diff --git a/clang/lib/APINotes/APINotesReader.cpp b/clang/lib/APINotes/APINotesReader.cpp
diff --git a/clang/lib/AST/Decl.cpp b/clang/lib/AST/Decl.cpp
@@ -4534,7 +4534,7 @@ unsigned FunctionDecl::getODRHash() {
   }
 
   class ODRHash Hash;
-  Hash.AddFunctionDecl(this, /*SkipBody=*/shouldSkipCheckingODR());
+  Hash.AddFunctionDecl(this);
   setHasODRHash(true);
   ODRHash = Hash.CalculateHash();
   return ODRHash;

diff --git a/clang/lib/AST/DeclBase.cpp b/clang/lib/AST/DeclBase.cpp
@@ -1106,11 +1106,6 @@ bool Decl::isFromExplicitGlobalModule() const {
   return getOwningModule() && getOwningModule()->isExplicitGlobalModule();
 }
 
-bool Decl::shouldSkipCheckingODR() const {
-  return getASTContext().getLangOpts().SkipODRCheckInGMF &&
-         isFromExplicitGlobalModule();
-}
-
 static Decl::Kind getKind(const Decl *D) { return D->getKind(); }
 static Decl::Kind getKind(const DeclContext *DC) { return DC->getDeclKind(); }
 

diff --git a/clang/lib/AST/Interp/ByteCodeExprGen.cpp b/clang/lib/AST/Interp/ByteCodeExprGen.cpp
@@ -262,7 +262,7 @@ bool ByteCodeExprGen<Emitter>::VisitCastExpr(const CastExpr *CE) {
       return this->discard(SubExpr);
 
     std::optional<PrimType> FromT = classify(SubExpr->getType());
-    std::optional<PrimType> ToT = classifyPrim(CE->getType());
+    std::optional<PrimType> ToT = classify(CE->getType());
     if (!FromT || !ToT)
       return false;
 
@@ -1251,17 +1251,46 @@ bool ByteCodeExprGen<Emitter>::VisitUnaryExprOrTypeTraitExpr(
     return this->emitConst(Size.getQuantity(), E);
   }
 
+  if (Kind == UETT_VectorElements) {
+    if (const auto *VT = E->getTypeOfArgument()->getAs<VectorType>())
+      return this->emitConst(VT->getNumElements(), E);
+
+    // FIXME: Apparently we need to catch the fact that a sizeless vector type
+    // has been passed and diagnose that (at run time).
+    assert(E->getTypeOfArgument()->isSizelessVectorType());
+  }
+
   return false;
 }
 
 template <class Emitter>
 bool ByteCodeExprGen<Emitter>::VisitMemberExpr(const MemberExpr *E) {
   // 'Base.Member'
   const Expr *Base = E->getBase();
+  const ValueDecl *Member = E->getMemberDecl();
 
   if (DiscardResult)
     return this->discard(Base);
 
+  // MemberExprs are almost always lvalues, in which case we don't need to
+  // do the load. But sometimes they aren't.
+  const auto maybeLoadValue = [&]() -> bool {
+    if (E->isGLValue())
+      return true;
+    if (std::optional<PrimType> T = classify(E))
+      return this->emitLoadPop(*T, E);
+    return false;
+  };
+
+  if (const auto *VD = dyn_cast<VarDecl>(Member)) {
+    // I am almost confident in saying that a var decl must be static
+    // and therefore registered as a global variable. But this will probably
+    // turn out to be wrong some time in the future, as always.
+    if (auto GlobalIndex = P.getGlobal(VD))
+      return this->emitGetPtrGlobal(*GlobalIndex, E) && maybeLoadValue();
+    return false;
+  }
+
   if (Initializing) {
     if (!this->delegate(Base))
       return false;
@@ -1271,16 +1300,14 @@ bool ByteCodeExprGen<Emitter>::VisitMemberExpr(const MemberExpr *E) {
   }
 
   // Base above gives us a pointer on the stack.
-  // TODO: Implement non-FieldDecl members.
-  const ValueDecl *Member = E->getMemberDecl();
   if (const auto *FD = dyn_cast<FieldDecl>(Member)) {
     const RecordDecl *RD = FD->getParent();
     const Record *R = getRecord(RD);
     const Record::Field *F = R->getField(FD);
     // Leave a pointer to the field on the stack.
     if (F->Decl->getType()->isReferenceType())
-      return this->emitGetFieldPop(PT_Ptr, F->Offset, E);
-    return this->emitGetPtrField(F->Offset, E);
+      return this->emitGetFieldPop(PT_Ptr, F->Offset, E) && maybeLoadValue();
+    return this->emitGetPtrField(F->Offset, E) && maybeLoadValue();
   }
 
   return false;
@@ -1615,7 +1642,7 @@ bool ByteCodeExprGen<Emitter>::VisitCompoundAssignOperator(
     return false;
   if (!this->emitLoad(*LT, E))
     return false;
-  if (*LT != *LHSComputationT) {
+  if (LT != LHSComputationT) {
     if (!this->emitCast(*LT, *LHSComputationT, E))
       return false;
   }
@@ -1671,7 +1698,7 @@ bool ByteCodeExprGen<Emitter>::VisitCompoundAssignOperator(
   }
 
   // And now cast from LHSComputationT to ResultT.
-  if (*ResultT != *LHSComputationT) {
+  if (ResultT != LHSComputationT) {
     if (!this->emitCast(*LHSComputationT, *ResultT, E))
       return false;
   }

diff --git a/clang/lib/AST/Interp/Disasm.cpp b/clang/lib/AST/Interp/Disasm.cpp
@@ -140,7 +140,7 @@ LLVM_DUMP_METHOD void Program::dump(llvm::raw_ostream &OS) const {
     const Descriptor *Desc = G->block()->getDescriptor();
     Pointer GP = getPtrGlobal(GI);
 
-    OS << GI << ": " << (void *)G->block() << " ";
+    OS << GI << ": " << (const void *)G->block() << " ";
     {
       ColorScope SC(OS, true,
                     GP.isInitialized()
@@ -264,3 +264,19 @@ LLVM_DUMP_METHOD void Record::dump(llvm::raw_ostream &OS, unsigned Indentation,
     ++I;
   }
 }
+
+LLVM_DUMP_METHOD void Block::dump(llvm::raw_ostream &OS) const {
+  {
+    ColorScope SC(OS, true, {llvm::raw_ostream::BRIGHT_BLUE, true});
+    OS << "Block " << (const void *)this << "\n";
+  }
+  unsigned NPointers = 0;
+  for (const Pointer *P = Pointers; P; P = P->Next) {
+    ++NPointers;
+  }
+  OS << "  Pointers: " << NPointers << "\n";
+  OS << "  Dead: " << IsDead << "\n";
+  OS << "  Static: " << IsStatic << "\n";
+  OS << "  Extern: " << IsExtern << "\n";
+  OS << "  Initialized: " << IsInitialized << "\n";
+}
diff --git a/clang/lib/AST/Interp/FunctionPointer.h b/clang/lib/AST/Interp/FunctionPointer.h
@@ -32,6 +32,7 @@ class FunctionPointer final {
 
   const Function *getFunction() const { return Func; }
   bool isZero() const { return !Func; }
+  bool isValid() const { return Valid; }
   bool isWeak() const {
     if (!Func || !Valid)
       return false;

diff --git a/clang/lib/AST/Interp/Interp.h b/clang/lib/AST/Interp/Interp.h
@@ -2236,6 +2236,10 @@ inline bool CallPtr(InterpState &S, CodePtr OpPC, uint32_t ArgSize,
         << const_cast<Expr *>(E) << E->getSourceRange();
     return false;
   }
+
+  if (!FuncPtr.isValid())
+    return false;
+
   assert(F);
 
   // Check argument nullability state.

diff --git a/clang/lib/AST/Interp/InterpBlock.h b/clang/lib/AST/Interp/InterpBlock.h
@@ -118,6 +118,9 @@ class Block final {
     IsInitialized = false;
   }
 
+  void dump() const { dump(llvm::errs()); }
+  void dump(llvm::raw_ostream &OS) const;
+
 protected:
   friend class Pointer;
   friend class DeadBlock;

diff --git a/clang/lib/AST/Interp/InterpBuiltin.cpp b/clang/lib/AST/Interp/InterpBuiltin.cpp
@@ -977,6 +977,117 @@ static bool interp__builtin_complex(InterpState &S, CodePtr OpPC,
   return true;
 }
 
+/// __builtin_is_aligned()
+/// __builtin_align_up()
+/// __builtin_align_down()
+/// The first parameter is either an integer or a pointer.
+/// The second parameter is the requested alignment as an integer.
+static bool interp__builtin_is_aligned_up_down(InterpState &S, CodePtr OpPC,
+                                               const InterpFrame *Frame,
+                                               const Function *Func,
+                                               const CallExpr *Call) {
+  unsigned BuiltinOp = Func->getBuiltinID();
+  unsigned CallSize = callArgSize(S, Call);
+
+  PrimType AlignmentT = *S.Ctx.classify(Call->getArg(1));
+  const APSInt &Alignment = peekToAPSInt(S.Stk, AlignmentT);
+
+  if (Alignment < 0 || !Alignment.isPowerOf2()) {
+    S.FFDiag(Call, diag::note_constexpr_invalid_alignment) << Alignment;
+    return false;
+  }
+  unsigned SrcWidth = S.getCtx().getIntWidth(Call->getArg(0)->getType());
+  APSInt MaxValue(APInt::getOneBitSet(SrcWidth, SrcWidth - 1));
+  if (APSInt::compareValues(Alignment, MaxValue) > 0) {
+    S.FFDiag(Call, diag::note_constexpr_alignment_too_big)
+        << MaxValue << Call->getArg(0)->getType() << Alignment;
+    return false;
+  }
+
+  // The first parameter is either an integer or a pointer (but not a function
+  // pointer).
+  PrimType FirstArgT = *S.Ctx.classify(Call->getArg(0));
+
+  if (isIntegralType(FirstArgT)) {
+    const APSInt &Src = peekToAPSInt(S.Stk, FirstArgT, CallSize);
+    APSInt Align = Alignment.extOrTrunc(Src.getBitWidth());
+    if (BuiltinOp == Builtin::BI__builtin_align_up) {
+      APSInt AlignedVal =
+          APSInt((Src + (Align - 1)) & ~(Align - 1), Src.isUnsigned());
+      pushInteger(S, AlignedVal, Call->getType());
+    } else if (BuiltinOp == Builtin::BI__builtin_align_down) {
+      APSInt AlignedVal = APSInt(Src & ~(Align - 1), Src.isUnsigned());
+      pushInteger(S, AlignedVal, Call->getType());
+    } else {
+      assert(*S.Ctx.classify(Call->getType()) == PT_Bool);
+      S.Stk.push<Boolean>((Src & (Align - 1)) == 0);
+    }
+    return true;
+  }
+
+  assert(FirstArgT == PT_Ptr);
+  const Pointer &Ptr = S.Stk.peek<Pointer>(CallSize);
+
+  unsigned PtrOffset = Ptr.getByteOffset();
+  PtrOffset = Ptr.getIndex();
+  CharUnits BaseAlignment =
+      S.getCtx().getDeclAlign(Ptr.getDeclDesc()->asValueDecl());
+  CharUnits PtrAlign =
+      BaseAlignment.alignmentAtOffset(CharUnits::fromQuantity(PtrOffset));
+
+  if (BuiltinOp == Builtin::BI__builtin_is_aligned) {
+    if (PtrAlign.getQuantity() >= Alignment) {
+      S.Stk.push<Boolean>(true);
+      return true;
+    }
+    // If the alignment is not known to be sufficient, some cases could still
+    // be aligned at run time. However, if the requested alignment is less or
+    // equal to the base alignment and the offset is not aligned, we know that
+    // the run-time value can never be aligned.
+    if (BaseAlignment.getQuantity() >= Alignment &&
+        PtrAlign.getQuantity() < Alignment) {
+      S.Stk.push<Boolean>(false);
+      return true;
+    }
+
+    S.FFDiag(Call->getArg(0), diag::note_constexpr_alignment_compute)
+        << Alignment;
+    return false;
+  }
+
+  assert(BuiltinOp == Builtin::BI__builtin_align_down ||
+         BuiltinOp == Builtin::BI__builtin_align_up);
+
+  // For align_up/align_down, we can return the same value if the alignment
+  // is known to be greater or equal to the requested value.
+  if (PtrAlign.getQuantity() >= Alignment) {
+    S.Stk.push<Pointer>(Ptr);
+    return true;
+  }
+
+  // The alignment could be greater than the minimum at run-time, so we cannot
+  // infer much about the resulting pointer value. One case is possible:
+  // For `_Alignas(32) char buf[N]; __builtin_align_down(&buf[idx], 32)` we
+  // can infer the correct index if the requested alignment is smaller than
+  // the base alignment so we can perform the computation on the offset.
+  if (BaseAlignment.getQuantity() >= Alignment) {
+    assert(Alignment.getBitWidth() <= 64 &&
+           "Cannot handle > 64-bit address-space");
+    uint64_t Alignment64 = Alignment.getZExtValue();
+    CharUnits NewOffset =
+        CharUnits::fromQuantity(BuiltinOp == Builtin::BI__builtin_align_down
+                                    ? llvm::alignDown(PtrOffset, Alignment64)
+                                    : llvm::alignTo(PtrOffset, Alignment64));
+
+    S.Stk.push<Pointer>(Ptr.atIndex(NewOffset.getQuantity()));
+    return true;
+  }
+
+  // Otherwise, we cannot constant-evaluate the result.
+  S.FFDiag(Call->getArg(0), diag::note_constexpr_alignment_adjust) << Alignment;
+  return false;
+}
+
 bool InterpretBuiltin(InterpState &S, CodePtr OpPC, const Function *F,
                       const CallExpr *Call) {
   const InterpFrame *Frame = S.Current;
@@ -1291,6 +1402,13 @@ bool InterpretBuiltin(InterpState &S, CodePtr OpPC, const Function *F,
       return false;
     break;
 
+  case Builtin::BI__builtin_is_aligned:
+  case Builtin::BI__builtin_align_up:
+  case Builtin::BI__builtin_align_down:
+    if (!interp__builtin_is_aligned_up_down(S, OpPC, Frame, F, Call))
+      return false;
+    break;
+
   default:
     S.FFDiag(S.Current->getLocation(OpPC),
              diag::note_invalid_subexpr_in_const_expr)

diff --git a/clang/lib/AST/Interp/InterpFrame.cpp b/clang/lib/AST/Interp/InterpFrame.cpp
@@ -152,6 +152,13 @@ void print(llvm::raw_ostream &OS, const Pointer &P, ASTContext &Ctx,
 }
 
 void InterpFrame::describe(llvm::raw_ostream &OS) const {
+  // We create frames for builtin functions as well, but we can't reliably
+  // diagnose them. The 'in call to' diagnostics for them add no value to the
+  // user _and_ it doesn't generally work since the argument types don't always
+  // match the function prototype. Just ignore them.
+  if (const auto *F = getFunction(); F && F->isBuiltin())
+    return;
+
   const FunctionDecl *F = getCallee();
   if (const auto *M = dyn_cast<CXXMethodDecl>(F);
       M && M->isInstance() && !isa<CXXConstructorDecl>(F)) {

diff --git a/clang/lib/AST/Interp/Pointer.h b/clang/lib/AST/Interp/Pointer.h
@@ -241,13 +241,10 @@ class Pointer {
 
   /// Checks if the pointer is null.
   bool isZero() const {
-    if (Offset != 0)
-      return false;
-
     if (isBlockPointer())
       return asBlockPointer().Pointee == nullptr;
     assert(isIntegralPointer());
-    return asIntPointer().Value == 0;
+    return asIntPointer().Value == 0 && Offset == 0;
   }
   /// Checks if the pointer is live.
   bool isLive() const {

diff --git a/clang/lib/AST/Interp/State.cpp b/clang/lib/AST/Interp/State.cpp
@@ -155,7 +155,8 @@ void State::addCallStack(unsigned Limit) {
     SmallString<128> Buffer;
     llvm::raw_svector_ostream Out(Buffer);
     F->describe(Out);
-    addDiag(CallRange.getBegin(), diag::note_constexpr_call_here)
-        << Out.str() << CallRange;
+    if (!Buffer.empty())
+      addDiag(CallRange.getBegin(), diag::note_constexpr_call_here)
+          << Out.str() << CallRange;
   }
 }
diff --git a/clang/lib/AST/StmtProfile.cpp b/clang/lib/AST/StmtProfile.cpp
@@ -2071,13 +2071,31 @@ StmtProfiler::VisitLambdaExpr(const LambdaExpr *S) {
   }
 
   CXXRecordDecl *Lambda = S->getLambdaClass();
-  ID.AddInteger(Lambda->getODRHash());
-
   for (const auto &Capture : Lambda->captures()) {
     ID.AddInteger(Capture.getCaptureKind());
     if (Capture.capturesVariable())
       VisitDecl(Capture.getCapturedVar());
   }
+
+  // Profiling the body of the lambda may be dangerous during deserialization.
+  // So we'd like only to profile the signature here.
+  ODRHash Hasher;
+  // FIXME: We can't get the operator call easily by
+  // `CXXRecordDecl::getLambdaCallOperator()` if we're in deserialization.
+  // So we have to do something raw here.
+  for (auto *SubDecl : Lambda->decls()) {
+    FunctionDecl *Call = nullptr;
+    if (auto *FTD = dyn_cast<FunctionTemplateDecl>(SubDecl))
+      Call = FTD->getTemplatedDecl();
+    else if (auto *FD = dyn_cast<FunctionDecl>(SubDecl))
+      Call = FD;
+
+    if (!Call)
+      continue;
+
+    Hasher.AddFunctionDecl(Call, /*SkipBody=*/true);
+  }
+  ID.AddInteger(Hasher.CalculateHash());
 }
 
 void

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -508,6 +508,11 @@ class ResultObjectVisitor : public RecursiveASTVisitor<ResultObjectVisitor> {
         isa<CXXStdInitializerListExpr>(E)) {
       return;
     }
+    if (auto *Op = dyn_cast<BinaryOperator>(E);
+        Op && Op->getOpcode() == BO_Cmp) {
+      // Builtin `<=>` returns a `std::strong_ordering` object.
+      return;
+    }
 
     if (auto *InitList = dyn_cast<InitListExpr>(E)) {
       if (!InitList->isSemanticForm())

diff --git a/clang/lib/Basic/Targets/SPIR.h b/clang/lib/Basic/Targets/SPIR.h
@@ -259,7 +259,7 @@ class LLVM_LIBRARY_VISIBILITY SPIR32TargetInfo : public SPIRTargetInfo {
     SizeType = TargetInfo::UnsignedInt;
     PtrDiffType = IntPtrType = TargetInfo::SignedInt;
     resetDataLayout("e-p:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024");
+                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,
@@ -276,7 +276,7 @@ class LLVM_LIBRARY_VISIBILITY SPIR64TargetInfo : public SPIRTargetInfo {
     SizeType = TargetInfo::UnsignedLong;
     PtrDiffType = IntPtrType = TargetInfo::SignedLong;
     resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024");
+                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,
@@ -336,7 +336,7 @@ class LLVM_LIBRARY_VISIBILITY SPIRV32TargetInfo : public BaseSPIRVTargetInfo {
     SizeType = TargetInfo::UnsignedInt;
     PtrDiffType = IntPtrType = TargetInfo::SignedInt;
     resetDataLayout("e-p:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024");
+                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,
@@ -357,7 +357,7 @@ class LLVM_LIBRARY_VISIBILITY SPIRV64TargetInfo : public BaseSPIRVTargetInfo {
     SizeType = TargetInfo::UnsignedLong;
     PtrDiffType = IntPtrType = TargetInfo::SignedLong;
     resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024");
+                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,

diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -18194,7 +18194,8 @@ Value *CodeGenFunction::EmitHLSLBuiltinExpr(unsigned BuiltinID,
     Value *Op0 = EmitScalarExpr(E->getArg(0));
     return Builder.CreateIntrinsic(
         /*ReturnType=*/llvm::Type::getInt1Ty(getLLVMContext()),
-        Intrinsic::dx_any, ArrayRef<Value *>{Op0}, nullptr, "dx.any");
+        CGM.getHLSLRuntime().getAnyIntrinsic(), ArrayRef<Value *>{Op0}, nullptr,
+        "hlsl.any");
   }
   case Builtin::BI__builtin_hlsl_elementwise_clamp: {
     Value *OpX = EmitScalarExpr(E->getArg(0));

diff --git a/clang/lib/CodeGen/CGCall.cpp b/clang/lib/CodeGen/CGCall.cpp
@@ -4694,11 +4694,11 @@ void CodeGenFunction::EmitCallArg(CallArgList &args, const Expr *E,
     AggValueSlot Slot = args.isUsingInAlloca()
         ? createPlaceholderSlot(*this, type) : CreateAggTemp(type, "agg.tmp");
 
-    bool DestroyedInCallee = true, NeedsEHCleanup = true;
+    bool DestroyedInCallee = true, NeedsCleanup = true;
     if (const auto *RD = type->getAsCXXRecordDecl())
       DestroyedInCallee = RD->hasNonTrivialDestructor();
     else
-      NeedsEHCleanup = needsEHCleanup(type.isDestructedType());
+      NeedsCleanup = type.isDestructedType();
 
     if (DestroyedInCallee)
       Slot.setExternallyDestructed();
@@ -4707,14 +4707,15 @@ void CodeGenFunction::EmitCallArg(CallArgList &args, const Expr *E,
     RValue RV = Slot.asRValue();
     args.add(RV, type);
 
-    if (DestroyedInCallee && NeedsEHCleanup) {
+    if (DestroyedInCallee && NeedsCleanup) {
       // Create a no-op GEP between the placeholder and the cleanup so we can
       // RAUW it successfully.  It also serves as a marker of the first
       // instruction where the cleanup is active.
-      pushFullExprCleanup<DestroyUnpassedArg>(EHCleanup, Slot.getAddress(),
-                                              type);
+      pushFullExprCleanup<DestroyUnpassedArg>(NormalAndEHCleanup,
+                                              Slot.getAddress(), type);
       // This unreachable is a temporary marker which will be removed later.
-      llvm::Instruction *IsActive = Builder.CreateUnreachable();
+      llvm::Instruction *IsActive =
+          Builder.CreateFlagLoad(llvm::Constant::getNullValue(Int8PtrTy));
       args.addArgCleanupDeactivation(EHStack.stable_begin(), IsActive);
     }
     return;

diff --git a/clang/lib/CodeGen/CGCleanup.cpp b/clang/lib/CodeGen/CGCleanup.cpp
@@ -634,12 +634,19 @@ static void destroyOptimisticNormalEntry(CodeGenFunction &CGF,
 /// Pops a cleanup block.  If the block includes a normal cleanup, the
 /// current insertion point is threaded through the cleanup, as are
 /// any branch fixups on the cleanup.
-void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
+void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough,
+                                      bool ForDeactivation) {
   assert(!EHStack.empty() && "cleanup stack is empty!");
   assert(isa<EHCleanupScope>(*EHStack.begin()) && "top not a cleanup!");
   EHCleanupScope &Scope = cast<EHCleanupScope>(*EHStack.begin());
   assert(Scope.getFixupDepth() <= EHStack.getNumBranchFixups());
 
+  // If we are deactivating a normal cleanup, we need to pretend that the
+  // fallthrough is unreachable. We restore this IP before returning.
+  CGBuilderTy::InsertPoint NormalDeactivateOrigIP;
+  if (ForDeactivation && (Scope.isNormalCleanup() || !getLangOpts().EHAsynch)) {
+    NormalDeactivateOrigIP = Builder.saveAndClearIP();
+  }
   // Remember activation information.
   bool IsActive = Scope.isActive();
   Address NormalActiveFlag =
@@ -729,6 +736,8 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
     EHStack.popCleanup(); // safe because there are no fixups
     assert(EHStack.getNumBranchFixups() == 0 ||
            EHStack.hasNormalCleanups());
+    if (NormalDeactivateOrigIP.isSet())
+      Builder.restoreIP(NormalDeactivateOrigIP);
     return;
   }
 
@@ -765,9 +774,16 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
   if (!RequiresNormalCleanup) {
     // Mark CPP scope end for passed-by-value Arg temp
     //   per Windows ABI which is "normally" Cleanup in callee
-    if (IsEHa && getInvokeDest() && Builder.GetInsertBlock()) {
-      if (Personality.isMSVCXXPersonality())
+    if (IsEHa && getInvokeDest()) {
+      // If we are deactivating a normal cleanup then we don't have a
+      // fallthrough. Restore original IP to emit CPP scope ends in the correct
+      // block.
+      if (NormalDeactivateOrigIP.isSet())
+        Builder.restoreIP(NormalDeactivateOrigIP);
+      if (Personality.isMSVCXXPersonality() && Builder.GetInsertBlock())
         EmitSehCppScopeEnd();
+      if (NormalDeactivateOrigIP.isSet())
+        NormalDeactivateOrigIP = Builder.saveAndClearIP();
     }
     destroyOptimisticNormalEntry(*this, Scope);
     Scope.MarkEmitted();
@@ -992,6 +1008,8 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
     }
   }
 
+  if (NormalDeactivateOrigIP.isSet())
+    Builder.restoreIP(NormalDeactivateOrigIP);
   assert(EHStack.hasNormalCleanups() || EHStack.getNumBranchFixups() == 0);
 
   // Emit the EH cleanup if required.
@@ -1281,17 +1299,8 @@ void CodeGenFunction::DeactivateCleanupBlock(EHScopeStack::stable_iterator C,
   // to the current RunCleanupsScope.
   if (C == EHStack.stable_begin() &&
       CurrentCleanupScopeDepth.strictlyEncloses(C)) {
-    // Per comment below, checking EHAsynch is not really necessary
-    // it's there to assure zero-impact w/o EHAsynch option
-    if (!Scope.isNormalCleanup() && getLangOpts().EHAsynch) {
-      PopCleanupBlock();
-    } else {
-      // If it's a normal cleanup, we need to pretend that the
-      // fallthrough is unreachable.
-      CGBuilderTy::InsertPoint SavedIP = Builder.saveAndClearIP();
-      PopCleanupBlock();
-      Builder.restoreIP(SavedIP);
-    }
+    PopCleanupBlock(/*FallthroughIsBranchThrough=*/false,
+                    /*ForDeactivation=*/true);
     return;
   }
 

diff --git a/clang/lib/CodeGen/CGDecl.cpp b/clang/lib/CodeGen/CGDecl.cpp
@@ -2216,8 +2216,11 @@ void CodeGenFunction::pushDestroyAndDeferDeactivation(
 void CodeGenFunction::pushDestroyAndDeferDeactivation(
     CleanupKind cleanupKind, Address addr, QualType type, Destroyer *destroyer,
     bool useEHCleanupForArray) {
-  pushCleanupAndDeferDeactivation<DestroyObject>(
-      cleanupKind, addr, type, destroyer, useEHCleanupForArray);
+  llvm::Instruction *DominatingIP =
+      Builder.CreateFlagLoad(llvm::Constant::getNullValue(Int8PtrTy));
+  pushDestroy(cleanupKind, addr, type, destroyer, useEHCleanupForArray);
+  DeferredDeactivationCleanupStack.push_back(
+      {EHStack.stable_begin(), DominatingIP});
 }
 
 void CodeGenFunction::pushStackRestore(CleanupKind Kind, Address SPMem) {

diff --git a/clang/lib/CodeGen/CGHLSLRuntime.h b/clang/lib/CodeGen/CGHLSLRuntime.h
@@ -73,6 +73,7 @@ class CGHLSLRuntime {
   //===----------------------------------------------------------------------===//
 
   GENERATE_HLSL_INTRINSIC_FUNCTION(All, all)
+  GENERATE_HLSL_INTRINSIC_FUNCTION(Any, any)
   GENERATE_HLSL_INTRINSIC_FUNCTION(ThreadId, thread_id)
 
   //===----------------------------------------------------------------------===//

diff --git a/clang/lib/CodeGen/CGRecordLayoutBuilder.cpp b/clang/lib/CodeGen/CGRecordLayoutBuilder.cpp
@@ -41,10 +41,11 @@ namespace {
 ///   contains enough information to determine where the runs break.  Microsoft
 ///   and Itanium follow different rules and use different codepaths.
 /// * It is desired that, when possible, bitfields use the appropriate iN type
-///   when lowered to llvm types.  For example unsigned x : 24 gets lowered to
+///   when lowered to llvm types. For example unsigned x : 24 gets lowered to
 ///   i24.  This isn't always possible because i24 has storage size of 32 bit
-///   and if it is possible to use that extra byte of padding we must use
-///   [i8 x 3] instead of i24.  The function clipTailPadding does this.
+///   and if it is possible to use that extra byte of padding we must use [i8 x
+///   3] instead of i24. This is computed when accumulating bitfields in
+///   accumulateBitfields.
 ///   C++ examples that require clipping:
 ///   struct { int a : 24; char b; }; // a must be clipped, b goes at offset 3
 ///   struct A { int a : 24; ~A(); }; // a must be clipped because:
@@ -62,11 +63,7 @@ namespace {
 ///   that the tail padding is not used in the complete class.) However,
 ///   because LLVM reads from the complete type it can generate incorrect code
 ///   if we do not clip the tail padding off of the bitfield in the complete
-///   layout.  This introduces a somewhat awkward extra unnecessary clip stage.
-///   The location of the clip is stored internally as a sentinel of type
-///   SCISSOR.  If LLVM were updated to read base types (which it probably
-///   should because locations of things such as VBases are bogus in the llvm
-///   type anyway) then we could eliminate the SCISSOR.
+///   layout.
 /// * Itanium allows nearly empty primary virtual bases.  These bases don't get
 ///   get their own storage because they're laid out as part of another base
 ///   or at the beginning of the structure.  Determining if a VBase actually
@@ -200,9 +197,7 @@ struct CGRecordLowering {
                      const CXXRecordDecl *Query) const;
   void calculateZeroInit();
   CharUnits calculateTailClippingOffset(bool isNonVirtualBaseType) const;
-  /// Lowers bitfield storage types to I8 arrays for bitfields with tail
-  /// padding that is or can potentially be used.
-  void clipTailPadding();
+  void checkBitfieldClipping() const;
   /// Determines if we need a packed llvm struct.
   void determinePacked(bool NVBaseType);
   /// Inserts padding everywhere it's needed.
@@ -305,7 +300,7 @@ void CGRecordLowering::lower(bool NVBaseType) {
   }
   llvm::stable_sort(Members);
   Members.push_back(StorageInfo(Size, getIntNType(8)));
-  clipTailPadding();
+  checkBitfieldClipping();
   determinePacked(NVBaseType);
   insertPadding();
   Members.pop_back();
@@ -531,6 +526,7 @@ CGRecordLowering::accumulateBitFields(bool isNonVirtualBaseType,
   // available padding characters.
   RecordDecl::field_iterator BestEnd = Begin;
   CharUnits BestEndOffset;
+  bool BestClipped; // Whether the representation must be in a byte array.
 
   for (;;) {
     // AtAlignedBoundary is true iff Field is the (potential) start of a new
@@ -593,10 +589,9 @@ CGRecordLowering::accumulateBitFields(bool isNonVirtualBaseType,
         // this is the best seen so far.
         BestEnd = Field;
         BestEndOffset = BeginOffset + AccessSize;
-        if (Types.getCodeGenOpts().FineGrainedBitfieldAccesses)
-          // Fine-grained access, so no merging of spans.
-          InstallBest = true;
-        else if (!BitSizeSinceBegin)
+        // Assume clipped until proven not below.
+        BestClipped = true;
+        if (!BitSizeSinceBegin)
           // A zero-sized initial span -- this will install nothing and reset
           // for another.
           InstallBest = true;
@@ -624,6 +619,12 @@ CGRecordLowering::accumulateBitFields(bool isNonVirtualBaseType,
             // The access unit is not at a naturally aligned offset within the
             // structure.
             InstallBest = true;
+
+          if (InstallBest && BestEnd == Field)
+            // We're installing the first span, whose clipping was presumed
+            // above. Compute it correctly.
+            if (getSize(Type) == AccessSize)
+              BestClipped = false;
         }
 
         if (!InstallBest) {
@@ -656,11 +657,15 @@ CGRecordLowering::accumulateBitFields(bool isNonVirtualBaseType,
             // access unit.
             BestEndOffset = BeginOffset + TypeSize;
             BestEnd = Field;
+            BestClipped = false;
           }
 
           if (Barrier)
             // The next field is a barrier that we cannot merge across.
             InstallBest = true;
+          else if (Types.getCodeGenOpts().FineGrainedBitfieldAccesses)
+            // Fine-grained access, so no merging of spans.
+            InstallBest = true;
           else
             // Otherwise, we're not installing. Update the bit size
             // of the current span to go all the way to LimitOffset, which is
@@ -679,7 +684,17 @@ CGRecordLowering::accumulateBitFields(bool isNonVirtualBaseType,
         // Add the storage member for the access unit to the record. The
         // bitfields get the offset of their storage but come afterward and
         // remain there after a stable sort.
-        llvm::Type *Type = getIntNType(Context.toBits(AccessSize));
+        llvm::Type *Type;
+        if (BestClipped) {
+          assert(getSize(getIntNType(Context.toBits(AccessSize))) >
+                     AccessSize &&
+                 "Clipped access need not be clipped");
+          Type = getByteArrayType(AccessSize);
+        } else {
+          Type = getIntNType(Context.toBits(AccessSize));
+          assert(getSize(Type) == AccessSize &&
+                 "Unclipped access must be clipped");
+        }
         Members.push_back(StorageInfo(BeginOffset, Type));
         for (; Begin != BestEnd; ++Begin)
           if (!Begin->isZeroLengthBitField(Context))
@@ -934,32 +949,21 @@ void CGRecordLowering::calculateZeroInit() {
   }
 }
 
-void CGRecordLowering::clipTailPadding() {
-  std::vector<MemberInfo>::iterator Prior = Members.begin();
-  CharUnits Tail = getSize(Prior->Data);
-  for (std::vector<MemberInfo>::iterator Member = Prior + 1,
-                                         MemberEnd = Members.end();
-       Member != MemberEnd; ++Member) {
+// Verify accumulateBitfields computed the correct storage representations.
+void CGRecordLowering::checkBitfieldClipping() const {
+#ifndef NDEBUG
+  auto Tail = CharUnits::Zero();
+  for (const auto &M : Members) {
     // Only members with data and the scissor can cut into tail padding.
-    if (!Member->Data && Member->Kind != MemberInfo::Scissor)
+    if (!M.Data && M.Kind != MemberInfo::Scissor)
       continue;
-    if (Member->Offset < Tail) {
-      assert(Prior->Kind == MemberInfo::Field &&
-             "Only storage fields have tail padding!");
-      if (!Prior->FD || Prior->FD->isBitField())
-        Prior->Data = getByteArrayType(bitsToCharUnits(llvm::alignTo(
-            cast<llvm::IntegerType>(Prior->Data)->getIntegerBitWidth(), 8)));
-      else {
-        assert(Prior->FD->hasAttr<NoUniqueAddressAttr>() &&
-               "should not have reused this field's tail padding");
-        Prior->Data = getByteArrayType(
-            Context.getTypeInfoDataSizeInChars(Prior->FD->getType()).Width);
-      }
-    }
-    if (Member->Data)
-      Prior = Member;
-    Tail = Prior->Offset + getSize(Prior->Data);
+
+    assert(M.Offset >= Tail && "Bitfield access unit is not clipped");
+    Tail = M.Offset;
+    if (M.Data)
+      Tail += getSize(M.Data);
   }
+#endif
 }
 
 void CGRecordLowering::determinePacked(bool NVBaseType) {

diff --git a/clang/lib/CodeGen/CodeGenFunction.h b/clang/lib/CodeGen/CodeGenFunction.h
@@ -957,7 +957,8 @@ class CodeGenFunction : public CodeGenTypeCache {
 
   /// PopCleanupBlock - Will pop the cleanup entry on the stack and
   /// process all branch fixups.
-  void PopCleanupBlock(bool FallThroughIsBranchThrough = false);
+  void PopCleanupBlock(bool FallThroughIsBranchThrough = false,
+                       bool ForDeactivation = false);
 
   /// DeactivateCleanupBlock - Deactivates the given cleanup block.
   /// The block cannot be reactivated.  Pops it if it's the top of the

diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -3952,9 +3952,20 @@ bool CodeGenModule::shouldEmitFunction(GlobalDecl GD) {
   // behavior may break ABI compatibility of the current unit.
   if (const Module *M = F->getOwningModule();
       M && M->getTopLevelModule()->isNamedModule() &&
-      getContext().getCurrentNamedModule() != M->getTopLevelModule() &&
-      !F->hasAttr<AlwaysInlineAttr>())
-    return false;
+      getContext().getCurrentNamedModule() != M->getTopLevelModule()) {
+    // There are practices to mark template member function as always-inline
+    // and mark the template as extern explicit instantiation but not give
+    // the definition for member function. So we have to emit the function
+    // from explicitly instantiation with always-inline.
+    //
+    // See https://github.com/llvm/llvm-project/issues/86893 for details.
+    //
+    // TODO: Maybe it is better to give it a warning if we call a non-inline
+    // function from other module units which is marked as always-inline.
+    if (!F->isTemplateInstantiation() || !F->hasAttr<AlwaysInlineAttr>()) {
+      return false;
+    }
+  }
 
   if (F->hasAttr<NoInlineAttr>())
     return false;

diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -346,11 +346,14 @@ static bool addExceptionArgs(const ArgList &Args, types::ID InputType,
   bool EH = Args.hasFlag(options::OPT_fexceptions, options::OPT_fno_exceptions,
                          false);
 
-  bool EHa = Args.hasFlag(options::OPT_fasync_exceptions,
-                          options::OPT_fno_async_exceptions, false);
-  if (EHa) {
-    CmdArgs.push_back("-fasync-exceptions");
-    EH = true;
+  // Async exceptions are Windows MSVC only.
+  if (Triple.isWindowsMSVCEnvironment()) {
+    bool EHa = Args.hasFlag(options::OPT_fasync_exceptions,
+                            options::OPT_fno_async_exceptions, false);
+    if (EHa) {
+      CmdArgs.push_back("-fasync-exceptions");
+      EH = true;
+    }
   }
 
   // Obj-C exceptions are enabled by default, regardless of -fexceptions. This
@@ -8102,7 +8105,8 @@ struct EHFlags {
 ///      The 'a' modifier is unimplemented and fundamentally hard in LLVM IR.
 /// - c: Assume that extern "C" functions are implicitly nounwind.
 /// The default is /EHs-c-, meaning cleanups are disabled.
-static EHFlags parseClangCLEHFlags(const Driver &D, const ArgList &Args) {
+static EHFlags parseClangCLEHFlags(const Driver &D, const ArgList &Args,
+                                   bool isWindowsMSVC) {
   EHFlags EH;
 
   std::vector<std::string> EHArgs =
@@ -8112,8 +8116,15 @@ static EHFlags parseClangCLEHFlags(const Driver &D, const ArgList &Args) {
       switch (EHVal[I]) {
       case 'a':
         EH.Asynch = maybeConsumeDash(EHVal, I);
-        if (EH.Asynch)
+        if (EH.Asynch) {
+          // Async exceptions are Windows MSVC only.
+          if (!isWindowsMSVC) {
+            EH.Asynch = false;
+            D.Diag(clang::diag::warn_drv_unused_argument) << "/EHa" << EHVal;
+            continue;
+          }
           EH.Synch = false;
+        }
         continue;
       case 'c':
         EH.NoUnwindC = maybeConsumeDash(EHVal, I);
@@ -8177,7 +8188,8 @@ void Clang::AddClangCLArgs(const ArgList &Args, types::ID InputType,
 
   const Driver &D = getToolChain().getDriver();
 
-  EHFlags EH = parseClangCLEHFlags(D, Args);
+  bool IsWindowsMSVC = getToolChain().getTriple().isWindowsMSVCEnvironment();
+  EHFlags EH = parseClangCLEHFlags(D, Args, IsWindowsMSVC);
   if (!isNVPTX && (EH.Synch || EH.Asynch)) {
     if (types::isCXX(InputType))
       CmdArgs.push_back("-fcxx-exceptions");

diff --git a/clang/lib/Index/USRGeneration.cpp b/clang/lib/Index/USRGeneration.cpp
@@ -267,10 +267,13 @@ void USRGenerator::VisitFunctionDecl(const FunctionDecl *D) {
     Out << '>';
   }
 
+  QualType CanonicalType = D->getType().getCanonicalType();
   // Mangle in type information for the arguments.
-  for (auto *PD : D->parameters()) {
-    Out << '#';
-    VisitType(PD->getType());
+  if (const auto *FPT = CanonicalType->getAs<FunctionProtoType>()) {
+    for (QualType PT : FPT->param_types()) {
+      Out << '#';
+      VisitType(PT);
+    }
   }
   if (D->isVariadic())
     Out << '.';

diff --git a/clang/lib/Parse/ParseDecl.cpp b/clang/lib/Parse/ParseDecl.cpp
@@ -2380,10 +2380,6 @@ Parser::DeclGroupPtrTy Parser::ParseDeclGroup(ParsingDeclSpec &DS,
       if (getLangOpts().CPlusPlus23) {
         auto &LastRecord = Actions.ExprEvalContexts.back();
         LastRecord.InLifetimeExtendingContext = true;
-
-        // Materialize non-`cv void` prvalue temporaries in discarded
-        // expressions. These materialized temporaries may be lifetime-extented.
-        LastRecord.InMaterializeTemporaryObjectContext = true;
       }
 
       if (getLangOpts().OpenMP)

diff --git a/clang/lib/Sema/CMakeLists.txt b/clang/lib/Sema/CMakeLists.txt
@@ -1,5 +1,6 @@
 set(LLVM_LINK_COMPONENTS
   Core
+  Demangle
   FrontendHLSL
   FrontendOpenMP
   MC

diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -45,6 +45,7 @@
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/STLForwardCompat.h"
 #include "llvm/ADT/StringExtras.h"
+#include "llvm/Demangle/Demangle.h"
 #include "llvm/IR/Assumptions.h"
 #include "llvm/MC/MCSectionMachO.h"
 #include "llvm/Support/Error.h"
@@ -1983,6 +1984,36 @@ static void handleWeakRefAttr(Sema &S, Decl *D, const ParsedAttr &AL) {
   D->addAttr(::new (S.Context) WeakRefAttr(S.Context, AL));
 }
 
+// Mark alias/ifunc target as used. Due to name mangling, we look up the
+// demangled name ignoring parameters (not supported by microsoftDemangle
+// https://github.com/llvm/llvm-project/issues/88825). This should handle the
+// majority of use cases while leaving namespace scope names unmarked.
+static void markUsedForAliasOrIfunc(Sema &S, Decl *D, const ParsedAttr &AL,
+                                    StringRef Str) {
+  std::unique_ptr<char, llvm::FreeDeleter> Demangled;
+  if (S.getASTContext().getCXXABIKind() != TargetCXXABI::Microsoft)
+    Demangled.reset(llvm::itaniumDemangle(Str, /*ParseParams=*/false));
+  std::unique_ptr<MangleContext> MC(S.Context.createMangleContext());
+  SmallString<256> Name;
+
+  const DeclarationNameInfo Target(
+      &S.Context.Idents.get(Demangled ? Demangled.get() : Str), AL.getLoc());
+  LookupResult LR(S, Target, Sema::LookupOrdinaryName);
+  if (S.LookupName(LR, S.TUScope)) {
+    for (NamedDecl *ND : LR) {
+      if (MC->shouldMangleDeclName(ND)) {
+        llvm::raw_svector_ostream Out(Name);
+        Name.clear();
+        MC->mangleName(GlobalDecl(ND), Out);
+      } else {
+        Name = ND->getIdentifier()->getName();
+      }
+      if (Name == Str)
+        ND->markUsed(S.Context);
+    }
+  }
+}
+
 static void handleIFuncAttr(Sema &S, Decl *D, const ParsedAttr &AL) {
   StringRef Str;
   if (!S.checkStringLiteralArgumentAttr(AL, 0, Str))
@@ -1995,6 +2026,7 @@ static void handleIFuncAttr(Sema &S, Decl *D, const ParsedAttr &AL) {
     return;
   }
 
+  markUsedForAliasOrIfunc(S, D, AL, Str);
   D->addAttr(::new (S.Context) IFuncAttr(S.Context, AL, Str));
 }
 
@@ -2029,17 +2061,7 @@ static void handleAliasAttr(Sema &S, Decl *D, const ParsedAttr &AL) {
     }
   }
 
-  // Mark target used to prevent unneeded-internal-declaration warnings.
-  if (!S.LangOpts.CPlusPlus) {
-    // FIXME: demangle Str for C++, as the attribute refers to the mangled
-    // linkage name, not the pre-mangled identifier.
-    const DeclarationNameInfo target(&S.Context.Idents.get(Str), AL.getLoc());
-    LookupResult LR(S, target, Sema::LookupOrdinaryName);
-    if (S.LookupQualifiedName(LR, S.getCurLexicalContext()))
-      for (NamedDecl *ND : LR)
-        ND->markUsed(S.Context);
-  }
-
+  markUsedForAliasOrIfunc(S, D, AL, Str);
   D->addAttr(::new (S.Context) AliasAttr(S.Context, AL, Str));
 }
 

diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
@@ -6314,7 +6314,6 @@ ExprResult Sema::BuildCXXDefaultArgExpr(SourceLocation CallLoc,
       // Pass down lifetime extending flag, and collect temporaries in
       // CreateMaterializeTemporaryExpr when we rewrite the call argument.
       keepInLifetimeExtendingContext();
-      keepInMaterializeTemporaryObjectContext();
       EnsureImmediateInvocationInDefaultArgs Immediate(*this);
       ExprResult Res;
       runWithSufficientStackSpace(CallLoc, [&] {
@@ -7740,7 +7739,8 @@ ExprResult Sema::BuildResolvedCallExpr(Expr *Fn, NamedDecl *NDecl,
   }
 
   if (CXXMethodDecl *Method = dyn_cast_or_null<CXXMethodDecl>(FDecl))
-    if (Method->isImplicitObjectMemberFunction())
+    if (!isa<RequiresExprBodyDecl>(CurContext) &&
+        Method->isImplicitObjectMemberFunction())
       return ExprError(Diag(LParenLoc, diag::err_member_call_without_object)
                        << Fn->getSourceRange() << 0);
 
@@ -18679,9 +18679,9 @@ void Sema::PopExpressionEvaluationContext() {
   // Append the collected materialized temporaries into previous context before
   // exit if the previous also is a lifetime extending context.
   auto &PrevRecord = ExprEvalContexts[ExprEvalContexts.size() - 2];
-  if (getLangOpts().CPlusPlus23 && isInLifetimeExtendingContext() &&
-      PrevRecord.InLifetimeExtendingContext && !ExprEvalContexts.empty()) {
-    auto &PrevRecord = ExprEvalContexts[ExprEvalContexts.size() - 2];
+  if (getLangOpts().CPlusPlus23 && Rec.InLifetimeExtendingContext &&
+      PrevRecord.InLifetimeExtendingContext &&
+      !Rec.ForRangeLifetimeExtendTemps.empty()) {
     PrevRecord.ForRangeLifetimeExtendTemps.append(
         Rec.ForRangeLifetimeExtendTemps);
   }

diff --git a/clang/lib/Sema/SemaExprCXX.cpp b/clang/lib/Sema/SemaExprCXX.cpp
@@ -8429,7 +8429,7 @@ ExprResult Sema::IgnoredValueConversions(Expr *E) {
     // unnecessary temporary objects. If we skip this step, IR generation is
     // able to synthesize the storage for itself in the aggregate case, and
     // adding the extra node to the AST is just clutter.
-    if (isInMaterializeTemporaryObjectContext() && getLangOpts().CPlusPlus17 &&
+    if (isInLifetimeExtendingContext() && getLangOpts().CPlusPlus17 &&
         E->isPRValue() && !E->getType()->isVoidType()) {
       ExprResult Res = TemporaryMaterializationConversion(E);
       if (Res.isInvalid())

diff --git a/clang/lib/Sema/SemaOverload.cpp b/clang/lib/Sema/SemaOverload.cpp
@@ -14395,9 +14395,16 @@ Sema::CreateOverloadedUnaryOp(SourceLocation OpLoc, UnaryOperatorKind Opc,
   ArrayRef<Expr *> ArgsArray(Args, NumArgs);
 
   if (Input->isTypeDependent()) {
+    ExprValueKind VK = ExprValueKind::VK_PRValue;
+    // [C++26][expr.unary.op][expr.pre.incr]
+    // The * operator yields an lvalue of type
+    // The pre/post increment operators yied an lvalue.
+    if (Opc == UO_PreDec || Opc == UO_PreInc || Opc == UO_Deref)
+      VK = VK_LValue;
+
     if (Fns.empty())
-      return UnaryOperator::Create(Context, Input, Opc, Context.DependentTy,
-                                   VK_PRValue, OK_Ordinary, OpLoc, false,
+      return UnaryOperator::Create(Context, Input, Opc, Context.DependentTy, VK,
+                                   OK_Ordinary, OpLoc, false,
                                    CurFPFeatureOverrides());
 
     CXXRecordDecl *NamingClass = nullptr; // lookup ignores member operators
@@ -14406,7 +14413,7 @@ Sema::CreateOverloadedUnaryOp(SourceLocation OpLoc, UnaryOperatorKind Opc,
     if (Fn.isInvalid())
       return ExprError();
     return CXXOperatorCallExpr::Create(Context, Op, Fn.get(), ArgsArray,
-                                       Context.DependentTy, VK_PRValue, OpLoc,
+                                       Context.DependentTy, VK, OpLoc,
                                        CurFPFeatureOverrides());
   }
 

diff --git a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
@@ -5483,7 +5483,6 @@ void Sema::InstantiateVariableInitializer(
         *this, Sema::ExpressionEvaluationContext::PotentiallyEvaluated, Var);
 
     keepInLifetimeExtendingContext();
-    keepInMaterializeTemporaryObjectContext();
     // Instantiate the initializer.
     ExprResult Init;
 

diff --git a/clang/lib/Sema/TreeTransform.h b/clang/lib/Sema/TreeTransform.h
@@ -4177,7 +4177,6 @@ ExprResult TreeTransform<Derived>::TransformInitializer(Expr *Init,
       getSema(), EnterExpressionEvaluationContext::InitList,
       Construct->isListInitialization());
 
-  getSema().keepInLifetimeExtendingContext();
   getSema().keepInLifetimeExtendingContext();
   SmallVector<Expr*, 8> NewArgs;
   bool ArgChanged = false;
@@ -8752,10 +8751,6 @@ TreeTransform<Derived>::TransformCXXForRangeStmt(CXXForRangeStmt *S) {
   if (getSema().getLangOpts().CPlusPlus23) {
     auto &LastRecord = getSema().ExprEvalContexts.back();
     LastRecord.InLifetimeExtendingContext = true;
-
-    // Materialize non-`cv void` prvalue temporaries in discarded
-    // expressions. These materialized temporaries may be lifetime-extented.
-    LastRecord.InMaterializeTemporaryObjectContext = true;
   }
   StmtResult Init =
       S->getInit() ? getDerived().TransformStmt(S->getInit()) : StmtResult();

diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp
@@ -915,10 +915,9 @@ ASTSelectorLookupTrait::ReadKey(const unsigned char* d, unsigned) {
   using namespace llvm::support;
 
   SelectorTable &SelTable = Reader.getContext().Selectors;
-  unsigned N =
-      endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
+  unsigned N = endian::readNext<uint16_t, llvm::endianness::little>(d);
   const IdentifierInfo *FirstII = Reader.getLocalIdentifier(
-      F, endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d));
+      F, endian::readNext<uint32_t, llvm::endianness::little>(d));
   if (N == 0)
     return SelTable.getNullarySelector(FirstII);
   else if (N == 1)
@@ -928,7 +927,7 @@ ASTSelectorLookupTrait::ReadKey(const unsigned char* d, unsigned) {
   Args.push_back(FirstII);
   for (unsigned I = 1; I != N; ++I)
     Args.push_back(Reader.getLocalIdentifier(
-        F, endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d)));
+        F, endian::readNext<uint32_t, llvm::endianness::little>(d)));
 
   return SelTable.getSelector(N, Args.data());
 }
@@ -941,11 +940,11 @@ ASTSelectorLookupTrait::ReadData(Selector, const unsigned char* d,
   data_type Result;
 
   Result.ID = Reader.getGlobalSelectorID(
-      F, endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d));
+      F, endian::readNext<uint32_t, llvm::endianness::little>(d));
   unsigned FullInstanceBits =
-      endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
+      endian::readNext<uint16_t, llvm::endianness::little>(d);
   unsigned FullFactoryBits =
-      endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
+      endian::readNext<uint16_t, llvm::endianness::little>(d);
   Result.InstanceBits = FullInstanceBits & 0x3;
   Result.InstanceHasMoreThanOneDecl = (FullInstanceBits >> 2) & 0x1;
   Result.FactoryBits = FullFactoryBits & 0x3;
@@ -956,16 +955,14 @@ ASTSelectorLookupTrait::ReadData(Selector, const unsigned char* d,
   // Load instance methods
   for (unsigned I = 0; I != NumInstanceMethods; ++I) {
     if (ObjCMethodDecl *Method = Reader.GetLocalDeclAs<ObjCMethodDecl>(
-            F,
-            endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d)))
+            F, endian::readNext<uint32_t, llvm::endianness::little>(d)))
       Result.Instance.push_back(Method);
   }
 
   // Load factory methods
   for (unsigned I = 0; I != NumFactoryMethods; ++I) {
     if (ObjCMethodDecl *Method = Reader.GetLocalDeclAs<ObjCMethodDecl>(
-            F,
-            endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d)))
+            F, endian::readNext<uint32_t, llvm::endianness::little>(d)))
       Result.Factory.push_back(Method);
   }
 
@@ -1009,8 +1006,7 @@ static bool readBit(unsigned &Bits) {
 IdentID ASTIdentifierLookupTrait::ReadIdentifierID(const unsigned char *d) {
   using namespace llvm::support;
 
-  unsigned RawID =
-      endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+  unsigned RawID = endian::readNext<uint32_t, llvm::endianness::little>(d);
   return Reader.getGlobalIdentifierID(F, RawID >> 1);
 }
 
@@ -1028,8 +1024,7 @@ IdentifierInfo *ASTIdentifierLookupTrait::ReadData(const internal_key_type& k,
                                                    unsigned DataLen) {
   using namespace llvm::support;
 
-  unsigned RawID =
-      endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+  unsigned RawID = endian::readNext<uint32_t, llvm::endianness::little>(d);
   bool IsInteresting = RawID & 0x01;
 
   // Wipe out the "is interesting" bit.
@@ -1053,9 +1048,8 @@ IdentifierInfo *ASTIdentifierLookupTrait::ReadData(const internal_key_type& k,
   }
 
   unsigned ObjCOrBuiltinID =
-      endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
-  unsigned Bits =
-      endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
+      endian::readNext<uint16_t, llvm::endianness::little>(d);
+  unsigned Bits = endian::readNext<uint16_t, llvm::endianness::little>(d);
   bool CPlusPlusOperatorKeyword = readBit(Bits);
   bool HasRevertedTokenIDToIdentifier = readBit(Bits);
   bool Poisoned = readBit(Bits);
@@ -1084,7 +1078,7 @@ IdentifierInfo *ASTIdentifierLookupTrait::ReadData(const internal_key_type& k,
   // definition.
   if (HadMacroDefinition) {
     uint32_t MacroDirectivesOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+        endian::readNext<uint32_t, llvm::endianness::little>(d);
     DataLen -= 4;
 
     Reader.addPendingMacro(II, &F, MacroDirectivesOffset);
@@ -1098,8 +1092,7 @@ IdentifierInfo *ASTIdentifierLookupTrait::ReadData(const internal_key_type& k,
     SmallVector<uint32_t, 4> DeclIDs;
     for (; DataLen > 0; DataLen -= 4)
       DeclIDs.push_back(Reader.getGlobalDeclID(
-          F,
-          endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d)));
+          F, endian::readNext<uint32_t, llvm::endianness::little>(d)));
     Reader.SetGloballyVisibleDecls(II, DeclIDs);
   }
 
@@ -1169,7 +1162,7 @@ ASTDeclContextNameLookupTrait::ReadFileRef(const unsigned char *&d) {
   using namespace llvm::support;
 
   uint32_t ModuleFileID =
-      endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+      endian::readNext<uint32_t, llvm::endianness::little>(d);
   return Reader.getLocalModuleFile(F, ModuleFileID);
 }
 
@@ -1189,18 +1182,15 @@ ASTDeclContextNameLookupTrait::ReadKey(const unsigned char *d, unsigned) {
   case DeclarationName::CXXLiteralOperatorName:
   case DeclarationName::CXXDeductionGuideName:
     Data = (uint64_t)Reader.getLocalIdentifier(
-        F, endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d));
+        F, endian::readNext<uint32_t, llvm::endianness::little>(d));
     break;
   case DeclarationName::ObjCZeroArgSelector:
   case DeclarationName::ObjCOneArgSelector:
   case DeclarationName::ObjCMultiArgSelector:
-    Data =
-        (uint64_t)Reader
-            .getLocalSelector(
-                F,
-                endian::readNext<uint32_t, llvm::endianness::little, unaligned>(
-                    d))
-            .getAsOpaquePtr();
+    Data = (uint64_t)Reader
+               .getLocalSelector(
+                   F, endian::readNext<uint32_t, llvm::endianness::little>(d))
+               .getAsOpaquePtr();
     break;
   case DeclarationName::CXXOperatorName:
     Data = *d++; // OverloadedOperatorKind
@@ -1223,8 +1213,7 @@ void ASTDeclContextNameLookupTrait::ReadDataInto(internal_key_type,
   using namespace llvm::support;
 
   for (unsigned NumDecls = DataLen / 4; NumDecls; --NumDecls) {
-    uint32_t LocalID =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+    uint32_t LocalID = endian::readNext<uint32_t, llvm::endianness::little>(d);
     Val.insert(Reader.getGlobalDeclID(F, LocalID));
   }
 }
@@ -2033,10 +2022,9 @@ HeaderFileInfoTrait::ReadKey(const unsigned char *d, unsigned) {
   using namespace llvm::support;
 
   internal_key_type ikey;
-  ikey.Size =
-      off_t(endian::readNext<uint64_t, llvm::endianness::little, unaligned>(d));
-  ikey.ModTime = time_t(
-      endian::readNext<uint64_t, llvm::endianness::little, unaligned>(d));
+  ikey.Size = off_t(endian::readNext<uint64_t, llvm::endianness::little>(d));
+  ikey.ModTime =
+      time_t(endian::readNext<uint64_t, llvm::endianness::little>(d));
   ikey.Filename = (const char *)d;
   ikey.Imported = true;
   return ikey;
@@ -2064,9 +2052,9 @@ HeaderFileInfoTrait::ReadData(internal_key_ref key, const unsigned char *d,
   HFI.DirInfo = (Flags >> 1) & 0x07;
   HFI.IndexHeaderMapHeader = Flags & 0x01;
   HFI.ControllingMacroID = Reader.getGlobalIdentifierID(
-      M, endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d));
+      M, endian::readNext<uint32_t, llvm::endianness::little>(d));
   if (unsigned FrameworkOffset =
-          endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d)) {
+          endian::readNext<uint32_t, llvm::endianness::little>(d)) {
     // The framework offset is 1 greater than the actual offset,
     // since 0 is used as an indicator for "no framework name".
     StringRef FrameworkName(FrameworkStrings + FrameworkOffset - 1);
@@ -2077,7 +2065,7 @@ HeaderFileInfoTrait::ReadData(internal_key_ref key, const unsigned char *d,
          "Wrong data length in HeaderFileInfo deserialization");
   while (d != End) {
     uint32_t LocalSMID =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+        endian::readNext<uint32_t, llvm::endianness::little>(d);
     auto HeaderRole = static_cast<ModuleMap::ModuleHeaderRole>(LocalSMID & 7);
     LocalSMID >>= 3;
 
@@ -4085,9 +4073,8 @@ void ASTReader::ReadModuleOffsetMap(ModuleFile &F) const {
     // how it goes...
     using namespace llvm::support;
     ModuleKind Kind = static_cast<ModuleKind>(
-        endian::readNext<uint8_t, llvm::endianness::little, unaligned>(Data));
-    uint16_t Len =
-        endian::readNext<uint16_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint8_t, llvm::endianness::little>(Data));
+    uint16_t Len = endian::readNext<uint16_t, llvm::endianness::little>(Data);
     StringRef Name = StringRef((const char*)Data, Len);
     Data += Len;
     ModuleFile *OM = (Kind == MK_PrebuiltModule || Kind == MK_ExplicitModule ||
@@ -4103,21 +4090,21 @@ void ASTReader::ReadModuleOffsetMap(ModuleFile &F) const {
     }
 
     SourceLocation::UIntTy SLocOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t IdentifierIDOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t MacroIDOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t PreprocessedEntityIDOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t SubmoduleIDOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t SelectorIDOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t DeclIDOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
     uint32_t TypeIndexOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Data);
+        endian::readNext<uint32_t, llvm::endianness::little>(Data);
 
     auto mapOffset = [&](uint32_t Offset, uint32_t BaseOffset,
                          RemapBuilder &Remap) {
@@ -9798,7 +9785,7 @@ void ASTReader::finishPendingActions() {
             !NonConstDefn->isLateTemplateParsed() &&
             // We only perform ODR checks for decls not in the explicit
             // global module fragment.
-            !FD->shouldSkipCheckingODR() &&
+            !shouldSkipCheckingODR(FD) &&
             FD->getODRHash() != NonConstDefn->getODRHash()) {
           if (!isa<CXXMethodDecl>(FD)) {
             PendingFunctionOdrMergeFailures[FD].push_back(NonConstDefn);

diff --git a/clang/lib/Serialization/ASTReaderDecl.cpp b/clang/lib/Serialization/ASTReaderDecl.cpp
@@ -826,7 +826,7 @@ void ASTDeclReader::VisitEnumDecl(EnumDecl *ED) {
       Reader.mergeDefinitionVisibility(OldDef, ED);
       // We don't want to check the ODR hash value for declarations from global
       // module fragment.
-      if (!ED->shouldSkipCheckingODR() &&
+      if (!shouldSkipCheckingODR(ED) &&
           OldDef->getODRHash() != ED->getODRHash())
         Reader.PendingEnumOdrMergeFailures[OldDef].push_back(ED);
     } else {
@@ -868,7 +868,7 @@ void ASTDeclReader::VisitRecordDecl(RecordDecl *RD) {
   VisitRecordDeclImpl(RD);
   // We should only reach here if we're in C/Objective-C. There is no
   // global module fragment.
-  assert(!RD->shouldSkipCheckingODR());
+  assert(!shouldSkipCheckingODR(RD));
   RD->setODRHash(Record.readInt());
 
   // Maintain the invariant of a redeclaration chain containing only
@@ -2155,7 +2155,7 @@ void ASTDeclReader::MergeDefinitionData(
   }
 
   // We don't want to check ODR for decls in the global module fragment.
-  if (MergeDD.Definition->shouldSkipCheckingODR())
+  if (shouldSkipCheckingODR(MergeDD.Definition))
     return;
 
   if (D->getODRHash() != MergeDD.ODRHash) {
@@ -3530,7 +3530,7 @@ ASTDeclReader::FindExistingResult ASTDeclReader::findExisting(NamedDecl *D) {
   // same template specialization into the same CXXRecordDecl.
   auto MergedDCIt = Reader.MergedDeclContexts.find(D->getLexicalDeclContext());
   if (MergedDCIt != Reader.MergedDeclContexts.end() &&
-      !D->shouldSkipCheckingODR() && MergedDCIt->second == D->getDeclContext())
+      !shouldSkipCheckingODR(D) && MergedDCIt->second == D->getDeclContext())
     Reader.PendingOdrMergeChecks.push_back(D);
 
   return FindExistingResult(Reader, D, /*Existing=*/nullptr,

diff --git a/clang/lib/Serialization/ASTWriter.cpp b/clang/lib/Serialization/ASTWriter.cpp
@@ -6188,7 +6188,7 @@ void ASTRecordWriter::AddCXXDefinitionData(const CXXRecordDecl *D) {
 
   BitsPacker DefinitionBits;
 
-  bool ShouldSkipCheckingODR = D->shouldSkipCheckingODR();
+  bool ShouldSkipCheckingODR = shouldSkipCheckingODR(D);
   DefinitionBits.addBit(ShouldSkipCheckingODR);
 
 #define FIELD(Name, Width, Merge)                                              \

diff --git a/clang/lib/Serialization/ASTWriterDecl.cpp b/clang/lib/Serialization/ASTWriterDecl.cpp
@@ -526,7 +526,7 @@ void ASTDeclWriter::VisitEnumDecl(EnumDecl *D) {
   BitsPacker EnumDeclBits;
   EnumDeclBits.addBits(D->getNumPositiveBits(), /*BitWidth=*/8);
   EnumDeclBits.addBits(D->getNumNegativeBits(), /*BitWidth=*/8);
-  bool ShouldSkipCheckingODR = D->shouldSkipCheckingODR();
+  bool ShouldSkipCheckingODR = shouldSkipCheckingODR(D);
   EnumDeclBits.addBit(ShouldSkipCheckingODR);
   EnumDeclBits.addBit(D->isScoped());
   EnumDeclBits.addBit(D->isScopedUsingClassTag());
@@ -552,7 +552,7 @@ void ASTDeclWriter::VisitEnumDecl(EnumDecl *D) {
       !D->isTopLevelDeclInObjCContainer() &&
       !CXXRecordDecl::classofKind(D->getKind()) &&
       !D->getIntegerTypeSourceInfo() && !D->getMemberSpecializationInfo() &&
-      !needsAnonymousDeclarationNumber(D) && !D->shouldSkipCheckingODR() &&
+      !needsAnonymousDeclarationNumber(D) && !shouldSkipCheckingODR(D) &&
       D->getDeclName().getNameKind() == DeclarationName::Identifier)
     AbbrevToUse = Writer.getDeclEnumAbbrev();
 
@@ -718,7 +718,7 @@ void ASTDeclWriter::VisitFunctionDecl(FunctionDecl *D) {
   // FIXME: stable encoding
   FunctionDeclBits.addBits(llvm::to_underlying(D->getLinkageInternal()), 3);
   FunctionDeclBits.addBits((uint32_t)D->getStorageClass(), /*BitWidth=*/3);
-  bool ShouldSkipCheckingODR = D->shouldSkipCheckingODR();
+  bool ShouldSkipCheckingODR = shouldSkipCheckingODR(D);
   FunctionDeclBits.addBit(ShouldSkipCheckingODR);
   FunctionDeclBits.addBit(D->isInlineSpecified());
   FunctionDeclBits.addBit(D->isInlined());
@@ -1559,7 +1559,7 @@ void ASTDeclWriter::VisitCXXMethodDecl(CXXMethodDecl *D) {
       D->getFirstDecl() == D->getMostRecentDecl() && !D->isInvalidDecl() &&
       !D->hasAttrs() && !D->isTopLevelDeclInObjCContainer() &&
       D->getDeclName().getNameKind() == DeclarationName::Identifier &&
-      !D->shouldSkipCheckingODR() && !D->hasExtInfo() &&
+      !shouldSkipCheckingODR(D) && !D->hasExtInfo() &&
       !D->isExplicitlyDefaulted()) {
     if (D->getTemplatedKind() == FunctionDecl::TK_NonTemplate ||
         D->getTemplatedKind() == FunctionDecl::TK_FunctionTemplate ||

diff --git a/clang/lib/Serialization/GlobalModuleIndex.cpp b/clang/lib/Serialization/GlobalModuleIndex.cpp
@@ -89,10 +89,8 @@ class IdentifierIndexReaderTrait {
   static std::pair<unsigned, unsigned>
   ReadKeyDataLength(const unsigned char*& d) {
     using namespace llvm::support;
-    unsigned KeyLen =
-        endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
-    unsigned DataLen =
-        endian::readNext<uint16_t, llvm::endianness::little, unaligned>(d);
+    unsigned KeyLen = endian::readNext<uint16_t, llvm::endianness::little>(d);
+    unsigned DataLen = endian::readNext<uint16_t, llvm::endianness::little>(d);
     return std::make_pair(KeyLen, DataLen);
   }
 
@@ -113,8 +111,7 @@ class IdentifierIndexReaderTrait {
 
     data_type Result;
     while (DataLen > 0) {
-      unsigned ID =
-          endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+      unsigned ID = endian::readNext<uint32_t, llvm::endianness::little>(d);
       Result.push_back(ID);
       DataLen -= 4;
     }
@@ -514,8 +511,7 @@ namespace {
       // The first bit indicates whether this identifier is interesting.
       // That's all we care about.
       using namespace llvm::support;
-      unsigned RawID =
-          endian::readNext<uint32_t, llvm::endianness::little, unaligned>(d);
+      unsigned RawID = endian::readNext<uint32_t, llvm::endianness::little>(d);
       bool IsInteresting = RawID & 0x01;
       return std::make_pair(k, IsInteresting);
     }

diff --git a/clang/lib/Serialization/MultiOnDiskHashTable.h b/clang/lib/Serialization/MultiOnDiskHashTable.h
@@ -200,11 +200,11 @@ template<typename Info> class MultiOnDiskHashTable {
     storage_type Ptr = Data;
 
     uint32_t BucketOffset =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Ptr);
+        endian::readNext<uint32_t, llvm::endianness::little>(Ptr);
 
     // Read the list of overridden files.
     uint32_t NumFiles =
-        endian::readNext<uint32_t, llvm::endianness::little, unaligned>(Ptr);
+        endian::readNext<uint32_t, llvm::endianness::little>(Ptr);
     // FIXME: Add a reserve() to TinyPtrVector so that we don't need to make
     // an additional copy.
     llvm::SmallVector<file_type, 16> OverriddenFiles;

diff --git a/clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
@@ -6,7 +6,11 @@
 //
 //===----------------------------------------------------------------------===//
 //
-// This checker evaluates clang builtin functions.
+// This checker evaluates "standalone" clang builtin functions that are not
+// just special-cased variants of well-known non-builtin functions.
+// Builtin functions like __builtin_memcpy and __builtin_alloca should be
+// evaluated by the same checker that handles their non-builtin variant to
+// ensure that the two variants are handled consistently.
 //
 //===----------------------------------------------------------------------===//
 
@@ -80,25 +84,6 @@ bool BuiltinFunctionChecker::evalCall(const CallEvent &Call,
     return true;
   }
 
-  case Builtin::BI__builtin_alloca_with_align:
-  case Builtin::BI__builtin_alloca: {
-    SValBuilder &SVB = C.getSValBuilder();
-    const loc::MemRegionVal R =
-        SVB.getAllocaRegionVal(CE, C.getLocationContext(), C.blockCount());
-
-    // Set the extent of the region in bytes. This enables us to use the SVal
-    // of the argument directly. If we saved the extent in bits, it'd be more
-    // difficult to reason about values like symbol*8.
-    auto Size = Call.getArgSVal(0);
-    if (auto DefSize = Size.getAs<DefinedOrUnknownSVal>()) {
-      // This `getAs()` is mostly paranoia, because core.CallAndMessage reports
-      // undefined function arguments (unless it's disabled somehow).
-      state = setDynamicExtent(state, R.getRegion(), *DefSize, SVB);
-    }
-    C.addTransition(state->BindExpr(CE, LCtx, R));
-    return true;
-  }
-
   case Builtin::BI__builtin_dynamic_object_size:
   case Builtin::BI__builtin_object_size:
   case Builtin::BI__builtin_constant_p: {

diff --git a/clang/lib/StaticAnalyzer/Checkers/MallocChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/MallocChecker.cpp
@@ -401,10 +401,11 @@ class MallocChecker
   };
 
   const CallDescriptionMap<CheckFn> FreeingMemFnMap{
-      {{{"free"}, 1}, &MallocChecker::checkFree},
-      {{{"if_freenameindex"}, 1}, &MallocChecker::checkIfFreeNameIndex},
-      {{{"kfree"}, 1}, &MallocChecker::checkFree},
-      {{{"g_free"}, 1}, &MallocChecker::checkFree},
+      {{CDM::CLibrary, {"free"}, 1}, &MallocChecker::checkFree},
+      {{CDM::CLibrary, {"if_freenameindex"}, 1},
+       &MallocChecker::checkIfFreeNameIndex},
+      {{CDM::CLibrary, {"kfree"}, 1}, &MallocChecker::checkFree},
+      {{CDM::CLibrary, {"g_free"}, 1}, &MallocChecker::checkFree},
   };
 
   bool isFreeingCall(const CallEvent &Call) const;
@@ -413,41 +414,46 @@ class MallocChecker
   friend class NoOwnershipChangeVisitor;
 
   CallDescriptionMap<CheckFn> AllocatingMemFnMap{
-      {{{"alloca"}, 1}, &MallocChecker::checkAlloca},
-      {{{"_alloca"}, 1}, &MallocChecker::checkAlloca},
-      {{{"malloc"}, 1}, &MallocChecker::checkBasicAlloc},
-      {{{"malloc"}, 3}, &MallocChecker::checkKernelMalloc},
-      {{{"calloc"}, 2}, &MallocChecker::checkCalloc},
-      {{{"valloc"}, 1}, &MallocChecker::checkBasicAlloc},
+      {{CDM::CLibrary, {"alloca"}, 1}, &MallocChecker::checkAlloca},
+      {{CDM::CLibrary, {"_alloca"}, 1}, &MallocChecker::checkAlloca},
+      // The line for "alloca" also covers "__builtin_alloca", but the
+      // _with_align variant must be listed separately because it takes an
+      // extra argument:
+      {{CDM::CLibrary, {"__builtin_alloca_with_align"}, 2},
+       &MallocChecker::checkAlloca},
+      {{CDM::CLibrary, {"malloc"}, 1}, &MallocChecker::checkBasicAlloc},
+      {{CDM::CLibrary, {"malloc"}, 3}, &MallocChecker::checkKernelMalloc},
+      {{CDM::CLibrary, {"calloc"}, 2}, &MallocChecker::checkCalloc},
+      {{CDM::CLibrary, {"valloc"}, 1}, &MallocChecker::checkBasicAlloc},
       {{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},
+      {{CDM::CLibrary, {"_strdup"}, 1}, &MallocChecker::checkStrdup},
+      {{CDM::CLibrary, {"kmalloc"}, 2}, &MallocChecker::checkKernelMalloc},
+      {{CDM::CLibrary, {"if_nameindex"}, 1}, &MallocChecker::checkIfNameIndex},
       {{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},
-      {{{"g_try_malloc0"}, 1}, &MallocChecker::checkGMalloc0},
-      {{{"g_memdup"}, 2}, &MallocChecker::checkGMemdup},
-      {{{"g_malloc_n"}, 2}, &MallocChecker::checkGMallocN},
-      {{{"g_malloc0_n"}, 2}, &MallocChecker::checkGMallocN0},
-      {{{"g_try_malloc_n"}, 2}, &MallocChecker::checkGMallocN},
-      {{{"g_try_malloc0_n"}, 2}, &MallocChecker::checkGMallocN0},
+      {{CDM::CLibrary, {"g_malloc"}, 1}, &MallocChecker::checkBasicAlloc},
+      {{CDM::CLibrary, {"g_malloc0"}, 1}, &MallocChecker::checkGMalloc0},
+      {{CDM::CLibrary, {"g_try_malloc"}, 1}, &MallocChecker::checkBasicAlloc},
+      {{CDM::CLibrary, {"g_try_malloc0"}, 1}, &MallocChecker::checkGMalloc0},
+      {{CDM::CLibrary, {"g_memdup"}, 2}, &MallocChecker::checkGMemdup},
+      {{CDM::CLibrary, {"g_malloc_n"}, 2}, &MallocChecker::checkGMallocN},
+      {{CDM::CLibrary, {"g_malloc0_n"}, 2}, &MallocChecker::checkGMallocN0},
+      {{CDM::CLibrary, {"g_try_malloc_n"}, 2}, &MallocChecker::checkGMallocN},
+      {{CDM::CLibrary, {"g_try_malloc0_n"}, 2}, &MallocChecker::checkGMallocN0},
   };
 
   CallDescriptionMap<CheckFn> ReallocatingMemFnMap{
-      {{{"realloc"}, 2},
+      {{CDM::CLibrary, {"realloc"}, 2},
        std::bind(&MallocChecker::checkRealloc, _1, _2, _3, false)},
-      {{{"reallocf"}, 2},
+      {{CDM::CLibrary, {"reallocf"}, 2},
        std::bind(&MallocChecker::checkRealloc, _1, _2, _3, true)},
-      {{{"g_realloc"}, 2},
+      {{CDM::CLibrary, {"g_realloc"}, 2},
        std::bind(&MallocChecker::checkRealloc, _1, _2, _3, false)},
-      {{{"g_try_realloc"}, 2},
+      {{CDM::CLibrary, {"g_try_realloc"}, 2},
        std::bind(&MallocChecker::checkRealloc, _1, _2, _3, false)},
-      {{{"g_realloc_n"}, 3}, &MallocChecker::checkReallocN},
-      {{{"g_try_realloc_n"}, 3}, &MallocChecker::checkReallocN},
+      {{CDM::CLibrary, {"g_realloc_n"}, 3}, &MallocChecker::checkReallocN},
+      {{CDM::CLibrary, {"g_try_realloc_n"}, 3}, &MallocChecker::checkReallocN},
 
       // NOTE: the following CallDescription also matches the C++ standard
       // library function std::getline(); the callback will filter it out.
@@ -1259,9 +1265,6 @@ static bool isStandardRealloc(const CallEvent &Call) {
   assert(FD);
   ASTContext &AC = FD->getASTContext();
 
-  if (isa<CXXMethodDecl>(FD))
-    return false;
-
   return FD->getDeclaredReturnType().getDesugaredType(AC) == AC.VoidPtrTy &&
          FD->getParamDecl(0)->getType().getDesugaredType(AC) == AC.VoidPtrTy &&
          FD->getParamDecl(1)->getType().getDesugaredType(AC) ==
@@ -1273,9 +1276,6 @@ static bool isGRealloc(const CallEvent &Call) {
   assert(FD);
   ASTContext &AC = FD->getASTContext();
 
-  if (isa<CXXMethodDecl>(FD))
-    return false;
-
   return FD->getDeclaredReturnType().getDesugaredType(AC) == AC.VoidPtrTy &&
          FD->getParamDecl(0)->getType().getDesugaredType(AC) == AC.VoidPtrTy &&
          FD->getParamDecl(1)->getType().getDesugaredType(AC) ==
@@ -1284,14 +1284,14 @@ static bool isGRealloc(const CallEvent &Call) {
 
 void MallocChecker::checkRealloc(const CallEvent &Call, CheckerContext &C,
                                  bool ShouldFreeOnFail) const {
-  // HACK: CallDescription currently recognizes non-standard realloc functions
-  // as standard because it doesn't check the type, or wether its a non-method
-  // function. This should be solved by making CallDescription smarter.
-  // Mind that this came from a bug report, and all other functions suffer from
-  // this.
-  // https://bugs.llvm.org/show_bug.cgi?id=46253
+  // Ignore calls to functions whose type does not match the expected type of
+  // either the standard realloc or g_realloc from GLib.
+  // FIXME: Should we perform this kind of checking consistently for each
+  // function? If yes, then perhaps extend the `CallDescription` interface to
+  // handle this.
   if (!isStandardRealloc(Call) && !isGRealloc(Call))
     return;
+
   ProgramStateRef State = C.getState();
   State = ReallocMemAux(C, Call, ShouldFreeOnFail, State, AF_Malloc);
   State = ProcessZeroAllocCheck(Call, 1, State);
@@ -1842,9 +1842,18 @@ static ProgramStateRef MallocUpdateRefState(CheckerContext &C, const Expr *E,
     return nullptr;
 
   SymbolRef Sym = RetVal->getAsLocSymbol();
+
   // This is a return value of a function that was not inlined, such as malloc()
   // or new(). We've checked that in the caller. Therefore, it must be a symbol.
   assert(Sym);
+  // FIXME: In theory this assertion should fail for `alloca()` calls (because
+  // `AllocaRegion`s are not symbolic); but in practice this does not happen.
+  // As the current code appears to work correctly, I'm not touching this issue
+  // now, but it would be good to investigate and clarify this.
+  // Also note that perhaps the special `AllocaRegion` should be replaced by
+  // `SymbolicRegion` (or turned into a subclass of `SymbolicRegion`) to enable
+  // proper tracking of memory allocated by `alloca()` -- and after that change
+  // this assertion would become valid again.
 
   // Set the symbol's state to Allocated.
   return State->set<RegionState>(Sym, RefState::getAllocated(Family, E));

diff --git a/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp b/clang/lib/Tooling/DependencyScanning/ModuleDepCollector.cpp
@@ -154,6 +154,26 @@ void ModuleDepCollector::addOutputPaths(CowCompilerInvocation &CI,
   }
 }
 
+void dependencies::resetBenignCodeGenOptions(frontend::ActionKind ProgramAction,
+                                             const LangOptions &LangOpts,
+                                             CodeGenOptions &CGOpts) {
+  // TODO: Figure out better way to set options to their default value.
+  if (ProgramAction == frontend::GenerateModule) {
+    CGOpts.MainFileName.clear();
+    CGOpts.DwarfDebugFlags.clear();
+  }
+  if (ProgramAction == frontend::GeneratePCH ||
+      (ProgramAction == frontend::GenerateModule && !LangOpts.ModulesCodegen)) {
+    CGOpts.DebugCompilationDir.clear();
+    CGOpts.CoverageCompilationDir.clear();
+    CGOpts.CoverageDataFile.clear();
+    CGOpts.CoverageNotesFile.clear();
+    CGOpts.ProfileInstrumentUsePath.clear();
+    CGOpts.SampleProfileFile.clear();
+    CGOpts.ProfileRemappingFile.clear();
+  }
+}
+
 static CowCompilerInvocation
 makeCommonInvocationForModuleBuild(CompilerInvocation CI) {
   CI.resetNonModularOptions();
@@ -167,18 +187,8 @@ makeCommonInvocationForModuleBuild(CompilerInvocation CI) {
   // LLVM options are not going to affect the AST
   CI.getFrontendOpts().LLVMArgs.clear();
 
-  // TODO: Figure out better way to set options to their default value.
-  CI.getCodeGenOpts().MainFileName.clear();
-  CI.getCodeGenOpts().DwarfDebugFlags.clear();
-  if (!CI.getLangOpts().ModulesCodegen) {
-    CI.getCodeGenOpts().DebugCompilationDir.clear();
-    CI.getCodeGenOpts().CoverageCompilationDir.clear();
-    CI.getCodeGenOpts().CoverageDataFile.clear();
-    CI.getCodeGenOpts().CoverageNotesFile.clear();
-    CI.getCodeGenOpts().ProfileInstrumentUsePath.clear();
-    CI.getCodeGenOpts().SampleProfileFile.clear();
-    CI.getCodeGenOpts().ProfileRemappingFile.clear();
-  }
+  resetBenignCodeGenOptions(frontend::GenerateModule, CI.getLangOpts(),
+                            CI.getCodeGenOpts());
 
   // Map output paths that affect behaviour to "-" so their existence is in the
   // context hash. The final path will be computed in addOutputPaths.
@@ -342,6 +352,8 @@ static bool needsModules(FrontendInputFile FIF) {
 
 void ModuleDepCollector::applyDiscoveredDependencies(CompilerInvocation &CI) {
   CI.clearImplicitModuleBuildOptions();
+  resetBenignCodeGenOptions(CI.getFrontendOpts().ProgramAction,
+                            CI.getLangOpts(), CI.getCodeGenOpts());
 
   if (llvm::any_of(CI.getFrontendOpts().Inputs, needsModules)) {
     Preprocessor &PP = ScanInstance.getPreprocessor();

diff --git a/clang/test/AST/Interp/builtin-align-cxx.cpp b/clang/test/AST/Interp/builtin-align-cxx.cpp
@@ -0,0 +1,258 @@
+// C++-specific checks for the alignment builtins
+// RUN: %clang_cc1 -triple=x86_64-unknown-unknown -std=c++11 %s -fsyntax-only -verify=expected,both -fexperimental-new-constant-interpreter
+// RUN: %clang_cc1 -triple=x86_64-unknown-unknown -std=c++11 %s -fsyntax-only -verify=ref,both
+
+
+/// This is just a copy of the one from test/SemaCXX/ with some of the
+/// diagnostic output adapted.
+/// Also, align32array has an initializer now, which means it's not just
+/// a dummy pointer for us and we do actually have type information for it.
+/// In the future, we need to retain type information for dummy pointers as
+/// well, so here is a test that will break once we do that:
+namespace {
+  _Alignas(32) char heh[4];
+  static_assert(!__builtin_is_aligned(&heh[1], 4), ""); // expected-error {{failed}}
+}
+
+
+// Check that we don't crash when using dependent types in __builtin_align:
+template <typename a, a b>
+void *c(void *d) { // both-note{{candidate template ignored}}
+  return __builtin_align_down(d, b);
+}
+
+struct x {};
+x foo;
+void test(void *value) {
+  c<int, 16>(value);
+  c<struct x, foo>(value); // both-error{{no matching function for call to 'c'}}
+}
+
+template <typename T, long Alignment, long ArraySize = 16>
+void test_templated_arguments() {
+  T array[ArraySize];                                                           // both-error{{variable has incomplete type 'fwddecl'}}
+  static_assert(__is_same(decltype(__builtin_align_up(array, Alignment)), T *), // both-error{{requested alignment is not a power of 2}}
+                "return type should be the decayed array type");
+  static_assert(__is_same(decltype(__builtin_align_down(array, Alignment)), T *),
+                "return type should be the decayed array type");
+  static_assert(__is_same(decltype(__builtin_is_aligned(array, Alignment)), bool),
+                "return type should be bool");
+  T *x1 = __builtin_align_up(array, Alignment);
+  T *x2 = __builtin_align_down(array, Alignment);
+  bool x3 = __builtin_align_up(array, Alignment);
+}
+
+void test() {
+  test_templated_arguments<int, 32>(); // fine
+  test_templated_arguments<struct fwddecl, 16>();
+  // both-note@-1{{in instantiation of function template specialization 'test_templated_arguments<fwddecl, 16L, 16L>'}}
+  // both-note@-2{{forward declaration of 'fwddecl'}}
+  test_templated_arguments<int, 7>(); // invalid alignment value
+  // both-note@-1{{in instantiation of function template specialization 'test_templated_arguments<int, 7L, 16L>'}}
+}
+
+template <typename T, long ArraySize>
+void test_incorrect_alignment_without_instatiation(T value) {
+  int array[32];
+  static_assert(__is_same(decltype(__builtin_align_up(array, 31)), int *), // both-error{{requested alignment is not a power of 2}}
+                "return type should be the decayed array type");
+  static_assert(__is_same(decltype(__builtin_align_down(array, 7)), int *), // both-error{{requested alignment is not a power of 2}}
+                "return type should be the decayed array type");
+  static_assert(__is_same(decltype(__builtin_is_aligned(array, -1)), bool), // both-error{{requested alignment must be 1 or greater}}
+                "return type should be bool");
+  __builtin_align_up(array);       // both-error{{too few arguments to function call, expected 2, have 1}}
+  __builtin_align_up(array, 31);   // both-error{{requested alignment is not a power of 2}}
+  __builtin_align_down(array, 31); // both-error{{requested alignment is not a power of 2}}
+  __builtin_align_up(array, 31);   // both-error{{requested alignment is not a power of 2}}
+  __builtin_align_up(value, 31);   // This shouldn't want since the type is dependent
+  __builtin_align_up(value);       // Same here
+
+  __builtin_align_up(array, sizeof(sizeof(value)) - 1); // both-error{{requested alignment is not a power of 2}}
+  __builtin_align_up(array, value); // no diagnostic as the alignment is value dependent.
+  (void)__builtin_align_up(array, ArraySize); // The same above here
+}
+
+// The original fix for the issue above broke some legitimate code.
+// Here is a regression test:
+typedef __SIZE_TYPE__ size_t;
+void *allocate_impl(size_t size);
+template <typename T>
+T *allocate() {
+  constexpr size_t allocation_size =
+      __builtin_align_up(sizeof(T), sizeof(void *));
+  return static_cast<T *>(
+      __builtin_assume_aligned(allocate_impl(allocation_size), sizeof(void *)));
+}
+struct Foo {
+  int value;
+};
+void *test2() {
+  return allocate<struct Foo>();
+}
+
+// Check that pointers-to-members cannot be used:
+class MemPtr {
+public:
+  int data;
+  void func();
+  virtual void vfunc();
+};
+void test_member_ptr() {
+  __builtin_align_up(&MemPtr::data, 64);    // both-error{{operand of type 'int MemPtr::*' where arithmetic or pointer type is required}}
+  __builtin_align_down(&MemPtr::func, 64);  // both-error{{operand of type 'void (MemPtr::*)()' where arithmetic or pointer type is required}}
+  __builtin_is_aligned(&MemPtr::vfunc, 64); // both-error{{operand of type 'void (MemPtr::*)()' where arithmetic or pointer type is required}}
+}
+
+void test_references(Foo &i) {
+  // Check that the builtins look at the referenced type rather than the reference itself.
+  (void)__builtin_align_up(i, 64);                            // both-error{{operand of type 'Foo' where arithmetic or pointer type is required}}
+  (void)__builtin_align_up(static_cast<Foo &>(i), 64);        // both-error{{operand of type 'Foo' where arithmetic or pointer type is required}}
+  (void)__builtin_align_up(static_cast<const Foo &>(i), 64);  // both-error{{operand of type 'const Foo' where arithmetic or pointer type is required}}
+  (void)__builtin_align_up(static_cast<Foo &&>(i), 64);       // both-error{{operand of type 'Foo' where arithmetic or pointer type is required}}
+  (void)__builtin_align_up(static_cast<const Foo &&>(i), 64); // both-error{{operand of type 'const Foo' where arithmetic or pointer type is required}}
+  (void)__builtin_align_up(&i, 64);
+}
+
+// Check that constexpr wrapper functions can be constant-evaluated.
+template <typename T>
+constexpr bool wrap_is_aligned(T ptr, long align) {
+  return __builtin_is_aligned(ptr, align);
+  // both-note@-1{{requested alignment -3 is not a positive power of two}}
+  // both-note@-2{{requested alignment 19 is not a positive power of two}}
+  // both-note@-3{{requested alignment must be 128 or less for type 'char'; 4194304 is invalid}}
+}
+template <typename T>
+constexpr T wrap_align_up(T ptr, long align) {
+  return __builtin_align_up(ptr, align);
+  // both-note@-1{{requested alignment -2 is not a positive power of two}}
+  // both-note@-2{{requested alignment 18 is not a positive power of two}}
+  // both-note@-3{{requested alignment must be 2147483648 or less for type 'int'; 8589934592 is invalid}}
+  // both-error@-4{{operand of type 'bool' where arithmetic or pointer type is required}}
+}
+
+template <typename T>
+constexpr T wrap_align_down(T ptr, long align) {
+  return __builtin_align_down(ptr, align);
+  // both-note@-1{{requested alignment -1 is not a positive power of two}}
+  // both-note@-2{{requested alignment 17 is not a positive power of two}}
+  // both-note@-3{{requested alignment must be 32768 or less for type 'short'; 1048576 is invalid}}
+}
+
+constexpr int a1 = wrap_align_up(22, 32);
+static_assert(a1 == 32, "");
+constexpr int a2 = wrap_align_down(22, 16);
+static_assert(a2 == 16, "");
+constexpr bool a3 = wrap_is_aligned(22, 32);
+static_assert(!a3, "");
+static_assert(wrap_align_down(wrap_align_up(22, 16), 32) == 32, "");
+static_assert(wrap_is_aligned(wrap_align_down(wrap_align_up(22, 16), 32), 32), "");
+static_assert(!wrap_is_aligned(wrap_align_down(wrap_align_up(22, 16), 32), 64), "");
+
+constexpr long const_value(long l) { return l; }
+// Check some invalid values during constant-evaluation
+static_assert(wrap_align_down(1, const_value(-1)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+static_assert(wrap_align_up(1, const_value(-2)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+static_assert(wrap_is_aligned(1, const_value(-3)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+static_assert(wrap_align_down(1, const_value(17)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+static_assert(wrap_align_up(1, const_value(18)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+static_assert(wrap_is_aligned(1, const_value(19)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+
+// Check invalid values for smaller types:
+static_assert(wrap_align_down(static_cast<short>(1), const_value(1 << 20)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to }}
+// Check invalid boolean type
+static_assert(wrap_align_up(static_cast<int>(1), const_value(1ull << 33)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+static_assert(wrap_is_aligned(static_cast<char>(1), const_value(1 << 22)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in call to}}
+
+// Check invalid boolean type
+static_assert(wrap_align_up(static_cast<bool>(1), const_value(1 << 21)), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{in instantiation of function template specialization 'wrap_align_up<bool>' requested here}}
+
+// Check constant evaluation for pointers:
+_Alignas(32) char align32array[128] = {};
+static_assert(&align32array[0] == &align32array[0], "");
+// __builtin_align_up/down can be constant evaluated as a no-op for values
+// that are known to have greater alignment:
+static_assert(__builtin_align_up(&align32array[0], 32) == &align32array[0], "");
+static_assert(__builtin_align_up(&align32array[0], 4) == &align32array[0], "");
+static_assert(__builtin_align_down(&align32array[0], 4) == __builtin_align_up(&align32array[0], 8), "");
+// But it can not be evaluated if the alignment is greater than the minimum
+// known alignment, since in that case the value might be the same if it happens
+// to actually be aligned to 64 bytes at run time.
+static_assert(&align32array[0] == __builtin_align_up(&align32array[0], 64), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{cannot constant evaluate the result of adjusting alignment to 64}}
+static_assert(__builtin_align_up(&align32array[0], 64) == __builtin_align_up(&align32array[0], 64), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{cannot constant evaluate the result of adjusting alignment to 64}}
+
+// However, we can compute in case the requested alignment is less than the
+// base alignment:
+static_assert(__builtin_align_up(&align32array[0], 4) == &align32array[0], "");
+static_assert(__builtin_align_up(&align32array[1], 4) == &align32array[4], "");
+static_assert(__builtin_align_up(&align32array[2], 4) == &align32array[4], "");
+static_assert(__builtin_align_up(&align32array[3], 4) == &align32array[4], "");
+static_assert(__builtin_align_up(&align32array[4], 4) == &align32array[4], "");
+static_assert(__builtin_align_up(&align32array[5], 4) == &align32array[8], "");
+static_assert(__builtin_align_up(&align32array[6], 4) == &align32array[8], "");
+static_assert(__builtin_align_up(&align32array[7], 4) == &align32array[8], "");
+static_assert(__builtin_align_up(&align32array[8], 4) == &align32array[8], "");
+
+static_assert(__builtin_align_down(&align32array[0], 4) == &align32array[0], "");
+static_assert(__builtin_align_down(&align32array[1], 4) == &align32array[0], "");
+static_assert(__builtin_align_down(&align32array[2], 4) == &align32array[0], "");
+static_assert(__builtin_align_down(&align32array[3], 4) == &align32array[0], "");
+static_assert(__builtin_align_down(&align32array[4], 4) == &align32array[4], "");
+static_assert(__builtin_align_down(&align32array[5], 4) == &align32array[4], "");
+static_assert(__builtin_align_down(&align32array[6], 4) == &align32array[4], "");
+static_assert(__builtin_align_down(&align32array[7], 4) == &align32array[4], "");
+static_assert(__builtin_align_down(&align32array[8], 4) == &align32array[8], "");
+
+// Achieving the same thing using casts to uintptr_t is not allowed:
+static_assert((char *)((__UINTPTR_TYPE__)&align32array[7] & ~3) == &align32array[4], ""); // both-error{{not an integral constant expression}} \
+                                                                                          // expected-note {{cast that performs the conversions of a reinterpret_cast is not allowed in a constant expression}}
+
+static_assert(__builtin_align_down(&align32array[1], 4) == &align32array[0], "");
+static_assert(__builtin_align_down(&align32array[1], 64) == &align32array[0], ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{cannot constant evaluate the result of adjusting alignment to 64}}
+
+// Add some checks for __builtin_is_aligned:
+static_assert(__builtin_is_aligned(&align32array[0], 32), "");
+static_assert(__builtin_is_aligned(&align32array[4], 4), "");
+// We cannot constant evaluate whether the array is aligned to > 32 since this
+// may well be true at run time.
+static_assert(!__builtin_is_aligned(&align32array[0], 64), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{cannot constant evaluate whether run-time alignment is at least 64}}
+
+// However, if the alignment being checked is less than the minimum alignment of
+// the base object we can check the low bits of the alignment:
+static_assert(__builtin_is_aligned(&align32array[0], 4), "");
+static_assert(!__builtin_is_aligned(&align32array[1], 4), "");
+static_assert(!__builtin_is_aligned(&align32array[2], 4), "");
+static_assert(!__builtin_is_aligned(&align32array[3], 4), "");
+static_assert(__builtin_is_aligned(&align32array[4], 4), "");
+
+// TODO: this should evaluate to true even though we can't evaluate the result
+//  of __builtin_align_up() to a concrete value
+static_assert(__builtin_is_aligned(__builtin_align_up(&align32array[0], 64), 64), ""); // both-error{{not an integral constant expression}}
+// both-note@-1{{cannot constant evaluate the result of adjusting alignment to 64}}
+
+// Check different source and alignment type widths are handled correctly.
+static_assert(!__builtin_is_aligned(static_cast<signed long>(7), static_cast<signed short>(4)), "");
+static_assert(!__builtin_is_aligned(static_cast<signed short>(7), static_cast<signed long>(4)), "");
+// Also check signed -- unsigned mismatch.
+static_assert(!__builtin_is_aligned(static_cast<signed long>(7), static_cast<signed long>(4)), "");
+static_assert(!__builtin_is_aligned(static_cast<unsigned long>(7), static_cast<unsigned long>(4)), "");
+static_assert(!__builtin_is_aligned(static_cast<signed long>(7), static_cast<unsigned long>(4)), "");
+static_assert(!__builtin_is_aligned(static_cast<unsigned long>(7), static_cast<signed long>(4)), "");
+static_assert(!__builtin_is_aligned(static_cast<signed long>(7), static_cast<unsigned short>(4)), "");
+static_assert(!__builtin_is_aligned(static_cast<unsigned short>(7), static_cast<signed long>(4)), "");
+
+// Check the diagnostic message
+_Alignas(void) char align_void_array[1]; // both-error {{invalid application of '_Alignas' to an incomplete type 'void'}}
diff --git a/clang/test/AST/Interp/builtin-functions.cpp b/clang/test/AST/Interp/builtin-functions.cpp
@@ -24,16 +24,13 @@ namespace strcmp {
   static_assert(__builtin_strcmp("abab", "abab\0banana") == 0, "");
   static_assert(__builtin_strcmp("abab\0banana", "abab\0canada") == 0, "");
   static_assert(__builtin_strcmp(0, "abab") == 0, ""); // both-error {{not an integral constant}} \
-                                                       // both-note {{dereferenced null}} \
-                                                       // expected-note {{in call to}}
+                                                       // both-note {{dereferenced null}}
   static_assert(__builtin_strcmp("abab", 0) == 0, ""); // both-error {{not an integral constant}} \
-                                                       // both-note {{dereferenced null}} \
-                                                       // expected-note {{in call to}}
+                                                       // both-note {{dereferenced null}}
 
   static_assert(__builtin_strcmp(kFoobar, kFoobazfoobar) == -1, "");
   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}}
+                                                                        // both-note {{dereferenced one-past-the-end}}
 
   /// Used to assert because we're passing a dummy pointer to
   /// __builtin_strcmp() when evaluating the return statement.
@@ -72,14 +69,11 @@ constexpr const char *a = "foo\0quux";
   static_assert(check(c), "");
 
   constexpr int over1 = __builtin_strlen(a + 9); // both-error {{constant expression}} \
-                                                 // both-note {{one-past-the-end}} \
-                                                 // expected-note {{in call to}}
+                                                 // both-note {{one-past-the-end}}
   constexpr int over2 = __builtin_strlen(b + 9); // both-error {{constant expression}} \
-                                                 // both-note {{one-past-the-end}} \
-                                                 // expected-note {{in call to}}
+                                                 // both-note {{one-past-the-end}}
   constexpr int over3 = __builtin_strlen(c + 9); // both-error {{constant expression}} \
-                                                 // both-note {{one-past-the-end}} \
-                                                 // expected-note {{in call to}}
+                                                 // both-note {{one-past-the-end}}
 
   constexpr int under1 = __builtin_strlen(a - 1); // both-error {{constant expression}} \
                                                   // both-note {{cannot refer to element -1}}
@@ -90,8 +84,7 @@ constexpr const char *a = "foo\0quux";
 
   constexpr char d[] = { 'f', 'o', 'o' }; // no nul terminator.
   constexpr int bad = __builtin_strlen(d); // both-error {{constant expression}} \
-                                           // both-note {{one-past-the-end}} \
-                                           // expected-note {{in call to}}
+                                           // both-note {{one-past-the-end}}
 }
 
 namespace nan {
@@ -114,8 +107,7 @@ namespace nan {
   /// FIXME: Current interpreter misses diagnostics.
   constexpr char f2[] = {'0', 'x', 'A', 'E'}; /// No trailing 0 byte.
   constexpr double NaN7 = __builtin_nan(f2); // both-error {{must be initialized by a constant expression}} \
-                                             // expected-note {{read of dereferenced one-past-the-end pointer}} \
-                                             // expected-note {{in call to}}
+                                             // expected-note {{read of dereferenced one-past-the-end pointer}}
   static_assert(!__builtin_issignaling(__builtin_nan("")), "");
   static_assert(__builtin_issignaling(__builtin_nans("")), "");
 }

diff --git a/clang/test/AST/Interp/cxx03.cpp b/clang/test/AST/Interp/cxx03.cpp
@@ -10,3 +10,17 @@ namespace NonInitializingMemberExpr {
                                                                                               // both-note {{required by}} \
                                                                                               // both-note {{subexpression not valid}}
 }
+
+
+namespace NonLValueMemberExpr {
+  struct PODType {
+    int value;
+  };
+
+#define ATTR __attribute__((require_constant_initialization))
+  struct TT1 {
+    ATTR static const int &subobj_init;
+  };
+
+  const int &TT1::subobj_init = PODType().value;
+}
diff --git a/clang/test/AST/Interp/functions.cpp b/clang/test/AST/Interp/functions.cpp
@@ -584,9 +584,20 @@ namespace VariadicOperator {
 namespace WeakCompare {
   [[gnu::weak]]void weak_method();
   static_assert(weak_method != nullptr, ""); // both-error {{not an integral constant expression}} \
-                                         // both-note {{comparison against address of weak declaration '&weak_method' can only be performed at runtim}}
+                                             // both-note {{comparison against address of weak declaration '&weak_method' can only be performed at runtim}}
 
   constexpr auto A = &weak_method;
   static_assert(A != nullptr, ""); // both-error {{not an integral constant expression}} \
-                               // both-note {{comparison against address of weak declaration '&weak_method' can only be performed at runtim}}
+                                   // both-note {{comparison against address of weak declaration '&weak_method' can only be performed at runtim}}
+}
+
+namespace FromIntegral {
+#if __cplusplus >= 202002L
+  typedef double (*DoubleFn)();
+  int a[(int)DoubleFn((void*)-1)()]; // both-error {{not allowed at file scope}} \
+                                    // both-warning {{variable length arrays}}
+  int b[(int)DoubleFn((void*)(-1 + 1))()]; // both-error {{not allowed at file scope}} \
+                                           // expected-note {{evaluates to a null function pointer}} \
+                                           // both-warning {{variable length arrays}}
+#endif
 }
diff --git a/clang/test/AST/Interp/records.cpp b/clang/test/AST/Interp/records.cpp
@@ -1309,3 +1309,11 @@ namespace pr18633 {
     func2<int>();
   }
 }
+
+namespace {
+  struct F {
+    static constexpr int Z = 12;
+  };
+  F f;
+  static_assert(f.Z == 12, "");
+}
diff --git a/clang/test/AST/Interp/vectors.cpp b/clang/test/AST/Interp/vectors.cpp
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -fexperimental-new-constant-interpreter -verify=expected,both %s
 // RUN: %clang_cc1 -verify=ref,both %s
 
-// both-no-diagnostics
+// ref-no-diagnostics
 
 typedef int __attribute__((vector_size(16))) VI4;
 constexpr VI4 A = {1,2,3,4};
@@ -13,10 +13,18 @@ namespace Vector {
     return VI4 { n * 3, n + 4, n - 5, n / 6 };
   }
   constexpr auto v1 = f(10);
+  static_assert(__builtin_vectorelements(v1) == (16 / sizeof(int)), "");
 
   typedef double __attribute__((vector_size(32))) VD4;
   constexpr VD4 g(int n) {
     return (VD4) { n / 2.0, n + 1.5, n - 5.4, n * 0.9 };
   }
   constexpr auto v2 = g(4);
+  static_assert(__builtin_vectorelements(v2) == (32 / sizeof(double)), "");
+}
+
+/// FIXME: We need to support BitCasts between vector types.
+namespace {
+  typedef float __attribute__((vector_size(16))) VI42;
+  constexpr VI42 A2 = A; // expected-error {{must be initialized by a constant expression}}
 }
diff --git a/clang/test/AST/ast-dump-attr-json.cpp b/clang/test/AST/ast-dump-attr-json.cpp
@@ -46,6 +46,7 @@ __thread __attribute__ ((tls_model ("local-exec"))) int tls_model_var;
 // CHECK-NEXT:    "tokLen": 11
 // CHECK-NEXT:   }
 // CHECK-NEXT:  },
+// CHECK-NEXT:  "isUsed": true,
 // CHECK-NEXT:  "name": "global_decl",
 // CHECK-NEXT:  "mangledName": "global_decl",
 // CHECK-NEXT:  "type": {

diff --git a/clang/test/Analysis/Inputs/system-header-simulator-cxx.h b/clang/test/Analysis/Inputs/system-header-simulator-cxx.h
@@ -1106,6 +1106,7 @@ using ostream = basic_ostream<char>;
 extern std::ostream cout;
 
 ostream &operator<<(ostream &, const string &);
+
 #if __cplusplus >= 202002L
 template <class T>
 ostream &operator<<(ostream &, const std::unique_ptr<T> &);
@@ -1122,11 +1123,12 @@ istream &getline(istream &, string &, char);
 istream &getline(istream &, string &);
 } // namespace std
 
-#ifdef TEST_INLINABLE_ALLOCATORS
 namespace std {
   void *malloc(size_t);
   void free(void *);
-}
+} // namespace std
+
+#ifdef TEST_INLINABLE_ALLOCATORS
 void* operator new(std::size_t size, const std::nothrow_t&) throw() { return std::malloc(size); }
 void* operator new[](std::size_t size, const std::nothrow_t&) throw() { return std::malloc(size); }
 void operator delete(void* ptr, const std::nothrow_t&) throw() { std::free(ptr); }

diff --git a/clang/test/Analysis/cxx-uninitialized-object-ptr-ref.cpp b/clang/test/Analysis/cxx-uninitialized-object-ptr-ref.cpp
@@ -1,9 +1,9 @@
-// RUN: %clang_analyze_cc1 -analyzer-checker=core,optin.cplusplus.UninitializedObject \
+// RUN: %clang_analyze_cc1 -analyzer-checker=core,unix.Malloc,optin.cplusplus.UninitializedObject \
 // RUN:   -analyzer-config optin.cplusplus.UninitializedObject:Pedantic=true -DPEDANTIC \
 // RUN:   -analyzer-config optin.cplusplus.UninitializedObject:CheckPointeeInitialization=true \
 // RUN:   -std=c++11 -verify  %s
 
-// RUN: %clang_analyze_cc1 -analyzer-checker=core,optin.cplusplus.UninitializedObject \
+// RUN: %clang_analyze_cc1 -analyzer-checker=core,unix.Malloc,optin.cplusplus.UninitializedObject \
 // RUN:   -analyzer-config optin.cplusplus.UninitializedObject:CheckPointeeInitialization=true \
 // RUN:   -std=c++11 -verify  %s
 
@@ -316,7 +316,10 @@ void fCyclicPointerTest2() {
 
 // Void pointer tests are mainly no-crash tests.
 
-void *malloc(int size);
+typedef __typeof(sizeof(int)) size_t;
+
+void *calloc(size_t nmemb, size_t size);
+void free(void *p);
 
 class VoidPointerTest1 {
   void *vptr;
@@ -328,8 +331,9 @@ class VoidPointerTest1 {
 };
 
 void fVoidPointerTest1() {
-  void *vptr = malloc(sizeof(int));
+  void *vptr = calloc(1, sizeof(int));
   VoidPointerTest1(vptr, char());
+  free(vptr);
 }
 
 class VoidPointerTest2 {
@@ -342,8 +346,9 @@ class VoidPointerTest2 {
 };
 
 void fVoidPointerTest2() {
-  void *vptr = malloc(sizeof(int));
+  void *vptr = calloc(1, sizeof(int));
   VoidPointerTest2(&vptr, char());
+  free(vptr);
 }
 
 class VoidPointerRRefTest1 {
@@ -359,8 +364,9 @@ upon returning to the caller.  This will be a dangling reference}}
 };
 
 void fVoidPointerRRefTest1() {
-  void *vptr = malloc(sizeof(int));
+  void *vptr = calloc(1, sizeof(int));
   VoidPointerRRefTest1(vptr, char());
+  free(vptr);
 }
 
 class VoidPointerRRefTest2 {
@@ -376,8 +382,9 @@ upon returning to the caller.  This will be a dangling reference}}
 };
 
 void fVoidPointerRRefTest2() {
-  void *vptr = malloc(sizeof(int));
+  void *vptr = calloc(1, sizeof(int));
   VoidPointerRRefTest2(&vptr, char());
+  free(vptr);
 }
 
 class VoidPointerLRefTest {
@@ -393,8 +400,9 @@ upon returning to the caller.  This will be a dangling reference}}
 };
 
 void fVoidPointerLRefTest() {
-  void *vptr = malloc(sizeof(int));
+  void *vptr = calloc(1, sizeof(int));
   VoidPointerLRefTest(vptr, char());
+  free(vptr);
 }
 
 struct CyclicVoidPointerTest {

diff --git a/clang/test/Analysis/exercise-ps.c b/clang/test/Analysis/exercise-ps.c
@@ -1,5 +1,5 @@
 // RUN: %clang_analyze_cc1 %s -verify -Wno-error=implicit-function-declaration \
-// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=core,unix.Malloc \
 // RUN:   -analyzer-config core.CallAndMessage:ArgPointeeInitializedness=true
 //
 // Just exercise the analyzer on code that has at one point caused issues

diff --git a/clang/test/Analysis/explain-svals.cpp b/clang/test/Analysis/explain-svals.cpp
@@ -1,7 +1,7 @@
 // RUN: %clang_analyze_cc1 -triple i386-apple-darwin10 -verify %s \
-// RUN:   -analyzer-checker=core.builtin \
 // RUN:   -analyzer-checker=debug.ExprInspection \
 // RUN:   -analyzer-checker=unix.cstring \
+// RUN:   -analyzer-checker=unix.Malloc \
 // RUN:   -analyzer-config display-checker-name=false
 
 typedef unsigned long size_t;

diff --git a/clang/test/Analysis/malloc-std-namespace.cpp b/clang/test/Analysis/malloc-std-namespace.cpp
@@ -0,0 +1,24 @@
+// RUN: %clang_analyze_cc1 -analyzer-checker=core,unix.Malloc -verify -analyzer-output=text %s
+
+// This file tests that unix.Malloc can handle C++ code where e.g. malloc and
+// free are declared within the namespace 'std' by the header <cstdlib>.
+
+#include "Inputs/system-header-simulator-cxx.h"
+
+void leak() {
+  int *p = static_cast<int*>(std::malloc(sizeof(int))); // expected-note{{Memory is allocated}}
+} // expected-warning{{Potential leak of memory pointed to by 'p'}}
+  // expected-note@-1{{Potential leak of memory pointed to by 'p'}}
+
+void no_leak() {
+  int *p = static_cast<int*>(std::malloc(sizeof(int)));
+  std::free(p); // no-warning
+}
+
+void invalid_free() {
+  int i;
+  int *p = &i;
+  //expected-note@+2{{Argument to free() is the address of the local variable 'i', which is not memory allocated by malloc()}}
+  //expected-warning@+1{{Argument to free() is the address of the local variable 'i', which is not memory allocated by malloc()}}
+  std::free(p);
+}
diff --git a/clang/test/Analysis/malloc.c b/clang/test/Analysis/malloc.c
@@ -740,6 +740,17 @@ void allocaFree(void) {
   free(p); // expected-warning {{Memory allocated by alloca() should not be deallocated}}
 }
 
+void allocaFreeBuiltin(void) {
+  int *p = __builtin_alloca(sizeof(int));
+  free(p); // expected-warning {{Memory allocated by alloca() should not be deallocated}}
+}
+
+void allocaFreeBuiltinAlign(void) {
+  int *p = __builtin_alloca_with_align(sizeof(int), 64);
+  free(p); // expected-warning {{Memory allocated by alloca() should not be deallocated}}
+}
+
+
 int* mallocEscapeRet(void) {
   int *p = malloc(12);
   return p; // no warning