diff --git a/clang/lib/AST/DynamicRecursiveASTVisitor.cpp b/clang/lib/AST/DynamicRecursiveASTVisitor.cpp
diff --git a/clang/lib/AST/Type.cpp b/clang/lib/AST/Type.cpp
@@ -43,6 +43,7 @@
 #include "llvm/ADT/APSInt.h"
 #include "llvm/ADT/ArrayRef.h"
 #include "llvm/ADT/FoldingSet.h"
+#include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/Support/Casting.h"
 #include "llvm/Support/ErrorHandling.h"
@@ -4774,7 +4775,10 @@ bool Type::canHaveNullability(bool ResultIfUnknown) const {
                 ->getTemplateName()
                 .getAsTemplateDecl())
       if (auto *CTD = dyn_cast<ClassTemplateDecl>(templateDecl))
-        return CTD->getTemplatedDecl()->hasAttr<TypeNullableAttr>();
+        return llvm::any_of(
+            CTD->redecls(), [](const RedeclarableTemplateDecl *RTD) {
+              return RTD->getTemplatedDecl()->hasAttr<TypeNullableAttr>();
+            });
     return ResultIfUnknown;
 
   case Type::Builtin:
@@ -4841,10 +4845,14 @@ bool Type::canHaveNullability(bool ResultIfUnknown) const {
     // For template specializations, look only at primary template attributes.
     // This is a consistent regardless of whether the instantiation is known.
     if (const auto *CTSD = dyn_cast<ClassTemplateSpecializationDecl>(RD))
-      return CTSD->getSpecializedTemplate()
-          ->getTemplatedDecl()
-          ->hasAttr<TypeNullableAttr>();
-    return RD->hasAttr<TypeNullableAttr>();
+      return llvm::any_of(
+          CTSD->getSpecializedTemplate()->redecls(),
+          [](const RedeclarableTemplateDecl *RTD) {
+            return RTD->getTemplatedDecl()->hasAttr<TypeNullableAttr>();
+          });
+    return llvm::any_of(RD->redecls(), [](const TagDecl *RD) {
+      return RD->hasAttr<TypeNullableAttr>();
+    });
   }
 
   // Non-pointer types.

diff --git a/clang/lib/Basic/DiagnosticIDs.cpp b/clang/lib/Basic/DiagnosticIDs.cpp
@@ -575,6 +575,12 @@ DiagnosticIDs::getDiagnosticSeverity(unsigned DiagID, SourceLocation Loc,
       DiagID != diag::fatal_too_many_errors && Diag.FatalsAsError)
     Result = diag::Severity::Error;
 
+  // Rest of the mappings are only applicable for diagnostics associated with a
+  // SourceLocation, bail out early for others.
+  if (!Diag.hasSourceManager())
+    return Result;
+
+  const auto &SM = Diag.getSourceManager();
   // Custom diagnostics always are emitted in system headers.
   bool ShowInSystemHeader =
       !GetDiagInfo(DiagID) || GetDiagInfo(DiagID)->WarnShowInSystemHeader;
@@ -583,15 +589,14 @@ DiagnosticIDs::getDiagnosticSeverity(unsigned DiagID, SourceLocation Loc,
   // because we also want to ignore extensions and warnings in -Werror and
   // -pedantic-errors modes, which *map* warnings/extensions to errors.
   if (State->SuppressSystemWarnings && !ShowInSystemHeader && Loc.isValid() &&
-      Diag.getSourceManager().isInSystemHeader(
-          Diag.getSourceManager().getExpansionLoc(Loc)))
+      SM.isInSystemHeader(SM.getExpansionLoc(Loc)))
     return diag::Severity::Ignored;
 
   // We also ignore warnings due to system macros
   bool ShowInSystemMacro =
       !GetDiagInfo(DiagID) || GetDiagInfo(DiagID)->WarnShowInSystemMacro;
   if (State->SuppressSystemWarnings && !ShowInSystemMacro && Loc.isValid() &&
-      Diag.getSourceManager().isInSystemMacro(Loc))
+      SM.isInSystemMacro(Loc))
     return diag::Severity::Ignored;
 
   return Result;

diff --git a/clang/lib/Basic/Targets/SPIR.h b/clang/lib/Basic/Targets/SPIR.h
@@ -310,8 +310,8 @@ class LLVM_LIBRARY_VISIBILITY SPIRVTargetInfo : public BaseSPIRVTargetInfo {
 
     // SPIR-V IDs are represented with a single 32-bit word.
     SizeType = TargetInfo::UnsignedInt;
-    resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1");
+    resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-"
+                    "v256:256-v512:512-v1024:1024-n8:16:32:64-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,
@@ -334,8 +334,8 @@ class LLVM_LIBRARY_VISIBILITY SPIRV32TargetInfo : public BaseSPIRVTargetInfo {
     // SPIR-V has core support for atomic ops, and Int32 is always available;
     // we take the maximum because it's possible the Host supports wider types.
     MaxAtomicInlineWidth = std::max<unsigned char>(MaxAtomicInlineWidth, 32);
-    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-G1");
+    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-n8:16:32:64-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,
@@ -358,8 +358,8 @@ class LLVM_LIBRARY_VISIBILITY SPIRV64TargetInfo : public BaseSPIRVTargetInfo {
     // SPIR-V has core support for atomic ops, and Int64 is always available;
     // we take the maximum because it's possible the Host supports wider types.
     MaxAtomicInlineWidth = std::max<unsigned char>(MaxAtomicInlineWidth, 64);
-    resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1");
+    resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-"
+                    "v256:256-v512:512-v1024:1024-n8:16:32:64-G1");
   }
 
   void getTargetDefines(const LangOptions &Opts,
@@ -384,8 +384,8 @@ class LLVM_LIBRARY_VISIBILITY SPIRV64AMDGCNTargetInfo final
     PtrDiffType = IntPtrType = TargetInfo::SignedLong;
     AddrSpaceMap = &SPIRDefIsGenMap;
 
-    resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-"
-                    "v96:128-v192:256-v256:256-v512:512-v1024:1024-G1-P4-A0");
+    resetDataLayout("e-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-"
+                    "v256:256-v512:512-v1024:1024-n32:64-S32-G1-P4-A0");
 
     BFloat16Width = BFloat16Align = 16;
     BFloat16Format = &llvm::APFloat::BFloat();

diff --git a/clang/lib/Basic/Targets/X86.cpp b/clang/lib/Basic/Targets/X86.cpp
@@ -430,6 +430,8 @@ bool X86TargetInfo::handleTargetFeatures(std::vector<std::string> &Features,
       HasAMXCOMPLEX = true;
     } else if (Feature == "+amx-fp8") {
       HasAMXFP8 = true;
+    } else if (Feature == "+amx-transpose") {
+      HasAMXTRANSPOSE = true;
     } else if (Feature == "+cmpccxadd") {
       HasCMPCCXADD = true;
     } else if (Feature == "+raoint") {
@@ -951,6 +953,8 @@ void X86TargetInfo::getTargetDefines(const LangOptions &Opts,
     Builder.defineMacro("__AMX_COMPLEX__");
   if (HasAMXFP8)
     Builder.defineMacro("__AMX_FP8__");
+  if (HasAMXTRANSPOSE)
+    Builder.defineMacro("__AMX_TRANSPOSE__");
   if (HasCMPCCXADD)
     Builder.defineMacro("__CMPCCXADD__");
   if (HasRAOINT)
@@ -1079,9 +1083,10 @@ bool X86TargetInfo::isValidFeatureName(StringRef Name) const {
       .Case("amx-bf16", true)
       .Case("amx-complex", true)
       .Case("amx-fp16", true)
+      .Case("amx-fp8", true)
       .Case("amx-int8", true)
       .Case("amx-tile", true)
-      .Case("amx-fp8", true)
+      .Case("amx-transpose", true)
       .Case("avx", true)
       .Case("avx10.1-256", true)
       .Case("avx10.1-512", true)
@@ -1198,9 +1203,10 @@ bool X86TargetInfo::hasFeature(StringRef Feature) const {
       .Case("amx-bf16", HasAMXBF16)
       .Case("amx-complex", HasAMXCOMPLEX)
       .Case("amx-fp16", HasAMXFP16)
+      .Case("amx-fp8", HasAMXFP8)
       .Case("amx-int8", HasAMXINT8)
       .Case("amx-tile", HasAMXTILE)
-      .Case("amx-fp8", HasAMXFP8)
+      .Case("amx-transpose", HasAMXTRANSPOSE)
       .Case("avx", SSELevel >= AVX)
       .Case("avx10.1-256", HasAVX10_1)
       .Case("avx10.1-512", HasAVX10_1_512)

diff --git a/clang/lib/Basic/Targets/X86.h b/clang/lib/Basic/Targets/X86.h
@@ -158,6 +158,7 @@ class LLVM_LIBRARY_VISIBILITY X86TargetInfo : public TargetInfo {
   bool HasAMXBF16 = false;
   bool HasAMXCOMPLEX = false;
   bool HasAMXFP8 = false;
+  bool HasAMXTRANSPOSE = false;
   bool HasSERIALIZE = false;
   bool HasTSXLDTRK = false;
   bool HasUSERMSR = false;

diff --git a/clang/lib/CodeGen/BackendUtil.cpp b/clang/lib/CodeGen/BackendUtil.cpp
@@ -674,7 +674,7 @@ static void addKCFIPass(const Triple &TargetTriple, const LangOptions &LangOpts,
 
   // Ensure we lower KCFI operand bundles with -O0.
   PB.registerOptimizerLastEPCallback(
-      [&](ModulePassManager &MPM, OptimizationLevel Level) {
+      [&](ModulePassManager &MPM, OptimizationLevel Level, ThinOrFullLTOPhase) {
         if (Level == OptimizationLevel::O0 &&
             LangOpts.Sanitize.has(SanitizerKind::KCFI))
           MPM.addPass(createModuleToFunctionPassAdaptor(KCFIPass()));
@@ -693,8 +693,8 @@ static void addKCFIPass(const Triple &TargetTriple, const LangOptions &LangOpts,
 static void addSanitizers(const Triple &TargetTriple,
                           const CodeGenOptions &CodeGenOpts,
                           const LangOptions &LangOpts, PassBuilder &PB) {
-  auto SanitizersCallback = [&](ModulePassManager &MPM,
-                                OptimizationLevel Level) {
+  auto SanitizersCallback = [&](ModulePassManager &MPM, OptimizationLevel Level,
+                                ThinOrFullLTOPhase) {
     if (CodeGenOpts.hasSanitizeCoverage()) {
       auto SancovOpts = getSancovOptsFromCGOpts(CodeGenOpts);
       MPM.addPass(SanitizerCoveragePass(
@@ -778,9 +778,10 @@ static void addSanitizers(const Triple &TargetTriple,
   };
   if (ClSanitizeOnOptimizerEarlyEP) {
     PB.registerOptimizerEarlyEPCallback(
-        [SanitizersCallback](ModulePassManager &MPM, OptimizationLevel Level) {
+        [SanitizersCallback](ModulePassManager &MPM, OptimizationLevel Level,
+                             ThinOrFullLTOPhase Phase) {
           ModulePassManager NewMPM;
-          SanitizersCallback(NewMPM, Level);
+          SanitizersCallback(NewMPM, Level, Phase);
           if (!NewMPM.isEmpty()) {
             // Sanitizers can abandon<GlobalsAA>.
             NewMPM.addPass(RequireAnalysisPass<GlobalsAA, llvm::Module>());
@@ -993,7 +994,8 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
                   createModuleToFunctionPassAdaptor(ObjCARCExpandPass()));
           });
       PB.registerPipelineEarlySimplificationEPCallback(
-          [](ModulePassManager &MPM, OptimizationLevel Level) {
+          [](ModulePassManager &MPM, OptimizationLevel Level,
+             ThinOrFullLTOPhase) {
             if (Level != OptimizationLevel::O0)
               MPM.addPass(ObjCARCAPElimPass());
           });
@@ -1057,11 +1059,12 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
     // TODO: Consider passing the MemoryProfileOutput to the pass builder via
     // the PGOOptions, and set this up there.
     if (!CodeGenOpts.MemoryProfileOutput.empty()) {
-      PB.registerOptimizerLastEPCallback(
-          [](ModulePassManager &MPM, OptimizationLevel Level) {
-            MPM.addPass(createModuleToFunctionPassAdaptor(MemProfilerPass()));
-            MPM.addPass(ModuleMemProfilerPass());
-          });
+      PB.registerOptimizerLastEPCallback([](ModulePassManager &MPM,
+                                            OptimizationLevel Level,
+                                            ThinOrFullLTOPhase) {
+        MPM.addPass(createModuleToFunctionPassAdaptor(MemProfilerPass()));
+        MPM.addPass(ModuleMemProfilerPass());
+      });
     }
 
     if (CodeGenOpts.FatLTO) {

diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -16994,6 +16994,58 @@ Value *CodeGenFunction::EmitX86BuiltinExpr(unsigned BuiltinID,
     // instruction, but it will create a memset that won't be optimized away.
     return Builder.CreateMemSet(Ops[0], Ops[1], Ops[2], Align(1), true);
   }
+  // Corresponding to intrisics which will return 2 tiles (tile0_tile1).
+  case X86::BI__builtin_ia32_t2rpntlvwz0_internal:
+  case X86::BI__builtin_ia32_t2rpntlvwz0t1_internal:
+  case X86::BI__builtin_ia32_t2rpntlvwz1_internal:
+  case X86::BI__builtin_ia32_t2rpntlvwz1t1_internal: {
+    Intrinsic::ID IID;
+    switch (BuiltinID) {
+    default:
+      llvm_unreachable("Unsupported intrinsic!");
+    case X86::BI__builtin_ia32_t2rpntlvwz0_internal:
+      IID = Intrinsic::x86_t2rpntlvwz0_internal;
+      break;
+    case X86::BI__builtin_ia32_t2rpntlvwz0t1_internal:
+      IID = Intrinsic::x86_t2rpntlvwz0t1_internal;
+      break;
+    case X86::BI__builtin_ia32_t2rpntlvwz1_internal:
+      IID = Intrinsic::x86_t2rpntlvwz1_internal;
+      break;
+    case X86::BI__builtin_ia32_t2rpntlvwz1t1_internal:
+      IID = Intrinsic::x86_t2rpntlvwz1t1_internal;
+      break;
+    }
+
+    // Ops = (Row0, Col0, Col1, DstPtr0, DstPtr1, SrcPtr, Stride)
+    Value *Call = Builder.CreateCall(CGM.getIntrinsic(IID),
+                                     {Ops[0], Ops[1], Ops[2], Ops[5], Ops[6]});
+
+    auto *PtrTy = E->getArg(3)->getType()->getAs<PointerType>();
+    assert(PtrTy && "arg3 must be of pointer type");
+    QualType PtreeTy = PtrTy->getPointeeType();
+    llvm::Type *TyPtee = ConvertType(PtreeTy);
+
+    // Bitcast amx type (x86_amx) to vector type (256 x i32)
+    // Then store tile0 into DstPtr0
+    Value *T0 = Builder.CreateExtractValue(Call, 0);
+    Value *VecT0 = Builder.CreateIntrinsic(Intrinsic::x86_cast_tile_to_vector,
+                                           {TyPtee}, {T0});
+    Builder.CreateDefaultAlignedStore(VecT0, Ops[3]);
+
+    // Then store tile1 into DstPtr1
+    Value *T1 = Builder.CreateExtractValue(Call, 1);
+    Value *VecT1 = Builder.CreateIntrinsic(Intrinsic::x86_cast_tile_to_vector,
+                                           {TyPtee}, {T1});
+    Value *Store = Builder.CreateDefaultAlignedStore(VecT1, Ops[4]);
+
+    // Note: Here we escape directly use x86_tilestored64_internal to store
+    // the results due to it can't make sure the Mem written scope. This may
+    // cause shapes reloads after first amx intrinsic, which current amx reg-
+    // ister allocation has no ability to handle it.
+
+    return Store;
+  }
   case X86::BI__ud2:
     // llvm.trap makes a ud2a instruction on x86.
     return EmitTrapCall(Intrinsic::trap);
@@ -18803,7 +18855,17 @@ Value *CodeGenFunction::EmitHLSLBuiltinExpr(unsigned BuiltinID,
         /*ReturnType=*/T0->getScalarType(),
         getDotProductIntrinsic(CGM.getHLSLRuntime(), VecTy0->getElementType()),
         ArrayRef<Value *>{Op0, Op1}, nullptr, "hlsl.dot");
-  } break;
+  }
+  case Builtin::BI__builtin_hlsl_dot4add_i8packed: {
+    Value *A = EmitScalarExpr(E->getArg(0));
+    Value *B = EmitScalarExpr(E->getArg(1));
+    Value *C = EmitScalarExpr(E->getArg(2));
+
+    Intrinsic::ID ID = CGM.getHLSLRuntime().getDot4AddI8PackedIntrinsic();
+    return Builder.CreateIntrinsic(
+        /*ReturnType=*/C->getType(), ID, ArrayRef<Value *>{A, B, C}, nullptr,
+        "hlsl.dot4add.i8packed");
+  }
   case Builtin::BI__builtin_hlsl_lerp: {
     Value *X = EmitScalarExpr(E->getArg(0));
     Value *Y = EmitScalarExpr(E->getArg(1));

diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
@@ -1941,6 +1941,10 @@ bool CodeGenFunction::EmitScalarRangeCheck(llvm::Value *Value, QualType Ty,
       cast<llvm::IntegerType>(Value->getType())->getBitWidth() == 1)
     return false;
 
+  if (NeedsEnumCheck &&
+      getContext().isTypeIgnoredBySanitizer(SanitizerKind::Enum, Ty))
+    return false;
+
   llvm::APInt Min, End;
   if (!getRangeForType(*this, Ty, Min, End, /*StrictEnums=*/true, IsBool))
     return true;

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -197,6 +197,18 @@ static bool CanElideOverflowCheck(const ASTContext &Ctx, const BinOpInfo &Op) {
   if (!Op.mayHaveIntegerOverflow())
     return true;
 
+  if (Op.Ty->isSignedIntegerType() &&
+      Ctx.isTypeIgnoredBySanitizer(SanitizerKind::SignedIntegerOverflow,
+                                   Op.Ty)) {
+    return true;
+  }
+
+  if (Op.Ty->isUnsignedIntegerType() &&
+      Ctx.isTypeIgnoredBySanitizer(SanitizerKind::UnsignedIntegerOverflow,
+                                   Op.Ty)) {
+    return true;
+  }
+
   const UnaryOperator *UO = dyn_cast<UnaryOperator>(Op.E);
 
   if (UO && UO->getOpcode() == UO_Minus &&
@@ -1125,6 +1137,10 @@ void ScalarExprEmitter::EmitIntegerTruncationCheck(Value *Src, QualType SrcType,
   if (!CGF.SanOpts.has(Check.second.second))
     return;
 
+  // Does some SSCL ignore this type?
+  if (CGF.getContext().isTypeIgnoredBySanitizer(Check.second.second, DstType))
+    return;
+
   llvm::Constant *StaticArgs[] = {
       CGF.EmitCheckSourceLocation(Loc), CGF.EmitCheckTypeDescriptor(SrcType),
       CGF.EmitCheckTypeDescriptor(DstType),
@@ -1235,6 +1251,15 @@ void ScalarExprEmitter::EmitIntegerSignChangeCheck(Value *Src, QualType SrcType,
     // Because here sign change check is interchangeable with truncation check.
     return;
   }
+  // Does an SSCL have an entry for the DstType under its respective sanitizer
+  // section?
+  if (DstSigned && CGF.getContext().isTypeIgnoredBySanitizer(
+                       SanitizerKind::ImplicitSignedIntegerTruncation, DstType))
+    return;
+  if (!DstSigned &&
+      CGF.getContext().isTypeIgnoredBySanitizer(
+          SanitizerKind::ImplicitUnsignedIntegerTruncation, DstType))
+    return;
   // That's it. We can't rule out any more cases with the data we have.
 
   CodeGenFunction::SanitizerScope SanScope(&CGF);
@@ -2784,10 +2809,11 @@ llvm::Value *ScalarExprEmitter::EmitIncDecConsiderOverflowBehavior(
       return Builder.CreateNSWAdd(InVal, Amount, Name);
     [[fallthrough]];
   case LangOptions::SOB_Trapping:
-    if (!E->canOverflow())
+    BinOpInfo Info = createBinOpInfoFromIncDec(
+        E, InVal, IsInc, E->getFPFeaturesInEffect(CGF.getLangOpts()));
+    if (!E->canOverflow() || CanElideOverflowCheck(CGF.getContext(), Info))
       return Builder.CreateNSWAdd(InVal, Amount, Name);
-    return EmitOverflowCheckedBinOp(createBinOpInfoFromIncDec(
-        E, InVal, IsInc, E->getFPFeaturesInEffect(CGF.getLangOpts())));
+    return EmitOverflowCheckedBinOp(Info);
   }
   llvm_unreachable("Unknown SignedOverflowBehaviorTy");
 }
@@ -2990,7 +3016,9 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
       value = EmitIncDecConsiderOverflowBehavior(E, value, isInc);
     } else if (E->canOverflow() && type->isUnsignedIntegerType() &&
                CGF.SanOpts.has(SanitizerKind::UnsignedIntegerOverflow) &&
-               !excludeOverflowPattern) {
+               !excludeOverflowPattern &&
+               !CGF.getContext().isTypeIgnoredBySanitizer(
+                   SanitizerKind::UnsignedIntegerOverflow, E->getType())) {
       value = EmitOverflowCheckedBinOp(createBinOpInfoFromIncDec(
           E, value, isInc, E->getFPFeaturesInEffect(CGF.getLangOpts())));
     } else {

diff --git a/clang/lib/CodeGen/CGHLSLRuntime.h b/clang/lib/CodeGen/CGHLSLRuntime.h
@@ -89,6 +89,7 @@ class CGHLSLRuntime {
   GENERATE_HLSL_INTRINSIC_FUNCTION(FDot, fdot)
   GENERATE_HLSL_INTRINSIC_FUNCTION(SDot, sdot)
   GENERATE_HLSL_INTRINSIC_FUNCTION(UDot, udot)
+  GENERATE_HLSL_INTRINSIC_FUNCTION(Dot4AddI8Packed, dot4add_i8packed)
   GENERATE_HLSL_INTRINSIC_FUNCTION(WaveIsFirstLane, wave_is_first_lane)
   GENERATE_HLSL_INTRINSIC_FUNCTION(WaveReadLaneAt, wave_readlane)
 

diff --git a/clang/lib/CodeGen/CGObjCMac.cpp b/clang/lib/CodeGen/CGObjCMac.cpp
@@ -365,7 +365,7 @@ class ObjCCommonTypesHelper {
   /// GcReadWeakFn -- LLVM objc_read_weak (id *src) function.
   llvm::FunctionCallee getGcReadWeakFn() {
     // id objc_read_weak (id *)
-    llvm::Type *args[] = { ObjectPtrTy->getPointerTo() };
+    llvm::Type *args[] = {CGM.UnqualPtrTy};
     llvm::FunctionType *FTy =
       llvm::FunctionType::get(ObjectPtrTy, args, false);
     return CGM.CreateRuntimeFunction(FTy, "objc_read_weak");
@@ -374,7 +374,7 @@ class ObjCCommonTypesHelper {
   /// GcAssignWeakFn -- LLVM objc_assign_weak function.
   llvm::FunctionCallee getGcAssignWeakFn() {
     // id objc_assign_weak (id, id *)
-    llvm::Type *args[] = { ObjectPtrTy, ObjectPtrTy->getPointerTo() };
+    llvm::Type *args[] = {ObjectPtrTy, CGM.UnqualPtrTy};
     llvm::FunctionType *FTy =
       llvm::FunctionType::get(ObjectPtrTy, args, false);
     return CGM.CreateRuntimeFunction(FTy, "objc_assign_weak");
@@ -383,7 +383,7 @@ class ObjCCommonTypesHelper {
   /// GcAssignGlobalFn -- LLVM objc_assign_global function.
   llvm::FunctionCallee getGcAssignGlobalFn() {
     // id objc_assign_global(id, id *)
-    llvm::Type *args[] = { ObjectPtrTy, ObjectPtrTy->getPointerTo() };
+    llvm::Type *args[] = {ObjectPtrTy, CGM.UnqualPtrTy};
     llvm::FunctionType *FTy =
       llvm::FunctionType::get(ObjectPtrTy, args, false);
     return CGM.CreateRuntimeFunction(FTy, "objc_assign_global");
@@ -392,7 +392,7 @@ class ObjCCommonTypesHelper {
   /// GcAssignThreadLocalFn -- LLVM objc_assign_threadlocal function.
   llvm::FunctionCallee getGcAssignThreadLocalFn() {
     // id objc_assign_threadlocal(id src, id * dest)
-    llvm::Type *args[] = { ObjectPtrTy, ObjectPtrTy->getPointerTo() };
+    llvm::Type *args[] = {ObjectPtrTy, CGM.UnqualPtrTy};
     llvm::FunctionType *FTy =
       llvm::FunctionType::get(ObjectPtrTy, args, false);
     return CGM.CreateRuntimeFunction(FTy, "objc_assign_threadlocal");
@@ -401,8 +401,7 @@ class ObjCCommonTypesHelper {
   /// GcAssignIvarFn -- LLVM objc_assign_ivar function.
   llvm::FunctionCallee getGcAssignIvarFn() {
     // id objc_assign_ivar(id, id *, ptrdiff_t)
-    llvm::Type *args[] = { ObjectPtrTy, ObjectPtrTy->getPointerTo(),
-                           CGM.PtrDiffTy };
+    llvm::Type *args[] = {ObjectPtrTy, CGM.UnqualPtrTy, CGM.PtrDiffTy};
     llvm::FunctionType *FTy =
       llvm::FunctionType::get(ObjectPtrTy, args, false);
     return CGM.CreateRuntimeFunction(FTy, "objc_assign_ivar");
@@ -419,7 +418,7 @@ class ObjCCommonTypesHelper {
   /// GcAssignStrongCastFn -- LLVM objc_assign_strongCast function.
   llvm::FunctionCallee getGcAssignStrongCastFn() {
     // id objc_assign_strongCast(id, id *)
-    llvm::Type *args[] = { ObjectPtrTy, ObjectPtrTy->getPointerTo() };
+    llvm::Type *args[] = {ObjectPtrTy, CGM.UnqualPtrTy};
     llvm::FunctionType *FTy =
       llvm::FunctionType::get(ObjectPtrTy, args, false);
     return CGM.CreateRuntimeFunction(FTy, "objc_assign_strongCast");
@@ -554,23 +553,23 @@ class ObjCTypesHelper : public ObjCCommonTypesHelper {
 
   /// ExceptionTryEnterFn - LLVM objc_exception_try_enter function.
   llvm::FunctionCallee getExceptionTryEnterFn() {
-    llvm::Type *params[] = { ExceptionDataTy->getPointerTo() };
+    llvm::Type *params[] = {CGM.UnqualPtrTy};
     return CGM.CreateRuntimeFunction(
       llvm::FunctionType::get(CGM.VoidTy, params, false),
       "objc_exception_try_enter");
   }
 
   /// ExceptionTryExitFn - LLVM objc_exception_try_exit function.
   llvm::FunctionCallee getExceptionTryExitFn() {
-    llvm::Type *params[] = { ExceptionDataTy->getPointerTo() };
+    llvm::Type *params[] = {CGM.UnqualPtrTy};
     return CGM.CreateRuntimeFunction(
       llvm::FunctionType::get(CGM.VoidTy, params, false),
       "objc_exception_try_exit");
   }
 
   /// ExceptionExtractFn - LLVM objc_exception_extract function.
   llvm::FunctionCallee getExceptionExtractFn() {
-    llvm::Type *params[] = { ExceptionDataTy->getPointerTo() };
+    llvm::Type *params[] = {CGM.UnqualPtrTy};
     return CGM.CreateRuntimeFunction(llvm::FunctionType::get(ObjectPtrTy,
                                                              params, false),
                                      "objc_exception_extract");
@@ -587,7 +586,7 @@ class ObjCTypesHelper : public ObjCCommonTypesHelper {
   /// SetJmpFn - LLVM _setjmp function.
   llvm::FunctionCallee getSetJmpFn() {
     // This is specifically the prototype for x86.
-    llvm::Type *params[] = { CGM.Int32Ty->getPointerTo() };
+    llvm::Type *params[] = {CGM.UnqualPtrTy};
     return CGM.CreateRuntimeFunction(
         llvm::FunctionType::get(CGM.Int32Ty, params, false), "_setjmp",
         llvm::AttributeList::get(CGM.getLLVMContext(),
@@ -6051,9 +6050,7 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul
       Int8PtrTy, PropertyListPtrTy);
 
   // ImpnfABITy - LLVM for id (*)(id, SEL, ...)
-  llvm::Type *params[] = { ObjectPtrTy, SelectorPtrTy };
-  ImpnfABITy = llvm::FunctionType::get(ObjectPtrTy, params, false)
-                 ->getPointerTo();
+  ImpnfABITy = CGM.UnqualPtrTy;
 
   // struct _class_t {
   //   struct _class_t *isa;
@@ -6469,8 +6466,7 @@ void CGObjCNonFragileABIMac::GenerateClass(const ObjCImplementationDecl *ID) {
                                    llvm::GlobalValue::ExternalLinkage, nullptr,
                                    "_objc_empty_vtable");
     else
-      ObjCEmptyVtableVar =
-        llvm::ConstantPointerNull::get(ObjCTypes.ImpnfABITy->getPointerTo());
+      ObjCEmptyVtableVar = llvm::ConstantPointerNull::get(CGM.UnqualPtrTy);
   }
 
   // FIXME: Is this correct (that meta class size is never computed)?

diff --git a/clang/lib/CodeGen/Targets/SPIR.cpp b/clang/lib/CodeGen/Targets/SPIR.cpp
@@ -52,6 +52,10 @@ class CommonSPIRTargetCodeGenInfo : public TargetCodeGenInfo {
 
   unsigned getOpenCLKernelCallingConv() const override;
   llvm::Type *getOpenCLType(CodeGenModule &CGM, const Type *T) const override;
+  llvm::Type *getHLSLType(CodeGenModule &CGM, const Type *Ty) const override;
+  llvm::Type *getSPIRVImageTypeFromHLSLResource(
+      const HLSLAttributedResourceType::Attributes &attributes,
+      llvm::Type *ElementType, llvm::LLVMContext &Ctx) const;
 };
 class SPIRVTargetCodeGenInfo : public CommonSPIRTargetCodeGenInfo {
 public:
@@ -323,6 +327,81 @@ llvm::Type *CommonSPIRTargetCodeGenInfo::getOpenCLType(CodeGenModule &CGM,
   return nullptr;
 }
 
+llvm::Type *CommonSPIRTargetCodeGenInfo::getHLSLType(CodeGenModule &CGM,
+                                                     const Type *Ty) const {
+  auto *ResType = dyn_cast<HLSLAttributedResourceType>(Ty);
+  if (!ResType)
+    return nullptr;
+
+  llvm::LLVMContext &Ctx = CGM.getLLVMContext();
+  const HLSLAttributedResourceType::Attributes &ResAttrs = ResType->getAttrs();
+  switch (ResAttrs.ResourceClass) {
+  case llvm::dxil::ResourceClass::UAV:
+  case llvm::dxil::ResourceClass::SRV: {
+    // TypedBuffer and RawBuffer both need element type
+    QualType ContainedTy = ResType->getContainedType();
+    if (ContainedTy.isNull())
+      return nullptr;
+
+    assert(!ResAttrs.RawBuffer &&
+           "Raw buffers handles are not implemented for SPIR-V yet");
+    assert(!ResAttrs.IsROV &&
+           "Rasterizer order views not implemented for SPIR-V yet");
+
+    // convert element type
+    llvm::Type *ElemType = CGM.getTypes().ConvertType(ContainedTy);
+    return getSPIRVImageTypeFromHLSLResource(ResAttrs, ElemType, Ctx);
+  }
+  case llvm::dxil::ResourceClass::CBuffer:
+    llvm_unreachable("CBuffer handles are not implemented for SPIR-V yet");
+    break;
+  case llvm::dxil::ResourceClass::Sampler:
+    return llvm::TargetExtType::get(Ctx, "spirv.Sampler");
+  }
+  return nullptr;
+}
+
+llvm::Type *CommonSPIRTargetCodeGenInfo::getSPIRVImageTypeFromHLSLResource(
+    const HLSLAttributedResourceType::Attributes &attributes,
+    llvm::Type *ElementType, llvm::LLVMContext &Ctx) const {
+
+  if (ElementType->isVectorTy())
+    ElementType = ElementType->getScalarType();
+
+  assert((ElementType->isIntegerTy() || ElementType->isFloatingPointTy()) &&
+         "The element type for a SPIR-V resource must be a scalar integer or "
+         "floating point type.");
+
+  // These parameters correspond to the operands to the OpTypeImage SPIR-V
+  // instruction. See
+  // https://registry.khronos.org/SPIR-V/specs/unified1/SPIRV.html#OpTypeImage.
+  SmallVector<unsigned, 6> IntParams(6, 0);
+
+  // Dim
+  // For now we assume everything is a buffer.
+  IntParams[0] = 5;
+
+  // Depth
+  // HLSL does not indicate if it is a depth texture or not, so we use unknown.
+  IntParams[1] = 2;
+
+  // Arrayed
+  IntParams[2] = 0;
+
+  // MS
+  IntParams[3] = 0;
+
+  // Sampled
+  IntParams[4] =
+      attributes.ResourceClass == llvm::dxil::ResourceClass::UAV ? 2 : 1;
+
+  // Image format.
+  // Setting to unknown for now.
+  IntParams[5] = 0;
+
+  return llvm::TargetExtType::get(Ctx, "spirv.Image", {ElementType}, IntParams);
+}
+
 std::unique_ptr<TargetCodeGenInfo>
 CodeGen::createCommonSPIRTargetCodeGenInfo(CodeGenModule &CGM) {
   return std::make_unique<CommonSPIRTargetCodeGenInfo>(CGM.getTypes());

diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
@@ -149,13 +149,9 @@ static std::optional<llvm::Triple>
 getHIPOffloadTargetTriple(const Driver &D, const ArgList &Args) {
   if (!Args.hasArg(options::OPT_offload_EQ)) {
     auto OffloadArchs = Args.getAllArgValues(options::OPT_offload_arch_EQ);
-    if (llvm::is_contained(OffloadArchs, "amdgcnspirv")) {
-      if (OffloadArchs.size() == 1)
-        return llvm::Triple("spirv64-amd-amdhsa");
-      // Mixing specific & SPIR-V compilation is not supported for now.
-      D.Diag(diag::err_drv_only_one_offload_target_supported);
-      return std::nullopt;
-    }
+    if (llvm::is_contained(OffloadArchs, "amdgcnspirv") &&
+        OffloadArchs.size() == 1)
+      return llvm::Triple("spirv64-amd-amdhsa");
     return llvm::Triple("amdgcn-amd-amdhsa"); // Default HIP triple.
   }
   auto TT = getOffloadTargetTriple(D, Args);
@@ -458,6 +454,7 @@ DerivedArgList *Driver::TranslateInputArgs(const InputArgList &Args) const {
     // some build systems. We don't try to be complete here because we don't
     // care to encourage this usage model.
     if (A->getOption().matches(options::OPT_Wp_COMMA) &&
+        A->getNumValues() > 0 &&
         (A->getValue(0) == StringRef("-MD") ||
          A->getValue(0) == StringRef("-MMD"))) {
       // Rewrite to -MD/-MMD along with -MF.
@@ -3477,9 +3474,11 @@ class OffloadingActionBuilder final {
       llvm::StringMap<bool> Features;
       // getHIPOffloadTargetTriple() is known to return valid value as it has
       // been called successfully in the CreateOffloadingDeviceToolChains().
-      auto ArchStr = parseTargetID(
-          *getHIPOffloadTargetTriple(C.getDriver(), C.getInputArgs()), IdStr,
-          &Features);
+      auto T =
+          (IdStr == "amdgcnspirv")
+              ? llvm::Triple("spirv64-amd-amdhsa")
+              : *getHIPOffloadTargetTriple(C.getDriver(), C.getInputArgs());
+      auto ArchStr = parseTargetID(T, IdStr, &Features);
       if (!ArchStr) {
         C.getDriver().Diag(clang::diag::err_drv_bad_target_id) << IdStr;
         C.setContainsError();
@@ -4035,7 +4034,7 @@ void Driver::handleArguments(Compilation &C, DerivedArgList &Args,
     if (C.getDefaultToolChain().getTriple().isWindowsMSVCEnvironment() &&
         LTOMode != LTOK_None &&
         !Args.getLastArgValue(options::OPT_fuse_ld_EQ)
-             .equals_insensitive("lld"))
+             .starts_with_insensitive("lld"))
       Diag(clang::diag::err_drv_lto_without_lld);
 
     // If -dumpdir is not specified, give a default prefix derived from the link
@@ -5755,7 +5754,7 @@ InputInfoList Driver::BuildJobsForActionNoCache(
     // We only have to generate a prefix for the host if this is not a top-level
     // action.
     std::string OffloadingPrefix = Action::GetOffloadingFileNamePrefix(
-        A->getOffloadingDeviceKind(), TC->getTriple().normalize(),
+        A->getOffloadingDeviceKind(), EffectiveTriple.normalize(),
         /*CreatePrefixForHost=*/isa<OffloadPackagerJobAction>(A) ||
             !(A->getOffloadingHostActiveKinds() == Action::OFK_None ||
               AtTopLevel));

diff --git a/clang/lib/Driver/ToolChain.cpp b/clang/lib/Driver/ToolChain.cpp
@@ -899,7 +899,9 @@ bool ToolChain::needsProfileRT(const ArgList &Args) {
          Args.hasArg(options::OPT_fprofile_instr_generate) ||
          Args.hasArg(options::OPT_fprofile_instr_generate_EQ) ||
          Args.hasArg(options::OPT_fcreate_profile) ||
-         Args.hasArg(options::OPT_forder_file_instrumentation);
+         Args.hasArg(options::OPT_forder_file_instrumentation) ||
+         Args.hasArg(options::OPT_fprofile_generate_cold_function_coverage) ||
+         Args.hasArg(options::OPT_fprofile_generate_cold_function_coverage_EQ);
 }
 
 bool ToolChain::needsGCovInstrumentation(const llvm::opt::ArgList &Args) {
@@ -1097,6 +1099,12 @@ std::string ToolChain::ComputeLLVMTriple(const ArgList &Args,
   }
   case llvm::Triple::aarch64_32:
     return getTripleString();
+  case llvm::Triple::amdgcn: {
+    llvm::Triple Triple = getTriple();
+    if (Args.getLastArgValue(options::OPT_mcpu_EQ) == "amdgcnspirv")
+      Triple.setArch(llvm::Triple::ArchType::spirv64);
+    return Triple.getTriple();
+  }
   case llvm::Triple::arm:
   case llvm::Triple::armeb:
   case llvm::Triple::thumb:

diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -632,6 +632,28 @@ static void addPGOAndCoverageFlags(const ToolChain &TC, Compilation &C,
     }
   }
 
+  if (auto *ColdFuncCoverageArg = Args.getLastArg(
+          options::OPT_fprofile_generate_cold_function_coverage,
+          options::OPT_fprofile_generate_cold_function_coverage_EQ)) {
+    SmallString<128> Path(
+        ColdFuncCoverageArg->getOption().matches(
+            options::OPT_fprofile_generate_cold_function_coverage_EQ)
+            ? ColdFuncCoverageArg->getValue()
+            : "");
+    llvm::sys::path::append(Path, "default_%m.profraw");
+    // FIXME: Idealy the file path should be passed through
+    // `-fprofile-instrument-path=`(InstrProfileOutput), however, this field is
+    // shared with other profile use path(see PGOOptions), we need to refactor
+    // PGOOptions to make it work.
+    CmdArgs.push_back("-mllvm");
+    CmdArgs.push_back(Args.MakeArgString(
+        Twine("--instrument-cold-function-only-path=") + Path));
+    CmdArgs.push_back("-mllvm");
+    CmdArgs.push_back("--pgo-instrument-cold-function-only");
+    CmdArgs.push_back("-mllvm");
+    CmdArgs.push_back("--pgo-function-entry-coverage");
+  }
+
   Arg *PGOGenArg = nullptr;
   if (PGOGenerateArg) {
     assert(!CSPGOGenerateArg);
@@ -1185,8 +1207,7 @@ void Clang::AddPreprocessingOptions(Compilation &C, const JobAction &JA,
 
   Args.addAllArgs(CmdArgs,
                   {options::OPT_D, options::OPT_U, options::OPT_I_Group,
-                   options::OPT_F, options::OPT_index_header_map,
-                   options::OPT_embed_dir_EQ});
+                   options::OPT_F, options::OPT_embed_dir_EQ});
 
   // Add -Wp, and -Xpreprocessor if using the preprocessor.
 

diff --git a/clang/lib/Driver/ToolChains/HIPAMD.cpp b/clang/lib/Driver/ToolChains/HIPAMD.cpp
@@ -205,7 +205,7 @@ void AMDGCN::Linker::ConstructJob(Compilation &C, const JobAction &JA,
   if (JA.getType() == types::TY_LLVM_BC)
     return constructLlvmLinkCommand(C, JA, Inputs, Output, Args);
 
-  if (getToolChain().getTriple().isSPIRV())
+  if (getToolChain().getEffectiveTriple().isSPIRV())
     return constructLinkAndEmitSpirvCommand(C, JA, Inputs, Output, Args);
 
   return constructLldCommand(C, JA, Inputs, Output, Args);
@@ -264,12 +264,14 @@ void HIPAMDToolChain::addClangTargetOptions(
     CC1Args.push_back("-fapply-global-visibility-to-externs");
   }
 
-  // For SPIR-V we embed the command-line into the generated binary, in order to
-  // retrieve it at JIT time and be able to do target specific compilation with
-  // options that match the user-supplied ones.
-  if (getTriple().isSPIRV() &&
-      !DriverArgs.hasArg(options::OPT_fembed_bitcode_marker))
-    CC1Args.push_back("-fembed-bitcode=marker");
+  if (getEffectiveTriple().isSPIRV()) {
+    // For SPIR-V we embed the command-line into the generated binary, in order
+    // to retrieve it at JIT time and be able to do target specific compilation
+    // with options that match the user-supplied ones.
+    if (!DriverArgs.hasArg(options::OPT_fembed_bitcode_marker))
+      CC1Args.push_back("-fembed-bitcode=marker");
+    return; // No DeviceLibs for SPIR-V.
+  }
 
   for (auto BCFile : getDeviceLibs(DriverArgs)) {
     CC1Args.push_back(BCFile.ShouldInternalize ? "-mlink-builtin-bitcode"
@@ -361,8 +363,7 @@ llvm::SmallVector<ToolChain::BitCodeLibraryInfo, 12>
 HIPAMDToolChain::getDeviceLibs(const llvm::opt::ArgList &DriverArgs) const {
   llvm::SmallVector<BitCodeLibraryInfo, 12> BCLibs;
   if (DriverArgs.hasArg(options::OPT_nogpulib) ||
-      (getTriple().getArch() == llvm::Triple::spirv64 &&
-       getTriple().getVendor() == llvm::Triple::AMD))
+      getGPUArch(DriverArgs) == "amdgcnspirv")
     return {};
   ArgStringList LibraryPaths;
 
@@ -437,8 +438,8 @@ HIPAMDToolChain::getDeviceLibs(const llvm::opt::ArgList &DriverArgs) const {
 void HIPAMDToolChain::checkTargetID(
     const llvm::opt::ArgList &DriverArgs) const {
   auto PTID = getParsedTargetID(DriverArgs);
-  if (PTID.OptionalTargetID && !PTID.OptionalGPUArch) {
+  if (PTID.OptionalTargetID && !PTID.OptionalGPUArch &&
+      PTID.OptionalTargetID != "amdgcnspirv")
     getDriver().Diag(clang::diag::err_drv_bad_target_id)
         << *PTID.OptionalTargetID;
-  }
 }
diff --git a/clang/lib/Driver/ToolChains/HIPUtility.cpp b/clang/lib/Driver/ToolChains/HIPUtility.cpp
@@ -304,10 +304,14 @@ void HIP::constructHIPFatbinCommand(Compilation &C, const JobAction &JA,
   for (const auto &II : Inputs) {
     const auto *A = II.getAction();
     auto ArchStr = llvm::StringRef(A->getOffloadingArch());
-    BundlerTargetArg +=
-        "," + OffloadKind + "-" + normalizeForBundler(TT, !ArchStr.empty());
+    BundlerTargetArg += ',' + OffloadKind + '-';
+    if (ArchStr == "amdgcnspirv")
+      BundlerTargetArg +=
+          normalizeForBundler(llvm::Triple("spirv64-amd-amdhsa"), true);
+    else
+      BundlerTargetArg += normalizeForBundler(TT, !ArchStr.empty());
     if (!ArchStr.empty())
-      BundlerTargetArg += "-" + ArchStr.str();
+      BundlerTargetArg += '-' + ArchStr.str();
   }
   BundlerArgs.push_back(Args.MakeArgString(BundlerTargetArg));
 

diff --git a/clang/lib/Driver/ToolChains/PS4CPU.cpp b/clang/lib/Driver/ToolChains/PS4CPU.cpp
@@ -248,8 +248,9 @@ void tools::PS5cpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
       Args.MakeArgString("--sysroot=" + TC.getSDKLibraryRootDir()));
 
   // Default to PIE for non-static executables.
-  const bool PIE = !Relocatable && !Shared && !Static;
-  if (Args.hasFlag(options::OPT_pie, options::OPT_no_pie, PIE))
+  const bool PIE = Args.hasFlag(options::OPT_pie, options::OPT_no_pie,
+                                !Relocatable && !Shared && !Static);
+  if (PIE)
     CmdArgs.push_back("-pie");
 
   if (!Relocatable) {
@@ -276,6 +277,16 @@ void tools::PS5cpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
     CmdArgs.push_back("-z");
     CmdArgs.push_back("start-stop-visibility=hidden");
 
+    // DT_DEBUG is not supported on PlayStation.
+    CmdArgs.push_back("-z");
+    CmdArgs.push_back("rodynamic");
+
+    CmdArgs.push_back("-z");
+    CmdArgs.push_back("common-page-size=0x4000");
+
+    CmdArgs.push_back("-z");
+    CmdArgs.push_back("max-page-size=0x4000");
+
     // Patch relocated regions of DWARF whose targets are eliminated at link
     // time with specific tombstones, such that they're recognisable by the
     // PlayStation debugger.
@@ -286,6 +297,16 @@ void tools::PS5cpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
         "dead-reloc-in-nonalloc=.debug_ranges=0xfffffffffffffffe");
     CmdArgs.push_back("-z");
     CmdArgs.push_back("dead-reloc-in-nonalloc=.debug_loc=0xfffffffffffffffe");
+
+    // The PlayStation loader expects linked objects to be laid out in a
+    // particular way. This is achieved by linker scripts that are supplied
+    // with the SDK. The scripts are inside <sdkroot>/target/lib, which is
+    // added as a search path elsewhere.
+    // "PRX" has long stood for "PlayStation Relocatable eXecutable".
+    CmdArgs.push_back("--default-script");
+    CmdArgs.push_back(Static   ? "static.script"
+                      : Shared ? "prx.script"
+                               : "main.script");
   }
 
   if (Static)
@@ -295,6 +316,11 @@ void tools::PS5cpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
   if (Shared)
     CmdArgs.push_back("--shared");
 
+  // Provide a base address for non-PIE executables. This includes cases where
+  // -static is supplied without -pie.
+  if (!Relocatable && !Shared && !PIE)
+    CmdArgs.push_back("--image-base=0x400000");
+
   assert((Output.isFilename() || Output.isNothing()) && "Invalid output.");
   if (Output.isFilename()) {
     CmdArgs.push_back("-o");

diff --git a/clang/lib/Format/TokenAnnotator.cpp b/clang/lib/Format/TokenAnnotator.cpp
@@ -1259,7 +1259,7 @@ class AnnotatingParser {
 
   bool parseConditional() {
     while (CurrentToken) {
-      if (CurrentToken->is(tok::colon)) {
+      if (CurrentToken->is(tok::colon) && CurrentToken->is(TT_Unknown)) {
         CurrentToken->setType(TT_ConditionalExpr);
         next();
         return true;

diff --git a/clang/lib/Format/UnwrappedLineParser.cpp b/clang/lib/Format/UnwrappedLineParser.cpp
@@ -2091,7 +2091,8 @@ void UnwrappedLineParser::parseStructuralElement(
     case tok::kw_switch:
       if (Style.Language == FormatStyle::LK_Java)
         parseSwitch(/*IsExpr=*/true);
-      nextToken();
+      else
+        nextToken();
       break;
     case tok::kw_case:
       // Proto: there are no switch/case statements.
@@ -2656,7 +2657,10 @@ bool UnwrappedLineParser::parseParens(TokenType AmpAmpTokenType) {
         nextToken();
       break;
     case tok::kw_switch:
-      parseSwitch(/*IsExpr=*/true);
+      if (Style.Language == FormatStyle::LK_Java)
+        parseSwitch(/*IsExpr=*/true);
+      else
+        nextToken();
       break;
     case tok::kw_requires: {
       auto RequiresToken = FormatTok;

diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -3190,24 +3190,17 @@ static void GenerateHeaderSearchArgs(const HeaderSearchOptions &Opts,
   auto It = Opts.UserEntries.begin();
   auto End = Opts.UserEntries.end();
 
-  // Add -I..., -F..., and -index-header-map options in order.
-  for (; It < End && Matches(*It, {frontend::IndexHeaderMap, frontend::Angled},
-                             std::nullopt, true);
+  // Add -I... and -F... options in order.
+  for (; It < End && Matches(*It, {frontend::Angled}, std::nullopt, true);
        ++It) {
     OptSpecifier Opt = [It, Matches]() {
-      if (Matches(*It, frontend::IndexHeaderMap, true, true))
-        return OPT_F;
-      if (Matches(*It, frontend::IndexHeaderMap, false, true))
-        return OPT_I;
       if (Matches(*It, frontend::Angled, true, true))
         return OPT_F;
       if (Matches(*It, frontend::Angled, false, true))
         return OPT_I;
       llvm_unreachable("Unexpected HeaderSearchOptions::Entry.");
     }();
 
-    if (It->Group == frontend::IndexHeaderMap)
-      GenerateArg(Consumer, OPT_index_header_map);
     GenerateArg(Consumer, Opt, It->Path);
   };
 
@@ -3319,8 +3312,7 @@ static bool ParseHeaderSearchArgs(HeaderSearchOptions &Opts, ArgList &Args,
         llvm::CachedHashString(MacroDef.split('=').first));
   }
 
-  // Add -I..., -F..., and -index-header-map options in order.
-  bool IsIndexHeaderMap = false;
+  // Add -I... and -F... options in order.
   bool IsSysrootSpecified =
       Args.hasArg(OPT__sysroot_EQ) || Args.hasArg(OPT_isysroot);
 
@@ -3339,20 +3331,10 @@ static bool ParseHeaderSearchArgs(HeaderSearchOptions &Opts, ArgList &Args,
     return A->getValue();
   };
 
-  for (const auto *A : Args.filtered(OPT_I, OPT_F, OPT_index_header_map)) {
-    if (A->getOption().matches(OPT_index_header_map)) {
-      // -index-header-map applies to the next -I or -F.
-      IsIndexHeaderMap = true;
-      continue;
-    }
-
-    frontend::IncludeDirGroup Group =
-        IsIndexHeaderMap ? frontend::IndexHeaderMap : frontend::Angled;
-
+  for (const auto *A : Args.filtered(OPT_I, OPT_F)) {
     bool IsFramework = A->getOption().matches(OPT_F);
-    Opts.AddPath(PrefixHeaderPath(A, IsFramework), Group, IsFramework,
-                 /*IgnoreSysroot*/ true);
-    IsIndexHeaderMap = false;
+    Opts.AddPath(PrefixHeaderPath(A, IsFramework), frontend::Angled,
+                 IsFramework, /*IgnoreSysroot=*/true);
   }
 
   // Add -iprefix/-iwithprefix/-iwithprefixbefore options.

diff --git a/clang/lib/Headers/CMakeLists.txt b/clang/lib/Headers/CMakeLists.txt
@@ -148,8 +148,9 @@ set(x86_files
   ammintrin.h
   amxcomplexintrin.h
   amxfp16intrin.h
-  amxintrin.h
   amxfp8intrin.h
+  amxintrin.h
+  amxtransposeintrin.h
   avx10_2_512bf16intrin.h
   avx10_2_512convertintrin.h
   avx10_2_512minmaxintrin.h

diff --git a/clang/lib/Headers/amxintrin.h b/clang/lib/Headers/amxintrin.h
@@ -232,6 +232,8 @@ static __inline__ void __DEFAULT_FN_ATTRS_TILE _tile_release(void) {
 /// bytes. Since there is no 2D type in llvm IR, we use vector type to
 /// represent 2D tile and the fixed size is maximum amx tile register size.
 typedef int _tile1024i __attribute__((__vector_size__(1024), __aligned__(64)));
+typedef int _tile1024i_1024a
+    __attribute__((__vector_size__(1024), __aligned__(1024)));
 
 /// This is internal intrinsic. C/C++ user should avoid calling it directly.
 static __inline__ _tile1024i __DEFAULT_FN_ATTRS_INT8

diff --git a/clang/lib/Headers/amxtransposeintrin.h b/clang/lib/Headers/amxtransposeintrin.h
@@ -0,0 +1,248 @@
+/* ===--- amxtransposeintrin.h - AMX_TRANSPOSE intrinsics -*- C++ -*---------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ * ===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error "Never use <amxtransposeintrin.h> directly; use <immintrin.h> instead."
+#endif /* __IMMINTRIN_H */
+
+#ifndef __AMX_TRANSPOSEINTRIN_H
+#define __AMX_TRANSPOSEINTRIN_H
+#ifdef __x86_64__
+
+#define __DEFAULT_FN_ATTRS_TRANSPOSE                                           \
+  __attribute__((__always_inline__, __nodebug__, __target__("amx-transpose")))
+
+#define _tile_2rpntlvwz0(tdst, base, stride)                                   \
+  __builtin_ia32_t2rpntlvwz0(tdst, base, stride)
+#define _tile_2rpntlvwz0t1(tdst, base, stride)                                 \
+  __builtin_ia32_t2rpntlvwz0t1(tdst, base, stride)
+#define _tile_2rpntlvwz1(tdst, base, stride)                                   \
+  __builtin_ia32_t2rpntlvwz1(tdst, base, stride)
+#define _tile_2rpntlvwz1t1(tdst, base, stride)                                 \
+  __builtin_ia32_t2rpntlvwz1t1(tdst, base, stride)
+
+/// Transpose 32-bit elements from \a src and write the result to \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// void _tile_transposed(__tile dst, __tile src);
+/// \endcode
+///
+/// This intrinsic corresponds to the <c> TTRANSPOSED </c> instruction.
+///
+/// \param dst
+/// 	The destination tile. Max size is 1024 Bytes.
+/// \param src
+/// 	The source tile. Max size is 1024 Bytes.
+///
+/// \code{.operation}
+///
+/// FOR i := 0 TO (dst.rows-1)
+/// 	tmp[511:0] := 0
+/// 	FOR j := 0 TO (dst.colsb/4-1)
+/// 		tmp.dword[j] := src.row[j].dword[i]
+/// 	ENDFOR
+/// 	dst.row[i] := tmp
+/// ENDFOR
+///
+/// zero_upper_rows(dst, dst.rows)
+/// zero_tileconfig_start()
+/// \endcode
+#define _tile_transposed(dst, src) __builtin_ia32_ttransposed(dst, src)
+
+static __inline__ void __DEFAULT_FN_ATTRS_TRANSPOSE _tile_2rpntlvwz0_internal(
+    unsigned short row, unsigned short col0, unsigned short col1,
+    _tile1024i *dst0, _tile1024i *dst1, const void *base,
+    __SIZE_TYPE__ stride) {
+  // Use __tile1024i_1024a* to escape the alignment check in
+  // clang/test/Headers/x86-intrinsics-headers-clean.cpp
+  __builtin_ia32_t2rpntlvwz0_internal(row, col0, col1, (_tile1024i_1024a *)dst0,
+                                      (_tile1024i_1024a *)dst1, base,
+                                      (__SIZE_TYPE__)(stride));
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS_TRANSPOSE _tile_2rpntlvwz0t1_internal(
+    unsigned short row, unsigned short col0, unsigned short col1,
+    _tile1024i *dst0, _tile1024i *dst1, const void *base,
+    __SIZE_TYPE__ stride) {
+  __builtin_ia32_t2rpntlvwz0t1_internal(
+      row, col0, col1, (_tile1024i_1024a *)dst0, (_tile1024i_1024a *)dst1, base,
+      (__SIZE_TYPE__)(stride));
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS_TRANSPOSE _tile_2rpntlvwz1_internal(
+    unsigned short row, unsigned short col0, unsigned short col1,
+    _tile1024i *dst0, _tile1024i *dst1, const void *base,
+    __SIZE_TYPE__ stride) {
+  __builtin_ia32_t2rpntlvwz1_internal(row, col0, col1, (_tile1024i_1024a *)dst0,
+                                      (_tile1024i_1024a *)dst1, base,
+                                      (__SIZE_TYPE__)(stride));
+}
+
+static __inline__ void __DEFAULT_FN_ATTRS_TRANSPOSE _tile_2rpntlvwz1t1_internal(
+    unsigned short row, unsigned short col0, unsigned short col1,
+    _tile1024i *dst0, _tile1024i *dst1, const void *base,
+    __SIZE_TYPE__ stride) {
+  __builtin_ia32_t2rpntlvwz1t1_internal(
+      row, col0, col1, (_tile1024i_1024a *)dst0, (_tile1024i_1024a *)dst1, base,
+      (__SIZE_TYPE__)(stride));
+}
+
+// This is internal intrinsic. C/C++ user should avoid calling it directly.
+static __inline__ _tile1024i __DEFAULT_FN_ATTRS_TRANSPOSE
+_tile_transposed_internal(unsigned short m, unsigned short n, _tile1024i src) {
+  return __builtin_ia32_ttransposed_internal(m, n, src);
+}
+
+/// Converts a pair of tiles from memory into VNNI format, and places the
+/// results in a pair of destinations specified by dst. The pair of tiles
+/// in memory is specified via a tsib; the second tile is after the first
+/// one, separated by the same stride that separates each row.
+/// The tile configuration for the destination tiles indicates the amount
+/// of data to read from memory. The instruction will load a number of rows
+/// that is equal to twice the number of rows in tmm1. The size of each row
+/// is equal to the average width of the destination tiles. If the second
+/// tile is configured with zero rows and columns, only the first tile will
+/// be written.
+/// Provides a hint to the implementation that the data will likely not be
+/// reused in the near future and the data caching can be optimized.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> T2RPNTLVWZ0 </c> instruction.
+///
+/// \param dst0
+///    First tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param dst1
+///    Second tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param base
+///    A pointer to base address.
+/// \param stride
+///    The stride between the rows' data to be loaded in memory.
+__DEFAULT_FN_ATTRS_TRANSPOSE
+static void __tile_2rpntlvwz0(__tile1024i *dst0, __tile1024i *dst1,
+                              const void *base, __SIZE_TYPE__ stride) {
+  _tile_2rpntlvwz0_internal(dst0->row, dst0->col, dst1->col, &dst0->tile,
+                            &dst1->tile, base, stride);
+}
+
+/// Converts a pair of tiles from memory into VNNI format, and places the
+/// results in a pair of destinations specified by dst. The pair of tiles
+/// in memory is specified via a tsib; the second tile is after the first
+/// one, separated by the same stride that separates each row.
+/// The tile configuration for the destination tiles indicates the amount
+/// of data to read from memory. The instruction will load a number of rows
+/// that is equal to twice the number of rows in tmm1. The size of each row
+/// is equal to the average width of the destination tiles. If the second
+/// tile is configured with zero rows and columns, only the first tile will
+/// be written.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> T2RPNTLVWZ0T1 </c> instruction.
+///
+/// \param dst0
+///    First tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param dst1
+///    Second tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param base
+///    A pointer to base address.
+/// \param stride
+///    The stride between the rows' data to be loaded in memory.
+__DEFAULT_FN_ATTRS_TRANSPOSE
+static void __tile_2rpntlvwz0t1(__tile1024i *dst0, __tile1024i *dst1,
+                                const void *base, __SIZE_TYPE__ stride) {
+  _tile_2rpntlvwz0t1_internal(dst0->row, dst0->col, dst1->col, &dst0->tile,
+                              &dst1->tile, base, stride);
+}
+
+/// Converts a pair of tiles from memory into VNNI format, and places the
+/// results in a pair of destinations specified by dst. The pair of tiles
+/// in memory is specified via a tsib; the second tile is after the first
+/// one, separated by the same stride that separates each row.
+/// The tile configuration for the destination tiles indicates the amount
+/// of data to read from memory. The instruction will load a number of rows
+/// that is equal to twice the number of rows in tmm1. The size of each row
+/// is equal to the average width of the destination tiles. If the second
+/// tile is configured with zero rows and columns, only the first tile will
+/// be written. The last row will be not be read from memory but instead
+/// filled with zeros.
+/// Provides a hint to the implementation that the data will likely not be
+/// reused in the near future and the data caching can be optimized.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> T2RPNTLVWZ1 </c> instruction.
+///
+/// \param dst0
+///    First tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param dst1
+///    Second tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param base
+///    A pointer to base address.
+/// \param stride
+///    The stride between the rows' data to be loaded in memory.
+__DEFAULT_FN_ATTRS_TRANSPOSE
+static void __tile_2rpntlvwz1(__tile1024i *dst0, __tile1024i *dst1,
+                              const void *base, __SIZE_TYPE__ stride) {
+  _tile_2rpntlvwz1_internal(dst0->row, dst0->col, dst1->col, &dst0->tile,
+                            &dst1->tile, base, stride);
+}
+
+/// Converts a pair of tiles from memory into VNNI format, and places the
+/// results in a pair of destinations specified by dst. The pair of tiles
+/// in memory is specified via a tsib; the second tile is after the first
+/// one, separated by the same stride that separates each row.
+/// The tile configuration for the destination tiles indicates the amount
+/// of data to read from memory. The instruction will load a number of rows
+/// that is equal to twice the number of rows in tmm1. The size of each row
+/// is equal to the average width of the destination tiles. If the second
+/// tile is configured with zero rows and columns, only the first tile will
+/// be written. The last row will be not be read from memory but instead
+/// filled with zeros.
+/// Provides a hint to the implementation that the data will likely not be
+/// reused in the near future and the data caching can be optimized.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> T2RPNTLVWZ1T1 </c> instruction.
+///
+/// \param dst0
+///    First tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param dst1
+///    Second tile of destination tile pair. Max size is 1024i*2 Bytes.
+/// \param base
+///    A pointer to base address.
+/// \param stride
+///    The stride between the rows' data to be loaded in memory.
+__DEFAULT_FN_ATTRS_TRANSPOSE
+static void __tile_2rpntlvwz1t1(__tile1024i *dst0, __tile1024i *dst1,
+                                const void *base, __SIZE_TYPE__ stride) {
+  _tile_2rpntlvwz1t1_internal(dst0->row, dst0->col, dst1->col, &dst0->tile,
+                              &dst1->tile, base, stride);
+}
+
+/// Transpose 32-bit elements from src and write the result to dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> TTRANSPOSED </c> instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param src
+///    The source tile. Max size is 1024 Bytes.
+__DEFAULT_FN_ATTRS_TRANSPOSE
+static void __tile_transposed(__tile1024i *dst, __tile1024i src) {
+  dst->tile = _tile_transposed_internal(dst->row, dst->col, src.tile);
+}
+
+#endif /* __x86_64__ */
+#endif /* __AMX_TRANSPOSEINTRIN_H */
diff --git a/clang/lib/Headers/cmpccxaddintrin.h b/clang/lib/Headers/cmpccxaddintrin.h
@@ -63,7 +63,7 @@ typedef enum {
                                     (int)(__D))))
 
 #define _cmpccxadd_epi64(__A, __B, __C, __D)                                   \
-  ((long long)(__builtin_ia32_cmpccxadd64((void *)(__A), (long long)(__B),     \
+  ((long long)(__builtin_ia32_cmpccxadd64((__A), (long long)(__B),             \
                                           (long long)(__C), (int)(__D))))
 
 #endif // __x86_64__

diff --git a/clang/lib/Headers/hlsl/hlsl_intrinsics.h b/clang/lib/Headers/hlsl/hlsl_intrinsics.h
@@ -934,6 +934,16 @@ uint64_t dot(uint64_t3, uint64_t3);
 _HLSL_BUILTIN_ALIAS(__builtin_hlsl_dot)
 uint64_t dot(uint64_t4, uint64_t4);
 
+//===----------------------------------------------------------------------===//
+// dot4add builtins
+//===----------------------------------------------------------------------===//
+
+/// \fn int dot4add_i8packed(uint A, uint B, int C)
+
+_HLSL_AVAILABILITY(shadermodel, 6.4)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_dot4add_i8packed)
+int dot4add_i8packed(unsigned int, unsigned int, int);
+
 //===----------------------------------------------------------------------===//
 // exp builtins
 //===----------------------------------------------------------------------===//

diff --git a/clang/lib/Headers/immintrin.h b/clang/lib/Headers/immintrin.h
@@ -652,6 +652,10 @@ _storebe_i64(void * __P, long long __D) {
 #include <amxfp8intrin.h>
 #endif
 
+#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_TRANSPOSE__)
+#include <amxtransposeintrin.h>
+#endif
+
 #if !defined(__SCE__) || __has_feature(modules) ||                             \
     defined(__AVX512VP2INTERSECT__)
 #include <avx512vp2intersectintrin.h>

diff --git a/clang/lib/Lex/HeaderSearch.cpp b/clang/lib/Lex/HeaderSearch.cpp
@@ -974,13 +974,9 @@ OptionalFileEntryRef HeaderSearch::LookupFile(
         const HeaderFileInfo *FromHFI = getExistingFileInfo(*Includer);
         assert(FromHFI && "includer without file info");
         unsigned DirInfo = FromHFI->DirInfo;
-        bool IndexHeaderMapHeader = FromHFI->IndexHeaderMapHeader;
-        StringRef Framework = FromHFI->Framework;
 
         HeaderFileInfo &ToHFI = getFileInfo(*FE);
         ToHFI.DirInfo = DirInfo;
-        ToHFI.IndexHeaderMapHeader = IndexHeaderMapHeader;
-        ToHFI.Framework = Framework;
 
         if (SearchPath) {
           StringRef SearchPathRef(IncluderAndDir.second.getName());
@@ -1122,23 +1118,6 @@ OptionalFileEntryRef HeaderSearch::LookupFile(
       }
     }
 
-    // Set the `Framework` info if this file is in a header map with framework
-    // style include spelling or found in a framework dir. The header map case
-    // is possible when building frameworks which use header maps.
-    if (CurDir->isHeaderMap() && isAngled) {
-      size_t SlashPos = Filename.find('/');
-      if (SlashPos != StringRef::npos)
-        HFI.Framework =
-            getUniqueFrameworkName(StringRef(Filename.begin(), SlashPos));
-      if (CurDir->isIndexHeaderMap())
-        HFI.IndexHeaderMapHeader = 1;
-    } else if (CurDir->isFramework()) {
-      size_t SlashPos = Filename.find('/');
-      if (SlashPos != StringRef::npos)
-        HFI.Framework =
-            getUniqueFrameworkName(StringRef(Filename.begin(), SlashPos));
-    }
-
     if (checkMSVCHeaderSearch(Diags, MSFE, &File->getFileEntry(), IncludeLoc)) {
       if (SuggestedModule)
         *SuggestedModule = MSSuggestedModule;
@@ -1156,41 +1135,6 @@ OptionalFileEntryRef HeaderSearch::LookupFile(
     return File;
   }
 
-  // If we are including a file with a quoted include "foo.h" from inside
-  // a header in a framework that is currently being built, and we couldn't
-  // resolve "foo.h" any other way, change the include to <Foo/foo.h>, where
-  // "Foo" is the name of the framework in which the including header was found.
-  if (!Includers.empty() && Includers.front().first && !isAngled &&
-      !Filename.contains('/')) {
-    const HeaderFileInfo *IncludingHFI =
-        getExistingFileInfo(*Includers.front().first);
-    assert(IncludingHFI && "includer without file info");
-    if (IncludingHFI->IndexHeaderMapHeader) {
-      SmallString<128> ScratchFilename;
-      ScratchFilename += IncludingHFI->Framework;
-      ScratchFilename += '/';
-      ScratchFilename += Filename;
-
-      OptionalFileEntryRef File = LookupFile(
-          ScratchFilename, IncludeLoc, /*isAngled=*/true, FromDir, &CurDir,
-          Includers.front(), SearchPath, RelativePath, RequestingModule,
-          SuggestedModule, IsMapped, /*IsFrameworkFound=*/nullptr);
-
-      if (checkMSVCHeaderSearch(Diags, MSFE,
-                                File ? &File->getFileEntry() : nullptr,
-                                IncludeLoc)) {
-        if (SuggestedModule)
-          *SuggestedModule = MSSuggestedModule;
-        return MSFE;
-      }
-
-      cacheLookupSuccess(LookupFileCache[Filename],
-                         LookupFileCache[ScratchFilename].HitIt, IncludeLoc);
-      // FIXME: SuggestedModule.
-      return File;
-    }
-  }
-
   if (checkMSVCHeaderSearch(Diags, MSFE, nullptr, IncludeLoc)) {
     if (SuggestedModule)
       *SuggestedModule = MSSuggestedModule;
@@ -1358,10 +1302,6 @@ static void mergeHeaderFileInfo(HeaderFileInfo &HFI,
   HFI.DirInfo = OtherHFI.DirInfo;
   HFI.External = (!HFI.IsValid || HFI.External);
   HFI.IsValid = true;
-  HFI.IndexHeaderMapHeader = OtherHFI.IndexHeaderMapHeader;
-
-  if (HFI.Framework.empty())
-    HFI.Framework = OtherHFI.Framework;
 }
 
 HeaderFileInfo &HeaderSearch::getFileInfo(FileEntryRef FE) {

diff --git a/clang/lib/Lex/InitHeaderSearch.cpp b/clang/lib/Lex/InitHeaderSearch.cpp
@@ -149,7 +149,7 @@ bool InitHeaderSearch::AddUnmappedPath(const Twine &Path, IncludeDirGroup Group,
 
   // Compute the DirectoryLookup type.
   SrcMgr::CharacteristicKind Type;
-  if (Group == Quoted || Group == Angled || Group == IndexHeaderMap) {
+  if (Group == Quoted || Group == Angled) {
     Type = SrcMgr::C_User;
   } else if (Group == ExternCSystem) {
     Type = SrcMgr::C_ExternCSystem;
@@ -170,9 +170,8 @@ bool InitHeaderSearch::AddUnmappedPath(const Twine &Path, IncludeDirGroup Group,
     if (auto FE = FM.getOptionalFileRef(MappedPathStr)) {
       if (const HeaderMap *HM = Headers.CreateHeaderMap(*FE)) {
         // It is a headermap, add it to the search path.
-        IncludePath.emplace_back(
-            Group, DirectoryLookup(HM, Type, Group == IndexHeaderMap),
-            UserEntryIdx);
+        IncludePath.emplace_back(Group, DirectoryLookup(HM, Type),
+                                 UserEntryIdx);
         return true;
       }
     }
@@ -488,7 +487,7 @@ void InitHeaderSearch::Realize(const LangOptions &Lang) {
   unsigned NumQuoted = SearchList.size();
 
   for (auto &Include : IncludePath)
-    if (Include.Group == Angled || Include.Group == IndexHeaderMap)
+    if (Include.Group == Angled)
       SearchList.push_back(Include);
 
   RemoveDuplicates(SearchList, NumQuoted, Verbose);

diff --git a/clang/lib/Lex/PPMacroExpansion.cpp b/clang/lib/Lex/PPMacroExpansion.cpp
@@ -323,84 +323,69 @@ void Preprocessor::dumpMacroInfo(const IdentifierInfo *II) {
   }
 }
 
-/// RegisterBuiltinMacro - Register the specified identifier in the identifier
-/// table and mark it as a builtin macro to be expanded.
-static IdentifierInfo *RegisterBuiltinMacro(Preprocessor &PP, const char *Name){
-  // Get the identifier.
-  IdentifierInfo *Id = PP.getIdentifierInfo(Name);
-
-  // Mark it as being a macro that is builtin.
-  MacroInfo *MI = PP.AllocateMacroInfo(SourceLocation());
-  MI->setIsBuiltinMacro();
-  PP.appendDefMacroDirective(Id, MI);
-  return Id;
-}
-
 /// RegisterBuiltinMacros - Register builtin macros, such as __LINE__ with the
 /// identifier table.
 void Preprocessor::RegisterBuiltinMacros() {
-  Ident__LINE__ = RegisterBuiltinMacro(*this, "__LINE__");
-  Ident__FILE__ = RegisterBuiltinMacro(*this, "__FILE__");
-  Ident__DATE__ = RegisterBuiltinMacro(*this, "__DATE__");
-  Ident__TIME__ = RegisterBuiltinMacro(*this, "__TIME__");
-  Ident__COUNTER__ = RegisterBuiltinMacro(*this, "__COUNTER__");
-  Ident_Pragma  = RegisterBuiltinMacro(*this, "_Pragma");
-  Ident__FLT_EVAL_METHOD__ = RegisterBuiltinMacro(*this, "__FLT_EVAL_METHOD__");
+  Ident__LINE__ = RegisterBuiltinMacro("__LINE__");
+  Ident__FILE__ = RegisterBuiltinMacro("__FILE__");
+  Ident__DATE__ = RegisterBuiltinMacro("__DATE__");
+  Ident__TIME__ = RegisterBuiltinMacro("__TIME__");
+  Ident__COUNTER__ = RegisterBuiltinMacro("__COUNTER__");
+  Ident_Pragma = RegisterBuiltinMacro("_Pragma");
+  Ident__FLT_EVAL_METHOD__ = RegisterBuiltinMacro("__FLT_EVAL_METHOD__");
 
   // C++ Standing Document Extensions.
   if (getLangOpts().CPlusPlus)
-    Ident__has_cpp_attribute =
-        RegisterBuiltinMacro(*this, "__has_cpp_attribute");
+    Ident__has_cpp_attribute = RegisterBuiltinMacro("__has_cpp_attribute");
   else
     Ident__has_cpp_attribute = nullptr;
 
   // GCC Extensions.
-  Ident__BASE_FILE__     = RegisterBuiltinMacro(*this, "__BASE_FILE__");
-  Ident__INCLUDE_LEVEL__ = RegisterBuiltinMacro(*this, "__INCLUDE_LEVEL__");
-  Ident__TIMESTAMP__     = RegisterBuiltinMacro(*this, "__TIMESTAMP__");
+  Ident__BASE_FILE__ = RegisterBuiltinMacro("__BASE_FILE__");
+  Ident__INCLUDE_LEVEL__ = RegisterBuiltinMacro("__INCLUDE_LEVEL__");
+  Ident__TIMESTAMP__ = RegisterBuiltinMacro("__TIMESTAMP__");
 
   // Microsoft Extensions.
   if (getLangOpts().MicrosoftExt) {
-    Ident__identifier = RegisterBuiltinMacro(*this, "__identifier");
-    Ident__pragma = RegisterBuiltinMacro(*this, "__pragma");
+    Ident__identifier = RegisterBuiltinMacro("__identifier");
+    Ident__pragma = RegisterBuiltinMacro("__pragma");
   } else {
     Ident__identifier = nullptr;
     Ident__pragma = nullptr;
   }
 
   // Clang Extensions.
-  Ident__FILE_NAME__      = RegisterBuiltinMacro(*this, "__FILE_NAME__");
-  Ident__has_feature      = RegisterBuiltinMacro(*this, "__has_feature");
-  Ident__has_extension    = RegisterBuiltinMacro(*this, "__has_extension");
-  Ident__has_builtin      = RegisterBuiltinMacro(*this, "__has_builtin");
+  Ident__FILE_NAME__ = RegisterBuiltinMacro("__FILE_NAME__");
+  Ident__has_feature = RegisterBuiltinMacro("__has_feature");
+  Ident__has_extension = RegisterBuiltinMacro("__has_extension");
+  Ident__has_builtin = RegisterBuiltinMacro("__has_builtin");
   Ident__has_constexpr_builtin =
-      RegisterBuiltinMacro(*this, "__has_constexpr_builtin");
-  Ident__has_attribute    = RegisterBuiltinMacro(*this, "__has_attribute");
+      RegisterBuiltinMacro("__has_constexpr_builtin");
+  Ident__has_attribute = RegisterBuiltinMacro("__has_attribute");
   if (!getLangOpts().CPlusPlus)
-    Ident__has_c_attribute = RegisterBuiltinMacro(*this, "__has_c_attribute");
+    Ident__has_c_attribute = RegisterBuiltinMacro("__has_c_attribute");
   else
     Ident__has_c_attribute = nullptr;
 
-  Ident__has_declspec = RegisterBuiltinMacro(*this, "__has_declspec_attribute");
-  Ident__has_embed = RegisterBuiltinMacro(*this, "__has_embed");
-  Ident__has_include      = RegisterBuiltinMacro(*this, "__has_include");
-  Ident__has_include_next = RegisterBuiltinMacro(*this, "__has_include_next");
-  Ident__has_warning      = RegisterBuiltinMacro(*this, "__has_warning");
-  Ident__is_identifier    = RegisterBuiltinMacro(*this, "__is_identifier");
-  Ident__is_target_arch   = RegisterBuiltinMacro(*this, "__is_target_arch");
-  Ident__is_target_vendor = RegisterBuiltinMacro(*this, "__is_target_vendor");
-  Ident__is_target_os     = RegisterBuiltinMacro(*this, "__is_target_os");
+  Ident__has_declspec = RegisterBuiltinMacro("__has_declspec_attribute");
+  Ident__has_embed = RegisterBuiltinMacro("__has_embed");
+  Ident__has_include = RegisterBuiltinMacro("__has_include");
+  Ident__has_include_next = RegisterBuiltinMacro("__has_include_next");
+  Ident__has_warning = RegisterBuiltinMacro("__has_warning");
+  Ident__is_identifier = RegisterBuiltinMacro("__is_identifier");
+  Ident__is_target_arch = RegisterBuiltinMacro("__is_target_arch");
+  Ident__is_target_vendor = RegisterBuiltinMacro("__is_target_vendor");
+  Ident__is_target_os = RegisterBuiltinMacro("__is_target_os");
   Ident__is_target_environment =
-      RegisterBuiltinMacro(*this, "__is_target_environment");
-  Ident__is_target_variant_os =
-      RegisterBuiltinMacro(*this, "__is_target_variant_os");
+      RegisterBuiltinMacro("__is_target_environment");
+  Ident__is_target_variant_os = RegisterBuiltinMacro("__is_target_variant_os");
   Ident__is_target_variant_environment =
-      RegisterBuiltinMacro(*this, "__is_target_variant_environment");
+      RegisterBuiltinMacro("__is_target_variant_environment");
 
   // Modules.
-  Ident__building_module  = RegisterBuiltinMacro(*this, "__building_module");
+  Ident__building_module = RegisterBuiltinMacro("__building_module");
   if (!getLangOpts().CurrentModule.empty())
-    Ident__MODULE__ = RegisterBuiltinMacro(*this, "__MODULE__");
+    Ident__MODULE__ = RegisterBuiltinMacro("__MODULE__");
   else
     Ident__MODULE__ = nullptr;
 }

diff --git a/clang/lib/Sema/CheckExprLifetime.cpp b/clang/lib/Sema/CheckExprLifetime.cpp
@@ -226,14 +226,14 @@ using LocalVisitor = llvm::function_ref<bool(IndirectLocalPath &Path, Local L,
                                              ReferenceKind RK)>;
 } // namespace
 
-static bool isVarOnPath(IndirectLocalPath &Path, VarDecl *VD) {
+static bool isVarOnPath(const IndirectLocalPath &Path, VarDecl *VD) {
   for (auto E : Path)
     if (E.Kind == IndirectLocalPathEntry::VarInit && E.D == VD)
       return true;
   return false;
 }
 
-static bool pathContainsInit(IndirectLocalPath &Path) {
+static bool pathContainsInit(const IndirectLocalPath &Path) {
   return llvm::any_of(Path, [=](IndirectLocalPathEntry E) {
     return E.Kind == IndirectLocalPathEntry::DefaultInit ||
            E.Kind == IndirectLocalPathEntry::VarInit;
@@ -1076,7 +1076,7 @@ static SourceRange nextPathEntryRange(const IndirectLocalPath &Path, unsigned I,
   return E->getSourceRange();
 }
 
-static bool pathOnlyHandlesGslPointer(IndirectLocalPath &Path) {
+static bool pathOnlyHandlesGslPointer(const IndirectLocalPath &Path) {
   for (const auto &It : llvm::reverse(Path)) {
     switch (It.Kind) {
     case IndirectLocalPathEntry::VarInit:
@@ -1124,24 +1124,23 @@ static void checkExprLifetimeImpl(Sema &SemaRef,
 
   // FIXME: consider moving the TemporaryVisitor and visitLocalsRetained*
   // functions to a dedicated class.
-  auto TemporaryVisitor = [&](IndirectLocalPath &Path, Local L,
+  auto TemporaryVisitor = [&](const IndirectLocalPath &Path, Local L,
                               ReferenceKind RK) -> bool {
     SourceRange DiagRange = nextPathEntryRange(Path, 0, L);
     SourceLocation DiagLoc = DiagRange.getBegin();
 
     auto *MTE = dyn_cast<MaterializeTemporaryExpr>(L);
 
     bool IsGslPtrValueFromGslTempOwner = false;
-    bool IsLocalGslOwner = false;
     if (pathOnlyHandlesGslPointer(Path)) {
       if (isa<DeclRefExpr>(L)) {
         // We do not want to follow the references when returning a pointer
         // originating from a local owner to avoid the following false positive:
         //   int &p = *localUniquePtr;
         //   someContainer.add(std::move(localUniquePtr));
         //   return p;
-        IsLocalGslOwner = isRecordWithAttr<OwnerAttr>(L->getType());
-        if (pathContainsInit(Path) || !IsLocalGslOwner)
+        if (pathContainsInit(Path) ||
+            !isRecordWithAttr<OwnerAttr>(L->getType()))
           return false;
       } else {
         IsGslPtrValueFromGslTempOwner =
@@ -1261,12 +1260,12 @@ static void checkExprLifetimeImpl(Sema &SemaRef,
         if (pathContainsInit(Path))
           return false;
 
+        auto *DRE = dyn_cast<DeclRefExpr>(L);
         // Suppress false positives for code like the one below:
-        //   Ctor(unique_ptr<T> up) : member(*up), member2(move(up)) {}
-        if (IsLocalGslOwner && pathOnlyHandlesGslPointer(Path))
+        //   Ctor(unique_ptr<T> up) : pointer(up.get()), owner(move(up)) {}
+        if (DRE && isRecordWithAttr<OwnerAttr>(DRE->getType()))
           return false;
 
-        auto *DRE = dyn_cast<DeclRefExpr>(L);
         auto *VD = DRE ? dyn_cast<VarDecl>(DRE->getDecl()) : nullptr;
         if (!VD) {
           // A member was initialized to a local block.

diff --git a/clang/lib/Sema/HLSLExternalSemaSource.cpp b/clang/lib/Sema/HLSLExternalSemaSource.cpp
@@ -14,8 +14,6 @@
 #include "clang/AST/Attr.h"
 #include "clang/AST/DeclCXX.h"
 #include "clang/AST/Type.h"
-#include "clang/Basic/AttrKinds.h"
-#include "clang/Basic/HLSLRuntime.h"
 #include "clang/Basic/SourceLocation.h"
 #include "clang/Sema/Lookup.h"
 #include "clang/Sema/Sema.h"
@@ -164,16 +162,7 @@ struct BuiltinTypeDeclBuilder {
                                VD, false, NameInfo, Ty, VK_PRValue);
   }
 
-  static Expr *emitResourceClassExpr(ASTContext &AST, ResourceClass RC) {
-    return IntegerLiteral::Create(
-        AST,
-        llvm::APInt(AST.getIntWidth(AST.UnsignedCharTy),
-                    static_cast<uint8_t>(RC)),
-        AST.UnsignedCharTy, SourceLocation());
-  }
-
-  BuiltinTypeDeclBuilder &addDefaultHandleConstructor(Sema &S,
-                                                      ResourceClass RC) {
+  BuiltinTypeDeclBuilder &addDefaultHandleConstructor(Sema &S) {
     if (Record->isCompleteDefinition())
       return *this;
     ASTContext &AST = Record->getASTContext();
@@ -480,7 +469,7 @@ static BuiltinTypeDeclBuilder setupBufferType(CXXRecordDecl *Decl, Sema &S,
                                               bool IsROV, bool RawBuffer) {
   return BuiltinTypeDeclBuilder(Decl)
       .addHandleMember(S, RC, RK, IsROV, RawBuffer)
-      .addDefaultHandleConstructor(S, RC);
+      .addDefaultHandleConstructor(S);
 }
 
 void HLSLExternalSemaSource::defineHLSLTypesWithForwardDeclarations() {

diff --git a/clang/lib/Sema/SemaAPINotes.cpp b/clang/lib/Sema/SemaAPINotes.cpp
@@ -12,6 +12,7 @@
 
 #include "clang/APINotes/APINotesReader.h"
 #include "clang/AST/Decl.h"
+#include "clang/AST/DeclCXX.h"
 #include "clang/AST/DeclObjC.h"
 #include "clang/Basic/SourceLocation.h"
 #include "clang/Lex/Lexer.h"
@@ -415,6 +416,13 @@ static void ProcessAPINotes(Sema &S, ParmVarDecl *D,
       return new (S.Context) NoEscapeAttr(S.Context, getPlaceholderAttrInfo());
     });
 
+  if (auto Lifetimebound = Info.isLifetimebound())
+    handleAPINotedAttribute<LifetimeBoundAttr>(
+        S, D, *Lifetimebound, Metadata, [&] {
+          return new (S.Context)
+              LifetimeBoundAttr(S.Context, getPlaceholderAttrInfo());
+        });
+
   // Retain count convention
   handleAPINotedRetainCountConvention(S, D, Metadata,
                                       Info.getRetainCountConvention());
@@ -860,13 +868,12 @@ void Sema::ProcessAPINotes(Decl *D) {
   if (!D)
     return;
 
+  auto *DC = D->getDeclContext();
   // Globals.
-  if (D->getDeclContext()->isFileContext() ||
-      D->getDeclContext()->isNamespace() ||
-      D->getDeclContext()->isExternCContext() ||
-      D->getDeclContext()->isExternCXXContext()) {
+  if (DC->isFileContext() || DC->isNamespace() || DC->isExternCContext() ||
+      DC->isExternCXXContext()) {
     std::optional<api_notes::Context> APINotesContext =
-        UnwindNamespaceContext(D->getDeclContext(), APINotes);
+        UnwindNamespaceContext(DC, APINotes);
     // Global variables.
     if (auto VD = dyn_cast<VarDecl>(D)) {
       for (auto Reader : APINotes.findAPINotes(D->getLocation())) {
@@ -967,8 +974,8 @@ void Sema::ProcessAPINotes(Decl *D) {
   }
 
   // Enumerators.
-  if (D->getDeclContext()->getRedeclContext()->isFileContext() ||
-      D->getDeclContext()->getRedeclContext()->isExternCContext()) {
+  if (DC->getRedeclContext()->isFileContext() ||
+      DC->getRedeclContext()->isExternCContext()) {
     if (auto EnumConstant = dyn_cast<EnumConstantDecl>(D)) {
       for (auto Reader : APINotes.findAPINotes(D->getLocation())) {
         auto Info = Reader->lookupEnumConstant(EnumConstant->getName());
@@ -979,7 +986,7 @@ void Sema::ProcessAPINotes(Decl *D) {
     }
   }
 
-  if (auto ObjCContainer = dyn_cast<ObjCContainerDecl>(D->getDeclContext())) {
+  if (auto ObjCContainer = dyn_cast<ObjCContainerDecl>(DC)) {
     // Location function that looks up an Objective-C context.
     auto GetContext = [&](api_notes::APINotesReader *Reader)
         -> std::optional<api_notes::ContextID> {
@@ -1063,7 +1070,7 @@ void Sema::ProcessAPINotes(Decl *D) {
     }
   }
 
-  if (auto TagContext = dyn_cast<TagDecl>(D->getDeclContext())) {
+  if (auto TagContext = dyn_cast<TagDecl>(DC)) {
     if (auto CXXMethod = dyn_cast<CXXMethodDecl>(D)) {
       if (!isa<CXXConstructorDecl>(CXXMethod) &&
           !isa<CXXDestructorDecl>(CXXMethod) &&

diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
@@ -2212,7 +2212,9 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl, unsigned BuiltinID,
     if (CheckBuiltinTargetInSupported(
             *this, TheCall,
             {llvm::Triple::x86, llvm::Triple::x86_64, llvm::Triple::arm,
-             llvm::Triple::thumb, llvm::Triple::aarch64, llvm::Triple::amdgcn}))
+             llvm::Triple::thumb, llvm::Triple::aarch64, llvm::Triple::amdgcn,
+             llvm::Triple::ppc, llvm::Triple::ppc64, llvm::Triple::ppcle,
+             llvm::Triple::ppc64le}))
       return ExprError();
     break;
 

diff --git a/clang/lib/Sema/SemaConcept.cpp b/clang/lib/Sema/SemaConcept.cpp
@@ -975,7 +975,7 @@ static const Expr *SubstituteConstraintExpressionWithoutSatisfaction(
   std::optional<LocalInstantiationScope> ScopeForParameters;
   if (const NamedDecl *ND = DeclInfo.getDecl();
       ND && ND->isFunctionOrFunctionTemplate()) {
-    ScopeForParameters.emplace(S);
+    ScopeForParameters.emplace(S, /*CombineWithOuterScope=*/true);
     const FunctionDecl *FD = ND->getAsFunction();
     for (auto *PVD : FD->parameters()) {
       if (!PVD->isParameterPack()) {

diff --git a/clang/lib/Sema/SemaDecl.cpp b/clang/lib/Sema/SemaDecl.cpp
@@ -4694,8 +4694,10 @@ void Sema::MergeVarDecl(VarDecl *New, LookupResult &Previous) {
 
   // Keep a chain of previous declarations.
   New->setPreviousDecl(Old);
-  if (NewTemplate)
+  if (NewTemplate) {
+    NewTemplate->mergePrevDecl(OldTemplate);
     NewTemplate->setPreviousDecl(OldTemplate);
+  }
 
   // Inherit access appropriately.
   New->setAccess(Old->getAccess());
@@ -12092,6 +12094,9 @@ bool Sema::CheckFunctionDeclaration(Scope *S, FunctionDecl *NewFD,
   if (LangOpts.OpenMP)
     OpenMP().ActOnFinishedFunctionDefinitionInOpenMPAssumeScope(NewFD);
 
+  if (LangOpts.isSYCL() && NewFD->hasAttr<SYCLKernelEntryPointAttr>())
+    getASTContext().registerSYCLEntryPointFunction(NewFD);
+
   // Semantic checking for this function declaration (in isolation).
 
   if (getLangOpts().CPlusPlus) {
@@ -17955,6 +17960,8 @@ Sema::ActOnTag(Scope *S, unsigned TagSpec, TagUseKind TUK, SourceLocation KWLoc,
             << Name;
         Invalid = true;
       }
+      if (TUK == TagUseKind::Declaration)
+        Invalid = true;
     } else if (!PrevDecl) {
       Diag(Loc, diag::warn_decl_in_param_list) << Context.getTagDeclType(New);
     }

diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -6620,6 +6620,9 @@ ProcessDeclAttribute(Sema &S, Scope *scope, Decl *D, const ParsedAttr &AL,
   case ParsedAttr::AT_SYCLKernel:
     S.SYCL().handleKernelAttr(D, AL);
     break;
+  case ParsedAttr::AT_SYCLKernelEntryPoint:
+    S.SYCL().handleKernelEntryPointAttr(D, AL);
+    break;
   case ParsedAttr::AT_SYCLSpecialClass:
     handleSimpleAttribute<SYCLSpecialClassAttr>(S, D, AL);
     break;

diff --git a/clang/lib/Sema/SemaExprCXX.cpp b/clang/lib/Sema/SemaExprCXX.cpp
@@ -5032,6 +5032,7 @@ static bool CheckUnaryTypeTraitTypeCompleteness(Sema &S, TypeTrait UTT,
   case UTT_IsScalar:
   case UTT_IsCompound:
   case UTT_IsMemberPointer:
+  case UTT_IsTypedResourceElementCompatible:
     // Fall-through
 
     // These traits are modeled on type predicates in C++0x [meta.unary.prop]
@@ -5714,6 +5715,15 @@ static bool EvaluateUnaryTypeTrait(Sema &Self, TypeTrait UTT,
                                   tok::kw___builtin_hlsl_is_intangible))
       return false;
     return T->isHLSLIntangibleType();
+
+  case UTT_IsTypedResourceElementCompatible:
+    assert(Self.getLangOpts().HLSL &&
+           "typed resource element compatible types are an HLSL-only feature");
+    if (Self.RequireCompleteType(TInfo->getTypeLoc().getBeginLoc(), T,
+                                 diag::err_incomplete_type))
+      return false;
+
+    return Self.HLSL().IsTypedResourceElementCompatible(T);
   }
 }
 

diff --git a/clang/lib/Sema/SemaFunctionEffects.cpp b/clang/lib/Sema/SemaFunctionEffects.cpp
@@ -971,6 +971,7 @@ class Analyzer {
     PendingFunctionAnalysis &CurrentFunction;
     CallableInfo &CurrentCaller;
     ViolationSite VSite;
+    const Expr *TrailingRequiresClause = nullptr;
 
     FunctionBodyASTVisitor(Analyzer &Outer,
                            PendingFunctionAnalysis &CurrentFunction,
@@ -985,6 +986,9 @@ class Analyzer {
       if (auto *Dtor = dyn_cast<CXXDestructorDecl>(CurrentCaller.CDecl))
         followDestructor(dyn_cast<CXXRecordDecl>(Dtor->getParent()), Dtor);
 
+      if (auto *FD = dyn_cast<FunctionDecl>(CurrentCaller.CDecl))
+        TrailingRequiresClause = FD->getTrailingRequiresClause();
+
       // Do an AST traversal of the function/block body
       TraverseDecl(const_cast<Decl *>(CurrentCaller.CDecl));
     }
@@ -1259,6 +1263,17 @@ class Analyzer {
       return true;
     }
 
+    bool TraverseStmt(Stmt *Statement) {
+      // If this statement is a `requires` clause from the top-level function
+      // being traversed, ignore it, since it's not generating runtime code.
+      // We skip the traversal of lambdas (beyond their captures, see
+      // TraverseLambdaExpr below), so just caching this from our constructor
+      // should suffice.
+      if (Statement != TrailingRequiresClause)
+        return Base::TraverseStmt(Statement);
+      return true;
+    }
+
     bool TraverseConstructorInitializer(CXXCtorInitializer *Init) {
       ViolationSite PrevVS = VSite;
       if (Init->isAnyMemberInitializer())
@@ -1297,6 +1312,7 @@ class Analyzer {
     }
 
     bool TraverseBlockExpr(BlockExpr * /*unused*/) {
+      // As with lambdas, don't traverse the block's body.
       // TODO: are the capture expressions (ctor call?) safe?
       return true;
     }

diff --git a/clang/lib/Sema/SemaHLSL.cpp b/clang/lib/Sema/SemaHLSL.cpp
@@ -2199,6 +2199,50 @@ static void BuildFlattenedTypeList(QualType BaseTy,
   }
 }
 
+bool SemaHLSL::IsTypedResourceElementCompatible(clang::QualType QT) {
+  if (QT.isNull())
+    return false;
+
+  // check if the outer type was an array type
+  if (QT->isArrayType())
+    return false;
+
+  llvm::SmallVector<QualType, 4> QTTypes;
+  BuildFlattenedTypeList(QT, QTTypes);
+
+  assert(QTTypes.size() > 0 &&
+         "expected at least one constituent type from non-null type");
+  QualType FirstQT = SemaRef.Context.getCanonicalType(QTTypes[0]);
+
+  // element count cannot exceed 4
+  if (QTTypes.size() > 4)
+    return false;
+
+  for (QualType TempQT : QTTypes) {
+    // ensure homogeneity
+    if (!getASTContext().hasSameUnqualifiedType(FirstQT, TempQT))
+      return false;
+  }
+
+  if (const BuiltinType *BT = FirstQT->getAs<BuiltinType>()) {
+    if (BT->isBooleanType() || BT->isEnumeralType())
+      return false;
+
+    // Check if it is an array type.
+    if (FirstQT->isArrayType())
+      return false;
+  }
+
+  // if the loop above completes without returning, then
+  // we've guaranteed homogeneity
+  int TotalSizeInBytes =
+      (SemaRef.Context.getTypeSize(FirstQT) / 8) * QTTypes.size();
+  if (TotalSizeInBytes > 16)
+    return false;
+
+  return true;
+}
+
 bool SemaHLSL::IsScalarizedLayoutCompatible(QualType T1, QualType T2) const {
   if (T1.isNull() || T2.isNull())
     return false;

diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp
@@ -2253,6 +2253,10 @@ bool InitListChecker::CheckFlexibleArrayInit(const InitializedEntity &Entity,
   return FlexArrayDiag != diag::ext_flexible_array_init;
 }
 
+static bool isInitializedStructuredList(const InitListExpr *StructuredList) {
+  return StructuredList && StructuredList->getNumInits() == 1U;
+}
+
 void InitListChecker::CheckStructUnionTypes(
     const InitializedEntity &Entity, InitListExpr *IList, QualType DeclType,
     CXXRecordDecl::base_class_const_range Bases, RecordDecl::field_iterator Field,
@@ -2499,8 +2503,7 @@ void InitListChecker::CheckStructUnionTypes(
                         StructuredList, StructuredIndex);
     InitializedSomething = true;
     InitializedFields.insert(*Field);
-
-    if (RD->isUnion() && StructuredList) {
+    if (RD->isUnion() && isInitializedStructuredList(StructuredList)) {
       // Initialize the first field within the union.
       StructuredList->setInitializedFieldInUnion(*Field);
     }
@@ -2585,7 +2588,7 @@ void InitListChecker::CheckStructUnionTypes(
     CheckImplicitInitList(MemberEntity, IList, Field->getType(), Index,
                           StructuredList, StructuredIndex);
 
-  if (RD->isUnion() && StructuredList) {
+  if (RD->isUnion() && isInitializedStructuredList(StructuredList)) {
     // Initialize the first field within the union.
     StructuredList->setInitializedFieldInUnion(*Field);
   }
@@ -9954,7 +9957,7 @@ QualType Sema::DeduceTemplateSpecializationFromInitializer(
     auto SynthesizeAggrGuide = [&](InitListExpr *ListInit) {
       auto *Pattern = Template;
       while (Pattern->getInstantiatedFromMemberTemplate()) {
-        if (Pattern->hasMemberSpecialization())
+        if (Pattern->isMemberSpecialization())
           break;
         Pattern = Pattern->getInstantiatedFromMemberTemplate();
       }

diff --git a/clang/lib/Sema/SemaOpenACC.cpp b/clang/lib/Sema/SemaOpenACC.cpp
@@ -1224,6 +1224,36 @@ OpenACCClause *SemaOpenACCClauseVisitor::VisitGangClause(
   if (Clause.getDirectiveKind() != OpenACCDirectiveKind::Loop)
     return isNotImplemented();
 
+  // OpenACC 3.3 Section 2.9.11: A reduction clause may not appear on a loop
+  // directive that has a gang clause and is within a compute construct that has
+  // a num_gangs clause with more than one explicit argument.
+  if (Clause.getDirectiveKind() == OpenACCDirectiveKind::Loop &&
+      SemaRef.getActiveComputeConstructInfo().Kind !=
+          OpenACCDirectiveKind::Invalid) {
+    // num_gangs clause on the active compute construct.
+    auto *NumGangsClauseItr =
+        llvm::find_if(SemaRef.getActiveComputeConstructInfo().Clauses,
+                      llvm::IsaPred<OpenACCNumGangsClause>);
+
+    auto *ReductionClauseItr =
+        llvm::find_if(ExistingClauses, llvm::IsaPred<OpenACCReductionClause>);
+
+    if (ReductionClauseItr != ExistingClauses.end() &&
+        NumGangsClauseItr !=
+            SemaRef.getActiveComputeConstructInfo().Clauses.end() &&
+        cast<OpenACCNumGangsClause>(*NumGangsClauseItr)->getIntExprs().size() >
+            1) {
+      SemaRef.Diag(Clause.getBeginLoc(),
+                   diag::err_acc_gang_reduction_numgangs_conflict)
+          << OpenACCClauseKind::Gang << OpenACCClauseKind::Reduction;
+      SemaRef.Diag((*ReductionClauseItr)->getBeginLoc(),
+                   diag::note_acc_previous_clause_here);
+      SemaRef.Diag((*NumGangsClauseItr)->getBeginLoc(),
+                   diag::note_acc_previous_clause_here);
+      return nullptr;
+    }
+  }
+
   llvm::SmallVector<OpenACCGangKind> GangKinds;
   llvm::SmallVector<Expr *> IntExprs;
 
@@ -1316,9 +1346,9 @@ OpenACCClause *SemaOpenACCClauseVisitor::VisitGangClause(
     return nullptr;
   }
 
-  return OpenACCGangClause::Create(Ctx, Clause.getBeginLoc(),
-                                   Clause.getLParenLoc(), GangKinds, IntExprs,
-                                   Clause.getEndLoc());
+  return SemaRef.CheckGangClause(ExistingClauses, Clause.getBeginLoc(),
+                                 Clause.getLParenLoc(), GangKinds, IntExprs,
+                                 Clause.getEndLoc());
 }
 
 OpenACCClause *SemaOpenACCClauseVisitor::VisitSeqClause(
@@ -1367,9 +1397,57 @@ OpenACCClause *SemaOpenACCClauseVisitor::VisitReductionClause(
   // Restrictions only properly implemented on 'compute' constructs, and
   // 'compute' constructs are the only construct that can do anything with
   // this yet, so skip/treat as unimplemented in this case.
-  if (!isOpenACCComputeDirectiveKind(Clause.getDirectiveKind()))
+  if (!isOpenACCComputeDirectiveKind(Clause.getDirectiveKind()) &&
+      Clause.getDirectiveKind() != OpenACCDirectiveKind::Loop)
     return isNotImplemented();
 
+  // OpenACC 3.3 Section 2.9.11: A reduction clause may not appear on a loop
+  // directive that has a gang clause and is within a compute construct that has
+  // a num_gangs clause with more than one explicit argument.
+  if (Clause.getDirectiveKind() == OpenACCDirectiveKind::Loop &&
+      SemaRef.getActiveComputeConstructInfo().Kind !=
+          OpenACCDirectiveKind::Invalid) {
+    // num_gangs clause on the active compute construct.
+    auto *NumGangsClauseItr =
+        llvm::find_if(SemaRef.getActiveComputeConstructInfo().Clauses,
+                      llvm::IsaPred<OpenACCNumGangsClause>);
+
+    auto *GangClauseItr =
+        llvm::find_if(ExistingClauses, llvm::IsaPred<OpenACCGangClause>);
+
+    if (GangClauseItr != ExistingClauses.end() &&
+        NumGangsClauseItr !=
+            SemaRef.getActiveComputeConstructInfo().Clauses.end() &&
+        cast<OpenACCNumGangsClause>(*NumGangsClauseItr)->getIntExprs().size() >
+            1) {
+      SemaRef.Diag(Clause.getBeginLoc(),
+                   diag::err_acc_gang_reduction_numgangs_conflict)
+          << OpenACCClauseKind::Reduction << OpenACCClauseKind::Gang;
+      SemaRef.Diag((*GangClauseItr)->getBeginLoc(),
+                   diag::note_acc_previous_clause_here);
+      SemaRef.Diag((*NumGangsClauseItr)->getBeginLoc(),
+                   diag::note_acc_previous_clause_here);
+      return nullptr;
+    }
+  }
+
+  // OpenACC3.3 Section 2.9.11: If a variable is involved in a reduction that
+  // spans multiple nested loops where two or more of those loops have
+  // associated loop directives, a reduction clause containing that variable
+  // must appear on each of those loop directives.
+  //
+  // This can't really be implemented in the CFE, as this requires a level of
+  // rechability/useage analysis that we're not really wanting to get into.
+  // Additionally, I'm alerted that this restriction is one that the middle-end
+  // can just 'figure out' as an extension and isn't really necessary.
+  //
+  // OpenACC3.3 Section 2.9.11: Every 'var' in a reduction clause appearing on
+  // an orphaned loop construct must be private.
+  //
+  // This again is something we cannot really diagnose, as it requires we see
+  // all the uses/scopes of all variables referenced.  The middle end/MLIR might
+  // be able to diagnose this.
+
   // OpenACC 3.3 Section 2.5.4:
   // A reduction clause may not appear on a parallel construct with a
   // num_gangs clause that has more than one argument.
@@ -1394,15 +1472,17 @@ OpenACCClause *SemaOpenACCClauseVisitor::VisitReductionClause(
   SmallVector<Expr *> ValidVars;
 
   for (Expr *Var : Clause.getVarList()) {
-    ExprResult Res = SemaRef.CheckReductionVar(Var);
+    ExprResult Res = SemaRef.CheckReductionVar(Clause.getDirectiveKind(),
+                                               Clause.getReductionOp(), Var);
 
     if (Res.isUsable())
       ValidVars.push_back(Res.get());
   }
 
-  return OpenACCReductionClause::Create(
-      Ctx, Clause.getBeginLoc(), Clause.getLParenLoc(), Clause.getReductionOp(),
-      ValidVars, Clause.getEndLoc());
+  return SemaRef.CheckReductionClause(
+      ExistingClauses, Clause.getDirectiveKind(), Clause.getBeginLoc(),
+      Clause.getLParenLoc(), Clause.getReductionOp(), ValidVars,
+      Clause.getEndLoc());
 }
 
 OpenACCClause *SemaOpenACCClauseVisitor::VisitCollapseClause(
@@ -1425,6 +1505,16 @@ OpenACCClause *SemaOpenACCClauseVisitor::VisitCollapseClause(
                                        LoopCount.get(), Clause.getEndLoc());
 }
 
+void CollectActiveReductionClauses(
+    llvm::SmallVector<OpenACCReductionClause *> &ActiveClauses,
+    ArrayRef<OpenACCClause *> CurClauses) {
+  for (auto *CurClause : CurClauses) {
+    if (auto *RedClause = dyn_cast<OpenACCReductionClause>(CurClause);
+        RedClause && !RedClause->getVarList().empty())
+      ActiveClauses.push_back(RedClause);
+  }
+}
+
 } // namespace
 
 SemaOpenACC::SemaOpenACC(Sema &S) : SemaBase(S) {}
@@ -1437,11 +1527,14 @@ SemaOpenACC::AssociatedStmtRAII::AssociatedStmtRAII(
       DirKind(DK), OldLoopGangClauseOnKernelLoc(S.LoopGangClauseOnKernelLoc),
       OldLoopWorkerClauseLoc(S.LoopWorkerClauseLoc),
       OldLoopVectorClauseLoc(S.LoopVectorClauseLoc),
+      ActiveReductionClauses(S.ActiveReductionClauses),
       LoopRAII(SemaRef, /*PreserveDepth=*/false) {
+
   // Compute constructs end up taking their 'loop'.
   if (DirKind == OpenACCDirectiveKind::Parallel ||
       DirKind == OpenACCDirectiveKind::Serial ||
       DirKind == OpenACCDirectiveKind::Kernels) {
+    CollectActiveReductionClauses(S.ActiveReductionClauses, Clauses);
     SemaRef.ActiveComputeConstructInfo.Kind = DirKind;
     SemaRef.ActiveComputeConstructInfo.Clauses = Clauses;
     SemaRef.ParentlessLoopConstructs.swap(ParentlessLoopConstructs);
@@ -1456,6 +1549,7 @@ SemaOpenACC::AssociatedStmtRAII::AssociatedStmtRAII(
     SemaRef.LoopWorkerClauseLoc = {};
     SemaRef.LoopVectorClauseLoc = {};
   } else if (DirKind == OpenACCDirectiveKind::Loop) {
+    CollectActiveReductionClauses(S.ActiveReductionClauses, Clauses);
     SetCollapseInfoBeforeAssociatedStmt(UnInstClauses, Clauses);
     SetTileInfoBeforeAssociatedStmt(UnInstClauses, Clauses);
 
@@ -1559,6 +1653,7 @@ SemaOpenACC::AssociatedStmtRAII::~AssociatedStmtRAII() {
   SemaRef.LoopGangClauseOnKernelLoc = OldLoopGangClauseOnKernelLoc;
   SemaRef.LoopWorkerClauseLoc = OldLoopWorkerClauseLoc;
   SemaRef.LoopVectorClauseLoc = OldLoopVectorClauseLoc;
+  SemaRef.ActiveReductionClauses.swap(ActiveReductionClauses);
 
   if (DirKind == OpenACCDirectiveKind::Parallel ||
       DirKind == OpenACCDirectiveKind::Serial ||
@@ -1610,14 +1705,78 @@ SemaOpenACC::ActOnClause(ArrayRef<const OpenACCClause *> ExistingClauses,
 
 }
 
+namespace {
+// Return true if the two vars refer to the same variable, for the purposes of
+// equality checking.
+bool areVarsEqual(Expr *VarExpr1, Expr *VarExpr2) {
+  if (VarExpr1->isInstantiationDependent() ||
+      VarExpr2->isInstantiationDependent())
+    return false;
+
+  VarExpr1 = VarExpr1->IgnoreParenCasts();
+  VarExpr2 = VarExpr2->IgnoreParenCasts();
+
+  // Legal expressions can be: Scalar variable reference, sub-array, array
+  // element, or composite variable member.
+
+  // Sub-array.
+  if (isa<ArraySectionExpr>(VarExpr1)) {
+    auto *Expr2AS = dyn_cast<ArraySectionExpr>(VarExpr2);
+    if (!Expr2AS)
+      return false;
+
+    auto *Expr1AS = cast<ArraySectionExpr>(VarExpr1);
+
+    if (!areVarsEqual(Expr1AS->getBase(), Expr2AS->getBase()))
+      return false;
+    // We could possibly check to see if the ranges aren't overlapping, but it
+    // isn't clear that the rules allow this.
+    return true;
+  }
+
+  // Array-element.
+  if (isa<ArraySubscriptExpr>(VarExpr1)) {
+    auto *Expr2AS = dyn_cast<ArraySubscriptExpr>(VarExpr2);
+    if (!Expr2AS)
+      return false;
+
+    auto *Expr1AS = cast<ArraySubscriptExpr>(VarExpr1);
+
+    if (!areVarsEqual(Expr1AS->getBase(), Expr2AS->getBase()))
+      return false;
+
+    // We could possibly check to see if the elements referenced aren't the
+    // same, but it isn't clear by reading of the standard that this is allowed
+    // (and that the 'var' refered to isn't the array).
+    return true;
+  }
+
+  // Scalar variable reference, or composite variable.
+  if (isa<DeclRefExpr>(VarExpr1)) {
+    auto *Expr2DRE = dyn_cast<DeclRefExpr>(VarExpr2);
+    if (!Expr2DRE)
+      return false;
+
+    auto *Expr1DRE = cast<DeclRefExpr>(VarExpr1);
+
+    return Expr1DRE->getDecl()->getMostRecentDecl() ==
+           Expr2DRE->getDecl()->getMostRecentDecl();
+  }
+
+  llvm_unreachable("Unknown variable type encountered");
+}
+} // namespace
+
 /// OpenACC 3.3 section 2.5.15:
 /// At a mininmum, the supported data types include ... the numerical data types
 /// in C, C++, and Fortran.
 ///
 /// If the reduction var is a composite variable, each
 /// member of the composite variable must be a supported datatype for the
 /// reduction operation.
-ExprResult SemaOpenACC::CheckReductionVar(Expr *VarExpr) {
+ExprResult SemaOpenACC::CheckReductionVar(OpenACCDirectiveKind DirectiveKind,
+                                          OpenACCReductionOperator ReductionOp,
+                                          Expr *VarExpr) {
   VarExpr = VarExpr->IgnoreParenCasts();
 
   auto TypeIsValid = [](QualType Ty) {
@@ -1667,6 +1826,28 @@ ExprResult SemaOpenACC::CheckReductionVar(Expr *VarExpr) {
     return ExprError();
   }
 
+  // OpenACC3.3: 2.9.11: Reduction clauses on nested constructs for the same
+  // reduction 'var' must have the same reduction operator.
+  if (!VarExpr->isInstantiationDependent()) {
+
+    for (const OpenACCReductionClause *RClause : ActiveReductionClauses) {
+      if (RClause->getReductionOp() == ReductionOp)
+        break;
+
+      for (Expr *OldVarExpr : RClause->getVarList()) {
+        if (OldVarExpr->isInstantiationDependent())
+          continue;
+
+        if (areVarsEqual(VarExpr, OldVarExpr)) {
+          Diag(VarExpr->getExprLoc(), diag::err_reduction_op_mismatch)
+              << ReductionOp << RClause->getReductionOp();
+          Diag(OldVarExpr->getExprLoc(), diag::note_acc_previous_clause_here);
+          return ExprError();
+        }
+      }
+    }
+  }
+
   return VarExpr;
 }
 
@@ -2223,6 +2404,81 @@ ExprResult SemaOpenACC::CheckGangExpr(OpenACCGangKind GK, Expr *E) {
   llvm_unreachable("Compute construct directive not handled?");
 }
 
+OpenACCClause *
+SemaOpenACC::CheckGangClause(ArrayRef<const OpenACCClause *> ExistingClauses,
+                             SourceLocation BeginLoc, SourceLocation LParenLoc,
+                             ArrayRef<OpenACCGangKind> GangKinds,
+                             ArrayRef<Expr *> IntExprs, SourceLocation EndLoc) {
+  // OpenACC 3.3 2.9.11: A reduction clause may not appear on a loop directive
+  // that has a gang clause with a dim: argument whose value is greater than 1.
+
+  const auto *ReductionItr =
+      llvm::find_if(ExistingClauses, llvm::IsaPred<OpenACCReductionClause>);
+
+  if (ReductionItr != ExistingClauses.end()) {
+    const auto GangZip = llvm::zip_equal(GangKinds, IntExprs);
+    const auto GangItr = llvm::find_if(GangZip, [](const auto &Tuple) {
+      return std::get<0>(Tuple) == OpenACCGangKind::Dim;
+    });
+
+    if (GangItr != GangZip.end()) {
+      const Expr *DimExpr = std::get<1>(*GangItr);
+
+      assert(
+          (DimExpr->isInstantiationDependent() || isa<ConstantExpr>(DimExpr)) &&
+          "Improperly formed gang argument");
+      if (const auto *DimVal = dyn_cast<ConstantExpr>(DimExpr);
+          DimVal && DimVal->getResultAsAPSInt() > 1) {
+        Diag(DimVal->getBeginLoc(), diag::err_acc_gang_reduction_conflict)
+            << /*gang/reduction=*/0;
+        Diag((*ReductionItr)->getBeginLoc(),
+             diag::note_acc_previous_clause_here);
+        return nullptr;
+      }
+    }
+  }
+
+  return OpenACCGangClause::Create(getASTContext(), BeginLoc, LParenLoc,
+                                   GangKinds, IntExprs, EndLoc);
+}
+
+OpenACCClause *SemaOpenACC::CheckReductionClause(
+    ArrayRef<const OpenACCClause *> ExistingClauses,
+    OpenACCDirectiveKind DirectiveKind, SourceLocation BeginLoc,
+    SourceLocation LParenLoc, OpenACCReductionOperator ReductionOp,
+    ArrayRef<Expr *> Vars, SourceLocation EndLoc) {
+  if (DirectiveKind == OpenACCDirectiveKind::Loop) {
+    // OpenACC 3.3 2.9.11: A reduction clause may not appear on a loop directive
+    // that has a gang clause with a dim: argument whose value is greater
+    // than 1.
+    const auto *GangItr =
+        llvm::find_if(ExistingClauses, llvm::IsaPred<OpenACCGangClause>);
+
+    while (GangItr != ExistingClauses.end()) {
+      auto *GangClause = cast<OpenACCGangClause>(*GangItr);
+      for (unsigned I = 0; I < GangClause->getNumExprs(); ++I) {
+        std::pair<OpenACCGangKind, const Expr *> EPair = GangClause->getExpr(I);
+        // We know there is only 1 on this gang, so move onto the next gang.
+        if (EPair.first != OpenACCGangKind::Dim)
+          break;
+
+        if (const auto *DimVal = dyn_cast<ConstantExpr>(EPair.second);
+            DimVal && DimVal->getResultAsAPSInt() > 1) {
+          Diag(BeginLoc, diag::err_acc_gang_reduction_conflict)
+              << /*reduction/gang=*/1;
+          Diag((*GangItr)->getBeginLoc(), diag::note_acc_previous_clause_here);
+          return nullptr;
+        }
+      }
+      ++GangItr;
+    }
+  }
+
+  auto *Ret = OpenACCReductionClause::Create(
+      getASTContext(), BeginLoc, LParenLoc, ReductionOp, Vars, EndLoc);
+  return Ret;
+}
+
 ExprResult SemaOpenACC::CheckTileSizeExpr(Expr *SizeExpr) {
   if (!SizeExpr)
     return ExprError();

diff --git a/clang/lib/Sema/SemaSYCL.cpp b/clang/lib/Sema/SemaSYCL.cpp
@@ -198,3 +198,12 @@ void SemaSYCL::handleKernelAttr(Decl *D, const ParsedAttr &AL) {
 
   handleSimpleAttribute<SYCLKernelAttr>(*this, D, AL);
 }
+
+void SemaSYCL::handleKernelEntryPointAttr(Decl *D, const ParsedAttr &AL) {
+  ParsedType PT = AL.getTypeArg();
+  TypeSourceInfo *TSI = nullptr;
+  (void)SemaRef.GetTypeFromParser(PT, &TSI);
+  assert(TSI && "no type source info for attribute argument");
+  D->addAttr(::new (SemaRef.Context)
+                 SYCLKernelEntryPointAttr(SemaRef.Context, AL, TSI));
+}
diff --git a/clang/lib/Sema/SemaTemplateInstantiate.cpp b/clang/lib/Sema/SemaTemplateInstantiate.cpp
@@ -343,7 +343,7 @@ struct TemplateInstantiationArgumentCollecter
       // If this function was instantiated from a specialized member that is
       // a function template, we're done.
       assert(FD->getPrimaryTemplate() && "No function template?");
-      if (FD->getPrimaryTemplate()->hasMemberSpecialization())
+      if (FD->getPrimaryTemplate()->isMemberSpecialization())
         return Done();
 
       // If this function is a generic lambda specialization, we are done.
@@ -442,11 +442,11 @@ struct TemplateInstantiationArgumentCollecter
         Specialized = CTSD->getSpecializedTemplateOrPartial();
     if (auto *CTPSD =
             Specialized.dyn_cast<ClassTemplatePartialSpecializationDecl *>()) {
-      if (CTPSD->hasMemberSpecialization())
+      if (CTPSD->isMemberSpecialization())
         return Done();
     } else {
       auto *CTD = Specialized.get<ClassTemplateDecl *>();
-      if (CTD->hasMemberSpecialization())
+      if (CTD->isMemberSpecialization())
         return Done();
     }
     return UseNextDecl(CTSD);
@@ -478,11 +478,11 @@ struct TemplateInstantiationArgumentCollecter
         Specialized = VTSD->getSpecializedTemplateOrPartial();
     if (auto *VTPSD =
             Specialized.dyn_cast<VarTemplatePartialSpecializationDecl *>()) {
-      if (VTPSD->hasMemberSpecialization())
+      if (VTPSD->isMemberSpecialization())
         return Done();
     } else {
       auto *VTD = Specialized.get<VarTemplateDecl *>();
-      if (VTD->hasMemberSpecialization())
+      if (VTD->isMemberSpecialization())
         return Done();
     }
     return UseNextDecl(VTSD);
@@ -4141,7 +4141,7 @@ getPatternForClassTemplateSpecialization(
   CXXRecordDecl *Pattern = nullptr;
   Specialized = ClassTemplateSpec->getSpecializedTemplateOrPartial();
   if (auto *CTD = Specialized.dyn_cast<ClassTemplateDecl *>()) {
-    while (!CTD->hasMemberSpecialization()) {
+    while (!CTD->isMemberSpecialization()) {
       if (auto *NewCTD = CTD->getInstantiatedFromMemberTemplate())
         CTD = NewCTD;
       else
@@ -4151,7 +4151,7 @@ getPatternForClassTemplateSpecialization(
   } else if (auto *CTPSD =
                  Specialized
                      .dyn_cast<ClassTemplatePartialSpecializationDecl *>()) {
-    while (!CTPSD->hasMemberSpecialization()) {
+    while (!CTPSD->isMemberSpecialization()) {
       if (auto *NewCTPSD = CTPSD->getInstantiatedFromMemberTemplate())
         CTPSD = NewCTPSD;
       else