isVectorTy())
+        break;
+      NumElts += DL.getTypeSizeInBits(ArgTy) / 64;
+    }
+    Info.memVT = EVT::getVectorVT(I.getType()->getContext(), MVT::i64, NumElts);
+    Info.ptrVal = I.getArgOperand(I.arg_size() - 1);
+    Info.offset = 0;
+    Info.align.reset();
+    // volatile stores with NEON intrinsics not supported
+    Info.flags = MachineMemOperand::MOStore;
+    return true;
+  }
   case Intrinsic::aarch64_neon_st2lane:
   case Intrinsic::aarch64_neon_st3lane:
   case Intrinsic::aarch64_neon_st4lane: {
     Info.opc = ISD::INTRINSIC_VOID;
-    // Conservatively set memVT to the entire set of vectors stored.
     unsigned NumElts = 0;
+    // all the vector type is same
+    Type *VecTy = I.getArgOperand(0)->getType();
+    MVT EleVT = MVT::getVT(VecTy).getVectorElementType();
+
     for (const Value *Arg : I.args()) {
       Type *ArgTy = Arg->getType();
       if (!ArgTy->isVectorTy())
         break;
-      NumElts += DL.getTypeSizeInBits(ArgTy) / 64;
+      NumElts += 1;
     }
-    Info.memVT = EVT::getVectorVT(I.getType()->getContext(), MVT::i64, NumElts);
+
+    Info.memVT = EVT::getVectorVT(I.getType()->getContext(), EleVT, NumElts);
     Info.ptrVal = I.getArgOperand(I.arg_size() - 1);
     Info.offset = 0;
     Info.align.reset();
diff --git a/llvm/lib/Transforms/Scalar/JumpThreading.cpp b/llvm/lib/Transforms/Scalar/JumpThreading.cpp
index 24390f1b54f6..5b8f1b00dc03 100644
--- a/llvm/lib/Transforms/Scalar/JumpThreading.cpp
+++ b/llvm/lib/Transforms/Scalar/JumpThreading.cpp
@@ -1269,6 +1269,7 @@ bool JumpThreadingPass::simplifyPartiallyRedundantLoad(LoadInst *LoadI) {
     if (IsLoadCSE) {
       LoadInst *NLoadI = cast<LoadInst>(AvailableVal);
       combineMetadataForCSE(NLoadI, LoadI, false);
+      LVI->forgetValue(NLoadI);
     };
 
     // If the returned value is the load itself, replace with poison. This can
@@ -1461,6 +1462,7 @@ bool JumpThreadingPass::simplifyPartiallyRedundantLoad(LoadInst *LoadI) {
 
   for (LoadInst *PredLoadI : CSELoads) {
     combineMetadataForCSE(PredLoadI, LoadI, true);
+    LVI->forgetValue(PredLoadI);
   }
 
   LoadI->replaceAllUsesWith(PN);
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index d7e40e8ef978..b603bbe55dc9 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -3781,10 +3781,44 @@ void InnerLoopVectorizer::fixCrossIterationPHIs(VPTransformState &State) {
   // the incoming edges.
   VPBasicBlock *Header =
       State.Plan->getVectorLoopRegion()->getEntryBasicBlock();
+
+  // Gather all VPReductionPHIRecipe and sort them so that Intermediate stores
+  // sank outside of the loop would keep the same order as they had in the
+  // original loop.
+  SmallVector<VPReductionPHIRecipe *> ReductionPHIList;
   for (VPRecipeBase &R : Header->phis()) {
     if (auto *ReductionPhi = dyn_cast<VPReductionPHIRecipe>(&R))
-      fixReduction(ReductionPhi, State);
-    else if (auto *FOR = dyn_cast<VPFirstOrderRecurrencePHIRecipe>(&R))
+      ReductionPHIList.emplace_back(ReductionPhi);
+  }
+  stable_sort(ReductionPHIList, [this](const VPReductionPHIRecipe *R1,
+                                       const VPReductionPHIRecipe *R2) {
+    auto *IS1 = R1->getRecurrenceDescriptor().IntermediateStore;
+    auto *IS2 = R2->getRecurrenceDescriptor().IntermediateStore;
+
+    // If neither of the recipes has an intermediate store, keep the order the
+    // same.
+    if (!IS1 && !IS2)
+      return false;
+
+    // If only one of the recipes has an intermediate store, then move it
+    // towards the beginning of the list.
+    if (IS1 && !IS2)
+      return true;
+
+    if (!IS1 && IS2)
+      return false;
+
+    // If both recipes have an intermediate store, then the recipe with the
+    // later store should be processed earlier. So it should go to the beginning
+    // of the list.
+    return DT->dominates(IS2, IS1);
+  });
+
+  for (VPReductionPHIRecipe *ReductionPhi : ReductionPHIList)
+    fixReduction(ReductionPhi, State);
+
+  for (VPRecipeBase &R : Header->phis()) {
+    if (auto *FOR = dyn_cast<VPFirstOrderRecurrencePHIRecipe>(&R))
       fixFixedOrderRecurrence(FOR, State);
   }
 }