diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
@@ -802,9 +802,11 @@ def B : JoinedOrSeparate<["-"], "B">, MetaVarName<"<prefix>">,
     HelpText<"Search $prefix$file for executables, libraries, and data files. "
     "If $prefix is a directory, search $prefix/$file">;
 def gcc_install_dir_EQ : Joined<["--"], "gcc-install-dir=">,
+  Visibility<[ClangOption, FlangOption]>,
   HelpText<"Use GCC installation in the specified directory. The directory ends with path components like 'lib{,32,64}/gcc{,-cross}/$triple/$version'. "
   "Note: executables (e.g. ld) used by the compiler are not overridden by the selected GCC installation">;
 def gcc_toolchain : Joined<["--"], "gcc-toolchain=">, Flags<[NoXarchOption]>,
+  Visibility<[ClangOption, FlangOption]>,
   HelpText<"Specify a directory where Clang can find 'include' and 'lib{,32,64}/gcc{,-cross}/$triple/$version'. "
   "Clang will use the GCC installation with the largest version">;
 def gcc_triple_EQ : Joined<["--"], "gcc-triple=">,

diff --git a/clang/include/clang/Frontend/FrontendOptions.h b/clang/include/clang/Frontend/FrontendOptions.h
@@ -580,7 +580,9 @@ class FrontendOptions {
         BuildingImplicitModuleUsesLock(true), ModulesEmbedAllFiles(false),
         IncludeTimestamps(true), UseTemporary(true),
         AllowPCMWithCompilerErrors(false), ModulesShareFileManager(true),
-        TimeTraceGranularity(500) {}
+        EmitSymbolGraph(false), EmitExtensionSymbolGraphs(false),
+        EmitSymbolGraphSymbolLabelsForTesting(false),
+        EmitPrettySymbolGraphs(false), TimeTraceGranularity(500) {}
 
   /// getInputKindForExtension - Return the appropriate input kind for a file
   /// extension. For example, "c" would return Language::C.

diff --git a/clang/lib/AST/TypePrinter.cpp b/clang/lib/AST/TypePrinter.cpp
@@ -1746,14 +1746,15 @@ void TypePrinter::printPackExpansionAfter(const PackExpansionType *T,
 static void printCountAttributedImpl(const CountAttributedType *T,
                                      raw_ostream &OS,
                                      const PrintingPolicy &Policy) {
+  OS << ' ';
   if (T->isCountInBytes() && T->isOrNull())
-    OS << " __sized_by_or_null(";
+    OS << "__sized_by_or_null(";
   else if (T->isCountInBytes())
-    OS << " __sized_by(";
+    OS << "__sized_by(";
   else if (T->isOrNull())
-    OS << " __counted_by_or_null(";
+    OS << "__counted_by_or_null(";
   else
-    OS << " __counted_by(";
+    OS << "__counted_by(";
   if (T->getCountExpr())
     T->getCountExpr()->printPretty(OS, nullptr, Policy);
   OS << ')';
@@ -1762,14 +1763,14 @@ static void printCountAttributedImpl(const CountAttributedType *T,
 void TypePrinter::printCountAttributedBefore(const CountAttributedType *T,
                                              raw_ostream &OS) {
   printBefore(T->desugar(), OS);
-  if (!T->desugar()->isArrayType())
+  if (!T->isArrayType())
     printCountAttributedImpl(T, OS, Policy);
 }
 
 void TypePrinter::printCountAttributedAfter(const CountAttributedType *T,
                                             raw_ostream &OS) {
   printAfter(T->desugar(), OS);
-  if (T->desugar()->isArrayType())
+  if (T->isArrayType())
     printCountAttributedImpl(T, OS, Policy);
 }
 

diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -166,17 +166,16 @@ static Value *joinDistinctValues(QualType Type, Value &Val1,
   return JoinedVal;
 }
 
-// When widening does not change `Current`, return value will equal `&Prev`.
-static Value &widenDistinctValues(QualType Type, Value &Prev,
-                                  const Environment &PrevEnv, Value &Current,
-                                  Environment &CurrentEnv,
-                                  Environment::ValueModel &Model) {
+static WidenResult widenDistinctValues(QualType Type, Value &Prev,
+                                       const Environment &PrevEnv,
+                                       Value &Current, Environment &CurrentEnv,
+                                       Environment::ValueModel &Model) {
   // Boolean-model widening.
   if (auto *PrevBool = dyn_cast<BoolValue>(&Prev)) {
-    // If previous value was already Top, re-use that to (implicitly) indicate
-    // that no change occurred.
     if (isa<TopBoolValue>(Prev))
-      return Prev;
+      // Safe to return `Prev` here, because Top is never dependent on the
+      // environment.
+      return {&Prev, LatticeEffect::Unchanged};
 
     // We may need to widen to Top, but before we do so, check whether both
     // values are implied to be either true or false in the current environment.
@@ -185,22 +184,24 @@ static Value &widenDistinctValues(QualType Type, Value &Prev,
     bool TruePrev = PrevEnv.proves(PrevBool->formula());
     bool TrueCur = CurrentEnv.proves(CurBool.formula());
     if (TruePrev && TrueCur)
-      return CurrentEnv.getBoolLiteralValue(true);
+      return {&CurrentEnv.getBoolLiteralValue(true), LatticeEffect::Unchanged};
     if (!TruePrev && !TrueCur &&
         PrevEnv.proves(PrevEnv.arena().makeNot(PrevBool->formula())) &&
         CurrentEnv.proves(CurrentEnv.arena().makeNot(CurBool.formula())))
-      return CurrentEnv.getBoolLiteralValue(false);
+      return {&CurrentEnv.getBoolLiteralValue(false), LatticeEffect::Unchanged};
 
-    return CurrentEnv.makeTopBoolValue();
+    return {&CurrentEnv.makeTopBoolValue(), LatticeEffect::Changed};
   }
 
   // FIXME: Add other built-in model widening.
 
   // Custom-model widening.
-  if (auto *W = Model.widen(Type, Prev, PrevEnv, Current, CurrentEnv))
-    return *W;
+  if (auto Result = Model.widen(Type, Prev, PrevEnv, Current, CurrentEnv))
+    return *Result;
 
-  return equateUnknownValues(Prev.getKind()) ? Prev : Current;
+  return {&Current, equateUnknownValues(Prev.getKind())
+                        ? LatticeEffect::Unchanged
+                        : LatticeEffect::Changed};
 }
 
 // Returns whether the values in `Map1` and `Map2` compare equal for those
@@ -271,7 +272,7 @@ llvm::MapVector<Key, Value *>
 widenKeyToValueMap(const llvm::MapVector<Key, Value *> &CurMap,
                    const llvm::MapVector<Key, Value *> &PrevMap,
                    Environment &CurEnv, const Environment &PrevEnv,
-                   Environment::ValueModel &Model, LatticeJoinEffect &Effect) {
+                   Environment::ValueModel &Model, LatticeEffect &Effect) {
   llvm::MapVector<Key, Value *> WidenedMap;
   for (auto &Entry : CurMap) {
     Key K = Entry.first;
@@ -290,11 +291,11 @@ widenKeyToValueMap(const llvm::MapVector<Key, Value *> &CurMap,
       continue;
     }
 
-    Value &WidenedVal = widenDistinctValues(K->getType(), *PrevIt->second,
-                                            PrevEnv, *Val, CurEnv, Model);
-    WidenedMap.insert({K, &WidenedVal});
-    if (&WidenedVal != PrevIt->second)
-      Effect = LatticeJoinEffect::Changed;
+    auto [WidenedVal, ValEffect] = widenDistinctValues(
+        K->getType(), *PrevIt->second, PrevEnv, *Val, CurEnv, Model);
+    WidenedMap.insert({K, WidenedVal});
+    if (ValEffect == LatticeEffect::Changed)
+      Effect = LatticeEffect::Changed;
   }
 
   return WidenedMap;
@@ -617,15 +618,15 @@ bool Environment::equivalentTo(const Environment &Other,
   return true;
 }
 
-LatticeJoinEffect Environment::widen(const Environment &PrevEnv,
-                                     Environment::ValueModel &Model) {
+LatticeEffect Environment::widen(const Environment &PrevEnv,
+                                 Environment::ValueModel &Model) {
   assert(DACtx == PrevEnv.DACtx);
   assert(ReturnVal == PrevEnv.ReturnVal);
   assert(ReturnLoc == PrevEnv.ReturnLoc);
   assert(ThisPointeeLoc == PrevEnv.ThisPointeeLoc);
   assert(CallStack == PrevEnv.CallStack);
 
-  auto Effect = LatticeJoinEffect::Unchanged;
+  auto Effect = LatticeEffect::Unchanged;
 
   // By the API, `PrevEnv` is a previous version of the environment for the same
   // block, so we have some guarantees about its shape. In particular, it will
@@ -646,7 +647,7 @@ LatticeJoinEffect Environment::widen(const Environment &PrevEnv,
       ExprToLoc.size() != PrevEnv.ExprToLoc.size() ||
       ExprToVal.size() != PrevEnv.ExprToVal.size() ||
       LocToVal.size() != PrevEnv.LocToVal.size())
-    Effect = LatticeJoinEffect::Changed;
+    Effect = LatticeEffect::Changed;
 
   return Effect;
 }

diff --git a/clang/lib/Basic/Targets/Mips.h b/clang/lib/Basic/Targets/Mips.h
@@ -318,6 +318,7 @@ class LLVM_LIBRARY_VISIBILITY MipsTargetInfo : public TargetInfo {
     FPMode = isFP64Default() ? FP64 : FPXX;
     NoOddSpreg = false;
     bool OddSpregGiven = false;
+    bool StrictAlign = false;
 
     for (const auto &Feature : Features) {
       if (Feature == "+single-float")
@@ -330,6 +331,10 @@ class LLVM_LIBRARY_VISIBILITY MipsTargetInfo : public TargetInfo {
         IsMicromips = true;
       else if (Feature == "+mips32r6" || Feature == "+mips64r6")
         HasUnalignedAccess = true;
+      // We cannot be sure that the order of strict-align vs mips32r6.
+      // Thus we need an extra variable here.
+      else if (Feature == "+strict-align")
+        StrictAlign = true;
       else if (Feature == "+dsp")
         DspRev = std::max(DspRev, DSP1);
       else if (Feature == "+dspr2")
@@ -368,6 +373,9 @@ class LLVM_LIBRARY_VISIBILITY MipsTargetInfo : public TargetInfo {
     if (FPMode == FPXX && !OddSpregGiven)
       NoOddSpreg = true;
 
+    if (StrictAlign)
+      HasUnalignedAccess = false;
+
     setDataLayout();
 
     return true;

diff --git a/clang/lib/CodeGen/BackendUtil.cpp b/clang/lib/CodeGen/BackendUtil.cpp
@@ -100,9 +100,6 @@ using namespace llvm;
 namespace llvm {
 extern cl::opt<bool> PrintPipelinePasses;
 
-static cl::opt<bool> ClRemoveTraps("clang-remove-traps", cl::Optional,
-                                   cl::desc("Insert remove-traps pass."));
-
 // Experiment to move sanitizers earlier.
 static cl::opt<bool> ClSanitizeOnOptimizerEarlyEP(
     "sanitizer-early-opt-ep", cl::Optional,
@@ -750,7 +747,7 @@ static void addSanitizers(const Triple &TargetTriple,
     PB.registerOptimizerLastEPCallback(SanitizersCallback);
   }
 
-  if (ClRemoveTraps) {
+  if (RemoveTrapsPass::IsRequested()) {
     // We can optimize after inliner, and PGO profile matching. The hook below
     // is called at the end `buildFunctionSimplificationPipeline`, which called
     // from `buildInlinerPipeline`, which called after profile matching.

diff --git a/clang/lib/CodeGen/CGBlocks.cpp b/clang/lib/CodeGen/CGBlocks.cpp
@@ -962,7 +962,7 @@ llvm::Value *CodeGenFunction::EmitBlockLiteral(const CGBlockInfo &blockInfo) {
       }
 
     // If it's a reference variable, copy the reference into the block field.
-    } else if (auto refType = type->getAs<ReferenceType>()) {
+    } else if (type->getAs<ReferenceType>()) {
       Builder.CreateStore(src.emitRawPointer(*this), blockField);
 
       // If type is const-qualified, copy the value into the block field.

diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -7281,8 +7281,6 @@ static const std::pair<unsigned, unsigned> NEONEquivalentIntrinsicMap[] = {
   { NEON::BI__builtin_neon_vabdq_f16, NEON::BI__builtin_neon_vabdq_v, },
   { NEON::BI__builtin_neon_vabs_f16, NEON::BI__builtin_neon_vabs_v, },
   { NEON::BI__builtin_neon_vabsq_f16, NEON::BI__builtin_neon_vabsq_v, },
-  { NEON::BI__builtin_neon_vbsl_f16, NEON::BI__builtin_neon_vbsl_v, },
-  { NEON::BI__builtin_neon_vbslq_f16, NEON::BI__builtin_neon_vbslq_v, },
   { NEON::BI__builtin_neon_vcage_f16, NEON::BI__builtin_neon_vcage_v, },
   { NEON::BI__builtin_neon_vcageq_f16, NEON::BI__builtin_neon_vcageq_v, },
   { NEON::BI__builtin_neon_vcagt_f16, NEON::BI__builtin_neon_vcagt_v, },
@@ -7301,8 +7299,6 @@ static const std::pair<unsigned, unsigned> NEONEquivalentIntrinsicMap[] = {
   { NEON::BI__builtin_neon_vclezq_f16, NEON::BI__builtin_neon_vclezq_v, },
   { NEON::BI__builtin_neon_vcltz_f16, NEON::BI__builtin_neon_vcltz_v, },
   { NEON::BI__builtin_neon_vcltzq_f16, NEON::BI__builtin_neon_vcltzq_v, },
-  { NEON::BI__builtin_neon_vext_f16, NEON::BI__builtin_neon_vext_v, },
-  { NEON::BI__builtin_neon_vextq_f16, NEON::BI__builtin_neon_vextq_v, },
   { NEON::BI__builtin_neon_vfma_f16, NEON::BI__builtin_neon_vfma_v, },
   { NEON::BI__builtin_neon_vfma_lane_f16, NEON::BI__builtin_neon_vfma_lane_v, },
   { NEON::BI__builtin_neon_vfma_laneq_f16, NEON::BI__builtin_neon_vfma_laneq_v, },
@@ -7405,12 +7401,6 @@ static const std::pair<unsigned, unsigned> NEONEquivalentIntrinsicMap[] = {
   { NEON::BI__builtin_neon_vst4_lane_bf16, NEON::BI__builtin_neon_vst4_lane_v },
   { NEON::BI__builtin_neon_vst4q_bf16, NEON::BI__builtin_neon_vst4q_v },
   { NEON::BI__builtin_neon_vst4q_lane_bf16, NEON::BI__builtin_neon_vst4q_lane_v },
-  { NEON::BI__builtin_neon_vtrn_f16, NEON::BI__builtin_neon_vtrn_v, },
-  { NEON::BI__builtin_neon_vtrnq_f16, NEON::BI__builtin_neon_vtrnq_v, },
-  { NEON::BI__builtin_neon_vuzp_f16, NEON::BI__builtin_neon_vuzp_v, },
-  { NEON::BI__builtin_neon_vuzpq_f16, NEON::BI__builtin_neon_vuzpq_v, },
-  { NEON::BI__builtin_neon_vzip_f16, NEON::BI__builtin_neon_vzip_v, },
-  { NEON::BI__builtin_neon_vzipq_f16, NEON::BI__builtin_neon_vzipq_v, },
   // The mangling rules cause us to have one ID for each type for vldap1(q)_lane
   // and vstl1(q)_lane, but codegen is equivalent for all of them. Choose an
   // arbitrary one to be handled as tha canonical variation.

diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
@@ -5580,44 +5580,11 @@ LValue CodeGenFunction::EmitBinaryOperatorLValue(const BinaryOperator *E) {
       break;
     }
 
-    // TODO: Can we de-duplicate this code with the corresponding code in
-    // CGExprScalar, similar to the way EmitCompoundAssignmentLValue works?
-    RValue RV;
-    llvm::Value *Previous = nullptr;
-    QualType SrcType = E->getRHS()->getType();
-    // Check if LHS is a bitfield, if RHS contains an implicit cast expression
-    // we want to extract that value and potentially (if the bitfield sanitizer
-    // is enabled) use it to check for an implicit conversion.
-    if (E->getLHS()->refersToBitField()) {
-      llvm::Value *RHS =
-          EmitWithOriginalRHSBitfieldAssignment(E, Previous, &SrcType);
-      RV = RValue::get(RHS);
-    } else
-      RV = EmitAnyExpr(E->getRHS());
-
+    RValue RV = EmitAnyExpr(E->getRHS());
     LValue LV = EmitCheckedLValue(E->getLHS(), TCK_Store);
-
     if (RV.isScalar())
       EmitNullabilityCheck(LV, RV.getScalarVal(), E->getExprLoc());
-
-    if (LV.isBitField()) {
-      llvm::Value *Result;
-      // If bitfield sanitizers are enabled we want to use the result
-      // to check whether a truncation or sign change has occurred.
-      if (SanOpts.has(SanitizerKind::ImplicitBitfieldConversion))
-        EmitStoreThroughBitfieldLValue(RV, LV, &Result);
-      else
-        EmitStoreThroughBitfieldLValue(RV, LV);
-
-      // If the expression contained an implicit conversion, make sure
-      // to use the value before the scalar conversion.
-      llvm::Value *Src = Previous ? Previous : RV.getScalarVal();
-      QualType DstType = E->getLHS()->getType();
-      EmitBitfieldConversionCheck(Src, SrcType, Result, DstType,
-                                  LV.getBitFieldInfo(), E->getExprLoc());
-    } else
-      EmitStoreThroughLValue(RV, LV);
-
+    EmitStoreThroughLValue(RV, LV);
     if (getLangOpts().OpenMP)
       CGM.getOpenMPRuntime().checkAndEmitLastprivateConditional(*this,
                                                                 E->getLHS());

diff --git a/clang/lib/CodeGen/CGExprComplex.cpp b/clang/lib/CodeGen/CGExprComplex.cpp
@@ -319,12 +319,12 @@ class ComplexExprEmitter
     // doubles the exponent of SmallerType.LargestFiniteVal)
     if (llvm::APFloat::semanticsMaxExponent(ElementTypeSemantics) * 2 + 1 <=
         llvm::APFloat::semanticsMaxExponent(HigherElementTypeSemantics)) {
+      FPHasBeenPromoted = true;
       return CGF.getContext().getComplexType(HigherElementType);
     } else {
-      FPHasBeenPromoted = true;
       DiagnosticsEngine &Diags = CGF.CGM.getDiags();
       Diags.Report(diag::warn_next_larger_fp_type_same_size_than_fp);
-      return CGF.getContext().getComplexType(ElementType);
+      return QualType();
     }
   }
 
@@ -1037,7 +1037,7 @@ ComplexPairTy ComplexExprEmitter::EmitBinDiv(const BinOpInfo &Op) {
       LHSi = llvm::Constant::getNullValue(RHSi->getType());
     if (Op.FPFeatures.getComplexRange() == LangOptions::CX_Improved ||
         (Op.FPFeatures.getComplexRange() == LangOptions::CX_Promoted &&
-         FPHasBeenPromoted))
+         !FPHasBeenPromoted))
       return EmitRangeReductionDiv(LHSr, LHSi, RHSr, RHSi);
     else if (Op.FPFeatures.getComplexRange() == LangOptions::CX_Basic ||
              Op.FPFeatures.getComplexRange() == LangOptions::CX_Promoted)

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.cpp b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -2648,9 +2648,9 @@ void CGOpenMPRuntime::emitDistributeStaticInit(
 void CGOpenMPRuntime::emitForStaticFinish(CodeGenFunction &CGF,
                                           SourceLocation Loc,
                                           OpenMPDirectiveKind DKind) {
-  assert(DKind == OMPD_distribute || DKind == OMPD_for ||
-         DKind == OMPD_sections &&
-             "Expected distribute, for, or sections directive kind");
+  assert((DKind == OMPD_distribute || DKind == OMPD_for ||
+          DKind == OMPD_sections) &&
+         "Expected distribute, for, or sections directive kind");
   if (!CGF.HaveInsertPoint())
     return;
   // Call __kmpc_for_static_fini(ident_t *loc, kmp_int32 tid);

diff --git a/clang/lib/CodeGen/CodeGenFunction.h b/clang/lib/CodeGen/CodeGenFunction.h
@@ -2786,21 +2786,6 @@ class CodeGenFunction : public CodeGenTypeCache {
   /// expression and compare the result against zero, returning an Int1Ty value.
   llvm::Value *EvaluateExprAsBool(const Expr *E);
 
-  /// Retrieve the implicit cast expression of the rhs in a binary operator
-  /// expression by passing pointers to Value and QualType
-  /// This is used for implicit bitfield conversion checks, which
-  /// must compare with the value before potential truncation.
-  llvm::Value *EmitWithOriginalRHSBitfieldAssignment(const BinaryOperator *E,
-                                                     llvm::Value *Previous,
-                                                     QualType *SrcType);
-
-  /// Emit a check that an [implicit] conversion of a bitfield. It is not UB,
-  /// so we use the value after conversion.
-  void EmitBitfieldConversionCheck(llvm::Value *Src, QualType SrcType,
-                                   llvm::Value *Dst, QualType DstType,
-                                   const CGBitFieldInfo &Info,
-                                   SourceLocation Loc);
-
   /// EmitIgnoredExpr - Emit an expression in a context which ignores the result.
   void EmitIgnoredExpr(const Expr *E);
 

diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -5882,7 +5882,8 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
       CM = "large";
     if (Triple.isAArch64(64)) {
       Ok = CM == "tiny" || CM == "small" || CM == "large";
-      if (CM == "large" && RelocationModel != llvm::Reloc::Static)
+      if (CM == "large" && !Triple.isOSBinFormatMachO() &&
+          RelocationModel != llvm::Reloc::Static)
         D.Diag(diag::err_drv_argument_only_allowed_with)
             << A->getAsString(Args) << "-fno-pic";
     } else if (Triple.isLoongArch()) {

diff --git a/clang/lib/Headers/__stddef_unreachable.h b/clang/lib/Headers/__stddef_unreachable.h
@@ -7,6 +7,8 @@
  *===-----------------------------------------------------------------------===
  */
 
+#ifndef __cplusplus
+
 /*
  * When -fbuiltin-headers-in-system-modules is set this is a non-modular header
  * and needs to behave as if it was textual.
@@ -15,3 +17,5 @@
     (__has_feature(modules) && !__building_module(_Builtin_stddef))
 #define unreachable() __builtin_unreachable()
 #endif
+
+#endif
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
@@ -2751,7 +2751,7 @@ Sema::ActOnIdExpression(Scope *S, CXXScopeSpec &SS,
         QualType type = VD->getType().getNonReferenceType();
         // This will eventually be translated into MemberExpr upon
         // the use of instantiated struct fields.
-        return BuildDeclRefExpr(VD, type, VK_PRValue, NameLoc);
+        return BuildDeclRefExpr(VD, type, VK_LValue, NameLoc);
       }
     }
   }

diff --git a/clang/lib/Sema/SemaTemplate.cpp b/clang/lib/Sema/SemaTemplate.cpp
@@ -2711,7 +2711,6 @@ SmallVector<unsigned> TemplateParamsReferencedInTemplateArgumentList(
         : TemplateParams(TemplateParams.begin(), TemplateParams.end()) {}
 
     bool VisitTemplateTypeParmType(TemplateTypeParmType *TTP) {
-      TTP->getIndex();
       MarkAppeared(TTP->getDecl());
       return true;
     }

diff --git a/clang/test/CodeGen/aarch64-v8.2a-neon-intrinsics-generic.c b/clang/test/CodeGen/aarch64-v8.2a-neon-intrinsics-generic.c
diff --git a/clang/test/CodeGen/aarch64-v8.2a-neon-intrinsics.c b/clang/test/CodeGen/aarch64-v8.2a-neon-intrinsics.c
diff --git a/clang/test/CodeGen/allow-ubsan-check.c b/clang/test/CodeGen/allow-ubsan-check.c
@@ -0,0 +1,207 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -emit-llvm -o - %s -fsanitize=signed-integer-overflow,integer-divide-by-zero,null | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -emit-llvm -o - %s -fsanitize=signed-integer-overflow,integer-divide-by-zero,null -fsanitize-trap=signed-integer-overflow,integer-divide-by-zero,null | FileCheck %s --check-prefixes=TRAP
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -emit-llvm -o - %s -fsanitize=signed-integer-overflow,integer-divide-by-zero,null -fsanitize-recover=signed-integer-overflow,integer-divide-by-zero,null | FileCheck %s --check-prefixes=RECOVER
+
+
+// CHECK-LABEL: define dso_local i32 @div(
+// CHECK-SAME: i32 noundef [[X:%.*]], i32 noundef [[Y:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[Y_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    store i32 [[X]], ptr [[X_ADDR]], align 4
+// CHECK-NEXT:    store i32 [[Y]], ptr [[Y_ADDR]], align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[X_ADDR]], align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[Y_ADDR]], align 4
+// CHECK-NEXT:    [[TMP2:%.*]] = icmp ne i32 [[TMP1]], 0, !nosanitize [[META2:![0-9]+]]
+// CHECK-NEXT:    [[TMP3:%.*]] = icmp ne i32 [[TMP0]], -2147483648, !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP4:%.*]] = icmp ne i32 [[TMP1]], -1, !nosanitize [[META2]]
+// CHECK-NEXT:    [[OR:%.*]] = or i1 [[TMP3]], [[TMP4]], !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP5:%.*]] = and i1 [[TMP2]], [[OR]], !nosanitize [[META2]]
+// CHECK-NEXT:    br i1 [[TMP5]], label [[CONT:%.*]], label [[HANDLER_DIVREM_OVERFLOW:%.*]], !prof [[PROF3:![0-9]+]], !nosanitize [[META2]]
+// CHECK:       handler.divrem_overflow:
+// CHECK-NEXT:    [[TMP6:%.*]] = zext i32 [[TMP0]] to i64, !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP1]] to i64, !nosanitize [[META2]]
+// CHECK-NEXT:    call void @__ubsan_handle_divrem_overflow_abort(ptr @[[GLOB1:[0-9]+]], i64 [[TMP6]], i64 [[TMP7]]) #[[ATTR3:[0-9]+]], !nosanitize [[META2]]
+// CHECK-NEXT:    unreachable, !nosanitize [[META2]]
+// CHECK:       cont:
+// CHECK-NEXT:    [[DIV:%.*]] = sdiv i32 [[TMP0]], [[TMP1]]
+// CHECK-NEXT:    ret i32 [[DIV]]
+//
+// TRAP-LABEL: define dso_local i32 @div(
+// TRAP-SAME: i32 noundef [[X:%.*]], i32 noundef [[Y:%.*]]) #[[ATTR0:[0-9]+]] {
+// TRAP-NEXT:  entry:
+// TRAP-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4
+// TRAP-NEXT:    [[Y_ADDR:%.*]] = alloca i32, align 4
+// TRAP-NEXT:    store i32 [[X]], ptr [[X_ADDR]], align 4
+// TRAP-NEXT:    store i32 [[Y]], ptr [[Y_ADDR]], align 4
+// TRAP-NEXT:    [[TMP0:%.*]] = load i32, ptr [[X_ADDR]], align 4
+// TRAP-NEXT:    [[TMP1:%.*]] = load i32, ptr [[Y_ADDR]], align 4
+// TRAP-NEXT:    [[TMP2:%.*]] = icmp ne i32 [[TMP1]], 0, !nosanitize [[META2:![0-9]+]]
+// TRAP-NEXT:    [[TMP3:%.*]] = icmp ne i32 [[TMP0]], -2147483648, !nosanitize [[META2]]
+// TRAP-NEXT:    [[TMP4:%.*]] = icmp ne i32 [[TMP1]], -1, !nosanitize [[META2]]
+// TRAP-NEXT:    [[OR:%.*]] = or i1 [[TMP3]], [[TMP4]], !nosanitize [[META2]]
+// TRAP-NEXT:    [[TMP5:%.*]] = and i1 [[TMP2]], [[OR]], !nosanitize [[META2]]
+// TRAP-NEXT:    br i1 [[TMP5]], label [[CONT:%.*]], label [[TRAP:%.*]], !nosanitize [[META2]]
+// TRAP:       trap:
+// TRAP-NEXT:    call void @llvm.ubsantrap(i8 3) #[[ATTR3:[0-9]+]], !nosanitize [[META2]]
+// TRAP-NEXT:    unreachable, !nosanitize [[META2]]
+// TRAP:       cont:
+// TRAP-NEXT:    [[DIV:%.*]] = sdiv i32 [[TMP0]], [[TMP1]]
+// TRAP-NEXT:    ret i32 [[DIV]]
+//
+// RECOVER-LABEL: define dso_local i32 @div(
+// RECOVER-SAME: i32 noundef [[X:%.*]], i32 noundef [[Y:%.*]]) #[[ATTR0:[0-9]+]] {
+// RECOVER-NEXT:  entry:
+// RECOVER-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4
+// RECOVER-NEXT:    [[Y_ADDR:%.*]] = alloca i32, align 4
+// RECOVER-NEXT:    store i32 [[X]], ptr [[X_ADDR]], align 4
+// RECOVER-NEXT:    store i32 [[Y]], ptr [[Y_ADDR]], align 4
+// RECOVER-NEXT:    [[TMP0:%.*]] = load i32, ptr [[X_ADDR]], align 4
+// RECOVER-NEXT:    [[TMP1:%.*]] = load i32, ptr [[Y_ADDR]], align 4
+// RECOVER-NEXT:    [[TMP2:%.*]] = icmp ne i32 [[TMP1]], 0, !nosanitize [[META2:![0-9]+]]
+// RECOVER-NEXT:    [[TMP3:%.*]] = icmp ne i32 [[TMP0]], -2147483648, !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP4:%.*]] = icmp ne i32 [[TMP1]], -1, !nosanitize [[META2]]
+// RECOVER-NEXT:    [[OR:%.*]] = or i1 [[TMP3]], [[TMP4]], !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP5:%.*]] = and i1 [[TMP2]], [[OR]], !nosanitize [[META2]]
+// RECOVER-NEXT:    br i1 [[TMP5]], label [[CONT:%.*]], label [[HANDLER_DIVREM_OVERFLOW:%.*]], !prof [[PROF3:![0-9]+]], !nosanitize [[META2]]
+// RECOVER:       handler.divrem_overflow:
+// RECOVER-NEXT:    [[TMP6:%.*]] = zext i32 [[TMP0]] to i64, !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP1]] to i64, !nosanitize [[META2]]
+// RECOVER-NEXT:    call void @__ubsan_handle_divrem_overflow(ptr @[[GLOB1:[0-9]+]], i64 [[TMP6]], i64 [[TMP7]]) #[[ATTR3:[0-9]+]], !nosanitize [[META2]]
+// RECOVER-NEXT:    br label [[CONT]], !nosanitize [[META2]]
+// RECOVER:       cont:
+// RECOVER-NEXT:    [[DIV:%.*]] = sdiv i32 [[TMP0]], [[TMP1]]
+// RECOVER-NEXT:    ret i32 [[DIV]]
+//
+int div(int x, int y) {
+  return x / y;
+}
+
+// CHECK-LABEL: define dso_local i32 @null(
+// CHECK-SAME: ptr noundef [[X:%.*]]) #[[ATTR0]] {
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[X_ADDR:%.*]] = alloca ptr, align 8
+// CHECK-NEXT:    store ptr [[X]], ptr [[X_ADDR]], align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[X_ADDR]], align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = icmp ne ptr [[TMP0]], null, !nosanitize [[META2]]
+// CHECK-NEXT:    br i1 [[TMP1]], label [[CONT:%.*]], label [[HANDLER_TYPE_MISMATCH:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
+// CHECK:       handler.type_mismatch:
+// CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[TMP0]] to i64, !nosanitize [[META2]]
+// CHECK-NEXT:    call void @__ubsan_handle_type_mismatch_v1_abort(ptr @[[GLOB2:[0-9]+]], i64 [[TMP2]]) #[[ATTR3]], !nosanitize [[META2]]
+// CHECK-NEXT:    unreachable, !nosanitize [[META2]]
+// CHECK:       cont:
+// CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr [[TMP0]], align 4
+// CHECK-NEXT:    ret i32 [[TMP3]]
+//
+// TRAP-LABEL: define dso_local i32 @null(
+// TRAP-SAME: ptr noundef [[X:%.*]]) #[[ATTR0]] {
+// TRAP-NEXT:  entry:
+// TRAP-NEXT:    [[X_ADDR:%.*]] = alloca ptr, align 8
+// TRAP-NEXT:    store ptr [[X]], ptr [[X_ADDR]], align 8
+// TRAP-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[X_ADDR]], align 8
+// TRAP-NEXT:    [[TMP1:%.*]] = icmp ne ptr [[TMP0]], null, !nosanitize [[META2]]
+// TRAP-NEXT:    br i1 [[TMP1]], label [[CONT:%.*]], label [[TRAP:%.*]], !nosanitize [[META2]]
+// TRAP:       trap:
+// TRAP-NEXT:    call void @llvm.ubsantrap(i8 22) #[[ATTR3]], !nosanitize [[META2]]
+// TRAP-NEXT:    unreachable, !nosanitize [[META2]]
+// TRAP:       cont:
+// TRAP-NEXT:    [[TMP2:%.*]] = load i32, ptr [[TMP0]], align 4
+// TRAP-NEXT:    ret i32 [[TMP2]]
+//
+// RECOVER-LABEL: define dso_local i32 @null(
+// RECOVER-SAME: ptr noundef [[X:%.*]]) #[[ATTR0]] {
+// RECOVER-NEXT:  entry:
+// RECOVER-NEXT:    [[X_ADDR:%.*]] = alloca ptr, align 8
+// RECOVER-NEXT:    store ptr [[X]], ptr [[X_ADDR]], align 8
+// RECOVER-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[X_ADDR]], align 8
+// RECOVER-NEXT:    [[TMP1:%.*]] = icmp ne ptr [[TMP0]], null, !nosanitize [[META2]]
+// RECOVER-NEXT:    br i1 [[TMP1]], label [[CONT:%.*]], label [[HANDLER_TYPE_MISMATCH:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
+// RECOVER:       handler.type_mismatch:
+// RECOVER-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[TMP0]] to i64, !nosanitize [[META2]]
+// RECOVER-NEXT:    call void @__ubsan_handle_type_mismatch_v1(ptr @[[GLOB2:[0-9]+]], i64 [[TMP2]]) #[[ATTR3]], !nosanitize [[META2]]
+// RECOVER-NEXT:    br label [[CONT]], !nosanitize [[META2]]
+// RECOVER:       cont:
+// RECOVER-NEXT:    [[TMP3:%.*]] = load i32, ptr [[TMP0]], align 4
+// RECOVER-NEXT:    ret i32 [[TMP3]]
+//
+int null(int* x) {
+  return *x;
+}
+
+// CHECK-LABEL: define dso_local i32 @overflow(
+// CHECK-SAME: i32 noundef [[X:%.*]], i32 noundef [[Y:%.*]]) #[[ATTR0]] {
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[Y_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    store i32 [[X]], ptr [[X_ADDR]], align 4
+// CHECK-NEXT:    store i32 [[Y]], ptr [[Y_ADDR]], align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[X_ADDR]], align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[Y_ADDR]], align 4
+// CHECK-NEXT:    [[TMP2:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP0]], i32 [[TMP1]]), !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP3:%.*]] = extractvalue { i32, i1 } [[TMP2]], 0, !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP4:%.*]] = extractvalue { i32, i1 } [[TMP2]], 1, !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP5:%.*]] = xor i1 [[TMP4]], true, !nosanitize [[META2]]
+// CHECK-NEXT:    br i1 [[TMP5]], label [[CONT:%.*]], label [[HANDLER_ADD_OVERFLOW:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
+// CHECK:       handler.add_overflow:
+// CHECK-NEXT:    [[TMP6:%.*]] = zext i32 [[TMP0]] to i64, !nosanitize [[META2]]
+// CHECK-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP1]] to i64, !nosanitize [[META2]]
+// CHECK-NEXT:    call void @__ubsan_handle_add_overflow_abort(ptr @[[GLOB3:[0-9]+]], i64 [[TMP6]], i64 [[TMP7]]) #[[ATTR3]], !nosanitize [[META2]]
+// CHECK-NEXT:    unreachable, !nosanitize [[META2]]
+// CHECK:       cont:
+// CHECK-NEXT:    ret i32 [[TMP3]]
+//
+// TRAP-LABEL: define dso_local i32 @overflow(
+// TRAP-SAME: i32 noundef [[X:%.*]], i32 noundef [[Y:%.*]]) #[[ATTR0]] {
+// TRAP-NEXT:  entry:
+// TRAP-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4
+// TRAP-NEXT:    [[Y_ADDR:%.*]] = alloca i32, align 4
+// TRAP-NEXT:    store i32 [[X]], ptr [[X_ADDR]], align 4
+// TRAP-NEXT:    store i32 [[Y]], ptr [[Y_ADDR]], align 4
+// TRAP-NEXT:    [[TMP0:%.*]] = load i32, ptr [[X_ADDR]], align 4
+// TRAP-NEXT:    [[TMP1:%.*]] = load i32, ptr [[Y_ADDR]], align 4
+// TRAP-NEXT:    [[TMP2:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP0]], i32 [[TMP1]]), !nosanitize [[META2]]
+// TRAP-NEXT:    [[TMP3:%.*]] = extractvalue { i32, i1 } [[TMP2]], 0, !nosanitize [[META2]]
+// TRAP-NEXT:    [[TMP4:%.*]] = extractvalue { i32, i1 } [[TMP2]], 1, !nosanitize [[META2]]
+// TRAP-NEXT:    [[TMP5:%.*]] = xor i1 [[TMP4]], true, !nosanitize [[META2]]
+// TRAP-NEXT:    br i1 [[TMP5]], label [[CONT:%.*]], label [[TRAP:%.*]], !nosanitize [[META2]]
+// TRAP:       trap:
+// TRAP-NEXT:    call void @llvm.ubsantrap(i8 0) #[[ATTR3]], !nosanitize [[META2]]
+// TRAP-NEXT:    unreachable, !nosanitize [[META2]]
+// TRAP:       cont:
+// TRAP-NEXT:    ret i32 [[TMP3]]
+//
+// RECOVER-LABEL: define dso_local i32 @overflow(
+// RECOVER-SAME: i32 noundef [[X:%.*]], i32 noundef [[Y:%.*]]) #[[ATTR0]] {
+// RECOVER-NEXT:  entry:
+// RECOVER-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4
+// RECOVER-NEXT:    [[Y_ADDR:%.*]] = alloca i32, align 4
+// RECOVER-NEXT:    store i32 [[X]], ptr [[X_ADDR]], align 4
+// RECOVER-NEXT:    store i32 [[Y]], ptr [[Y_ADDR]], align 4
+// RECOVER-NEXT:    [[TMP0:%.*]] = load i32, ptr [[X_ADDR]], align 4
+// RECOVER-NEXT:    [[TMP1:%.*]] = load i32, ptr [[Y_ADDR]], align 4
+// RECOVER-NEXT:    [[TMP2:%.*]] = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 [[TMP0]], i32 [[TMP1]]), !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP3:%.*]] = extractvalue { i32, i1 } [[TMP2]], 0, !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP4:%.*]] = extractvalue { i32, i1 } [[TMP2]], 1, !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP5:%.*]] = xor i1 [[TMP4]], true, !nosanitize [[META2]]
+// RECOVER-NEXT:    br i1 [[TMP5]], label [[CONT:%.*]], label [[HANDLER_ADD_OVERFLOW:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
+// RECOVER:       handler.add_overflow:
+// RECOVER-NEXT:    [[TMP6:%.*]] = zext i32 [[TMP0]] to i64, !nosanitize [[META2]]
+// RECOVER-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP1]] to i64, !nosanitize [[META2]]
+// RECOVER-NEXT:    call void @__ubsan_handle_add_overflow(ptr @[[GLOB3:[0-9]+]], i64 [[TMP6]], i64 [[TMP7]]) #[[ATTR3]], !nosanitize [[META2]]
+// RECOVER-NEXT:    br label [[CONT]], !nosanitize [[META2]]
+// RECOVER:       cont:
+// RECOVER-NEXT:    ret i32 [[TMP3]]
+//
+int overflow(int x, int y) {
+  return x + y;
+}
+//.
+// CHECK: [[META2]] = !{}
+// CHECK: [[PROF3]] = !{!"branch_weights", i32 1048575, i32 1}
+//.
+// TRAP: [[META2]] = !{}
+//.
+// RECOVER: [[META2]] = !{}
+// RECOVER: [[PROF3]] = !{!"branch_weights", i32 1048575, i32 1}
+//.
diff --git a/clang/test/CodeGen/arm-v8.2a-neon-intrinsics-generic.c b/clang/test/CodeGen/arm-v8.2a-neon-intrinsics-generic.c
diff --git a/clang/test/CodeGen/arm-v8.2a-neon-intrinsics.c b/clang/test/CodeGen/arm-v8.2a-neon-intrinsics.c
@@ -817,181 +817,3 @@ float16x4_t test_vmul_n_f16(float16x4_t a, float16_t b) {
 float16x8_t test_vmulq_n_f16(float16x8_t a, float16_t b) {
   return vmulq_n_f16(a, b);
 }
-
-// CHECK-LABEL: test_vbsl_f16
-// CHECK:  [[TMP0:%.*]] = bitcast <4 x i16> %a to <8 x i8>
-// CHECK:  [[TMP1:%.*]] = bitcast <4 x half> %b to <8 x i8>
-// CHECK:  [[TMP2:%.*]] = bitcast <4 x half> %c to <8 x i8>
-// CHECK:  [[VBSL:%.*]] = call <8 x i8> @llvm.arm.neon.vbsl.v8i8(<8 x i8> [[TMP0]], <8 x i8> [[TMP1]], <8 x i8> [[TMP2]])
-// CHECK:  [[TMP3:%.*]] = bitcast <8 x i8> [[VBSL]] to <4 x half>
-// CHECK:  ret <4 x half> [[TMP3]]
-float16x4_t test_vbsl_f16(uint16x4_t a, float16x4_t b, float16x4_t c) {
-  return vbsl_f16(a, b, c);
-}
-
-// CHECK-LABEL: test_vbslq_f16
-// CHECK:  [[TMP0:%.*]] = bitcast <8 x i16> %a to <16 x i8>
-// CHECK:  [[TMP1:%.*]] = bitcast <8 x half> %b to <16 x i8>
-// CHECK:  [[TMP2:%.*]] = bitcast <8 x half> %c to <16 x i8>
-// CHECK:  [[VBSL:%.*]] = call <16 x i8> @llvm.arm.neon.vbsl.v16i8(<16 x i8> [[TMP0]], <16 x i8> [[TMP1]], <16 x i8> [[TMP2]])
-// CHECK:  [[TMP3:%.*]] = bitcast <16 x i8> [[VBSL]] to <8 x half>
-// CHECK:  ret <8 x half> [[TMP3]]
-float16x8_t test_vbslq_f16(uint16x8_t a, float16x8_t b, float16x8_t c) {
-  return vbslq_f16(a, b, c);
-}
-
-// CHECK-LABEL: test_vzip_f16
-// CHECK:  [[VZIP0:%.*]] = shufflevector <4 x half> %a, <4 x half> %b, <4 x i32> <i32 0, i32 4, i32 1, i32 5>
-// CHECK:  store <4 x half> [[VZIP0]], ptr [[addr1:%.*]]
-// CHECK:  [[VZIP1:%.*]] = shufflevector <4 x half> %a, <4 x half> %b, <4 x i32> <i32 2, i32 6, i32 3, i32 7>
-// CHECK:  store <4 x half> [[VZIP1]], ptr [[addr2:%.*]]
-float16x4x2_t test_vzip_f16(float16x4_t a, float16x4_t b) {
-  return vzip_f16(a, b);
-}
-
-// CHECK-LABEL: test_vzipq_f16
-// CHECK:  [[VZIP0:%.*]] = shufflevector <8 x half> %a, <8 x half> %b, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
-// CHECK:  store <8 x half> [[VZIP0]], ptr [[addr1:%.*]]
-// CHECK:  [[VZIP1:%.*]] = shufflevector <8 x half> %a, <8 x half> %b, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
-// CHECK:  store <8 x half> [[VZIP1]], ptr [[addr2:%.*]]
-float16x8x2_t test_vzipq_f16(float16x8_t a, float16x8_t b) {
-  return vzipq_f16(a, b);
-}
-
-// CHECK-LABEL: test_vuzp_f16
-// CHECK:  [[VUZP0:%.*]] = shufflevector <4 x half> %a, <4 x half> %b, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
-// CHECK:  store <4 x half> [[VUZP0]], ptr [[addr1:%.*]]
-// CHECK:  [[VUZP1:%.*]] = shufflevector <4 x half> %a, <4 x half> %b, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
-// CHECK:  store <4 x half> [[VUZP1]], ptr [[addr1:%.*]]
-float16x4x2_t test_vuzp_f16(float16x4_t a, float16x4_t b) {
-  return vuzp_f16(a, b);
-}
-
-// CHECK-LABEL: test_vuzpq_f16
-// CHECK:   [[VUZP0:%.*]] = shufflevector <8 x half> %a, <8 x half> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
-// CHECK:   store <8 x half> [[VUZP0]], ptr [[addr1:%.*]]
-// CHECK:   [[VUZP1:%.*]] = shufflevector <8 x half> %a, <8 x half> %b, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
-// CHECK:   store <8 x half> [[VUZP1]], ptr [[addr2:%.*]]
-float16x8x2_t test_vuzpq_f16(float16x8_t a, float16x8_t b) {
-  return vuzpq_f16(a, b);
-}
-
-// CHECK-LABEL: test_vtrn_f16
-// CHECK:   [[VTRN0:%.*]] = shufflevector <4 x half> %a, <4 x half> %b, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
-// CHECK:   store <4 x half> [[VTRN0]], ptr [[addr1:%.*]]
-// CHECK:   [[VTRN1:%.*]] = shufflevector <4 x half> %a, <4 x half> %b, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
-// CHECK:   store <4 x half> [[VTRN1]], ptr [[addr2:%.*]]
-float16x4x2_t test_vtrn_f16(float16x4_t a, float16x4_t b) {
-  return vtrn_f16(a, b);
-}
-
-// CHECK-LABEL: test_vtrnq_f16
-// CHECK:   [[VTRN0:%.*]] = shufflevector <8 x half> %a, <8 x half> %b, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
-// CHECK:   store <8 x half> [[VTRN0]], ptr [[addr1:%.*]]
-// CHECK:   [[VTRN1:%.*]] = shufflevector <8 x half> %a, <8 x half> %b, <8 x i32>  <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15>
-// CHECK:   store <8 x half> [[VTRN1]], ptr [[addr2:%.*]]
-float16x8x2_t test_vtrnq_f16(float16x8_t a, float16x8_t b) {
-  return vtrnq_f16(a, b);
-}
-
-// CHECK-LABEL: test_vmov_n_f16
-// CHECK:   [[TMP0:%.*]] = insertelement <4 x half> poison, half [[ARG:%.*]], i32 0
-// CHECK:   [[TMP1:%.*]] = insertelement <4 x half> [[TMP0]], half [[ARG]], i32 1
-// CHECK:   [[TMP2:%.*]] = insertelement <4 x half> [[TMP1]], half [[ARG]], i32 2
-// CHECK:   [[TMP3:%.*]] = insertelement <4 x half> [[TMP2]], half [[ARG]], i32 3
-// CHECK:   ret <4 x half> [[TMP3]]
-float16x4_t test_vmov_n_f16(float16_t a) {
-  return vmov_n_f16(a);
-}
-
-// CHECK-LABEL: test_vmovq_n_f16
-// CHECK:   [[TMP0:%.*]] = insertelement <8 x half> poison, half [[ARG:%.*]], i32 0
-// CHECK:   [[TMP1:%.*]] = insertelement <8 x half> [[TMP0]], half [[ARG]], i32 1
-// CHECK:   [[TMP2:%.*]] = insertelement <8 x half> [[TMP1]], half [[ARG]], i32 2
-// CHECK:   [[TMP3:%.*]] = insertelement <8 x half> [[TMP2]], half [[ARG]], i32 3
-// CHECK:   [[TMP4:%.*]] = insertelement <8 x half> [[TMP3]], half [[ARG]], i32 4
-// CHECK:   [[TMP5:%.*]] = insertelement <8 x half> [[TMP4]], half [[ARG]], i32 5
-// CHECK:   [[TMP6:%.*]] = insertelement <8 x half> [[TMP5]], half [[ARG]], i32 6
-// CHECK:   [[TMP7:%.*]] = insertelement <8 x half> [[TMP6]], half [[ARG]], i32 7
-// CHECK:   ret <8 x half> [[TMP7]]
-float16x8_t test_vmovq_n_f16(float16_t a) {
-  return vmovq_n_f16(a);
-}
-
-// CHECK-LABEL: test_vdup_n_f16
-// CHECK:   [[TMP0:%.*]] = insertelement <4 x half> poison, half [[ARG:%.*]], i32 0
-// CHECK:   [[TMP1:%.*]] = insertelement <4 x half> [[TMP0]], half [[ARG]], i32 1
-// CHECK:   [[TMP2:%.*]] = insertelement <4 x half> [[TMP1]], half [[ARG]], i32 2
-// CHECK:   [[TMP3:%.*]] = insertelement <4 x half> [[TMP2]], half [[ARG]], i32 3
-// CHECK:   ret <4 x half> [[TMP3]]
-float16x4_t test_vdup_n_f16(float16_t a) {
-  return vdup_n_f16(a);
-}
-
-// CHECK-LABEL: test_vdupq_n_f16
-// CHECK:   [[TMP0:%.*]] = insertelement <8 x half> poison, half [[ARG:%.*]], i32 0
-// CHECK:   [[TMP1:%.*]] = insertelement <8 x half> [[TMP0]], half [[ARG]], i32 1
-// CHECK:   [[TMP2:%.*]] = insertelement <8 x half> [[TMP1]], half [[ARG]], i32 2
-// CHECK:   [[TMP3:%.*]] = insertelement <8 x half> [[TMP2]], half [[ARG]], i32 3
-// CHECK:   [[TMP4:%.*]] = insertelement <8 x half> [[TMP3]], half [[ARG]], i32 4
-// CHECK:   [[TMP5:%.*]] = insertelement <8 x half> [[TMP4]], half [[ARG]], i32 5
-// CHECK:   [[TMP6:%.*]] = insertelement <8 x half> [[TMP5]], half [[ARG]], i32 6
-// CHECK:   [[TMP7:%.*]] = insertelement <8 x half> [[TMP6]], half [[ARG]], i32 7
-// CHECK:   ret <8 x half> [[TMP7]]
-float16x8_t test_vdupq_n_f16(float16_t a) {
-  return vdupq_n_f16(a);
-}
-
-// CHECK-LABEL: test_vdup_lane_f16
-// CHECK:   [[TMP0:%.*]] = bitcast <4 x half> [[A:%.*]] to <8 x i8>
-// CHECK:   [[TMP1:%.*]] = bitcast <8 x i8> [[TMP0]] to <4 x half>
-// CHECK:   [[LANE:%.*]] = shufflevector <4 x half> [[TMP1]], <4 x half> [[TMP1]], <4 x i32> <i32 3, i32 3, i32 3, i32 3>
-// CHECK:   ret <4 x half> [[LANE]]
-float16x4_t test_vdup_lane_f16(float16x4_t a) {
-  return vdup_lane_f16(a, 3);
-}
-
-// CHECK-LABEL: test_vdupq_lane_f16
-// CHECK:   [[TMP0:%.*]] = bitcast <4 x half> [[A:%.*]] to <8 x i8>
-// CHECK:   [[TMP1:%.*]] = bitcast <8 x i8> [[TMP0]] to <4 x half>
-// CHECK:   [[LANE:%.*]] = shufflevector <4 x half> [[TMP1]], <4 x half> [[TMP1]], <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3>
-// CHECK:   ret <8 x half> [[LANE]]
-float16x8_t test_vdupq_lane_f16(float16x4_t a) {
-  return vdupq_lane_f16(a, 3);
-}
-
-// CHECK-LABEL: @test_vext_f16(
-// CHECK:   [[TMP0:%.*]] = bitcast <4 x half> %a to <8 x i8>
-// CHECK:   [[TMP1:%.*]] = bitcast <4 x half> %b to <8 x i8>
-// CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP0]] to <4 x half>
-// CHECK:   [[TMP3:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x half>
-// CHECK:   [[VEXT:%.*]] = shufflevector <4 x half> [[TMP2]], <4 x half> [[TMP3]], <4 x i32> <i32 2, i32 3, i32 4, i32 5>
-// CHECK:   ret <4 x half> [[VEXT]]
-float16x4_t test_vext_f16(float16x4_t a, float16x4_t b) {
-  return vext_f16(a, b, 2);
-}
-
-// CHECK-LABEL: @test_vextq_f16(
-// CHECK:   [[TMP0:%.*]] = bitcast <8 x half> %a to <16 x i8>
-// CHECK:   [[TMP1:%.*]] = bitcast <8 x half> %b to <16 x i8>
-// CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP0]] to <8 x half>
-// CHECK:   [[TMP3:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x half>
-// CHECK:   [[VEXT:%.*]] = shufflevector <8 x half> [[TMP2]], <8 x half> [[TMP3]], <8 x i32> <i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12>
-// CHECK:   ret <8 x half> [[VEXT]]
-float16x8_t test_vextq_f16(float16x8_t a, float16x8_t b) {
-  return vextq_f16(a, b, 5);
-}
-
-// CHECK-LABEL: @test_vrev64_f16(
-// CHECK:   [[SHFL:%.*]] = shufflevector <4 x half> %a, <4 x half> %a, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
-// CHECK:   ret <4 x half> [[SHFL]]
-float16x4_t test_vrev64_f16(float16x4_t a) {
-  return vrev64_f16(a);
-}
-
-// CHECK-LABEL: @test_vrev64q_f16(
-// CHECK:   [[SHFL:%.*]] = shufflevector <8 x half> %a, <8 x half> %a, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4>
-// CHECK:   ret <8 x half> [[SHFL]]
-float16x8_t test_vrev64q_f16(float16x8_t a) {
-  return vrev64q_f16(a);
-}
diff --git a/clang/test/CodeGen/cx-complex-range.c b/clang/test/CodeGen/cx-complex-range.c
diff --git a/clang/test/CodeGen/pseudo-probe-emit.c b/clang/test/CodeGen/pseudo-probe-emit.c
@@ -10,9 +10,9 @@ void foo(int x) {
   // CHECK: call void @llvm.pseudoprobe(i64 [[#GUID:]], i64 1, i32 0, i64 -1)
   if (x == 0)
     // CHECK: call void @llvm.pseudoprobe(i64 [[#GUID]], i64 2, i32 0, i64 -1)
-    bar();
+    bar();  // probe id : 3
   else
-    // CHECK: call void @llvm.pseudoprobe(i64 [[#GUID]], i64 3, i32 0, i64 -1)
-    go();
-  // CHECK: call void @llvm.pseudoprobe(i64 [[#GUID]], i64 4, i32 0, i64 -1)
+    // CHECK: call void @llvm.pseudoprobe(i64 [[#GUID]], i64 4, i32 0, i64 -1)
+    go(); // probe id : 5
+  // CHECK: call void @llvm.pseudoprobe(i64 [[#GUID]], i64 6, i32 0, i64 -1)
 }
diff --git a/clang/test/CodeGen/remote-traps.c b/clang/test/CodeGen/remote-traps.c
@@ -1,5 +1,5 @@
 // RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow -fsanitize-trap=signed-integer-overflow %s -o - | FileCheck %s 
-// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow -fsanitize-trap=signed-integer-overflow -mllvm -clang-remove-traps -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call void @llvm.ubsantrap" --check-prefixes=REMOVE
+// RUN: %clang_cc1 -O1 -emit-llvm -fsanitize=signed-integer-overflow -fsanitize-trap=signed-integer-overflow -mllvm -remove-traps-random-rate=1 %s -o - | FileCheck %s --implicit-check-not="call void @llvm.ubsantrap" --check-prefixes=REMOVE
 
 int test(int x) {
   return x + 123;

diff --git a/clang/test/CodeGen/tbaa-struct-bitfield-endianness.cpp b/clang/test/CodeGen/tbaa-struct-bitfield-endianness.cpp
@@ -0,0 +1,39 @@
+// RUN: %clang_cc1 -triple aarch64_be-apple-darwin -emit-llvm -o - -O1 %s | \
+// RUN:     FileCheck -check-prefixes=CHECK,CHECK-BE %s
+// RUN: %clang_cc1 -triple aarch64-apple-darwin -emit-llvm -o - -O1 %s | \
+// RUN:     FileCheck -check-prefixes=CHECK,CHECK-LE %s
+//
+// Check that TBAA metadata for structs containing bitfields is
+// consistent between big and little endian layouts.
+//
+// FIXME: The metadata below is invalid for the big endian layout: the
+// start offset of 2 is incorrect.
+
+struct NamedBitfields {
+  int f1 : 8;
+  int f2 : 8;
+  unsigned f3 : 1;
+  unsigned f4 : 15;
+  int f5;
+  double f6;
+};
+
+// CHECK-LABEL: _Z4copyP14NamedBitfieldsS0_
+// CHECK-SAME: ptr nocapture noundef writeonly [[A1:%.*]], ptr nocapture noundef readonly [[A2:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    tail call void @llvm.memcpy.p0.p0.i64(ptr noundef nonnull align 8 dereferenceable(16) [[A1]], ptr noundef nonnull align 8 dereferenceable(16) [[A2]], i64 16, i1 false), !tbaa.struct [[TBAA_STRUCT2:![0-9]+]]
+// CHECK-NEXT:    ret void
+//
+void copy(NamedBitfields *a1, NamedBitfields *a2) {
+  *a1 = *a2;
+}
+
+// CHECK-BE: [[TBAA_STRUCT2]] = !{i64 2, i64 4, [[META3:![0-9]+]], i64 4, i64 4, [[META6:![0-9]+]], i64 8, i64 8, [[META8:![0-9]+]]}
+// CHECK-LE: [[TBAA_STRUCT2]] = !{i64 0, i64 4, [[META3:![0-9]+]], i64 4, i64 4, [[META6:![0-9]+]], i64 8, i64 8, [[META8:![0-9]+]]}
+// CHECK: [[META3]] = !{[[META4:![0-9]+]], [[META4]], i64 0}
+// CHECK: [[META4]] = !{!"omnipotent char", [[META5:![0-9]+]], i64 0}
+// CHECK: [[META5]] = !{!"Simple C++ TBAA"}
+// CHECK: [[META6]] = !{[[META7:![0-9]+]], [[META7]], i64 0}
+// CHECK: [[META7]] = !{!"int", [[META4]], i64 0}
+// CHECK: [[META8]] = !{[[META9:![0-9]+]], [[META9]], i64 0}
+// CHECK: [[META9]] = !{!"double", [[META4]], i64 0}
diff --git a/clang/test/CodeGen/ubsan-bitfield-conversion.c b/clang/test/CodeGen/ubsan-bitfield-conversion.c
diff --git a/clang/test/CodeGenCXX/ubsan-bitfield-conversion.cpp b/clang/test/CodeGenCXX/ubsan-bitfield-conversion.cpp
diff --git a/clang/test/Driver/fsanitize.c b/clang/test/Driver/fsanitize.c
@@ -35,20 +35,20 @@
 // RUN: %clang --target=%itanium_abi_triple -fsanitize=integer %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-INTEGER -implicit-check-not="-fsanitize-address-use-after-scope"
 // CHECK-INTEGER: "-fsanitize={{((signed-integer-overflow|unsigned-integer-overflow|integer-divide-by-zero|shift-base|shift-exponent|implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change|unsigned-shift-base),?){9}"}}
 
-// RUN: %clang -fsanitize=implicit-integer-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-integer-conversion,CHECK-implicit-integer-conversion-RECOVER
-// RUN: %clang -fsanitize=implicit-integer-conversion -fsanitize-recover=implicit-integer-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-integer-conversion,CHECK-implicit-integer-conversion-RECOVER
-// RUN: %clang -fsanitize=implicit-integer-conversion -fno-sanitize-recover=implicit-integer-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-integer-conversion,CHECK-implicit-integer-conversion-NORECOVER
-// RUN: %clang -fsanitize=implicit-integer-conversion -fsanitize-trap=implicit-integer-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-integer-conversion,CHECK-implicit-integer-conversion-TRAP
-// CHECK-implicit-integer-conversion: "-fsanitize={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-RECOVER: "-fsanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-RECOVER-NOT: "-fno-sanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-RECOVER-NOT: "-fsanitize-trap={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-NORECOVER-NOT: "-fno-sanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}} // ???
-// CHECK-implicit-integer-conversion-NORECOVER-NOT: "-fsanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-NORECOVER-NOT: "-fsanitize-trap={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-TRAP: "-fsanitize-trap={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-TRAP-NOT: "-fsanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
-// CHECK-implicit-integer-conversion-TRAP-NOT: "-fno-sanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// RUN: %clang -fsanitize=implicit-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-conversion,CHECK-implicit-conversion-RECOVER
+// RUN: %clang -fsanitize=implicit-conversion -fsanitize-recover=implicit-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-conversion,CHECK-implicit-conversion-RECOVER
+// RUN: %clang -fsanitize=implicit-conversion -fno-sanitize-recover=implicit-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-conversion,CHECK-implicit-conversion-NORECOVER
+// RUN: %clang -fsanitize=implicit-conversion -fsanitize-trap=implicit-conversion %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-conversion,CHECK-implicit-conversion-TRAP
+// CHECK-implicit-conversion: "-fsanitize={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-RECOVER: "-fsanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-RECOVER-NOT: "-fno-sanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-RECOVER-NOT: "-fsanitize-trap={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-NORECOVER-NOT: "-fno-sanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}} // ???
+// CHECK-implicit-conversion-NORECOVER-NOT: "-fsanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-NORECOVER-NOT: "-fsanitize-trap={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-TRAP: "-fsanitize-trap={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-TRAP-NOT: "-fsanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
+// CHECK-implicit-conversion-TRAP-NOT: "-fno-sanitize-recover={{((implicit-unsigned-integer-truncation|implicit-signed-integer-truncation|implicit-integer-sign-change),?){3}"}}
 
 // RUN: %clang -fsanitize=implicit-integer-arithmetic-value-change %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-integer-arithmetic-value-change,CHECK-implicit-integer-arithmetic-value-change-RECOVER
 // RUN: %clang -fsanitize=implicit-integer-arithmetic-value-change -fsanitize-recover=implicit-integer-arithmetic-value-change %s -### 2>&1 | FileCheck %s --check-prefixes=CHECK-implicit-integer-arithmetic-value-change,CHECK-implicit-integer-arithmetic-value-change-RECOVER

diff --git a/clang/test/Driver/mcmodel.c b/clang/test/Driver/mcmodel.c
@@ -11,6 +11,7 @@
 // RUN: FileCheck --check-prefix=AIX-MCMEDIUM-OVERRIDE %s < %t.log
 // RUN: not %clang -### -c -mcmodel=lager %s 2>&1 | FileCheck --check-prefix=INVALID %s
 // RUN: %clang --target=aarch64 -### -S -mcmodel=large -fno-pic %s 2>&1 | FileCheck --check-prefix=LARGE %s
+// RUN: %clang --target=aarch64-apple-macosx -### -S -mcmodel=large %s 2>&1 | FileCheck --check-prefix=LARGE %s
 // RUN: not %clang --target=aarch64 -### -S -mcmodel=large -fpic %s 2>&1 | FileCheck --check-prefix=AARCH64-PIC-LARGE %s
 // RUN: not %clang -### -c --target=aarch64 -mcmodel=medium %s 2>&1 | FileCheck --check-prefix=ERR-MEDIUM %s
 // RUN: not %clang -### -c --target=aarch64 -mcmodel=kernel %s 2>&1 | FileCheck --check-prefix=ERR-KERNEL %s

diff --git a/clang/test/InstallAPI/driver-invalid-options.test b/clang/test/InstallAPI/driver-invalid-options.test
@@ -7,3 +7,9 @@
 // RUN: not clang-installapi -target x86_64-apple-ios-simulator  %s -o tmp.tbd 2> %t 
 // RUN: FileCheck --check-prefix INVALID_INSTALL_NAME -input-file %t %s
 // INVALID_INSTALL_NAME: error: no install name specified: add -install_name <path>
+
+/// Check invalid verification mode.
+// RUN: not clang-installapi -install_name Foo -target arm64-apple-ios13 \
+// RUN: --verify-mode=Invalid -o tmp.tbd 2> %t
+// RUN: FileCheck --check-prefix INVALID_VERIFY_MODE -input-file %t %s 
+// INVALID_VERIFY_MODE: error: invalid value 'Invalid' in '--verify-mode=Invalid'
diff --git a/clang/test/Interpreter/inline-asm.cpp b/clang/test/Interpreter/inline-asm.cpp
@@ -0,0 +1,17 @@
+// REQUIRES: host-supports-jit, x86_64-linux
+// UNSUPPORTED: system-aix
+//
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file %s %t
+//
+// RUN: cat %t/inline-asm.txt | clang-repl -Xcc="-I%t"
+
+//--- inline-asm.cpp
+__asm(".globl _ZSt21ios_base_library_initv");
+int x;
+
+//--- inline-asm.txt
+#include "inline-asm.cpp"
+x = 10;
+%quit
diff --git a/clang/test/Parser/c2x-typeof-ext-warns.c b/clang/test/Parser/c2x-typeof-ext-warns.c
@@ -12,9 +12,12 @@
 // standards before C23, and Clang has followed suit. Neither compiler exposes
 // 'typeof_unqual' as a non-conforming extension.
 
-// Show what happens with the underscored version of the keyword, which is a
-// conforming extension.
+// Show what happens with the underscored version of the keywords, which are
+// conforming extensions.
 __typeof__(int) i = 12;
+__typeof(int) _i = 12;
+__typeof_unqual__(int) u = 12;
+__typeof_unqual(int) _u = 12;
 
 // Show what happens with a regular 'typeof' use.
 typeof(i) j = 12; // c11-error {{expected function body after function declarator}} \

diff --git a/clang/test/SemaCXX/typeof_unqual.cpp b/clang/test/SemaCXX/typeof_unqual.cpp
@@ -0,0 +1,5 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+
+typeof_unqual(int) u = 12; // expected-error {{expected function body after function declarator}}
+__typeof_unqual(int) _u = 12;
+__typeof_unqual__(int) __u = 12;
diff --git a/clang/tools/clang-repl/ClangRepl.cpp b/clang/tools/clang-repl/ClangRepl.cpp
@@ -152,6 +152,7 @@ int main(int argc, const char **argv) {
   llvm::InitializeAllTargets();
   llvm::InitializeAllTargetMCs();
   llvm::InitializeAllAsmPrinters();
+  llvm::InitializeAllAsmParsers();
 
   if (OptHostSupportsJit) {
     auto J = llvm::orc::LLJITBuilder().create();

diff --git a/clang/unittests/Analysis/FlowSensitive/TypeErasedDataflowAnalysisTest.cpp b/clang/unittests/Analysis/FlowSensitive/TypeErasedDataflowAnalysisTest.cpp
@@ -805,6 +805,25 @@ class NullPointerAnalysis final
     else
       JoinedVal.setProperty("is_null", JoinedEnv.makeTopBoolValue());
   }
+
+  std::optional<WidenResult> widen(QualType Type, Value &Prev,
+                                   const Environment &PrevEnv, Value &Current,
+                                   Environment &CurrentEnv) override {
+    switch (compare(Type, Prev, PrevEnv, Current, CurrentEnv)) {
+    case ComparisonResult::Same:
+      return WidenResult{&Current, LatticeJoinEffect::Unchanged};
+    case ComparisonResult::Different: {
+      auto &CurPtr = cast<PointerValue>(Current);
+      auto &WidenedPtr =
+          CurrentEnv.create<PointerValue>(CurPtr.getPointeeLoc());
+      WidenedPtr.setProperty("is_null", CurrentEnv.makeTopBoolValue());
+      return WidenResult{&WidenedPtr, LatticeJoinEffect::Changed};
+    }
+    case ComparisonResult::Unknown:
+      return std::nullopt;
+    }
+    llvm_unreachable("all cases in switch covered");
+  }
 };
 
 class WideningTest : public Test {
@@ -846,7 +865,6 @@ TEST_F(WideningTest, JoinDistinctValuesWithDistinctProperties) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(), UnorderedElementsAre("p1", "p2", "p3"));
         const Environment &Env1 = getEnvironmentAtAnnotation(Results, "p1");
         const Environment &Env2 = getEnvironmentAtAnnotation(Results, "p2");
         const Environment &Env3 = getEnvironmentAtAnnotation(Results, "p3");
@@ -889,8 +907,6 @@ TEST_F(WideningTest, JoinDistinctValuesWithSameProperties) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(),
-                    UnorderedElementsAre("p1", "p2", "p3", "p4"));
         const Environment &Env1 = getEnvironmentAtAnnotation(Results, "p1");
         const Environment &Env2 = getEnvironmentAtAnnotation(Results, "p2");
         const Environment &Env3 = getEnvironmentAtAnnotation(Results, "p3");
@@ -929,19 +945,11 @@ TEST_F(WideningTest, DistinctPointersToTheSameLocationAreEquivalent) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(), UnorderedElementsAre("p"));
         const Environment &Env = getEnvironmentAtAnnotation(Results, "p");
-
-        const ValueDecl *FooDecl = findValueDecl(ASTCtx, "Foo");
-        ASSERT_THAT(FooDecl, NotNull());
-
-        const ValueDecl *BarDecl = findValueDecl(ASTCtx, "Bar");
-        ASSERT_THAT(BarDecl, NotNull());
-
-        const auto *FooLoc =
-            cast<ScalarStorageLocation>(Env.getStorageLocation(*FooDecl));
-        const auto *BarVal = cast<PointerValue>(Env.getValue(*BarDecl));
-        EXPECT_EQ(&BarVal->getPointeeLoc(), FooLoc);
+        const auto &FooLoc =
+            getLocForDecl<ScalarStorageLocation>(ASTCtx, Env, "Foo");
+        const auto &BarVal = getValueForDecl<PointerValue>(ASTCtx, Env, "Bar");
+        EXPECT_EQ(&BarVal.getPointeeLoc(), &FooLoc);
       });
 }
 
@@ -963,18 +971,38 @@ TEST_F(WideningTest, DistinctValuesWithSamePropertiesAreEquivalent) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(), UnorderedElementsAre("p"));
         const Environment &Env = getEnvironmentAtAnnotation(Results, "p");
-
-        const ValueDecl *FooDecl = findValueDecl(ASTCtx, "Foo");
-        ASSERT_THAT(FooDecl, NotNull());
-
-        const auto *FooVal = Env.getValue(*FooDecl);
-        EXPECT_EQ(FooVal->getProperty("is_null"),
+        const auto &FooVal = getValueForDecl<Value>(ASTCtx, Env, "Foo");
+        EXPECT_EQ(FooVal.getProperty("is_null"),
                   &Env.getBoolLiteralValue(false));
       });
 }
 
+TEST_F(WideningTest, DistinctValuesWithDifferentPropertiesWidenedToTop) {
+  std::string Code = R"(
+    void target(bool Cond) {
+      int *Foo;
+      int i = 0;
+      Foo = nullptr;
+      while (Cond) {
+        Foo = &i;
+      }
+      (void)0;
+      /*[[p]]*/
+    }
+  )";
+  runDataflow(
+      Code,
+      [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
+         ASTContext &ASTCtx) {
+        const Environment &Env = getEnvironmentAtAnnotation(Results, "p");
+        const auto &FooVal = getValueForDecl<Value>(ASTCtx, Env, "Foo");
+        ASSERT_THAT(FooVal.getProperty("is_null"), NotNull());
+        EXPECT_TRUE(areEquivalentValues(*FooVal.getProperty("is_null"),
+                                        Env.makeTopBoolValue()));
+      });
+}
+
 class FlowConditionTest : public Test {
 protected:
   template <typename Matcher>

diff --git a/clang/www/c_status.html b/clang/www/c_status.html
@@ -750,11 +750,6 @@ <h2 id="c2x">C23 implementation status</h2>
       <td><a href="https://www.open-std.org/jtc1/sc22/wg14/www/docs/n2379.htm">N2379</a></td>
       <td class="unknown" align="center">Unknown</td>
     </tr>
-    <tr>
-      <td>Floating-point negation and conversion</td>
-      <td><a href="https://www.open-std.org/jtc1/sc22/wg14/www/docs/n2416.pdf">N2416</a></td>
-      <td class="unknown" align="center">Unknown</td>
-    </tr>
     <tr>
       <td>Annex F.8 update for implementation extensions and rounding</td>
       <td><a href="https://www.open-std.org/jtc1/sc22/wg14/www/docs/n2384.pdf">N2384</a></td>

diff --git a/compiler-rt/lib/hwasan/hwasan_thread_list.h b/compiler-rt/lib/hwasan/hwasan_thread_list.h
@@ -18,7 +18,7 @@
 // * Start of the shadow memory region is aligned to 2**kShadowBaseAlignment.
 // * All stack ring buffers are located within (2**kShadowBaseAlignment)
 // sized region below and adjacent to the shadow region.
-// * Each ring buffer has a size of (2**N)*4096 where N is in [0, 8), and is
+// * Each ring buffer has a size of (2**N)*4096 where N is in [0, 7), and is
 // aligned to twice its size. The value of N can be different for each buffer.
 //
 // These constrains guarantee that, given an address A of any element of the
@@ -55,7 +55,10 @@ static uptr RingBufferSize() {
   uptr desired_bytes = flags()->stack_history_size * sizeof(uptr);
   // FIXME: increase the limit to 8 once this bug is fixed:
   // https://bugs.llvm.org/show_bug.cgi?id=39030
-  for (int shift = 1; shift < 7; ++shift) {
+  // Note that we *cannot* do that on Android, as the runtime will indefinitely
+  // have to support code that is compiled with ashr, which only works with
+  // shifts up to 6.
+  for (int shift = 0; shift < 7; ++shift) {
     uptr size = 4096 * (1ULL << shift);
     if (size >= desired_bytes)
       return size;

diff --git a/compiler-rt/lib/ubsan/ubsan_handlers.cpp b/compiler-rt/lib/ubsan/ubsan_handlers.cpp
@@ -555,11 +555,13 @@ static void handleImplicitConversion(ImplicitConversionData *Data,
                                      ReportOptions Opts, ValueHandle Src,
                                      ValueHandle Dst) {
   SourceLocation Loc = Data->Loc.acquire();
+  ErrorType ET = ErrorType::GenericUB;
+
   const TypeDescriptor &SrcTy = Data->FromType;
   const TypeDescriptor &DstTy = Data->ToType;
+
   bool SrcSigned = SrcTy.isSignedIntegerTy();
   bool DstSigned = DstTy.isSignedIntegerTy();
-  ErrorType ET = ErrorType::GenericUB;
 
   switch (Data->Kind) {
   case ICCK_IntegerTruncation: { // Legacy, no longer used.
@@ -592,23 +594,14 @@ static void handleImplicitConversion(ImplicitConversionData *Data,
 
   ScopedReport R(Opts, Loc, ET);
 
-  // In the case we have a bitfield, we want to explicitly say so in the
-  // error message.
   // FIXME: is it possible to dump the values as hex with fixed width?
-  if (Data->BitfieldBits)
-    Diag(Loc, DL_Error, ET,
-         "implicit conversion from type %0 of value %1 (%2-bit, %3signed) to "
-         "type %4 changed the value to %5 (%6-bit bitfield, %7signed)")
-        << SrcTy << Value(SrcTy, Src) << SrcTy.getIntegerBitWidth()
-        << (SrcSigned ? "" : "un") << DstTy << Value(DstTy, Dst)
-        << Data->BitfieldBits << (DstSigned ? "" : "un");
-  else
-    Diag(Loc, DL_Error, ET,
-         "implicit conversion from type %0 of value %1 (%2-bit, %3signed) to "
-         "type %4 changed the value to %5 (%6-bit, %7signed)")
-        << SrcTy << Value(SrcTy, Src) << SrcTy.getIntegerBitWidth()
-        << (SrcSigned ? "" : "un") << DstTy << Value(DstTy, Dst)
-        << DstTy.getIntegerBitWidth() << (DstSigned ? "" : "un");
+
+  Diag(Loc, DL_Error, ET,
+       "implicit conversion from type %0 of value %1 (%2-bit, %3signed) to "
+       "type %4 changed the value to %5 (%6-bit, %7signed)")
+      << SrcTy << Value(SrcTy, Src) << SrcTy.getIntegerBitWidth()
+      << (SrcSigned ? "" : "un") << DstTy << Value(DstTy, Dst)
+      << DstTy.getIntegerBitWidth() << (DstSigned ? "" : "un");
 }
 
 void __ubsan::__ubsan_handle_implicit_conversion(ImplicitConversionData *Data,

diff --git a/compiler-rt/lib/ubsan/ubsan_handlers.h b/compiler-rt/lib/ubsan/ubsan_handlers.h
@@ -147,7 +147,6 @@ struct ImplicitConversionData {
   const TypeDescriptor &FromType;
   const TypeDescriptor &ToType;
   /* ImplicitConversionCheckKind */ unsigned char Kind;
-  unsigned int BitfieldBits;
 };
 
 /// \brief Implict conversion that changed the value.

diff --git a/flang/.gitignore b/flang/.gitignore
@@ -5,7 +5,6 @@ build
 root
 tags
 TAGS
-*.o
 .nfs*
 *.sw?
 *~

diff --git a/flang/include/flang/Common/windows-include.h b/flang/include/flang/Common/windows-include.h
@@ -0,0 +1,25 @@
+//===-- include/flang/Common/windows-include.h ------------------*- 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
+//
+//===----------------------------------------------------------------------===//
+//
+// Wrapper around windows.h that works around the name conflicts.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef FORTRAN_COMMON_WINDOWS_INCLUDE_H_
+#define FORTRAN_COMMON_WINDOWS_INCLUDE_H_
+
+#ifdef _WIN32
+
+#define WIN32_LEAN_AND_MEAN
+#define NOMINMAX
+
+#include <windows.h>
+
+#endif // _WIN32
+
+#endif // FORTRAN_COMMON_WINDOWS_INCLUDE_H_
diff --git a/flang/lib/Lower/ConvertCall.cpp b/flang/lib/Lower/ConvertCall.cpp
@@ -1340,15 +1340,6 @@ static PreparedDummyArgument preparePresentUserCallActualArgument(
   } else {
     addr = hlfir::genVariableRawAddress(loc, builder, entity);
   }
-  // The last extent created for assumed-rank descriptors must be -1 (18.5.3
-  // point 5.). This should be done when creating the assumed-size shape for
-  // consistency.
-  if (auto baseBoxDummy = mlir::dyn_cast<fir::BaseBoxType>(dummyType))
-    if (baseBoxDummy.isAssumedRank())
-      if (const Fortran::semantics::Symbol *sym =
-              Fortran::evaluate::UnwrapWholeSymbolDataRef(*arg.entity))
-        if (Fortran::semantics::IsAssumedSizeArray(sym->GetUltimate()))
-          TODO(loc, "passing assumed-size to assumed-rank array");
 
   // For ranked actual passed to assumed-rank dummy, the cast to assumed-rank
   // box is inserted when building the fir.call op. Inserting it here would

diff --git a/flang/lib/Lower/OpenMP/ReductionProcessor.cpp b/flang/lib/Lower/OpenMP/ReductionProcessor.cpp
@@ -13,6 +13,7 @@
 #include "ReductionProcessor.h"
 
 #include "flang/Lower/AbstractConverter.h"
+#include "flang/Lower/SymbolMap.h"
 #include "flang/Optimizer/Builder/HLFIRTools.h"
 #include "flang/Optimizer/Builder/Todo.h"
 #include "flang/Optimizer/Dialect/FIRType.h"
@@ -522,12 +523,20 @@ void ReductionProcessor::addDeclareReduction(
     if (reductionSymbols)
       reductionSymbols->push_back(symbol);
     mlir::Value symVal = converter.getSymbolAddress(*symbol);
-    auto redType = mlir::cast<fir::ReferenceType>(symVal.getType());
+    mlir::Type eleType;
+    auto refType = mlir::dyn_cast_or_null<fir::ReferenceType>(symVal.getType());
+    if (refType)
+      eleType = refType.getEleTy();
+    else
+      eleType = symVal.getType();
 
     // all arrays must be boxed so that we have convenient access to all the
     // information needed to iterate over the array
-    if (mlir::isa<fir::SequenceType>(redType.getEleTy())) {
-      hlfir::Entity entity{symVal};
+    if (mlir::isa<fir::SequenceType>(eleType)) {
+      // For Host associated symbols, use `SymbolBox` instead
+      Fortran::lower::SymbolBox symBox =
+          converter.lookupOneLevelUpSymbol(*symbol);
+      hlfir::Entity entity{symBox.getAddr()};
       entity = genVariableBox(currentLocation, builder, entity);
       mlir::Value box = entity.getBase();
 
@@ -538,11 +547,25 @@ void ReductionProcessor::addDeclareReduction(
       builder.create<fir::StoreOp>(currentLocation, box, alloca);
 
       symVal = alloca;
-      redType = mlir::cast<fir::ReferenceType>(symVal.getType());
+    } else if (mlir::isa<fir::BaseBoxType>(symVal.getType())) {
+      // boxed arrays are passed as values not by reference. Unfortunately,
+      // we can't pass a box by value to omp.redution_declare, so turn it
+      // into a reference
+
+      auto alloca =
+          builder.create<fir::AllocaOp>(currentLocation, symVal.getType());
+      builder.create<fir::StoreOp>(currentLocation, symVal, alloca);
+      symVal = alloca;
     } else if (auto declOp = symVal.getDefiningOp<hlfir::DeclareOp>()) {
       symVal = declOp.getBase();
     }
 
+    // this isn't the same as the by-val and by-ref passing later in the
+    // pipeline. Both styles assume that the variable is a reference at
+    // this point
+    assert(mlir::isa<fir::ReferenceType>(symVal.getType()) &&
+           "reduction input var is a reference");
+
     reductionVars.push_back(symVal);
   }
   const bool isByRef = doReductionByRef(reductionVars);

diff --git a/flang/lib/Lower/Runtime.cpp b/flang/lib/Lower/Runtime.cpp
@@ -55,6 +55,8 @@ static void genUnreachable(fir::FirOpBuilder &builder, mlir::Location loc) {
 void Fortran::lower::genStopStatement(
     Fortran::lower::AbstractConverter &converter,
     const Fortran::parser::StopStmt &stmt) {
+  const bool isError = std::get<Fortran::parser::StopStmt::Kind>(stmt.t) ==
+                       Fortran::parser::StopStmt::Kind::ErrorStop;
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
   mlir::Location loc = converter.getCurrentLocation();
   Fortran::lower::StatementContext stmtCtx;
@@ -94,13 +96,12 @@ void Fortran::lower::genStopStatement(
   } else {
     callee = fir::runtime::getRuntimeFunc<mkRTKey(StopStatement)>(loc, builder);
     calleeType = callee.getFunctionType();
-    operands.push_back(
-        builder.createIntegerConstant(loc, calleeType.getInput(0), 0));
+    // Default to values are advised in F'2023 11.4 p2.
+    operands.push_back(builder.createIntegerConstant(
+        loc, calleeType.getInput(0), isError ? 1 : 0));
   }
 
   // Second operand indicates ERROR STOP
-  bool isError = std::get<Fortran::parser::StopStmt::Kind>(stmt.t) ==
-                 Fortran::parser::StopStmt::Kind::ErrorStop;
   operands.push_back(builder.createIntegerConstant(
       loc, calleeType.getInput(operands.size()), isError));
 

diff --git a/flang/lib/Optimizer/Builder/IntrinsicCall.cpp b/flang/lib/Optimizer/Builder/IntrinsicCall.cpp
@@ -5259,9 +5259,12 @@ mlir::Value IntrinsicLibrary::genModulo(mlir::Type resultType,
                                                  remainder);
   }
 
+  auto fastMathFlags = builder.getFastMathFlags();
   // F128 arith::RemFOp may be lowered to a runtime call that may be unsupported
   // on the target, so generate a call to Fortran Runtime's ModuloReal16.
-  if (resultType == mlir::FloatType::getF128(builder.getContext()))
+  if (resultType == mlir::FloatType::getF128(builder.getContext()) ||
+      (fastMathFlags & mlir::arith::FastMathFlags::ninf) ==
+          mlir::arith::FastMathFlags::none)
     return builder.createConvert(
         loc, resultType,
         fir::runtime::genModulo(builder, loc, args[0], args[1]));

diff --git a/flang/lib/Optimizer/Builder/Runtime/Numeric.cpp b/flang/lib/Optimizer/Builder/Runtime/Numeric.cpp
@@ -118,6 +118,20 @@ struct ForcedMod16 {
   }
 };
 
+/// Placeholder for real*10 version of Modulo Intrinsic
+struct ForcedModulo10 {
+  static constexpr const char *name = ExpandAndQuoteKey(RTNAME(ModuloReal10));
+  static constexpr fir::runtime::FuncTypeBuilderFunc getTypeModel() {
+    return [](mlir::MLIRContext *ctx) {
+      auto fltTy = mlir::FloatType::getF80(ctx);
+      auto strTy = fir::ReferenceType::get(mlir::IntegerType::get(ctx, 8));
+      auto intTy = mlir::IntegerType::get(ctx, 8 * sizeof(int));
+      return mlir::FunctionType::get(ctx, {fltTy, fltTy, strTy, intTy},
+                                     {fltTy});
+    };
+  }
+};
+
 /// Placeholder for real*16 version of Modulo Intrinsic
 struct ForcedModulo16 {
   static constexpr const char *name = ExpandAndQuoteKey(RTNAME(ModuloReal16));
@@ -349,7 +363,13 @@ mlir::Value fir::runtime::genModulo(fir::FirOpBuilder &builder,
 
   // MODULO is lowered into math operations in intrinsics lowering,
   // so genModulo() should only be used for F128 data type now.
-  if (fltTy.isF128())
+  if (fltTy.isF32())
+    func = fir::runtime::getRuntimeFunc<mkRTKey(ModuloReal4)>(loc, builder);
+  else if (fltTy.isF64())
+    func = fir::runtime::getRuntimeFunc<mkRTKey(ModuloReal8)>(loc, builder);
+  else if (fltTy.isF80())
+    func = fir::runtime::getRuntimeFunc<ForcedModulo10>(loc, builder);
+  else if (fltTy.isF128())
     func = fir::runtime::getRuntimeFunc<ForcedModulo16>(loc, builder);
   else
     fir::intrinsicTypeTODO(builder, fltTy, loc, "MODULO");

diff --git a/flang/lib/Semantics/check-declarations.cpp b/flang/lib/Semantics/check-declarations.cpp
@@ -948,6 +948,11 @@ void CheckHelper::CheckObjectEntity(
             "Component '%s' with ATTRIBUTES(DEVICE) must also be allocatable"_err_en_US,
             symbol.name());
       }
+      if (IsAssumedSizeArray(symbol)) {
+        messages_.Say(
+            "Object '%s' with ATTRIBUTES(DEVICE) may not be assumed size"_err_en_US,
+            symbol.name());
+      }
       break;
     case common::CUDADataAttr::Managed:
       if (!IsAutomatic(symbol) && !IsAllocatable(symbol) &&

diff --git a/flang/runtime/command.cpp b/flang/runtime/command.cpp
@@ -16,9 +16,7 @@
 #include <limits>
 
 #ifdef _WIN32
-#define WIN32_LEAN_AND_MEAN
-#define NOMINMAX
-#include <windows.h>
+#include "flang/Common/windows-include.h"
 
 // On Windows GetCurrentProcessId returns a DWORD aka uint32_t
 #include <processthreadsapi.h>

diff --git a/flang/runtime/execute.cpp b/flang/runtime/execute.cpp
@@ -16,9 +16,7 @@
 #include <future>
 #include <limits>
 #ifdef _WIN32
-#define LEAN_AND_MEAN
-#define NOMINMAX
-#include <windows.h>
+#include "flang/Common/windows-include.h"
 #else
 #include <signal.h>
 #include <sys/wait.h>

diff --git a/flang/runtime/file.cpp b/flang/runtime/file.cpp
@@ -17,9 +17,8 @@
 #include <stdlib.h>
 #include <sys/stat.h>
 #ifdef _WIN32
-#define NOMINMAX
+#include "flang/Common/windows-include.h"
 #include <io.h>
-#include <windows.h>
 #else
 #include <unistd.h>
 #endif

diff --git a/flang/runtime/lock.h b/flang/runtime/lock.h
@@ -25,9 +25,7 @@
 #if USE_PTHREADS
 #include <pthread.h>
 #elif defined(_WIN32)
-// Do not define macros for "min" and "max"
-#define NOMINMAX
-#include <windows.h>
+#include "flang/Common/windows-include.h"
 #else
 #include <mutex>
 #endif

diff --git a/flang/runtime/numeric-templates.h b/flang/runtime/numeric-templates.h
@@ -237,8 +237,12 @@ inline RT_API_ATTRS T RealMod(
   if (ISNANTy<T>::compute(a) || ISNANTy<T>::compute(p) ||
       ISINFTy<T>::compute(a)) {
     return QNANTy<T>::compute();
-  } else if (ISINFTy<T>::compute(p)) {
-    return a;
+  } else if (IS_MODULO && ISINFTy<T>::compute(p)) {
+    // Other compilers behave consistently for MOD(x, +/-INF)
+    // and always return x. This is probably related to
+    // implementation of std::fmod(). Stick to this behavior
+    // for MOD, but return NaN for MODULO(x, +/-INF).
+    return QNANTy<T>::compute();
   }
   T aAbs{ABSTy<T>::compute(a)};
   T pAbs{ABSTy<T>::compute(p)};
@@ -248,8 +252,19 @@ inline RT_API_ATTRS T RealMod(
       if (auto pInt{static_cast<std::int64_t>(p)}; p == pInt) {
         // Fast exact case for integer operands
         auto mod{aInt - (aInt / pInt) * pInt};
-        if (IS_MODULO && (aInt > 0) != (pInt > 0)) {
-          mod += pInt;
+        if constexpr (IS_MODULO) {
+          if (mod == 0) {
+            // Return properly signed zero.
+            return pInt > 0 ? T{0} : -T{0};
+          }
+          if ((aInt > 0) != (pInt > 0)) {
+            mod += pInt;
+          }
+        } else {
+          if (mod == 0) {
+            // Return properly signed zero.
+            return aInt > 0 ? T{0} : -T{0};
+          }
         }
         return static_cast<T>(mod);
       }
@@ -297,7 +312,11 @@ inline RT_API_ATTRS T RealMod(
     }
     if constexpr (IS_MODULO) {
       if ((a < 0) != (p < 0)) {
-        tmp += p;
+        if (tmp == 0.) {
+          tmp = -tmp;
+        } else {
+          tmp += p;
+        }
       }
     }
     return tmp;

diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-as b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-as
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-ld b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-ld
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-ld.bfd b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-ld.bfd
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-ld.gold b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/i386-unknown-linux-gnu-ld.gold
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-as b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-as
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-ld b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-ld
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-ld.bfd b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-ld.bfd
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-ld.gold b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/bin/x86_64-unknown-linux-gnu-ld.gold
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/as b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/as
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/ld b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/ld
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/ld.bfd b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/ld.bfd
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/ld.gold b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/bin/ld.gold
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/lib/.keep b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/i386-unknown-linux-gnu/lib/.keep
diff --git a/...Driver/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/crtbegin.o b/...Driver/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/crtbegin.o
diff --git a/...iver/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/crtbegin.o b/...iver/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/crtbegin.o
diff --git a/...ver/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/crtbeginT.o b/...ver/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/crtbeginT.o
diff --git a/...r/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/crtfastmath.o b/...r/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/crtfastmath.o
diff --git a/.../Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/x32/crtbegin.o b/.../Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/x32/crtbegin.o
diff --git a/...Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/x32/crtbeginT.o b/...Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/x32/crtbeginT.o
diff --git a/...puts/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/x32/crtfastmath.o b/...puts/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/x32/crtfastmath.o
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/as b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/as
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld.bfd b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld.bfd
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld.gold b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld.gold
@@ -0,0 +1 @@
+#!/bin/true
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld.lld b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/bin/ld.lld
diff --git a/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/lib/.keep b/flang/test/Driver/Inputs/basic_cross_linux_tree/usr/x86_64-unknown-linux-gnu/lib/.keep
diff --git a/flang/test/Driver/driver-help-hidden.f90 b/flang/test/Driver/driver-help-hidden.f90
@@ -104,6 +104,9 @@
 ! CHECK-NEXT: -fversion-loops-for-stride
 ! CHECK-NEXT:                         Create unit-strided versions of loops
 ! CHECK-NEXT: -fxor-operator          Enable .XOR. as a synonym of .NEQV.
+! CHECK-NEXT: --gcc-install-dir=<value>
+! CHECK-NEXT:                         Use GCC installation in the specified directory. The directory ends with path components like 'lib{,32,64}/gcc{,-cross}/$triple/$version'. Note: executables (e.g. ld) used by the compiler are not overridden by the selected GCC installation
+! CHECK-NEXT: --gcc-toolchain=<value> Specify a directory where Clang can find 'include' and 'lib{,32,64}/gcc{,-cross}/$triple/$version'. Clang will use the GCC installation with the largest version
 ! CHECK-NEXT: -gline-directives-only  Emit debug line info directives only
 ! CHECK-NEXT: -gline-tables-only      Emit debug line number tables only
 ! CHECK-NEXT: -gpulibc                Link the LLVM C Library for GPUs

diff --git a/flang/test/Driver/driver-help.f90 b/flang/test/Driver/driver-help.f90
@@ -92,6 +92,9 @@
 ! HELP-NEXT: -fversion-loops-for-stride
 ! HELP-NEXT:                         Create unit-strided versions of loops
 ! HELP-NEXT: -fxor-operator          Enable .XOR. as a synonym of .NEQV.
+! HELP-NEXT: --gcc-install-dir=<value>
+! HELP-NEXT:                         Use GCC installation in the specified directory. The directory ends with path components like 'lib{,32,64}/gcc{,-cross}/$triple/$version'. Note: executables (e.g. ld) used by the compiler are not overridden by the selected GCC installation
+! HELP-NEXT: --gcc-toolchain=<value> Specify a directory where Clang can find 'include' and 'lib{,32,64}/gcc{,-cross}/$triple/$version'. Clang will use the GCC installation with the largest version
 ! HELP-NEXT: -gline-directives-only  Emit debug line info directives only
 ! HELP-NEXT: -gline-tables-only      Emit debug line number tables only
 ! HELP-NEXT: -gpulibc                Link the LLVM C Library for GPUs

diff --git a/flang/test/Driver/gcc-toolchain-install-dir.f90 b/flang/test/Driver/gcc-toolchain-install-dir.f90
@@ -0,0 +1,21 @@
+!! Test that --gcc-toolchain and --gcc-install-dir options are working as expected.
+!! It does not test cross-compiling (--sysroot), so crtbegin.o, libgcc/compiler-rt, libc, libFortranRuntime, etc. are not supposed to be affected.
+!! PREFIX is captured twice because the driver escapes backslashes (occuring in Windows paths) in the -### output, but not on the "Selected GCC installation:" line.
+
+! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=i386-unknown-linux-gnu --gcc-install-dir=%S/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0 | FileCheck %s --check-prefix=CHECK-I386
+! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=i386-unknown-linux-gnu --gcc-toolchain=%S/Inputs/basic_cross_linux_tree/usr                                         | FileCheck %s --check-prefix=CHECK-I386
+! CHECK-I386:      Selected GCC installation: [[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0
+! CHECK-I386:      "-fc1" "-triple" "i386-unknown-linux-gnu" 
+! CHECK-I386:      "[[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/../../../../i386-unknown-linux-gnu/bin{{/|\\\\}}as"
+! CHECK-I386:      "[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/../../../../i386-unknown-linux-gnu/bin{{/|\\\\}}ld" {{.*}} "-m" "elf_i386"
+! CHECK-I386-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0" 
+! CHECK-I386-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/i386-unknown-linux-gnu/10.2.0/../../../../i386-unknown-linux-gnu/lib"
+
+! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=x86_64-unknown-linux-gnu --gcc-install-dir=%S/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0 | FileCheck %s --check-prefix=CHECK-X86-64
+! RUN: %flang 2>&1 -### -v -o %t %s -no-integrated-as -fuse-ld=ld --target=x86_64-unknown-linux-gnu --gcc-toolchain=%S/Inputs/basic_cross_linux_tree/usr                                           | FileCheck %s --check-prefix=CHECK-X86-64
+! CHECK-X86-64:      Selected GCC installation: [[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0
+! CHECK-X86-64:      "-fc1" "-triple" "x86_64-unknown-linux-gnu"
+! CHECK-X86-64:      "[[PREFIX:[^"]+]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/../../../../x86_64-unknown-linux-gnu/bin{{/|\\\\}}as" "--64"
+! CHECK-X86-64:      "[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/../../../../x86_64-unknown-linux-gnu/bin{{/|\\\\}}ld" {{.*}} "-m" "elf_x86_64"
+! CHECK-X86-64-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0" 
+! CHECK-X86-64-SAME: "-L[[PREFIX]]/Inputs/basic_cross_linux_tree/usr/lib/gcc/x86_64-unknown-linux-gnu/10.2.0/../../../../x86_64-unknown-linux-gnu/lib"
diff --git a/flang/test/Lower/HLFIR/assumed-rank-iface.f90 b/flang/test/Lower/HLFIR/assumed-rank-iface.f90
@@ -133,9 +133,20 @@ subroutine int_allocatable_to_assumed_rank_opt(x)
 ! CHECK:           %[[VAL_11:.*]] = fir.convert %[[VAL_7]] : (!fir.box<!fir.array<?x?xi32>>) -> !fir.box<!fir.array<*:i32>>
 ! CHECK:           fir.call @_QPint_opt_assumed_rank(%[[VAL_11]]) fastmath<contract> : (!fir.box<!fir.array<*:i32>>) -> ()
 
-! TODO: set assumed size last extent to -1.
-!subroutine int_r2_assumed_size_to_assumed_rank(x)
-!  use ifaces, only : int_assumed_rank
-!  integer :: x(10, *)
-!  call int_assumed_rank(x)
-!end subroutine
+subroutine int_r2_assumed_size_to_assumed_rank(x)
+  use ifaces, only : int_assumed_rank
+  integer :: x(10, *)
+  call int_assumed_rank(x)
+end subroutine
+! CHECK-LABEL:   func.func @_QPint_r2_assumed_size_to_assumed_rank(
+! CHECK-SAME:                                                      %[[VAL_0:.*]]: !fir.ref<!fir.array<10x?xi32>> {fir.bindc_name = "x"}) {
+! CHECK:           %[[VAL_1:.*]] = arith.constant 10 : i64
+! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_1]] : (i64) -> index
+! CHECK:           %[[VAL_3:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_4:.*]] = arith.cmpi sgt, %[[VAL_2]], %[[VAL_3]] : index
+! CHECK:           %[[VAL_5:.*]] = arith.select %[[VAL_4]], %[[VAL_2]], %[[VAL_3]] : index
+! CHECK:           %[[VAL_6:.*]] = arith.constant -1 : index
+! CHECK:           %[[VAL_7:.*]] = fir.shape %[[VAL_5]], %[[VAL_6]] : (index, index) -> !fir.shape<2>
+! CHECK:           %[[VAL_8:.*]]:2 = hlfir.declare %[[VAL_0]](%[[VAL_7]]) {uniq_name = "_QFint_r2_assumed_size_to_assumed_rankEx"} : (!fir.ref<!fir.array<10x?xi32>>, !fir.shape<2>) -> (!fir.box<!fir.array<10x?xi32>>, !fir.ref<!fir.array<10x?xi32>>)
+! CHECK:           %[[VAL_9:.*]] = fir.convert %[[VAL_8]]#0 : (!fir.box<!fir.array<10x?xi32>>) -> !fir.box<!fir.array<*:i32>>
+! CHECK:           fir.call @_QPint_assumed_rank(%[[VAL_9]]) fastmath<contract> : (!fir.box<!fir.array<*:i32>>) -> ()
diff --git a/flang/test/Lower/Intrinsics/modulo.f90 b/flang/test/Lower/Intrinsics/modulo.f90
@@ -1,11 +1,13 @@
-! RUN: bbc -emit-fir -hlfir=false %s -o - | FileCheck %s
+! RUN: bbc -emit-fir -hlfir=false %s -o - | FileCheck %s -check-prefixes=HONORINF,ALL
+! RUN: flang-new -fc1 -menable-no-infs -emit-fir -flang-deprecated-no-hlfir %s -o - | FileCheck %s -check-prefixes=CHECK,ALL
 
-! CHECK-LABEL: func @_QPmodulo_testr(
-! CHECK-SAME: %[[arg0:.*]]: !fir.ref<f64>{{.*}}, %[[arg1:.*]]: !fir.ref<f64>{{.*}}, %[[arg2:.*]]: !fir.ref<f64>{{.*}}) {
+! ALL-LABEL: func @_QPmodulo_testr(
+! ALL-SAME: %[[arg0:.*]]: !fir.ref<f64>{{.*}}, %[[arg1:.*]]: !fir.ref<f64>{{.*}}, %[[arg2:.*]]: !fir.ref<f64>{{.*}}) {
 subroutine modulo_testr(r, a, p)
   real(8) :: r, a, p
-  ! CHECK-DAG: %[[a:.*]] = fir.load %[[arg1]] : !fir.ref<f64>
-  ! CHECK-DAG: %[[p:.*]] = fir.load %[[arg2]] : !fir.ref<f64>
+  ! ALL-DAG: %[[a:.*]] = fir.load %[[arg1]] : !fir.ref<f64>
+  ! ALL-DAG: %[[p:.*]] = fir.load %[[arg2]] : !fir.ref<f64>
+  ! HONORINF: %[[res:.*]] = fir.call @_FortranAModuloReal8(%[[a]], %[[p]]
   ! CHECK-DAG: %[[rem:.*]] = arith.remf %[[a]], %[[p]] {{.*}}: f64
   ! CHECK-DAG: %[[zero:.*]] = arith.constant 0.000000e+00 : f64
   ! CHECK-DAG: %[[remNotZero:.*]] = arith.cmpf une, %[[rem]], %[[zero]] {{.*}} : f64
@@ -15,12 +17,12 @@ subroutine modulo_testr(r, a, p)
   ! CHECK-DAG: %[[mustAddP:.*]] = arith.andi %[[remNotZero]], %[[signDifferent]] : i1
   ! CHECK-DAG: %[[remPlusP:.*]] = arith.addf %[[rem]], %[[p]] {{.*}}: f64
   ! CHECK: %[[res:.*]] = arith.select %[[mustAddP]], %[[remPlusP]], %[[rem]] : f64
-  ! CHECK: fir.store %[[res]] to %[[arg0]] : !fir.ref<f64>
+  ! ALL: fir.store %[[res]] to %[[arg0]] : !fir.ref<f64>
   r = modulo(a, p)
 end subroutine
 
-! CHECK-LABEL: func @_QPmodulo_testi(
-! CHECK-SAME: %[[arg0:.*]]: !fir.ref<i64>{{.*}}, %[[arg1:.*]]: !fir.ref<i64>{{.*}}, %[[arg2:.*]]: !fir.ref<i64>{{.*}}) {
+! ALL-LABEL: func @_QPmodulo_testi(
+! ALL-SAME: %[[arg0:.*]]: !fir.ref<i64>{{.*}}, %[[arg1:.*]]: !fir.ref<i64>{{.*}}, %[[arg2:.*]]: !fir.ref<i64>{{.*}}) {
 subroutine modulo_testi(r, a, p)
   integer(8) :: r, a, p
   ! CHECK-DAG: %[[a:.*]] = fir.load %[[arg1]] : !fir.ref<i64>

diff --git a/flang/test/Lower/OpenMP/parallel-reduction-array.f90 b/flang/test/Lower/OpenMP/parallel-reduction-array.f90
@@ -50,7 +50,7 @@ program reduce
 ! CHECK:           %[[VAL_1:.*]] = arith.constant 3 : index
 ! CHECK:           %[[VAL_2:.*]] = fir.shape %[[VAL_1]] : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_3:.*]]:2 = hlfir.declare %[[VAL_0]](%[[VAL_2]]) {uniq_name = "_QFEi"} : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> (!fir.ref<!fir.array<3xi32>>, !fir.ref<!fir.array<3xi32>>)
-! CHECK:           %[[VAL_4:.*]] = fir.embox %[[VAL_3]]#1(%[[VAL_2]]) : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<3xi32>>
+! CHECK:           %[[VAL_4:.*]] = fir.embox %[[VAL_3]]#0(%[[VAL_2]]) : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<3xi32>>
 ! CHECK:           %[[VAL_5:.*]] = fir.alloca !fir.box<!fir.array<3xi32>>
 ! CHECK:           fir.store %[[VAL_4]] to %[[VAL_5]] : !fir.ref<!fir.box<!fir.array<3xi32>>>
 ! CHECK:           omp.parallel byref reduction(@add_reduction_byref_box_3xi32 %[[VAL_5]] -> %[[VAL_6:.*]] : !fir.ref<!fir.box<!fir.array<3xi32>>>) {

diff --git a/flang/test/Lower/OpenMP/parallel-reduction-array2.f90 b/flang/test/Lower/OpenMP/parallel-reduction-array2.f90
@@ -50,7 +50,7 @@ program reduce
 ! CHECK:           %[[VAL_1:.*]] = arith.constant 3 : index
 ! CHECK:           %[[VAL_2:.*]] = fir.shape %[[VAL_1]] : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_3:.*]]:2 = hlfir.declare %[[VAL_0]](%[[VAL_2]]) {uniq_name = "_QFEi"} : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> (!fir.ref<!fir.array<3xi32>>, !fir.ref<!fir.array<3xi32>>)
-! CHECK:           %[[VAL_4:.*]] = fir.embox %[[VAL_3]]#1(%[[VAL_2]]) : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<3xi32>>
+! CHECK:           %[[VAL_4:.*]] = fir.embox %[[VAL_3]]#0(%[[VAL_2]]) : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<3xi32>>
 ! CHECK:           %[[VAL_5:.*]] = fir.alloca !fir.box<!fir.array<3xi32>>
 ! CHECK:           fir.store %[[VAL_4]] to %[[VAL_5]] : !fir.ref<!fir.box<!fir.array<3xi32>>>
 ! CHECK:           omp.parallel byref reduction(@add_reduction_byref_box_3xi32 %[[VAL_5]] -> %[[VAL_6:.*]] : !fir.ref<!fir.box<!fir.array<3xi32>>>) {

diff --git a/flang/test/Lower/OpenMP/parallel-reduction3.f90 b/flang/test/Lower/OpenMP/parallel-reduction3.f90
@@ -0,0 +1,125 @@
+! NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
+
+! The script is designed to make adding checks to
+! a test case fast, it is *not* designed to be authoritative
+! about what constitutes a good test! The CHECK should be
+! minimized and named to reflect the test intent.
+
+! RUN: bbc -emit-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
+
+
+
+! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_box_Uxi32 : !fir.ref<!fir.box<!fir.array<?xi32>>> init {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<?xi32>>>):
+! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:           %[[VAL_3:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_4:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_3]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]]#1 : (index) -> !fir.shape<1>
+! CHECK:           %[[VAL_6:.*]] = fir.alloca !fir.array<?xi32>, %[[VAL_4]]#1 {bindc_name = ".tmp"}
+! CHECK:           %[[VAL_7:.*]]:2 = hlfir.declare %[[VAL_6]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.ref<!fir.array<?xi32>>, !fir.shape<1>) -> (!fir.box<!fir.array<?xi32>>, !fir.ref<!fir.array<?xi32>>)
+! CHECK:           hlfir.assign %[[VAL_1]] to %[[VAL_7]]#0 : i32, !fir.box<!fir.array<?xi32>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<?xi32>>
+! CHECK:           fir.store %[[VAL_7]]#0 to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xi32>>>)
+
+! CHECK-LABEL:   } combiner {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<?xi32>>>, %[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.array<?xi32>>>):
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_5:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_4]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_6:.*]] = fir.shape_shift %[[VAL_5]]#0, %[[VAL_5]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[VAL_7:.*]] = arith.constant 1 : index
+! CHECK:           fir.do_loop %[[VAL_8:.*]] = %[[VAL_7]] to %[[VAL_5]]#1 step %[[VAL_7]] unordered {
+! CHECK:             %[[VAL_9:.*]] = fir.array_coor %[[VAL_2]](%[[VAL_6]]) %[[VAL_8]] : (!fir.box<!fir.array<?xi32>>, !fir.shapeshift<1>, index) -> !fir.ref<i32>
+! CHECK:             %[[VAL_10:.*]] = fir.array_coor %[[VAL_3]](%[[VAL_6]]) %[[VAL_8]] : (!fir.box<!fir.array<?xi32>>, !fir.shapeshift<1>, index) -> !fir.ref<i32>
+! CHECK:             %[[VAL_11:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
+! CHECK:             %[[VAL_12:.*]] = fir.load %[[VAL_10]] : !fir.ref<i32>
+! CHECK:             %[[VAL_13:.*]] = arith.addi %[[VAL_11]], %[[VAL_12]] : i32
+! CHECK:             fir.store %[[VAL_13]] to %[[VAL_9]] : !fir.ref<i32>
+! CHECK:           }
+! CHECK:           omp.yield(%[[VAL_0]] : !fir.ref<!fir.box<!fir.array<?xi32>>>)
+! CHECK:         }
+
+! CHECK-LABEL:   func.func @_QPs(
+! CHECK-SAME:                    %[[VAL_0:.*]]: !fir.ref<i32> {fir.bindc_name = "x"}) {
+! CHECK:           %[[VAL_1:.*]]:2 = hlfir.declare %[[VAL_0]] {uniq_name = "_QFsEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:           %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsEi"}
+! CHECK:           %[[VAL_3:.*]]:2 = hlfir.declare %[[VAL_2]] {uniq_name = "_QFsEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:           %[[VAL_4:.*]] = fir.load %[[VAL_1]]#0 : !fir.ref<i32>
+! CHECK:           %[[VAL_5:.*]] = fir.convert %[[VAL_4]] : (i32) -> i64
+! CHECK:           %[[VAL_6:.*]] = fir.convert %[[VAL_5]] : (i64) -> index
+! CHECK:           %[[VAL_7:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_8:.*]] = arith.cmpi sgt, %[[VAL_6]], %[[VAL_7]] : index
+! CHECK:           %[[VAL_9:.*]] = arith.select %[[VAL_8]], %[[VAL_6]], %[[VAL_7]] : index
+! CHECK:           %[[VAL_10:.*]] = fir.alloca !fir.array<?xi32>, %[[VAL_9]] {bindc_name = "c", uniq_name = "_QFsEc"}
+! CHECK:           %[[VAL_11:.*]] = fir.shape %[[VAL_9]] : (index) -> !fir.shape<1>
+! CHECK:           %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]](%[[VAL_11]]) {uniq_name = "_QFsEc"} : (!fir.ref<!fir.array<?xi32>>, !fir.shape<1>) -> (!fir.box<!fir.array<?xi32>>, !fir.ref<!fir.array<?xi32>>)
+! CHECK:           %[[VAL_13:.*]] = arith.constant 0 : i32
+! CHECK:           hlfir.assign %[[VAL_13]] to %[[VAL_12]]#0 : i32, !fir.box<!fir.array<?xi32>>
+! CHECK:           omp.parallel {
+! CHECK:             %[[VAL_14:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
+! CHECK:             %[[VAL_15:.*]]:2 = hlfir.declare %[[VAL_14]] {uniq_name = "_QFsEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:             %[[VAL_16:.*]] = arith.constant 1 : i32
+! CHECK:             %[[VAL_17:.*]] = arith.constant 100 : i32
+! CHECK:             %[[VAL_18:.*]] = arith.constant 1 : i32
+! CHECK:             %[[VAL_19:.*]] = fir.alloca !fir.box<!fir.array<?xi32>>
+! CHECK:             fir.store %[[VAL_12]]#0 to %[[VAL_19]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_Uxi32 %[[VAL_19]] -> %[[VAL_20:.*]] : !fir.ref<!fir.box<!fir.array<?xi32>>>)  for  (%[[VAL_21:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
+! CHECK:               fir.store %[[VAL_21]] to %[[VAL_15]]#1 : !fir.ref<i32>
+! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFsEc"} : (!fir.ref<!fir.box<!fir.array<?xi32>>>) -> (!fir.ref<!fir.box<!fir.array<?xi32>>>, !fir.ref<!fir.box<!fir.array<?xi32>>>)
+! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:               %[[VAL_25:.*]] = arith.constant 0 : index
+! CHECK:               %[[VAL_26:.*]]:3 = fir.box_dims %[[VAL_23]], %[[VAL_25]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:               %[[VAL_27:.*]] = fir.shape %[[VAL_26]]#1 : (index) -> !fir.shape<1>
+! CHECK:               %[[VAL_28:.*]] = hlfir.elemental %[[VAL_27]] unordered : (!fir.shape<1>) -> !hlfir.expr<?xi32> {
+! CHECK:               ^bb0(%[[VAL_29:.*]]: index):
+! CHECK:                 %[[VAL_30:.*]] = arith.constant 0 : index
+! CHECK:                 %[[VAL_31:.*]]:3 = fir.box_dims %[[VAL_23]], %[[VAL_30]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:                 %[[VAL_32:.*]] = arith.constant 1 : index
+! CHECK:                 %[[VAL_33:.*]] = arith.subi %[[VAL_31]]#0, %[[VAL_32]] : index
+! CHECK:                 %[[VAL_34:.*]] = arith.addi %[[VAL_29]], %[[VAL_33]] : index
+! CHECK:                 %[[VAL_35:.*]] = hlfir.designate %[[VAL_23]] (%[[VAL_34]])  : (!fir.box<!fir.array<?xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_37:.*]] = arith.addi %[[VAL_36]], %[[VAL_24]] : i32
+! CHECK:                 hlfir.yield_element %[[VAL_37]] : i32
+! CHECK:               }
+! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:               hlfir.assign %[[VAL_28]] to %[[VAL_38]] : !hlfir.expr<?xi32>, !fir.box<!fir.array<?xi32>>
+! CHECK:               hlfir.destroy %[[VAL_28]] : !hlfir.expr<?xi32>
+! CHECK:               omp.yield
+! CHECK:             }
+! CHECK:             omp.terminator
+! CHECK:           }
+! CHECK:           %[[VAL_39:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_40:.*]] = hlfir.designate %[[VAL_12]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xi32>>, index) -> !fir.ref<i32>
+! CHECK:           %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<i32>
+! CHECK:           %[[VAL_42:.*]] = arith.constant 5050 : i32
+! CHECK:           %[[VAL_43:.*]] = arith.cmpi ne, %[[VAL_41]], %[[VAL_42]] : i32
+! CHECK:           cf.cond_br %[[VAL_43]], ^bb1, ^bb2
+! CHECK:         ^bb1:
+! CHECK:           %[[VAL_44:.*]] = arith.constant 1 : i32
+! CHECK:           %[[VAL_45:.*]] = arith.constant false
+! CHECK:           %[[VAL_46:.*]] = arith.constant false
+! CHECK:           %[[VAL_47:.*]] = fir.call @_FortranAStopStatement(%[[VAL_44]], %[[VAL_45]], %[[VAL_46]]) fastmath<contract> : (i32, i1, i1) -> none
+! CHECK:           fir.unreachable
+! CHECK:         ^bb2:
+! CHECK:           return
+! CHECK:         }
+! CHECK:         func.func private @_FortranAStopStatement(i32, i1, i1) -> none attributes {fir.runtime}
+
+subroutine s(x)
+    integer :: x
+    integer :: c(x)
+    c = 0
+    !$omp parallel do reduction(+:c)
+    do i = 1, 100
+        c = c + i
+    end do
+    !$omp end parallel do
+
+    if (c(1) /= 5050) stop 1
+end subroutine s
-Original file line number
+Diff line change
@@ Expand Up / @@ -5,7 +5,6 @@ build @@
     root
     tags
     TAGS
-    *.o
     .nfs*
     *.sw?
     *~
@@ Expand Down @@