Vendor import of clang trunk r301441:vendor/clang/clang-trunk-r301441

https://llvm.org/svn/llvm-project/cfe/trunk@301441

160 files changed, 7052 insertions, 767 deletions

diff --git a/cmake/caches/Fuchsia-stage2.cmake b/cmake/caches/Fuchsia-stage2.cmake
index 2e1e9a155e14..ca43e603a631 100644
--- a/cmake/caches/Fuchsia-stage2.cmake
+++ b/cmake/caches/Fuchsia-stage2.cmake
@@ -1,7 +1,7 @@
 # This file sets up a CMakeCache for the second stage of a Fuchsia toolchain
 # build.
 
-set(LLVM_TARGETS_TO_BUILD X86;AArch64 CACHE STRING "")
+set(LLVM_TARGETS_TO_BUILD X86;ARM;AArch64 CACHE STRING "")
 
 set(PACKAGE_VENDOR Fuchsia CACHE STRING "")
 
@@ -36,6 +36,7 @@ set(BUILTINS_aarch64-fuchsia-none_CMAKE_SYSTEM_NAME Fuchsia CACHE STRING "")
 # Setup toolchain.
 set(LLVM_INSTALL_TOOLCHAIN_ONLY ON CACHE BOOL "")
 set(LLVM_TOOLCHAIN_TOOLS
+  llc
   llvm-ar
   llvm-cov
   llvm-cxxfilt
@@ -49,6 +50,7 @@ set(LLVM_TOOLCHAIN_TOOLS
   llvm-readobj
   llvm-size
   llvm-symbolizer
+  opt
   CACHE STRING "")
 
 set(LLVM_DISTRIBUTION_COMPONENTS
diff --git a/docs/ClangFormat.rst b/docs/ClangFormat.rst
index dc1e2b9a7639..ed0e58e452d3 100644
--- a/docs/ClangFormat.rst
+++ b/docs/ClangFormat.rst
@@ -162,8 +162,9 @@ Download the latest Visual Studio extension from the `alpha build site
 Script for patch reformatting
 =============================
 
-The python script `clang/tools/clang-format-diff.py` parses the output of
-a unified diff and reformats all contained lines with :program:`clang-format`.
+The python script `clang/tools/clang-format/clang-format-diff.py` parses the
+output of a unified diff and reformats all contained lines with
+:program:`clang-format`.
 
 .. code-block:: console
 
diff --git a/include/clang-c/Index.h b/include/clang-c/Index.h
index 1ac2b7f802c2..9acc9b75bffb 100644
--- a/include/clang-c/Index.h
+++ b/include/clang-c/Index.h
@@ -5600,7 +5600,8 @@ typedef enum {
   CXIdxEntityLang_None = 0,
   CXIdxEntityLang_C    = 1,
   CXIdxEntityLang_ObjC = 2,
-  CXIdxEntityLang_CXX  = 3
+  CXIdxEntityLang_CXX  = 3,
+  CXIdxEntityLang_Swift  = 4
 } CXIdxEntityLanguage;
 
 /**
diff --git a/include/clang/AST/CommentSema.h b/include/clang/AST/CommentSema.h
index 6a803836e848..230e52739f24 100644
--- a/include/clang/AST/CommentSema.h
+++ b/include/clang/AST/CommentSema.h
@@ -208,6 +208,10 @@ public:
   /// \returns \c true if declaration that this comment is attached to declares
   /// a function pointer.
   bool isFunctionPointerVarDecl();
+  /// \returns \c true if the declaration that this comment is attached to
+  /// declares a variable or a field whose type is a function or a block
+  /// pointer.
+  bool isFunctionOrBlockPointerVarLikeDecl();
   bool isFunctionOrMethodVariadic();
   bool isObjCMethodDecl();
   bool isObjCPropertyDecl();
diff --git a/include/clang/ASTMatchers/ASTMatchers.h b/include/clang/ASTMatchers/ASTMatchers.h
index fb029470c830..f11469b8fc20 100644
--- a/include/clang/ASTMatchers/ASTMatchers.h
+++ b/include/clang/ASTMatchers/ASTMatchers.h
@@ -1406,7 +1406,7 @@ const internal::VariadicDynCastAllOfMatcher<
 ///
 /// Example: Given
 /// \code
-///   struct T {void func()};
+///   struct T {void func();};
 ///   T f();
 ///   void g(T);
 /// \endcode
diff --git a/include/clang/Basic/Diagnostic.td b/include/clang/Basic/Diagnostic.td
index 39da0060ddbd..f25068eca132 100644
--- a/include/clang/Basic/Diagnostic.td
+++ b/include/clang/Basic/Diagnostic.td
@@ -12,6 +12,8 @@
 //
 //===----------------------------------------------------------------------===//
 
+// See the Internals Manual, section The Diagnostics Subsystem for an overview.
+
 // Define the diagnostic severities.
 class Severity<string N> {
   string Name = N;
@@ -100,10 +102,20 @@ class SuppressInSystemHeader {
 class Error<string str>     : Diagnostic<str, CLASS_ERROR, SEV_Error>, SFINAEFailure {
   bit ShowInSystemHeader = 1;
 }
+// Warnings default to on (but can be default-off'd with DefaultIgnore).
+// This is used for warnings about questionable code; warnings about
+// accepted language extensions should use Extension or ExtWarn below instead.
 class Warning<string str>   : Diagnostic<str, CLASS_WARNING, SEV_Warning>;
+// Remarks can be turned on with -R flags and provide commentary, e.g. on
+// optimizer decisions.
 class Remark<string str>    : Diagnostic<str, CLASS_REMARK, SEV_Ignored>;
+// Extensions are warnings about accepted language extensions.
+// Extension warnings are default-off but enabled by -pedantic.
 class Extension<string str> : Diagnostic<str, CLASS_EXTENSION, SEV_Ignored>;
+// ExtWarns are warnings about accepted language extensions.
+// ExtWarn warnings are default-on.
 class ExtWarn<string str>   : Diagnostic<str, CLASS_EXTENSION, SEV_Warning>;
+// Notes can provide supplementary information on errors, warnings, and remarks.
 class Note<string str>      : Diagnostic<str, CLASS_NOTE, SEV_Fatal/*ignored*/>;
 
 
diff --git a/include/clang/Basic/DiagnosticParseKinds.td b/include/clang/Basic/DiagnosticParseKinds.td
index d95e43c10c55..f04ed8ed4ce6 100644
--- a/include/clang/Basic/DiagnosticParseKinds.td
+++ b/include/clang/Basic/DiagnosticParseKinds.td
@@ -1114,14 +1114,12 @@ def err_pragma_cannot_end_force_cuda_host_device : Error<
 } // end of Parse Issue category.
 
 let CategoryName = "Modules Issue" in {
-def err_expected_module_interface_decl : Error<
-  "expected module declaration at start of module interface">;
 def err_unexpected_module_decl : Error<
-  "module declaration must be the first declaration in the translation unit">;
+  "module declaration can only appear at the top level">;
 def err_module_expected_ident : Error<
-  "expected a module name after module%select{| import}0">;
-def err_unexpected_module_kind : Error<
-  "unexpected module kind %0; expected 'implementation' or 'partition'">;
+  "expected a module name after '%select{module|import}0'">;
+def err_module_implementation_partition : Error<
+  "module partition must be declared 'export'">;
 def err_attribute_not_module_attr : Error<
   "%0 attribute cannot be applied to a module">;
 def err_attribute_not_import_attr : Error<
diff --git a/include/clang/Basic/DiagnosticSemaKinds.td b/include/clang/Basic/DiagnosticSemaKinds.td
index 6cb872cc27c5..b5fed1f218b6 100644
--- a/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/include/clang/Basic/DiagnosticSemaKinds.td
@@ -8801,9 +8801,11 @@ def err_invalid_type_for_program_scope_var : Error<
 }
 
 let CategoryName = "Modules Issue" in {
+def err_module_decl_in_module_map_module : Error<
+  "'module' declaration found while building module from module map">;
 def err_module_interface_implementation_mismatch : Error<
-  "%select{'module'|'module partition'|'module implementation'}0 declaration "
-  "found while %select{not |not |}0building module interface">;
+  "missing 'export' specifier in module declaration while "
+  "building module interface">;
 def err_current_module_name_mismatch : Error<
   "module name '%0' specified on command line does not match name of module">;
 def err_module_redefinition : Error<
@@ -8846,8 +8848,13 @@ def err_module_self_import : Error<
   "import of module '%0' appears within same top-level module '%1'">;
 def err_module_import_in_implementation : Error<
   "@import of module '%0' in implementation of '%1'; use #import">;
+
+// C++ Modules TS
 def err_export_within_export : Error<
   "export declaration appears within another export declaration">;
+def err_export_not_in_module_interface : Error<
+  "export declaration can only be used within a module interface unit after "
+  "the module declaration">;
 
 def ext_equivalent_internal_linkage_decl_in_modules : ExtWarn<
   "ambiguous use of internal linkage declaration %0 defined in multiple modules">,
diff --git a/include/clang/Basic/Module.h b/include/clang/Basic/Module.h
index 8fcb0e8b056a..28aa7db52992 100644
--- a/include/clang/Basic/Module.h
+++ b/include/clang/Basic/Module.h
@@ -62,6 +62,18 @@ public:
   /// \brief The location of the module definition.
   SourceLocation DefinitionLoc;
 
+  enum ModuleKind {
+    /// \brief This is a module that was defined by a module map and built out
+    /// of header files.
+    ModuleMapModule,
+
+    /// \brief This is a C++ Modules TS module interface unit.
+    ModuleInterfaceUnit
+  };
+
+  /// \brief The kind of this module.
+  ModuleKind Kind = ModuleMapModule;
+
   /// \brief The parent of this module. This will be NULL for the top-level
   /// module.
   Module *Parent;
diff --git a/include/clang/Basic/Sanitizers.def b/include/clang/Basic/Sanitizers.def
index c574045e139a..f20d326e08f8 100644
--- a/include/clang/Basic/Sanitizers.def
+++ b/include/clang/Basic/Sanitizers.def
@@ -47,6 +47,9 @@ SANITIZER("kernel-address", KernelAddress)
 // MemorySanitizer
 SANITIZER("memory", Memory)
 
+// libFuzzer
+SANITIZER("fuzzer", Fuzzer)
+
 // ThreadSanitizer
 SANITIZER("thread", Thread)
 
diff --git a/include/clang/Driver/CC1Options.td b/include/clang/Driver/CC1Options.td
index ba09bccfe1c5..63348c094c93 100644
--- a/include/clang/Driver/CC1Options.td
+++ b/include/clang/Driver/CC1Options.td
@@ -573,6 +573,8 @@ def fblocks_runtime_optional : Flag<["-"], "fblocks-runtime-optional">,
   HelpText<"Weakly link in the blocks runtime">;
 def fexternc_nounwind : Flag<["-"], "fexternc-nounwind">,
   HelpText<"Assume all functions with C linkage do not unwind">;
+def enable_split_dwarf : Flag<["-"], "enable-split-dwarf">,
+  HelpText<"Use split dwarf/Fission">;
 def split_dwarf_file : Separate<["-"], "split-dwarf-file">,
   HelpText<"File name to use for split dwarf debug info output">;
 def fno_wchar : Flag<["-"], "fno-wchar">,
diff --git a/include/clang/Driver/SanitizerArgs.h b/include/clang/Driver/SanitizerArgs.h
index 2df8077d6da6..c7b3e8006dd5 100644
--- a/include/clang/Driver/SanitizerArgs.h
+++ b/include/clang/Driver/SanitizerArgs.h
@@ -50,6 +50,7 @@ class SanitizerArgs {
   bool needsSharedAsanRt() const { return AsanSharedRuntime; }
   bool needsTsanRt() const { return Sanitizers.has(SanitizerKind::Thread); }
   bool needsMsanRt() const { return Sanitizers.has(SanitizerKind::Memory); }
+  bool needsFuzzer() const { return Sanitizers.has(SanitizerKind::Fuzzer); }
   bool needsLsanRt() const {
     return Sanitizers.has(SanitizerKind::Leak) &&
            !Sanitizers.has(SanitizerKind::Address);
diff --git a/include/clang/Format/Format.h b/include/clang/Format/Format.h
index 05daf4f897f4..9bed253baca2 100644
--- a/include/clang/Format/Format.h
+++ b/include/clang/Format/Format.h
@@ -1512,6 +1512,18 @@ llvm::Expected<tooling::Replacements>
 cleanupAroundReplacements(StringRef Code, const tooling::Replacements &Replaces,
                           const FormatStyle &Style);
 
+/// \brief Represents the status of a formatting attempt.
+struct FormattingAttemptStatus {
+  /// \brief A value of ``false`` means that any of the affected ranges were not
+  /// formatted due to a non-recoverable syntax error.
+  bool FormatComplete = true;
+
+  /// \brief If ``FormatComplete`` is false, ``Line`` records a one-based
+  /// original line number at which a syntax error might have occurred. This is
+  /// based on a best-effort analysis and could be imprecise.
+  unsigned Line = 0;
+};
+
 /// \brief Reformats the given \p Ranges in \p Code.
 ///
 /// Each range is extended on either end to its next bigger logic unit, i.e.
@@ -1521,13 +1533,20 @@ cleanupAroundReplacements(StringRef Code, const tooling::Replacements &Replaces,
 /// Returns the ``Replacements`` necessary to make all \p Ranges comply with
 /// \p Style.
 ///
-/// If ``IncompleteFormat`` is non-null, its value will be set to true if any
-/// of the affected ranges were not formatted due to a non-recoverable syntax
-/// error.
+/// If ``Status`` is non-null, its value will be populated with the status of
+/// this formatting attempt. See \c FormattingAttemptStatus.
 tooling::Replacements reformat(const FormatStyle &Style, StringRef Code,
                                ArrayRef<tooling::Range> Ranges,
                                StringRef FileName = "<stdin>",
-                               bool *IncompleteFormat = nullptr);
+                               FormattingAttemptStatus *Status = nullptr);
+
+/// \brief Same as above, except if ``IncompleteFormat`` is non-null, its value
+/// will be set to true if any of the affected ranges were not formatted due to
+/// a non-recoverable syntax error.
+tooling::Replacements reformat(const FormatStyle &Style, StringRef Code,
+                               ArrayRef<tooling::Range> Ranges,
+                               StringRef FileName,
+                               bool *IncompleteFormat);
 
 /// \brief Clean up any erroneous/redundant code in the given \p Ranges in \p
 /// Code.
diff --git a/include/clang/Frontend/CodeGenOptions.def b/include/clang/Frontend/CodeGenOptions.def
index 9565b22e7059..0b7d466ab3d4 100644
--- a/include/clang/Frontend/CodeGenOptions.def
+++ b/include/clang/Frontend/CodeGenOptions.def
@@ -199,6 +199,7 @@ CODEGENOPT(DebugTypeExtRefs, 1, 0) ///< Whether or not debug info should contain
 CODEGENOPT(DebugExplicitImport, 1, 0)  ///< Whether or not debug info should
                                        ///< contain explicit imports for
                                        ///< anonymous namespaces
+CODEGENOPT(EnableSplitDwarf, 1, 0) ///< Whether to enable split DWARF
 CODEGENOPT(SplitDwarfInlining, 1, 1) ///< Whether to include inlining info in the
                                      ///< skeleton CU to allow for symbolication
 				     ///< of inline stack frames without .dwo files.
diff --git a/include/clang/Index/IndexSymbol.h b/include/clang/Index/IndexSymbol.h
index bc34938fb405..abb132f9e4ce 100644
--- a/include/clang/Index/IndexSymbol.h
+++ b/include/clang/Index/IndexSymbol.h
@@ -59,6 +59,7 @@ enum class SymbolLanguage {
   C,
   ObjC,
   CXX,
+  Swift,
 };
 
 /// Language specific sub-kinds.
diff --git a/include/clang/Index/USRGeneration.h b/include/clang/Index/USRGeneration.h
index 61f2c9d1ff13..8c661bd63cd8 100644
--- a/include/clang/Index/USRGeneration.h
+++ b/include/clang/Index/USRGeneration.h
@@ -30,10 +30,14 @@ static inline StringRef getUSRSpacePrefix() {
 bool generateUSRForDecl(const Decl *D, SmallVectorImpl<char> &Buf);
 
 /// \brief Generate a USR fragment for an Objective-C class.
-void generateUSRForObjCClass(StringRef Cls, raw_ostream &OS);
+void generateUSRForObjCClass(StringRef Cls, raw_ostream &OS,
+                             StringRef ExtSymbolDefinedIn = "",
+                             StringRef CategoryContextExtSymbolDefinedIn = "");
 
 /// \brief Generate a USR fragment for an Objective-C class category.
-void generateUSRForObjCCategory(StringRef Cls, StringRef Cat, raw_ostream &OS);
+void generateUSRForObjCCategory(StringRef Cls, StringRef Cat, raw_ostream &OS,
+                                StringRef ClsExtSymbolDefinedIn = "",
+                                StringRef CatExtSymbolDefinedIn = "");
 
 /// \brief Generate a USR fragment for an Objective-C instance variable.  The
 /// complete USR can be created by concatenating the USR for the
@@ -48,7 +52,15 @@ void generateUSRForObjCMethod(StringRef Sel, bool IsInstanceMethod,
 void generateUSRForObjCProperty(StringRef Prop, bool isClassProp, raw_ostream &OS);
 
 /// \brief Generate a USR fragment for an Objective-C protocol.
-void generateUSRForObjCProtocol(StringRef Prot, raw_ostream &OS);
+void generateUSRForObjCProtocol(StringRef Prot, raw_ostream &OS,
+                                StringRef ExtSymbolDefinedIn = "");
+
+/// Generate USR fragment for a global (non-nested) enum.
+void generateUSRForGlobalEnum(StringRef EnumName, raw_ostream &OS,
+                              StringRef ExtSymbolDefinedIn = "");
+
+/// Generate a USR fragment for an enum constant.
+void generateUSRForEnumConstant(StringRef EnumConstantName, raw_ostream &OS);
 
 /// \brief Generate a USR for a macro, including the USR prefix.
 ///
diff --git a/include/clang/Sema/Sema.h b/include/clang/Sema/Sema.h
index bd68842c9f73..0f16cb975b2a 100644
--- a/include/clang/Sema/Sema.h
+++ b/include/clang/Sema/Sema.h
@@ -1934,7 +1934,8 @@ public:
 
   /// The parser has processed a module-declaration that begins the definition
   /// of a module interface or implementation.
-  DeclGroupPtrTy ActOnModuleDecl(SourceLocation ModuleLoc, ModuleDeclKind MDK,
+  DeclGroupPtrTy ActOnModuleDecl(SourceLocation StartLoc,
+                                 SourceLocation ModuleLoc, ModuleDeclKind MDK,
                                  ModuleIdPath Path);
 
   /// \brief The parser has processed a module import declaration.
@@ -8326,6 +8327,12 @@ private:
   /// Returns OpenMP nesting level for current directive.
   unsigned getOpenMPNestingLevel() const;
 
+  /// Push new OpenMP function region for non-capturing function.
+  void pushOpenMPFunctionRegion();
+
+  /// Pop OpenMP function region for non-capturing function.
+  void popOpenMPFunctionRegion(const sema::FunctionScopeInfo *OldFSI);
+
   /// Checks if a type or a declaration is disabled due to the owning extension
   /// being disabled, and emits diagnostic messages if it is disabled.
   /// \param D type or declaration to be checked.
diff --git a/lib/AST/Comment.cpp b/lib/AST/Comment.cpp
index 7a7d3dd8304e..dfa2a1665d40 100644
--- a/lib/AST/Comment.cpp
+++ b/lib/AST/Comment.cpp
@@ -116,6 +116,9 @@ bool ParagraphComment::isWhitespaceNoCache() const {
 static TypeLoc lookThroughTypedefOrTypeAliasLocs(TypeLoc &SrcTL) {
   TypeLoc TL = SrcTL.IgnoreParens();
 
+  // Look through attribute types.
+  if (AttributedTypeLoc AttributeTL = TL.getAs<AttributedTypeLoc>())
+    return AttributeTL.getModifiedLoc();
   // Look through qualified types.
   if (QualifiedTypeLoc QualifiedTL = TL.getAs<QualifiedTypeLoc>())
     return QualifiedTL.getUnqualifiedLoc();
@@ -280,8 +283,25 @@ void DeclInfo::fill() {
   case Decl::EnumConstant:
   case Decl::ObjCIvar:
   case Decl::ObjCAtDefsField:
+  case Decl::ObjCProperty: {
+    const TypeSourceInfo *TSI;
+    if (const auto *VD = dyn_cast<DeclaratorDecl>(CommentDecl))
+      TSI = VD->getTypeSourceInfo();
+    else if (const auto *PD = dyn_cast<ObjCPropertyDecl>(CommentDecl))
+      TSI = PD->getTypeSourceInfo();
+    else
+      TSI = nullptr;
+    if (TSI) {
+      TypeLoc TL = TSI->getTypeLoc().getUnqualifiedLoc();
+      FunctionTypeLoc FTL;
+      if (getFunctionTypeLoc(TL, FTL)) {
+        ParamVars = FTL.getParams();
+        ReturnType = FTL.getReturnLoc().getType();
+      }
+    }
     Kind = VariableKind;
     break;
+  }
   case Decl::Namespace:
     Kind = NamespaceKind;
     break;
diff --git a/lib/AST/CommentSema.cpp b/lib/AST/CommentSema.cpp
index d39a9b26b2a8..403454d3ab7e 100644
--- a/lib/AST/CommentSema.cpp
+++ b/lib/AST/CommentSema.cpp
@@ -86,7 +86,7 @@ ParamCommandComment *Sema::actOnParamCommandStart(
       new (Allocator) ParamCommandComment(LocBegin, LocEnd, CommandID,
                                           CommandMarker);
 
-  if (!isFunctionDecl())
+  if (!isFunctionDecl() && !isFunctionOrBlockPointerVarLikeDecl())
     Diag(Command->getLocation(),
          diag::warn_doc_param_not_attached_to_a_function_decl)
       << CommandMarker
@@ -584,7 +584,11 @@ void Sema::checkReturnsCommand(const BlockCommandComment *Command) {
 
   assert(ThisDeclInfo && "should not call this check on a bare comment");
 
-  if (isFunctionDecl()) {
+  // We allow the return command for all @properties because it can be used
+  // to document the value that the property getter returns.
+  if (isObjCPropertyDecl())
+    return;
+  if (isFunctionDecl() || isFunctionOrBlockPointerVarLikeDecl()) {
     if (ThisDeclInfo->ReturnType->isVoidType()) {
       unsigned DiagKind;
       switch (ThisDeclInfo->CommentDecl->getKind()) {
@@ -610,8 +614,6 @@ void Sema::checkReturnsCommand(const BlockCommandComment *Command) {
     }
     return;
   }
-  else if (isObjCPropertyDecl())
-    return;
 
   Diag(Command->getLocation(),
        diag::warn_doc_returns_not_attached_to_a_function_decl)
@@ -844,6 +846,30 @@ bool Sema::isFunctionPointerVarDecl() {
   return false;
 }
 
+bool Sema::isFunctionOrBlockPointerVarLikeDecl() {
+  if (!ThisDeclInfo)
+    return false;
+  if (!ThisDeclInfo->IsFilled)
+    inspectThisDecl();
+  if (ThisDeclInfo->getKind() != DeclInfo::VariableKind ||
+      !ThisDeclInfo->CurrentDecl)
+    return false;
+  QualType QT;
+  if (const auto *VD = dyn_cast<DeclaratorDecl>(ThisDeclInfo->CurrentDecl))
+    QT = VD->getType();
+  else if (const auto *PD =
+               dyn_cast<ObjCPropertyDecl>(ThisDeclInfo->CurrentDecl))
+    QT = PD->getType();
+  else
+    return false;
+  // We would like to warn about the 'returns'/'param' commands for
+  // variables that don't directly specify the function type, so type aliases
+  // can be ignored.
+  if (QT->getAs<TypedefType>())
+    return false;
+  return QT->isFunctionPointerType() || QT->isBlockPointerType();
+}
+
 bool Sema::isObjCPropertyDecl() {
   if (!ThisDeclInfo)
     return false;
diff --git a/lib/AST/Decl.cpp b/lib/AST/Decl.cpp
index 094e8dcff088..0f2558e24ba5 100644
--- a/lib/AST/Decl.cpp
+++ b/lib/AST/Decl.cpp
@@ -2251,6 +2251,14 @@ bool VarDecl::checkInitIsICE() const {
   return Eval->IsICE;
 }
 
+template<typename DeclT>
+static DeclT *getDefinitionOrSelf(DeclT *D) {
+  assert(D);
+  if (auto *Def = D->getDefinition())
+    return Def;
+  return D;
+}
+
 VarDecl *VarDecl::getTemplateInstantiationPattern() const {
   // If it's a variable template specialization, find the template or partial
   // specialization from which it was instantiated.
@@ -2262,7 +2270,7 @@ VarDecl *VarDecl::getTemplateInstantiationPattern() const {
           break;
         VTD = NewVTD;
       }
-      return VTD->getTemplatedDecl()->getDefinition();
+      return getDefinitionOrSelf(VTD->getTemplatedDecl());
     }
     if (auto *VTPSD =
             From.dyn_cast<VarTemplatePartialSpecializationDecl *>()) {
@@ -2271,7 +2279,7 @@ VarDecl *VarDecl::getTemplateInstantiationPattern() const {
           break;
         VTPSD = NewVTPSD;
       }
-      return VTPSD->getDefinition();
+      return getDefinitionOrSelf<VarDecl>(VTPSD);
     }
   }
 
@@ -2280,23 +2288,18 @@ VarDecl *VarDecl::getTemplateInstantiationPattern() const {
       VarDecl *VD = getInstantiatedFromStaticDataMember();
       while (auto *NewVD = VD->getInstantiatedFromStaticDataMember())
         VD = NewVD;
-      return VD->getDefinition();
+      return getDefinitionOrSelf(VD);
     }
   }
 
   if (VarTemplateDecl *VarTemplate = getDescribedVarTemplate()) {
-
     while (VarTemplate->getInstantiatedFromMemberTemplate()) {
       if (VarTemplate->isMemberSpecialization())
         break;
       VarTemplate = VarTemplate->getInstantiatedFromMemberTemplate();
     }
 
-    assert((!VarTemplate->getTemplatedDecl() ||
-            !isTemplateInstantiation(getTemplateSpecializationKind())) &&
-           "couldn't find pattern for variable instantiation");
-
-    return VarTemplate->getTemplatedDecl();
+    return getDefinitionOrSelf(VarTemplate->getTemplatedDecl());
   }
   return nullptr;
 }
@@ -3198,9 +3201,12 @@ bool FunctionDecl::isTemplateInstantiation() const {
    
 FunctionDecl *FunctionDecl::getTemplateInstantiationPattern() const {
   // Handle class scope explicit specialization special case.
-  if (getTemplateSpecializationKind() == TSK_ExplicitSpecialization)
-    return getClassScopeSpecializationPattern();
-  
+  if (getTemplateSpecializationKind() == TSK_ExplicitSpecialization) {
+    if (auto *Spec = getClassScopeSpecializationPattern())
+      return getDefinitionOrSelf(Spec);
+    return nullptr;
+  }
+
   // If this is a generic lambda call operator specialization, its 
   // instantiation pattern is always its primary template's pattern
   // even if its primary template was instantiated from another 
@@ -3212,16 +3218,10 @@ FunctionDecl *FunctionDecl::getTemplateInstantiationPattern() const {
 
   if (isGenericLambdaCallOperatorSpecialization(
           dyn_cast<CXXMethodDecl>(this))) {
-    assert(getPrimaryTemplate() && "A generic lambda specialization must be "
-                                   "generated from a primary call operator "
-                                   "template");
-    assert(getPrimaryTemplate()->getTemplatedDecl()->getBody() &&
-           "A generic lambda call operator template must always have a body - "
-           "even if instantiated from a prototype (i.e. as written) member "
-           "template");
-    return getPrimaryTemplate()->getTemplatedDecl();
+    assert(getPrimaryTemplate() && "not a generic lambda call operator?");
+    return getDefinitionOrSelf(getPrimaryTemplate()->getTemplatedDecl());
   }
-  
+
   if (FunctionTemplateDecl *Primary = getPrimaryTemplate()) {
     while (Primary->getInstantiatedFromMemberTemplate()) {
       // If we have hit a point where the user provided a specialization of
@@ -3230,11 +3230,14 @@ FunctionDecl *FunctionDecl::getTemplateInstantiationPattern() const {
         break;
       Primary = Primary->getInstantiatedFromMemberTemplate();
     }
-    
-    return Primary->getTemplatedDecl();
+
+    return getDefinitionOrSelf(Primary->getTemplatedDecl());
   } 
-    
-  return getInstantiatedFromMemberFunction();
+
+  if (auto *MFD = getInstantiatedFromMemberFunction())
+    return getDefinitionOrSelf(MFD);
+
+  return nullptr;
 }
 
 FunctionTemplateDecl *FunctionDecl::getPrimaryTemplate() const {
@@ -3778,7 +3781,7 @@ EnumDecl *EnumDecl::getTemplateInstantiationPattern() const {
       EnumDecl *ED = getInstantiatedFromMemberEnum();
       while (auto *NewED = ED->getInstantiatedFromMemberEnum())
         ED = NewED;
-      return ED;
+      return getDefinitionOrSelf(ED);
     }
   }
 
diff --git a/lib/AST/DeclCXX.cpp b/lib/AST/DeclCXX.cpp
index 2e5cec9c108f..dd8f768c5711 100644
--- a/lib/AST/DeclCXX.cpp
+++ b/lib/AST/DeclCXX.cpp
@@ -1364,6 +1364,13 @@ CXXRecordDecl::setTemplateSpecializationKind(TemplateSpecializationKind TSK) {
 }
 
 const CXXRecordDecl *CXXRecordDecl::getTemplateInstantiationPattern() const {
+  auto GetDefinitionOrSelf =
+      [](const CXXRecordDecl *D) -> const CXXRecordDecl * {
+    if (auto *Def = D->getDefinition())
+      return Def;
+    return D;
+  };
+
   // If it's a class template specialization, find the template or partial
   // specialization from which it was instantiated.
   if (auto *TD = dyn_cast<ClassTemplateSpecializationDecl>(this)) {
@@ -1374,7 +1381,7 @@ const CXXRecordDecl *CXXRecordDecl::getTemplateInstantiationPattern() const {
           break;
         CTD = NewCTD;
       }
-      return CTD->getTemplatedDecl()->getDefinition();
+      return GetDefinitionOrSelf(CTD->getTemplatedDecl());
     }
     if (auto *CTPSD =
             From.dyn_cast<ClassTemplatePartialSpecializationDecl *>()) {
@@ -1383,7 +1390,7 @@ const CXXRecordDecl *CXXRecordDecl::getTemplateInstantiationPattern() const {
           break;
         CTPSD = NewCTPSD;
       }
-      return CTPSD->getDefinition();
+      return GetDefinitionOrSelf(CTPSD);
     }
   }
 
@@ -1392,7 +1399,7 @@ const CXXRecordDecl *CXXRecordDecl::getTemplateInstantiationPattern() const {
       const CXXRecordDecl *RD = this;
       while (auto *NewRD = RD->getInstantiatedFromMemberClass())
         RD = NewRD;
-      return RD->getDefinition();
+      return GetDefinitionOrSelf(RD);
     }
   }
 
diff --git a/lib/AST/DeclObjC.cpp b/lib/AST/DeclObjC.cpp
index 9218eb537a60..a12a38033c4a 100644
--- a/lib/AST/DeclObjC.cpp
+++ b/lib/AST/DeclObjC.cpp
@@ -539,9 +539,18 @@ void ObjCInterfaceDecl::getDesignatedInitializers(
 
 bool ObjCInterfaceDecl::isDesignatedInitializer(Selector Sel,
                                       const ObjCMethodDecl **InitMethod) const {
+  bool HasCompleteDef = isThisDeclarationADefinition();
+  // During deserialization the data record for the ObjCInterfaceDecl could
+  // be made invariant by reusing the canonical decl. Take this into account
+  // when checking for the complete definition.
+  if (!HasCompleteDef && getCanonicalDecl()->hasDefinition() &&
+      getCanonicalDecl()->getDefinition() == getDefinition())
+    HasCompleteDef = true;
+
   // Check for a complete definition and recover if not so.
-  if (!isThisDeclarationADefinition())
+  if (!HasCompleteDef)
     return false;
+
   if (data().ExternallyCompleted)
     LoadExternalDefinition();
 
diff --git a/lib/CodeGen/BackendUtil.cpp b/lib/CodeGen/BackendUtil.cpp
index 20059d922f90..85788b427208 100644
--- a/lib/CodeGen/BackendUtil.cpp
+++ b/lib/CodeGen/BackendUtil.cpp
@@ -129,16 +129,20 @@ public:
 // that we add to the PassManagerBuilder.
 class PassManagerBuilderWrapper : public PassManagerBuilder {
 public:
-  PassManagerBuilderWrapper(const CodeGenOptions &CGOpts,
+  PassManagerBuilderWrapper(const Triple &TargetTriple,
+                            const CodeGenOptions &CGOpts,
                             const LangOptions &LangOpts)
-      : PassManagerBuilder(), CGOpts(CGOpts), LangOpts(LangOpts) {}
+      : PassManagerBuilder(), TargetTriple(TargetTriple), CGOpts(CGOpts),
+        LangOpts(LangOpts) {}
+  const Triple &getTargetTriple() const { return TargetTriple; }
   const CodeGenOptions &getCGOpts() const { return CGOpts; }
   const LangOptions &getLangOpts() const { return LangOpts; }
+
 private:
+  const Triple &TargetTriple;
   const CodeGenOptions &CGOpts;
   const LangOptions &LangOpts;
 };
-
 }
 
 static void addObjCARCAPElimPass(const PassManagerBuilder &Builder, PassManagerBase &PM) {
@@ -185,16 +189,35 @@ static void addSanitizerCoveragePass(const PassManagerBuilder &Builder,
   PM.add(createSanitizerCoverageModulePass(Opts));
 }
 
+// Check if ASan should use GC-friendly instrumentation for globals.
+// First of all, there is no point if -fdata-sections is off (expect for MachO,
+// where this is not a factor). Also, on ELF this feature requires an assembler
+// extension that only works with -integrated-as at the moment.
+static bool asanUseGlobalsGC(const Triple &T, const CodeGenOptions &CGOpts) {
+  switch (T.getObjectFormat()) {
+  case Triple::MachO:
+  case Triple::COFF:
+    return true;
+  case Triple::ELF:
+    return CGOpts.DataSections && !CGOpts.DisableIntegratedAS;
+  default:
+    return false;
+  }
+}
+
 static void addAddressSanitizerPasses(const PassManagerBuilder &Builder,
                                       legacy::PassManagerBase &PM) {
   const PassManagerBuilderWrapper &BuilderWrapper =
       static_cast<const PassManagerBuilderWrapper&>(Builder);
+  const Triple &T = BuilderWrapper.getTargetTriple();
   const CodeGenOptions &CGOpts = BuilderWrapper.getCGOpts();
   bool Recover = CGOpts.SanitizeRecover.has(SanitizerKind::Address);
   bool UseAfterScope = CGOpts.SanitizeAddressUseAfterScope;
+  bool UseGlobalsGC = asanUseGlobalsGC(T, CGOpts);
   PM.add(createAddressSanitizerFunctionPass(/*CompileKernel*/ false, Recover,
                                             UseAfterScope));
-  PM.add(createAddressSanitizerModulePass(/*CompileKernel*/false, Recover));
+  PM.add(createAddressSanitizerModulePass(/*CompileKernel*/ false, Recover,
+                                          UseGlobalsGC));
 }
 
 static void addKernelAddressSanitizerPasses(const PassManagerBuilder &Builder,
@@ -407,6 +430,8 @@ static void initTargetOptions(llvm::TargetOptions &Options,
   Options.EmulatedTLS = CodeGenOpts.EmulatedTLS;
   Options.DebuggerTuning = CodeGenOpts.getDebuggerTuning();
 
+  if (CodeGenOpts.EnableSplitDwarf)
+    Options.MCOptions.SplitDwarfFile = CodeGenOpts.SplitDwarfFile;
   Options.MCOptions.MCRelaxAll = CodeGenOpts.RelaxAll;
   Options.MCOptions.MCSaveTempLabels = CodeGenOpts.SaveTempLabels;
   Options.MCOptions.MCUseDwarfDirectory = !CodeGenOpts.NoDwarfDirectoryAsm;
@@ -434,8 +459,6 @@ void EmitAssemblyHelper::CreatePasses(legacy::PassManager &MPM,
   if (CodeGenOpts.DisableLLVMPasses)
     return;
 
-  PassManagerBuilderWrapper PMBuilder(CodeGenOpts, LangOpts);
-
   // Figure out TargetLibraryInfo.  This needs to be added to MPM and FPM
   // manually (and not via PMBuilder), since some passes (eg. InstrProfiling)
   // are inserted before PMBuilder ones - they'd get the default-constructed
@@ -444,6 +467,8 @@ void EmitAssemblyHelper::CreatePasses(legacy::PassManager &MPM,
   std::unique_ptr<TargetLibraryInfoImpl> TLII(
       createTLII(TargetTriple, CodeGenOpts));
 
+  PassManagerBuilderWrapper PMBuilder(TargetTriple, CodeGenOpts, LangOpts);
+
   // At O0 and O1 we only run the always inliner which is more efficient. At
   // higher optimization levels we run the normal inliner.
   if (CodeGenOpts.OptimizationLevel <= 1) {
diff --git a/lib/CodeGen/CGDebugInfo.cpp b/lib/CodeGen/CGDebugInfo.cpp
index 2f6a2b95fb61..dd32a44393c6 100644
--- a/lib/CodeGen/CGDebugInfo.cpp
+++ b/lib/CodeGen/CGDebugInfo.cpp
@@ -528,12 +528,14 @@ void CGDebugInfo::CreateCompileUnit() {
   // Create new compile unit.
   // FIXME - Eliminate TheCU.
   TheCU = DBuilder.createCompileUnit(
-      LangTag, DBuilder.createFile(remapDIPath(MainFileName),
-                                   remapDIPath(getCurrentDirname()), CSKind,
-                                   Checksum),
+      LangTag,
+      DBuilder.createFile(remapDIPath(MainFileName),
+                          remapDIPath(getCurrentDirname()), CSKind, Checksum),
       Producer, LO.Optimize, CGM.getCodeGenOpts().DwarfDebugFlags, RuntimeVers,
-      CGM.getCodeGenOpts().SplitDwarfFile, EmissionKind, 0 /* DWOid */,
-      CGM.getCodeGenOpts().SplitDwarfInlining,
+      CGM.getCodeGenOpts().EnableSplitDwarf
+          ? ""
+          : CGM.getCodeGenOpts().SplitDwarfFile,
+      EmissionKind, 0 /* DWOid */, CGM.getCodeGenOpts().SplitDwarfInlining,
       CGM.getCodeGenOpts().DebugInfoForProfiling);
 }
 
diff --git a/lib/CodeGen/CGExpr.cpp b/lib/CodeGen/CGExpr.cpp
index 719147a58e08..d0aacf65428f 100644
--- a/lib/CodeGen/CGExpr.cpp
+++ b/lib/CodeGen/CGExpr.cpp
@@ -533,15 +533,6 @@ bool CodeGenFunction::sanitizePerformTypeCheck() const {
          SanOpts.has(SanitizerKind::Vptr);
 }
 
-/// Check if a runtime null check for \p Ptr can be omitted.
-static bool canOmitPointerNullCheck(llvm::Value *Ptr) {
-  // Note: do not perform any constant-folding in this function. That is best
-  // left to the IR builder.
-
-  // Pointers to alloca'd memory are non-null.
-  return isa<llvm::AllocaInst>(Ptr->stripPointerCastsNoFollowAliases());
-}
-
 void CodeGenFunction::EmitTypeCheck(TypeCheckKind TCK, SourceLocation Loc,
                                     llvm::Value *Ptr, QualType Ty,
                                     CharUnits Alignment,
@@ -560,11 +551,16 @@ void CodeGenFunction::EmitTypeCheck(TypeCheckKind TCK, SourceLocation Loc,
   SmallVector<std::pair<llvm::Value *, SanitizerMask>, 3> Checks;
   llvm::BasicBlock *Done = nullptr;
 
+  // Quickly determine whether we have a pointer to an alloca. It's possible
+  // to skip null checks, and some alignment checks, for these pointers. This
+  // can reduce compile-time significantly.
+  auto PtrToAlloca =
+      dyn_cast<llvm::AllocaInst>(Ptr->stripPointerCastsNoFollowAliases());
+
   bool AllowNullPointers = TCK == TCK_DowncastPointer || TCK == TCK_Upcast ||
                            TCK == TCK_UpcastToVirtualBase;
   if ((SanOpts.has(SanitizerKind::Null) || AllowNullPointers) &&
-      !SkippedChecks.has(SanitizerKind::Null) &&
-      !canOmitPointerNullCheck(Ptr)) {
+      !SkippedChecks.has(SanitizerKind::Null) && !PtrToAlloca) {
     // The glvalue must not be an empty glvalue.
     llvm::Value *IsNonNull = Builder.CreateIsNotNull(Ptr);
 
@@ -617,7 +613,8 @@ void CodeGenFunction::EmitTypeCheck(TypeCheckKind TCK, SourceLocation Loc,
       AlignVal = getContext().getTypeAlignInChars(Ty).getQuantity();
 
     // The glvalue must be suitably aligned.
-    if (AlignVal > 1) {
+    if (AlignVal > 1 &&
+        (!PtrToAlloca || PtrToAlloca->getAlignment() < AlignVal)) {
       llvm::Value *Align =
           Builder.CreateAnd(Builder.CreatePtrToInt(Ptr, IntPtrTy),
                             llvm::ConstantInt::get(IntPtrTy, AlignVal - 1));
diff --git a/lib/CodeGen/CGOpenMPRuntime.cpp b/lib/CodeGen/CGOpenMPRuntime.cpp
index 874b6a69e513..d1a706b8821e 100644
--- a/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -2466,16 +2466,14 @@ static int addMonoNonMonoModifier(OpenMPSchedType Schedule,
   return Schedule | Modifier;
 }
 
-void CGOpenMPRuntime::emitForDispatchInit(CodeGenFunction &CGF,
-                                          SourceLocation Loc,
-                                          const OpenMPScheduleTy &ScheduleKind,
-                                          unsigned IVSize, bool IVSigned,
-                                          bool Ordered, llvm::Value *UB,
-                                          llvm::Value *Chunk) {
+void CGOpenMPRuntime::emitForDispatchInit(
+    CodeGenFunction &CGF, SourceLocation Loc,
+    const OpenMPScheduleTy &ScheduleKind, unsigned IVSize, bool IVSigned,
+    bool Ordered, const DispatchRTInput &DispatchValues) {
   if (!CGF.HaveInsertPoint())
     return;
-  OpenMPSchedType Schedule =
-      getRuntimeSchedule(ScheduleKind.Schedule, Chunk != nullptr, Ordered);
+  OpenMPSchedType Schedule = getRuntimeSchedule(
+      ScheduleKind.Schedule, DispatchValues.Chunk != nullptr, Ordered);
   assert(Ordered ||
          (Schedule != OMP_sch_static && Schedule != OMP_sch_static_chunked &&
           Schedule != OMP_ord_static && Schedule != OMP_ord_static_chunked &&
@@ -2486,14 +2484,14 @@ void CGOpenMPRuntime::emitForDispatchInit(CodeGenFunction &CGF,
   //          kmp_int[32|64] stride, kmp_int[32|64] chunk);
 
   // If the Chunk was not specified in the clause - use default value 1.
-  if (Chunk == nullptr)
-    Chunk = CGF.Builder.getIntN(IVSize, 1);
+  llvm::Value *Chunk = DispatchValues.Chunk ? DispatchValues.Chunk
+                                            : CGF.Builder.getIntN(IVSize, 1);
   llvm::Value *Args[] = {
       emitUpdateLocation(CGF, Loc), getThreadID(CGF, Loc),
       CGF.Builder.getInt32(addMonoNonMonoModifier(
           Schedule, ScheduleKind.M1, ScheduleKind.M2)), // Schedule type
-      CGF.Builder.getIntN(IVSize, 0),                   // Lower
-      UB,                                               // Upper
+      DispatchValues.LB,                                // Lower
+      DispatchValues.UB,                                // Upper
       CGF.Builder.getIntN(IVSize, 1),                   // Stride
       Chunk                                             // Chunk
   };
diff --git a/lib/CodeGen/CGOpenMPRuntime.h b/lib/CodeGen/CGOpenMPRuntime.h
index 7901a6b7a8fc..6f460f121791 100644
--- a/lib/CodeGen/CGOpenMPRuntime.h
+++ b/lib/CodeGen/CGOpenMPRuntime.h
@@ -672,16 +672,50 @@ public:
   ///
   virtual bool isDynamic(OpenMPScheduleClauseKind ScheduleKind) const;
 
+  /// struct with the values to be passed to the dispatch runtime function
+  struct DispatchRTInput {
+    /// Loop lower bound
+    llvm::Value *LB = nullptr;
+    /// Loop upper bound
+    llvm::Value *UB = nullptr;
+    /// Chunk size specified using 'schedule' clause (nullptr if chunk
+    /// was not specified)
+    llvm::Value *Chunk = nullptr;
+    DispatchRTInput() = default;
+    DispatchRTInput(llvm::Value *LB, llvm::Value *UB, llvm::Value *Chunk)
+        : LB(LB), UB(UB), Chunk(Chunk) {}
+  };
+
+  /// Call the appropriate runtime routine to initialize it before start
+  /// of loop.
+
+  /// This is used for non static scheduled types and when the ordered
+  /// clause is present on the loop construct.
+  /// Depending on the loop schedule, it is necessary to call some runtime
+  /// routine before start of the OpenMP loop to get the loop upper / lower
+  /// bounds \a LB and \a UB and stride \a ST.
+  ///
+  /// \param CGF Reference to current CodeGenFunction.
+  /// \param Loc Clang source location.
+  /// \param ScheduleKind Schedule kind, specified by the 'schedule' clause.
+  /// \param IVSize Size of the iteration variable in bits.
+  /// \param IVSigned Sign of the interation variable.
+  /// \param Ordered true if loop is ordered, false otherwise.
+  /// \param DispatchValues struct containing llvm values for lower bound, upper
+  /// bound, and chunk expression.
+  /// For the default (nullptr) value, the chunk 1 will be used.
+  ///
   virtual void emitForDispatchInit(CodeGenFunction &CGF, SourceLocation Loc,
                                    const OpenMPScheduleTy &ScheduleKind,
                                    unsigned IVSize, bool IVSigned, bool Ordered,
-                                   llvm::Value *UB,
-                                   llvm::Value *Chunk = nullptr);
+                                   const DispatchRTInput &DispatchValues);
 
   /// \brief Call the appropriate runtime routine to initialize it before start
   /// of loop.
   ///
-  /// Depending on the loop schedule, it is nesessary to call some runtime
+  /// This is used only in case of static schedule, when the user did not
+  /// specify a ordered clause on the loop construct.
+  /// Depending on the loop schedule, it is necessary to call some runtime
   /// routine before start of the OpenMP loop to get the loop upper / lower
   /// bounds \a LB and \a UB and stride \a ST.
   ///
diff --git a/lib/CodeGen/CGStmtOpenMP.cpp b/lib/CodeGen/CGStmtOpenMP.cpp
index 22269e42c7a0..f738dd0750fa 100644
--- a/lib/CodeGen/CGStmtOpenMP.cpp
+++ b/lib/CodeGen/CGStmtOpenMP.cpp
@@ -87,7 +87,8 @@ public:
 class OMPParallelScope final : public OMPLexicalScope {
   bool EmitPreInitStmt(const OMPExecutableDirective &S) {
     OpenMPDirectiveKind Kind = S.getDirectiveKind();
-    return !isOpenMPTargetExecutionDirective(Kind) &&
+    return !(isOpenMPTargetExecutionDirective(Kind) ||
+             isOpenMPLoopBoundSharingDirective(Kind)) &&
            isOpenMPParallelDirective(Kind);
   }
 
@@ -1249,10 +1250,20 @@ static void emitPostUpdateForReductionClause(
     CGF.EmitBlock(DoneBB, /*IsFinished=*/true);
 }
 
-static void emitCommonOMPParallelDirective(CodeGenFunction &CGF,
-                                           const OMPExecutableDirective &S,
-                                           OpenMPDirectiveKind InnermostKind,
-                                           const RegionCodeGenTy &CodeGen) {
+namespace {
+/// Codegen lambda for appending distribute lower and upper bounds to outlined
+/// parallel function. This is necessary for combined constructs such as
+/// 'distribute parallel for'
+typedef llvm::function_ref<void(CodeGenFunction &,
+                                const OMPExecutableDirective &,
+                                llvm::SmallVectorImpl<llvm::Value *> &)>
+    CodeGenBoundParametersTy;
+} // anonymous namespace
+
+static void emitCommonOMPParallelDirective(
+    CodeGenFunction &CGF, const OMPExecutableDirective &S,
+    OpenMPDirectiveKind InnermostKind, const RegionCodeGenTy &CodeGen,
+    const CodeGenBoundParametersTy &CodeGenBoundParameters) {
   const CapturedStmt *CS = S.getCapturedStmt(OMPD_parallel);
   auto OutlinedFn = CGF.CGM.getOpenMPRuntime().emitParallelOutlinedFunction(
       S, *CS->getCapturedDecl()->param_begin(), InnermostKind, CodeGen);
@@ -1279,11 +1290,20 @@ static void emitCommonOMPParallelDirective(CodeGenFunction &CGF,
 
   OMPParallelScope Scope(CGF, S);
   llvm::SmallVector<llvm::Value *, 16> CapturedVars;
+  // Combining 'distribute' with 'for' requires sharing each 'distribute' chunk
+  // lower and upper bounds with the pragma 'for' chunking mechanism.
+  // The following lambda takes care of appending the lower and upper bound
+  // parameters when necessary
+  CodeGenBoundParameters(CGF, S, CapturedVars);
   CGF.GenerateOpenMPCapturedVars(*CS, CapturedVars);
   CGF.CGM.getOpenMPRuntime().emitParallelCall(CGF, S.getLocStart(), OutlinedFn,
                                               CapturedVars, IfCond);
 }
 
+static void emitEmptyBoundParameters(CodeGenFunction &,
+                                     const OMPExecutableDirective &,
+                                     llvm::SmallVectorImpl<llvm::Value *> &) {}
+
 void CodeGenFunction::EmitOMPParallelDirective(const OMPParallelDirective &S) {
   // Emit parallel region as a standalone region.
   auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
@@ -1304,7 +1324,8 @@ void CodeGenFunction::EmitOMPParallelDirective(const OMPParallelDirective &S) {
     CGF.EmitStmt(cast<CapturedStmt>(S.getAssociatedStmt())->getCapturedStmt());
     CGF.EmitOMPReductionClauseFinal(S, /*ReductionKind=*/OMPD_parallel);
   };
-  emitCommonOMPParallelDirective(*this, S, OMPD_parallel, CodeGen);
+  emitCommonOMPParallelDirective(*this, S, OMPD_parallel, CodeGen,
+                                 emitEmptyBoundParameters);
   emitPostUpdateForReductionClause(
       *this, S, [](CodeGenFunction &) -> llvm::Value * { return nullptr; });
 }
@@ -1649,6 +1670,13 @@ void CodeGenFunction::EmitOMPSimdFinal(
     EmitBlock(DoneBB, /*IsFinished=*/true);
 }
 
+static void emitOMPLoopBodyWithStopPoint(CodeGenFunction &CGF,
+                                         const OMPLoopDirective &S,
+                                         CodeGenFunction::JumpDest LoopExit) {
+  CGF.EmitOMPLoopBody(S, LoopExit);
+  CGF.EmitStopPoint(&S);
+};
+
 void CodeGenFunction::EmitOMPSimdDirective(const OMPSimdDirective &S) {
   auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
     OMPLoopScope PreInitScope(CGF, S);
@@ -1731,9 +1759,12 @@ void CodeGenFunction::EmitOMPSimdDirective(const OMPSimdDirective &S) {
   CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_simd, CodeGen);
 }
 
-void CodeGenFunction::EmitOMPOuterLoop(bool DynamicOrOrdered, bool IsMonotonic,
-    const OMPLoopDirective &S, OMPPrivateScope &LoopScope, bool Ordered,
-    Address LB, Address UB, Address ST, Address IL, llvm::Value *Chunk) {
+void CodeGenFunction::EmitOMPOuterLoop(
+    bool DynamicOrOrdered, bool IsMonotonic, const OMPLoopDirective &S,
+    CodeGenFunction::OMPPrivateScope &LoopScope,
+    const CodeGenFunction::OMPLoopArguments &LoopArgs,
+    const CodeGenFunction::CodeGenLoopTy &CodeGenLoop,
+    const CodeGenFunction::CodeGenOrderedTy &CodeGenOrdered) {
   auto &RT = CGM.getOpenMPRuntime();
 
   const Expr *IVExpr = S.getIterationVariable();
@@ -1751,15 +1782,18 @@ void CodeGenFunction::EmitOMPOuterLoop(bool DynamicOrOrdered, bool IsMonotonic,
 
   llvm::Value *BoolCondVal = nullptr;
   if (!DynamicOrOrdered) {
-    // UB = min(UB, GlobalUB)
-    EmitIgnoredExpr(S.getEnsureUpperBound());
+    // UB = min(UB, GlobalUB) or
+    // UB = min(UB, PrevUB) for combined loop sharing constructs (e.g.
+    // 'distribute parallel for')
+    EmitIgnoredExpr(LoopArgs.EUB);
     // IV = LB
-    EmitIgnoredExpr(S.getInit());
+    EmitIgnoredExpr(LoopArgs.Init);
     // IV < UB
-    BoolCondVal = EvaluateExprAsBool(S.getCond());
+    BoolCondVal = EvaluateExprAsBool(LoopArgs.Cond);
   } else {
-    BoolCondVal = RT.emitForNext(*this, S.getLocStart(), IVSize, IVSigned, IL,
-                                 LB, UB, ST);
+    BoolCondVal =
+        RT.emitForNext(*this, S.getLocStart(), IVSize, IVSigned, LoopArgs.IL,
+                       LoopArgs.LB, LoopArgs.UB, LoopArgs.ST);
   }
 
   // If there are any cleanups between here and the loop-exit scope,
@@ -1779,7 +1813,7 @@ void CodeGenFunction::EmitOMPOuterLoop(bool DynamicOrOrdered, bool IsMonotonic,
   // Emit "IV = LB" (in case of static schedule, we have already calculated new
   // LB for loop condition and emitted it above).
   if (DynamicOrOrdered)
-    EmitIgnoredExpr(S.getInit());
+    EmitIgnoredExpr(LoopArgs.Init);
 
   // Create a block for the increment.
   auto Continue = getJumpDestInCurrentScope("omp.dispatch.inc");
@@ -1793,24 +1827,27 @@ void CodeGenFunction::EmitOMPOuterLoop(bool DynamicOrOrdered, bool IsMonotonic,
     EmitOMPSimdInit(S, IsMonotonic);
 
   SourceLocation Loc = S.getLocStart();
-  EmitOMPInnerLoop(S, LoopScope.requiresCleanups(), S.getCond(), S.getInc(),
-                   [&S, LoopExit](CodeGenFunction &CGF) {
-                     CGF.EmitOMPLoopBody(S, LoopExit);
-                     CGF.EmitStopPoint(&S);
-                   },
-                   [Ordered, IVSize, IVSigned, Loc](CodeGenFunction &CGF) {
-                     if (Ordered) {
-                       CGF.CGM.getOpenMPRuntime().emitForOrderedIterationEnd(
-                           CGF, Loc, IVSize, IVSigned);
-                     }
-                   });
+
+  // when 'distribute' is not combined with a 'for':
+  // while (idx <= UB) { BODY; ++idx; }
+  // when 'distribute' is combined with a 'for'
+  // (e.g. 'distribute parallel for')
+  // while (idx <= UB) { <CodeGen rest of pragma>; idx += ST; }
+  EmitOMPInnerLoop(
+      S, LoopScope.requiresCleanups(), LoopArgs.Cond, LoopArgs.IncExpr,
+      [&S, LoopExit, &CodeGenLoop](CodeGenFunction &CGF) {
+        CodeGenLoop(CGF, S, LoopExit);
+      },
+      [IVSize, IVSigned, Loc, &CodeGenOrdered](CodeGenFunction &CGF) {
+        CodeGenOrdered(CGF, Loc, IVSize, IVSigned);
+      });
 
   EmitBlock(Continue.getBlock());
   BreakContinueStack.pop_back();
   if (!DynamicOrOrdered) {
     // Emit "LB = LB + Stride", "UB = UB + Stride".
-    EmitIgnoredExpr(S.getNextLowerBound());
-    EmitIgnoredExpr(S.getNextUpperBound());
+    EmitIgnoredExpr(LoopArgs.NextLB);
+    EmitIgnoredExpr(LoopArgs.NextUB);
   }
 
   EmitBranch(CondBlock);
@@ -1829,7 +1866,8 @@ void CodeGenFunction::EmitOMPOuterLoop(bool DynamicOrOrdered, bool IsMonotonic,
 void CodeGenFunction::EmitOMPForOuterLoop(
     const OpenMPScheduleTy &ScheduleKind, bool IsMonotonic,
     const OMPLoopDirective &S, OMPPrivateScope &LoopScope, bool Ordered,
-    Address LB, Address UB, Address ST, Address IL, llvm::Value *Chunk) {
+    const OMPLoopArguments &LoopArgs,
+    const CodeGenDispatchBoundsTy &CGDispatchBounds) {
   auto &RT = CGM.getOpenMPRuntime();
 
   // Dynamic scheduling of the outer loop (dynamic, guided, auto, runtime).
@@ -1838,7 +1876,7 @@ void CodeGenFunction::EmitOMPForOuterLoop(
 
   assert((Ordered ||
           !RT.isStaticNonchunked(ScheduleKind.Schedule,
-                                 /*Chunked=*/Chunk != nullptr)) &&
+                                 LoopArgs.Chunk != nullptr)) &&
          "static non-chunked schedule does not need outer loop");
 
   // Emit outer loop.
@@ -1896,22 +1934,46 @@ void CodeGenFunction::EmitOMPForOuterLoop(
   const bool IVSigned = IVExpr->getType()->hasSignedIntegerRepresentation();
 
   if (DynamicOrOrdered) {
-    llvm::Value *UBVal = EmitScalarExpr(S.getLastIteration());
+    auto DispatchBounds = CGDispatchBounds(*this, S, LoopArgs.LB, LoopArgs.UB);
+    llvm::Value *LBVal = DispatchBounds.first;
+    llvm::Value *UBVal = DispatchBounds.second;
+    CGOpenMPRuntime::DispatchRTInput DipatchRTInputValues = {LBVal, UBVal,
+                                                             LoopArgs.Chunk};
     RT.emitForDispatchInit(*this, S.getLocStart(), ScheduleKind, IVSize,
-                           IVSigned, Ordered, UBVal, Chunk);
+                           IVSigned, Ordered, DipatchRTInputValues);
   } else {
     RT.emitForStaticInit(*this, S.getLocStart(), ScheduleKind, IVSize, IVSigned,
-                         Ordered, IL, LB, UB, ST, Chunk);
+                         Ordered, LoopArgs.IL, LoopArgs.LB, LoopArgs.UB,
+                         LoopArgs.ST, LoopArgs.Chunk);
   }
 
-  EmitOMPOuterLoop(DynamicOrOrdered, IsMonotonic, S, LoopScope, Ordered, LB, UB,
-                   ST, IL, Chunk);
+  auto &&CodeGenOrdered = [Ordered](CodeGenFunction &CGF, SourceLocation Loc,
+                                    const unsigned IVSize,
+                                    const bool IVSigned) {
+    if (Ordered) {
+      CGF.CGM.getOpenMPRuntime().emitForOrderedIterationEnd(CGF, Loc, IVSize,
+                                                            IVSigned);
+    }
+  };
+
+  OMPLoopArguments OuterLoopArgs(LoopArgs.LB, LoopArgs.UB, LoopArgs.ST,
+                                 LoopArgs.IL, LoopArgs.Chunk, LoopArgs.EUB);
+  OuterLoopArgs.IncExpr = S.getInc();
+  OuterLoopArgs.Init = S.getInit();
+  OuterLoopArgs.Cond = S.getCond();
+  OuterLoopArgs.NextLB = S.getNextLowerBound();
+  OuterLoopArgs.NextUB = S.getNextUpperBound();
+  EmitOMPOuterLoop(DynamicOrOrdered, IsMonotonic, S, LoopScope, OuterLoopArgs,
+                   emitOMPLoopBodyWithStopPoint, CodeGenOrdered);
 }
 
+static void emitEmptyOrdered(CodeGenFunction &, SourceLocation Loc,
+                             const unsigned IVSize, const bool IVSigned) {}
+
 void CodeGenFunction::EmitOMPDistributeOuterLoop(
-    OpenMPDistScheduleClauseKind ScheduleKind,
-    const OMPDistributeDirective &S, OMPPrivateScope &LoopScope,
-    Address LB, Address UB, Address ST, Address IL, llvm::Value *Chunk) {
+    OpenMPDistScheduleClauseKind ScheduleKind, const OMPLoopDirective &S,
+    OMPPrivateScope &LoopScope, const OMPLoopArguments &LoopArgs,
+    const CodeGenLoopTy &CodeGenLoopContent) {
 
   auto &RT = CGM.getOpenMPRuntime();
 
@@ -1924,26 +1986,159 @@ void CodeGenFunction::EmitOMPDistributeOuterLoop(
   const unsigned IVSize = getContext().getTypeSize(IVExpr->getType());
   const bool IVSigned = IVExpr->getType()->hasSignedIntegerRepresentation();
 
-  RT.emitDistributeStaticInit(*this, S.getLocStart(), ScheduleKind,
-                              IVSize, IVSigned, /* Ordered = */ false,
-                              IL, LB, UB, ST, Chunk);
+  RT.emitDistributeStaticInit(*this, S.getLocStart(), ScheduleKind, IVSize,
+                              IVSigned, /* Ordered = */ false, LoopArgs.IL,
+                              LoopArgs.LB, LoopArgs.UB, LoopArgs.ST,
+                              LoopArgs.Chunk);
 
-  EmitOMPOuterLoop(/* DynamicOrOrdered = */ false, /* IsMonotonic = */ false,
-                   S, LoopScope, /* Ordered = */ false, LB, UB, ST, IL, Chunk);
+  // for combined 'distribute' and 'for' the increment expression of distribute
+  // is store in DistInc. For 'distribute' alone, it is in Inc.
+  Expr *IncExpr;
+  if (isOpenMPLoopBoundSharingDirective(S.getDirectiveKind()))
+    IncExpr = S.getDistInc();
+  else
+    IncExpr = S.getInc();
+
+  // this routine is shared by 'omp distribute parallel for' and
+  // 'omp distribute': select the right EUB expression depending on the
+  // directive
+  OMPLoopArguments OuterLoopArgs;
+  OuterLoopArgs.LB = LoopArgs.LB;
+  OuterLoopArgs.UB = LoopArgs.UB;
+  OuterLoopArgs.ST = LoopArgs.ST;
+  OuterLoopArgs.IL = LoopArgs.IL;
+  OuterLoopArgs.Chunk = LoopArgs.Chunk;
+  OuterLoopArgs.EUB = isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                          ? S.getCombinedEnsureUpperBound()
+                          : S.getEnsureUpperBound();
+  OuterLoopArgs.IncExpr = IncExpr;
+  OuterLoopArgs.Init = isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                           ? S.getCombinedInit()
+                           : S.getInit();
+  OuterLoopArgs.Cond = isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                           ? S.getCombinedCond()
+                           : S.getCond();
+  OuterLoopArgs.NextLB = isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                             ? S.getCombinedNextLowerBound()
+                             : S.getNextLowerBound();
+  OuterLoopArgs.NextUB = isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                             ? S.getCombinedNextUpperBound()
+                             : S.getNextUpperBound();
+
+  EmitOMPOuterLoop(/* DynamicOrOrdered = */ false, /* IsMonotonic = */ false, S,
+                   LoopScope, OuterLoopArgs, CodeGenLoopContent,
+                   emitEmptyOrdered);
+}
+
+/// Emit a helper variable and return corresponding lvalue.
+static LValue EmitOMPHelperVar(CodeGenFunction &CGF,
+                               const DeclRefExpr *Helper) {
+  auto VDecl = cast<VarDecl>(Helper->getDecl());
+  CGF.EmitVarDecl(*VDecl);
+  return CGF.EmitLValue(Helper);
+}
+
+static std::pair<LValue, LValue>
+emitDistributeParallelForInnerBounds(CodeGenFunction &CGF,
+                                     const OMPExecutableDirective &S) {
+  const OMPLoopDirective &LS = cast<OMPLoopDirective>(S);
+  LValue LB =
+      EmitOMPHelperVar(CGF, cast<DeclRefExpr>(LS.getLowerBoundVariable()));
+  LValue UB =
+      EmitOMPHelperVar(CGF, cast<DeclRefExpr>(LS.getUpperBoundVariable()));
+
+  // When composing 'distribute' with 'for' (e.g. as in 'distribute
+  // parallel for') we need to use the 'distribute'
+  // chunk lower and upper bounds rather than the whole loop iteration
+  // space. These are parameters to the outlined function for 'parallel'
+  // and we copy the bounds of the previous schedule into the
+  // the current ones.
+  LValue PrevLB = CGF.EmitLValue(LS.getPrevLowerBoundVariable());
+  LValue PrevUB = CGF.EmitLValue(LS.getPrevUpperBoundVariable());
+  llvm::Value *PrevLBVal = CGF.EmitLoadOfScalar(PrevLB, SourceLocation());
+  PrevLBVal = CGF.EmitScalarConversion(
+      PrevLBVal, LS.getPrevLowerBoundVariable()->getType(),
+      LS.getIterationVariable()->getType(), SourceLocation());
+  llvm::Value *PrevUBVal = CGF.EmitLoadOfScalar(PrevUB, SourceLocation());
+  PrevUBVal = CGF.EmitScalarConversion(
+      PrevUBVal, LS.getPrevUpperBoundVariable()->getType(),
+      LS.getIterationVariable()->getType(), SourceLocation());
+
+  CGF.EmitStoreOfScalar(PrevLBVal, LB);
+  CGF.EmitStoreOfScalar(PrevUBVal, UB);
+
+  return {LB, UB};
+}
+
+/// if the 'for' loop has a dispatch schedule (e.g. dynamic, guided) then
+/// we need to use the LB and UB expressions generated by the worksharing
+/// code generation support, whereas in non combined situations we would
+/// just emit 0 and the LastIteration expression
+/// This function is necessary due to the difference of the LB and UB
+/// types for the RT emission routines for 'for_static_init' and
+/// 'for_dispatch_init'
+static std::pair<llvm::Value *, llvm::Value *>
+emitDistributeParallelForDispatchBounds(CodeGenFunction &CGF,
+                                        const OMPExecutableDirective &S,
+                                        Address LB, Address UB) {
+  const OMPLoopDirective &LS = cast<OMPLoopDirective>(S);
+  const Expr *IVExpr = LS.getIterationVariable();
+  // when implementing a dynamic schedule for a 'for' combined with a
+  // 'distribute' (e.g. 'distribute parallel for'), the 'for' loop
+  // is not normalized as each team only executes its own assigned
+  // distribute chunk
+  QualType IteratorTy = IVExpr->getType();
+  llvm::Value *LBVal = CGF.EmitLoadOfScalar(LB, /*Volatile=*/false, IteratorTy,
+                                            SourceLocation());
+  llvm::Value *UBVal = CGF.EmitLoadOfScalar(UB, /*Volatile=*/false, IteratorTy,
+                                            SourceLocation());
+  return {LBVal, UBVal};
+};
+
+static void emitDistributeParallelForDistributeInnerBoundParams(
+    CodeGenFunction &CGF, const OMPExecutableDirective &S,
+    llvm::SmallVectorImpl<llvm::Value *> &CapturedVars) {
+  const auto &Dir = cast<OMPLoopDirective>(S);
+  LValue LB =
+      CGF.EmitLValue(cast<DeclRefExpr>(Dir.getCombinedLowerBoundVariable()));
+  auto LBCast = CGF.Builder.CreateIntCast(
+      CGF.Builder.CreateLoad(LB.getAddress()), CGF.SizeTy, /*isSigned=*/false);
+  CapturedVars.push_back(LBCast);
+  LValue UB =
+      CGF.EmitLValue(cast<DeclRefExpr>(Dir.getCombinedUpperBoundVariable()));
+
+  auto UBCast = CGF.Builder.CreateIntCast(
+      CGF.Builder.CreateLoad(UB.getAddress()), CGF.SizeTy, /*isSigned=*/false);
+  CapturedVars.push_back(UBCast);
+};
+
+static void
+emitInnerParallelForWhenCombined(CodeGenFunction &CGF,
+                                 const OMPLoopDirective &S,
+                                 CodeGenFunction::JumpDest LoopExit) {
+  auto &&CGInlinedWorksharingLoop = [&S](CodeGenFunction &CGF,
+                                         PrePostActionTy &) {
+    CGF.EmitOMPWorksharingLoop(S, S.getPrevEnsureUpperBound(),
+                               emitDistributeParallelForInnerBounds,
+                               emitDistributeParallelForDispatchBounds);
+  };
+
+  emitCommonOMPParallelDirective(
+      CGF, S, OMPD_for, CGInlinedWorksharingLoop,
+      emitDistributeParallelForDistributeInnerBoundParams);
 }
 
 void CodeGenFunction::EmitOMPDistributeParallelForDirective(
     const OMPDistributeParallelForDirective &S) {
+  auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
+    CGF.EmitOMPDistributeLoop(S, emitInnerParallelForWhenCombined,
+                              S.getDistInc());
+  };
   OMPLexicalScope Scope(*this, S, /*AsInlined=*/true);
-  CGM.getOpenMPRuntime().emitInlinedDirective(
-      *this, OMPD_distribute_parallel_for,
-      [&S](CodeGenFunction &CGF, PrePostActionTy &) {
-        OMPLoopScope PreInitScope(CGF, S);
-        OMPCancelStackRAII CancelRegion(CGF, OMPD_distribute_parallel_for,
-                                        /*HasCancel=*/false);
-        CGF.EmitStmt(
-            cast<CapturedStmt>(S.getAssociatedStmt())->getCapturedStmt());
-      });
+  OMPCancelStackRAII CancelRegion(*this, OMPD_distribute_parallel_for,
+                                  /*HasCancel=*/false);
+  CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_distribute, CodeGen,
+                                              /*HasCancel=*/false);
 }
 
 void CodeGenFunction::EmitOMPDistributeParallelForSimdDirective(
@@ -2081,14 +2276,6 @@ void CodeGenFunction::EmitOMPTargetTeamsDistributeSimdDirective(
       });
 }
 
-/// \brief Emit a helper variable and return corresponding lvalue.
-static LValue EmitOMPHelperVar(CodeGenFunction &CGF,
-                               const DeclRefExpr *Helper) {
-  auto VDecl = cast<VarDecl>(Helper->getDecl());
-  CGF.EmitVarDecl(*VDecl);
-  return CGF.EmitLValue(Helper);
-}
-
 namespace {
   struct ScheduleKindModifiersTy {
     OpenMPScheduleClauseKind Kind;
@@ -2101,7 +2288,10 @@ namespace {
   };
 } // namespace
 
-bool CodeGenFunction::EmitOMPWorksharingLoop(const OMPLoopDirective &S) {
+bool CodeGenFunction::EmitOMPWorksharingLoop(
+    const OMPLoopDirective &S, Expr *EUB,
+    const CodeGenLoopBoundsTy &CodeGenLoopBounds,
+    const CodeGenDispatchBoundsTy &CGDispatchBounds) {
   // Emit the loop iteration variable.
   auto IVExpr = cast<DeclRefExpr>(S.getIterationVariable());
   auto IVDecl = cast<VarDecl>(IVExpr->getDecl());
@@ -2151,10 +2341,10 @@ bool CodeGenFunction::EmitOMPWorksharingLoop(const OMPLoopDirective &S) {
     emitAlignedClause(*this, S);
     EmitOMPLinearClauseInit(S);
     // Emit helper vars inits.
-    LValue LB =
-        EmitOMPHelperVar(*this, cast<DeclRefExpr>(S.getLowerBoundVariable()));
-    LValue UB =
-        EmitOMPHelperVar(*this, cast<DeclRefExpr>(S.getUpperBoundVariable()));
+
+    std::pair<LValue, LValue> Bounds = CodeGenLoopBounds(*this, S);
+    LValue LB = Bounds.first;
+    LValue UB = Bounds.second;
     LValue ST =
         EmitOMPHelperVar(*this, cast<DeclRefExpr>(S.getStrideVariable()));
     LValue IL =
@@ -2240,9 +2430,11 @@ bool CodeGenFunction::EmitOMPWorksharingLoop(const OMPLoopDirective &S) {
             ScheduleKind.M2 == OMPC_SCHEDULE_MODIFIER_monotonic;
         // Emit the outer loop, which requests its work chunk [LB..UB] from
         // runtime and runs the inner loop to process it.
+        const OMPLoopArguments LoopArguments(LB.getAddress(), UB.getAddress(),
+                                             ST.getAddress(), IL.getAddress(),
+                                             Chunk, EUB);
         EmitOMPForOuterLoop(ScheduleKind, IsMonotonic, S, LoopScope, Ordered,
-                            LB.getAddress(), UB.getAddress(), ST.getAddress(),
-                            IL.getAddress(), Chunk);
+                            LoopArguments, CGDispatchBounds);
       }
       if (isOpenMPSimdDirective(S.getDirectiveKind())) {
         EmitOMPSimdFinal(S,
@@ -2280,12 +2472,42 @@ bool CodeGenFunction::EmitOMPWorksharingLoop(const OMPLoopDirective &S) {
   return HasLastprivateClause;
 }
 
+/// The following two functions generate expressions for the loop lower
+/// and upper bounds in case of static and dynamic (dispatch) schedule
+/// of the associated 'for' or 'distribute' loop.
+static std::pair<LValue, LValue>
+emitForLoopBounds(CodeGenFunction &CGF, const OMPExecutableDirective &S) {
+  const OMPLoopDirective &LS = cast<OMPLoopDirective>(S);
+  LValue LB =
+      EmitOMPHelperVar(CGF, cast<DeclRefExpr>(LS.getLowerBoundVariable()));
+  LValue UB =
+      EmitOMPHelperVar(CGF, cast<DeclRefExpr>(LS.getUpperBoundVariable()));
+  return {LB, UB};
+}
+
+/// When dealing with dispatch schedules (e.g. dynamic, guided) we do not
+/// consider the lower and upper bound expressions generated by the
+/// worksharing loop support, but we use 0 and the iteration space size as
+/// constants
+static std::pair<llvm::Value *, llvm::Value *>
+emitDispatchForLoopBounds(CodeGenFunction &CGF, const OMPExecutableDirective &S,
+                          Address LB, Address UB) {
+  const OMPLoopDirective &LS = cast<OMPLoopDirective>(S);
+  const Expr *IVExpr = LS.getIterationVariable();
+  const unsigned IVSize = CGF.getContext().getTypeSize(IVExpr->getType());
+  llvm::Value *LBVal = CGF.Builder.getIntN(IVSize, 0);
+  llvm::Value *UBVal = CGF.EmitScalarExpr(LS.getLastIteration());
+  return {LBVal, UBVal};
+}
+
 void CodeGenFunction::EmitOMPForDirective(const OMPForDirective &S) {
   bool HasLastprivates = false;
   auto &&CodeGen = [&S, &HasLastprivates](CodeGenFunction &CGF,
                                           PrePostActionTy &) {
     OMPCancelStackRAII CancelRegion(CGF, OMPD_for, S.hasCancel());
-    HasLastprivates = CGF.EmitOMPWorksharingLoop(S);
+    HasLastprivates = CGF.EmitOMPWorksharingLoop(S, S.getEnsureUpperBound(),
+                                                 emitForLoopBounds,
+                                                 emitDispatchForLoopBounds);
   };
   {
     OMPLexicalScope Scope(*this, S, /*AsInlined=*/true);
@@ -2303,7 +2525,9 @@ void CodeGenFunction::EmitOMPForSimdDirective(const OMPForSimdDirective &S) {
   bool HasLastprivates = false;
   auto &&CodeGen = [&S, &HasLastprivates](CodeGenFunction &CGF,
                                           PrePostActionTy &) {
-    HasLastprivates = CGF.EmitOMPWorksharingLoop(S);
+    HasLastprivates = CGF.EmitOMPWorksharingLoop(S, S.getEnsureUpperBound(),
+                                                 emitForLoopBounds,
+                                                 emitDispatchForLoopBounds);
   };
   {
     OMPLexicalScope Scope(*this, S, /*AsInlined=*/true);
@@ -2554,9 +2778,11 @@ void CodeGenFunction::EmitOMPParallelForDirective(
   // directives: 'parallel' with 'for' directive.
   auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
     OMPCancelStackRAII CancelRegion(CGF, OMPD_parallel_for, S.hasCancel());
-    CGF.EmitOMPWorksharingLoop(S);
+    CGF.EmitOMPWorksharingLoop(S, S.getEnsureUpperBound(), emitForLoopBounds,
+                               emitDispatchForLoopBounds);
   };
-  emitCommonOMPParallelDirective(*this, S, OMPD_for, CodeGen);
+  emitCommonOMPParallelDirective(*this, S, OMPD_for, CodeGen,
+                                 emitEmptyBoundParameters);
 }
 
 void CodeGenFunction::EmitOMPParallelForSimdDirective(
@@ -2564,9 +2790,11 @@ void CodeGenFunction::EmitOMPParallelForSimdDirective(
   // Emit directive as a combined directive that consists of two implicit
   // directives: 'parallel' with 'for' directive.
   auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
-    CGF.EmitOMPWorksharingLoop(S);
+    CGF.EmitOMPWorksharingLoop(S, S.getEnsureUpperBound(), emitForLoopBounds,
+                               emitDispatchForLoopBounds);
   };
-  emitCommonOMPParallelDirective(*this, S, OMPD_simd, CodeGen);
+  emitCommonOMPParallelDirective(*this, S, OMPD_simd, CodeGen,
+                                 emitEmptyBoundParameters);
 }
 
 void CodeGenFunction::EmitOMPParallelSectionsDirective(
@@ -2576,7 +2804,8 @@ void CodeGenFunction::EmitOMPParallelSectionsDirective(
   auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
     CGF.EmitSections(S);
   };
-  emitCommonOMPParallelDirective(*this, S, OMPD_sections, CodeGen);
+  emitCommonOMPParallelDirective(*this, S, OMPD_sections, CodeGen,
+                                 emitEmptyBoundParameters);
 }
 
 void CodeGenFunction::EmitOMPTaskBasedDirective(const OMPExecutableDirective &S,
@@ -2794,7 +3023,9 @@ void CodeGenFunction::EmitOMPFlushDirective(const OMPFlushDirective &S) {
   }(), S.getLocStart());
 }
 
-void CodeGenFunction::EmitOMPDistributeLoop(const OMPDistributeDirective &S) {
+void CodeGenFunction::EmitOMPDistributeLoop(const OMPLoopDirective &S,
+                                            const CodeGenLoopTy &CodeGenLoop,
+                                            Expr *IncExpr) {
   // Emit the loop iteration variable.
   auto IVExpr = cast<DeclRefExpr>(S.getIterationVariable());
   auto IVDecl = cast<VarDecl>(IVExpr->getDecl());
@@ -2835,10 +3066,17 @@ void CodeGenFunction::EmitOMPDistributeLoop(const OMPDistributeDirective &S) {
     // Emit 'then' code.
     {
       // Emit helper vars inits.
-      LValue LB =
-          EmitOMPHelperVar(*this, cast<DeclRefExpr>(S.getLowerBoundVariable()));
-      LValue UB =
-          EmitOMPHelperVar(*this, cast<DeclRefExpr>(S.getUpperBoundVariable()));
+
+      LValue LB = EmitOMPHelperVar(
+          *this, cast<DeclRefExpr>(
+                     (isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                          ? S.getCombinedLowerBoundVariable()
+                          : S.getLowerBoundVariable())));
+      LValue UB = EmitOMPHelperVar(
+          *this, cast<DeclRefExpr>(
+                     (isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                          ? S.getCombinedUpperBoundVariable()
+                          : S.getUpperBoundVariable())));
       LValue ST =
           EmitOMPHelperVar(*this, cast<DeclRefExpr>(S.getStrideVariable()));
       LValue IL =
@@ -2890,15 +3128,25 @@ void CodeGenFunction::EmitOMPDistributeLoop(const OMPDistributeDirective &S) {
         auto LoopExit =
             getJumpDestInCurrentScope(createBasicBlock("omp.loop.exit"));
         // UB = min(UB, GlobalUB);
-        EmitIgnoredExpr(S.getEnsureUpperBound());
+        EmitIgnoredExpr(isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                            ? S.getCombinedEnsureUpperBound()
+                            : S.getEnsureUpperBound());
         // IV = LB;
-        EmitIgnoredExpr(S.getInit());
+        EmitIgnoredExpr(isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                            ? S.getCombinedInit()
+                            : S.getInit());
+
+        Expr *Cond = isOpenMPLoopBoundSharingDirective(S.getDirectiveKind())
+                         ? S.getCombinedCond()
+                         : S.getCond();
+
+        // for distribute alone,  codegen
         // while (idx <= UB) { BODY; ++idx; }
-        EmitOMPInnerLoop(S, LoopScope.requiresCleanups(), S.getCond(),
-                         S.getInc(),
-                         [&S, LoopExit](CodeGenFunction &CGF) {
-                           CGF.EmitOMPLoopBody(S, LoopExit);
-                           CGF.EmitStopPoint(&S);
+        // when combined with 'for' (e.g. as in 'distribute parallel for')
+        // while (idx <= UB) { <CodeGen rest of pragma>; idx += ST; }
+        EmitOMPInnerLoop(S, LoopScope.requiresCleanups(), Cond, IncExpr,
+                         [&S, LoopExit, &CodeGenLoop](CodeGenFunction &CGF) {
+                           CodeGenLoop(CGF, S, LoopExit);
                          },
                          [](CodeGenFunction &) {});
         EmitBlock(LoopExit.getBlock());
@@ -2907,9 +3155,11 @@ void CodeGenFunction::EmitOMPDistributeLoop(const OMPDistributeDirective &S) {
       } else {
         // Emit the outer loop, which requests its work chunk [LB..UB] from
         // runtime and runs the inner loop to process it.
-        EmitOMPDistributeOuterLoop(ScheduleKind, S, LoopScope,
-                            LB.getAddress(), UB.getAddress(), ST.getAddress(),
-                            IL.getAddress(), Chunk);
+        const OMPLoopArguments LoopArguments = {
+            LB.getAddress(), UB.getAddress(), ST.getAddress(), IL.getAddress(),
+            Chunk};
+        EmitOMPDistributeOuterLoop(ScheduleKind, S, LoopScope, LoopArguments,
+                                   CodeGenLoop);
       }
 
       // Emit final copy of the lastprivate variables if IsLastIter != 0.
@@ -2931,7 +3181,8 @@ void CodeGenFunction::EmitOMPDistributeLoop(const OMPDistributeDirective &S) {
 void CodeGenFunction::EmitOMPDistributeDirective(
     const OMPDistributeDirective &S) {
   auto &&CodeGen = [&S](CodeGenFunction &CGF, PrePostActionTy &) {
-    CGF.EmitOMPDistributeLoop(S);
+
+    CGF.EmitOMPDistributeLoop(S, emitOMPLoopBodyWithStopPoint, S.getInc());
   };
   OMPLexicalScope Scope(*this, S, /*AsInlined=*/true);
   CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_distribute, CodeGen,
@@ -3840,7 +4091,8 @@ static void emitTargetParallelRegion(CodeGenFunction &CGF,
     CGF.EmitStmt(CS->getCapturedStmt());
     CGF.EmitOMPReductionClauseFinal(S, /*ReductionKind=*/OMPD_parallel);
   };
-  emitCommonOMPParallelDirective(CGF, S, OMPD_parallel, CodeGen);
+  emitCommonOMPParallelDirective(CGF, S, OMPD_parallel, CodeGen,
+                                 emitEmptyBoundParameters);
   emitPostUpdateForReductionClause(
       CGF, S, [](CodeGenFunction &) -> llvm::Value * { return nullptr; });
 }
diff --git a/lib/CodeGen/CodeGenFunction.h b/lib/CodeGen/CodeGenFunction.h
index fa72019eb08b..1ded824ba5b0 100644
--- a/lib/CodeGen/CodeGenFunction.h
+++ b/lib/CodeGen/CodeGenFunction.h
@@ -175,6 +175,25 @@ public:
   // because of jumps.
   VarBypassDetector Bypasses;
 
+  // CodeGen lambda for loops and support for ordered clause
+  typedef llvm::function_ref<void(CodeGenFunction &, const OMPLoopDirective &,
+                                  JumpDest)>
+      CodeGenLoopTy;
+  typedef llvm::function_ref<void(CodeGenFunction &, SourceLocation,
+                                  const unsigned, const bool)>
+      CodeGenOrderedTy;
+
+  // Codegen lambda for loop bounds in worksharing loop constructs
+  typedef llvm::function_ref<std::pair<LValue, LValue>(
+      CodeGenFunction &, const OMPExecutableDirective &S)>
+      CodeGenLoopBoundsTy;
+
+  // Codegen lambda for loop bounds in dispatch-based loop implementation
+  typedef llvm::function_ref<std::pair<llvm::Value *, llvm::Value *>(
+      CodeGenFunction &, const OMPExecutableDirective &S, Address LB,
+      Address UB)>
+      CodeGenDispatchBoundsTy;
+
   /// \brief CGBuilder insert helper. This function is called after an
   /// instruction is created using Builder.
   void InsertHelper(llvm::Instruction *I, const llvm::Twine &Name,
@@ -2756,7 +2775,6 @@ public:
   void EmitOMPTaskLoopDirective(const OMPTaskLoopDirective &S);
   void EmitOMPTaskLoopSimdDirective(const OMPTaskLoopSimdDirective &S);
   void EmitOMPDistributeDirective(const OMPDistributeDirective &S);
-  void EmitOMPDistributeLoop(const OMPDistributeDirective &S);
   void EmitOMPDistributeParallelForDirective(
       const OMPDistributeParallelForDirective &S);
   void EmitOMPDistributeParallelForSimdDirective(
@@ -2813,32 +2831,78 @@ public:
   void EmitOMPPrivateLoopCounters(const OMPLoopDirective &S,
                                   OMPPrivateScope &LoopScope);
 
+  /// Helper for the OpenMP loop directives.
+  void EmitOMPLoopBody(const OMPLoopDirective &D, JumpDest LoopExit);
+
+  /// \brief Emit code for the worksharing loop-based directive.
+  /// \return true, if this construct has any lastprivate clause, false -
+  /// otherwise.
+  bool EmitOMPWorksharingLoop(const OMPLoopDirective &S, Expr *EUB,
+                              const CodeGenLoopBoundsTy &CodeGenLoopBounds,
+                              const CodeGenDispatchBoundsTy &CGDispatchBounds);
+
 private:
   /// Helpers for blocks
   llvm::Value *EmitBlockLiteral(const CGBlockInfo &Info);
 
   /// Helpers for the OpenMP loop directives.
-  void EmitOMPLoopBody(const OMPLoopDirective &D, JumpDest LoopExit);
   void EmitOMPSimdInit(const OMPLoopDirective &D, bool IsMonotonic = false);
   void EmitOMPSimdFinal(
       const OMPLoopDirective &D,
       const llvm::function_ref<llvm::Value *(CodeGenFunction &)> &CondGen);
-  /// \brief Emit code for the worksharing loop-based directive.
-  /// \return true, if this construct has any lastprivate clause, false -
-  /// otherwise.
-  bool EmitOMPWorksharingLoop(const OMPLoopDirective &S);
-  void EmitOMPOuterLoop(bool IsMonotonic, bool DynamicOrOrdered,
-      const OMPLoopDirective &S, OMPPrivateScope &LoopScope, bool Ordered,
-      Address LB, Address UB, Address ST, Address IL, llvm::Value *Chunk);
+
+  void EmitOMPDistributeLoop(const OMPLoopDirective &S,
+                             const CodeGenLoopTy &CodeGenLoop, Expr *IncExpr);
+
+  /// struct with the values to be passed to the OpenMP loop-related functions
+  struct OMPLoopArguments {
+    /// loop lower bound
+    Address LB = Address::invalid();
+    /// loop upper bound
+    Address UB = Address::invalid();
+    /// loop stride
+    Address ST = Address::invalid();
+    /// isLastIteration argument for runtime functions
+    Address IL = Address::invalid();
+    /// Chunk value generated by sema
+    llvm::Value *Chunk = nullptr;
+    /// EnsureUpperBound
+    Expr *EUB = nullptr;
+    /// IncrementExpression
+    Expr *IncExpr = nullptr;
+    /// Loop initialization
+    Expr *Init = nullptr;
+    /// Loop exit condition
+    Expr *Cond = nullptr;
+    /// Update of LB after a whole chunk has been executed
+    Expr *NextLB = nullptr;
+    /// Update of UB after a whole chunk has been executed
+    Expr *NextUB = nullptr;
+    OMPLoopArguments() = default;
+    OMPLoopArguments(Address LB, Address UB, Address ST, Address IL,
+                     llvm::Value *Chunk = nullptr, Expr *EUB = nullptr,
+                     Expr *IncExpr = nullptr, Expr *Init = nullptr,
+                     Expr *Cond = nullptr, Expr *NextLB = nullptr,
+                     Expr *NextUB = nullptr)
+        : LB(LB), UB(UB), ST(ST), IL(IL), Chunk(Chunk), EUB(EUB),
+          IncExpr(IncExpr), Init(Init), Cond(Cond), NextLB(NextLB),
+          NextUB(NextUB) {}
+  };
+  void EmitOMPOuterLoop(bool DynamicOrOrdered, bool IsMonotonic,
+                        const OMPLoopDirective &S, OMPPrivateScope &LoopScope,
+                        const OMPLoopArguments &LoopArgs,
+                        const CodeGenLoopTy &CodeGenLoop,
+                        const CodeGenOrderedTy &CodeGenOrdered);
   void EmitOMPForOuterLoop(const OpenMPScheduleTy &ScheduleKind,
                            bool IsMonotonic, const OMPLoopDirective &S,
-                           OMPPrivateScope &LoopScope, bool Ordered, Address LB,
-                           Address UB, Address ST, Address IL,
-                           llvm::Value *Chunk);
-  void EmitOMPDistributeOuterLoop(
-      OpenMPDistScheduleClauseKind ScheduleKind,
-      const OMPDistributeDirective &S, OMPPrivateScope &LoopScope,
-      Address LB, Address UB, Address ST, Address IL, llvm::Value *Chunk);
+                           OMPPrivateScope &LoopScope, bool Ordered,
+                           const OMPLoopArguments &LoopArgs,
+                           const CodeGenDispatchBoundsTy &CGDispatchBounds);
+  void EmitOMPDistributeOuterLoop(OpenMPDistScheduleClauseKind ScheduleKind,
+                                  const OMPLoopDirective &S,
+                                  OMPPrivateScope &LoopScope,
+                                  const OMPLoopArguments &LoopArgs,
+                                  const CodeGenLoopTy &CodeGenLoopContent);
   /// \brief Emit code for sections directive.
   void EmitSections(const OMPExecutableDirective &S);
 
diff --git a/lib/CodeGen/CodeGenModule.cpp b/lib/CodeGen/CodeGenModule.cpp
index 19203973ff1b..25d32f19d0e5 100644
--- a/lib/CodeGen/CodeGenModule.cpp
+++ b/lib/CodeGen/CodeGenModule.cpp
@@ -565,12 +565,8 @@ void CodeGenModule::DecorateInstructionWithTBAA(llvm::Instruction *Inst,
 
 void CodeGenModule::DecorateInstructionWithInvariantGroup(
     llvm::Instruction *I, const CXXRecordDecl *RD) {
-  llvm::Metadata *MD = CreateMetadataIdentifierForType(QualType(RD->getTypeForDecl(), 0));
-  auto *MetaDataNode = dyn_cast<llvm::MDNode>(MD);
-  // Check if we have to wrap MDString in MDNode.
-  if (!MetaDataNode)
-    MetaDataNode = llvm::MDNode::get(getLLVMContext(), MD);
-  I->setMetadata(llvm::LLVMContext::MD_invariant_group, MetaDataNode);
+  I->setMetadata(llvm::LLVMContext::MD_invariant_group,
+                 llvm::MDNode::get(getLLVMContext(), {}));
 }
 
 void CodeGenModule::Error(SourceLocation loc, StringRef message) {
diff --git a/lib/CodeGen/CodeGenPGO.cpp b/lib/CodeGen/CodeGenPGO.cpp
index 6acedc033a6e..9e193531d0f6 100644
--- a/lib/CodeGen/CodeGenPGO.cpp
+++ b/lib/CodeGen/CodeGenPGO.cpp
@@ -666,7 +666,7 @@ void CodeGenPGO::mapRegionCounters(const Decl *D) {
 }
 
 bool CodeGenPGO::skipRegionMappingForDecl(const Decl *D) {
-  if (SkipCoverageMapping)
+  if (!D->getBody())
     return true;
 
   // Don't map the functions in system headers.
diff --git a/lib/CodeGen/CodeGenPGO.h b/lib/CodeGen/CodeGenPGO.h
index 0026df570bce..0759e65388b8 100644
--- a/lib/CodeGen/CodeGenPGO.h
+++ b/lib/CodeGen/CodeGenPGO.h
@@ -40,14 +40,11 @@ private:
   std::unique_ptr<llvm::InstrProfRecord> ProfRecord;
   std::vector<uint64_t> RegionCounts;
   uint64_t CurrentRegionCount;
-  /// \brief A flag that is set to true when this function doesn't need
-  /// to have coverage mapping data.
-  bool SkipCoverageMapping;
 
 public:
   CodeGenPGO(CodeGenModule &CGM)
-      : CGM(CGM), NumValueSites({{0}}), NumRegionCounters(0),
-        FunctionHash(0), CurrentRegionCount(0), SkipCoverageMapping(false) {}
+      : CGM(CGM), NumValueSites({{0}}), NumRegionCounters(0), FunctionHash(0),
+        CurrentRegionCount(0) {}
 
   /// Whether or not we have PGO region data for the current function. This is
   /// false both when we have no data at all and when our data has been
diff --git a/lib/Driver/SanitizerArgs.cpp b/lib/Driver/SanitizerArgs.cpp
index 8e61aadbf326..c9561367a3a8 100644
--- a/lib/Driver/SanitizerArgs.cpp
+++ b/lib/Driver/SanitizerArgs.cpp
@@ -265,6 +265,10 @@ SanitizerArgs::SanitizerArgs(const ToolChain &TC,
       Add &= ~InvalidTrappingKinds;
       Add &= Supported;
 
+      // Enable coverage if the fuzzing flag is set.
+      if (Add & Fuzzer)
+        CoverageFeatures |= CoverageTracePCGuard | CoverageIndirCall | CoverageTraceCmp;
+
       Kinds |= Add;
     } else if (Arg->getOption().matches(options::OPT_fno_sanitize_EQ)) {
       Arg->claim();
diff --git a/lib/Driver/ToolChains/Clang.cpp b/lib/Driver/ToolChains/Clang.cpp
index 49708e7d7242..6e1e4ccf44f0 100644
--- a/lib/Driver/ToolChains/Clang.cpp
+++ b/lib/Driver/ToolChains/Clang.cpp
@@ -2778,8 +2778,7 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
       CmdArgs.push_back("-fno-split-dwarf-inlining");
     if (DebugInfoKind == codegenoptions::NoDebugInfo)
       DebugInfoKind = codegenoptions::LimitedDebugInfo;
-    CmdArgs.push_back("-backend-option");
-    CmdArgs.push_back("-split-dwarf=Enable");
+    CmdArgs.push_back("-enable-split-dwarf");
   }
 
   // After we've dealt with all combinations of things that could
diff --git a/lib/Driver/ToolChains/CommonArgs.cpp b/lib/Driver/ToolChains/CommonArgs.cpp
index 93b66eb6954a..5e360f62e21a 100644
--- a/lib/Driver/ToolChains/CommonArgs.cpp
+++ b/lib/Driver/ToolChains/CommonArgs.cpp
@@ -577,6 +577,17 @@ collectSanitizerRuntimes(const ToolChain &TC, const ArgList &Args,
     StaticRuntimes.push_back("esan");
 }
 
+static void addLibFuzzerRuntime(const ToolChain &TC,
+                                const ArgList &Args,
+                                ArgStringList &CmdArgs) {
+    StringRef ParentDir = llvm::sys::path::parent_path(TC.getDriver().InstalledDir);
+    SmallString<128> P(ParentDir);
+    llvm::sys::path::append(P, "lib", "libLLVMFuzzer.a");
+    CmdArgs.push_back(Args.MakeArgString(P));
+    TC.AddCXXStdlibLibArgs(Args, CmdArgs);
+}
+
+
 // Should be called before we add system libraries (C++ ABI, libstdc++/libc++,
 // C runtime, etc). Returns true if sanitizer system deps need to be linked in.
 bool tools::addSanitizerRuntimes(const ToolChain &TC, const ArgList &Args,
@@ -586,6 +597,11 @@ bool tools::addSanitizerRuntimes(const ToolChain &TC, const ArgList &Args,
   collectSanitizerRuntimes(TC, Args, SharedRuntimes, StaticRuntimes,
                            NonWholeStaticRuntimes, HelperStaticRuntimes,
                            RequiredSymbols);
+  // Inject libfuzzer dependencies.
+  if (TC.getSanitizerArgs().needsFuzzer()) {
+    addLibFuzzerRuntime(TC, Args, CmdArgs);
+  }
+
   for (auto RT : SharedRuntimes)
     addSanitizerRuntime(TC, Args, CmdArgs, RT, true, false);
   for (auto RT : HelperStaticRuntimes)
diff --git a/lib/Driver/ToolChains/Darwin.cpp b/lib/Driver/ToolChains/Darwin.cpp
index 009a12da3015..e41b50c40b28 100644
--- a/lib/Driver/ToolChains/Darwin.cpp
+++ b/lib/Driver/ToolChains/Darwin.cpp
@@ -930,6 +930,18 @@ void MachO::AddLinkRuntimeLib(const ArgList &Args, ArgStringList &CmdArgs,
   }
 }
 
+void MachO::AddFuzzerLinkArgs(const ArgList &Args, ArgStringList &CmdArgs) const {
+
+  // Go up one directory from Clang to find the libfuzzer archive file.
+  StringRef ParentDir = llvm::sys::path::parent_path(getDriver().InstalledDir);
+  SmallString<128> P(ParentDir);
+  llvm::sys::path::append(P, "lib", "libLLVMFuzzer.a");
+  CmdArgs.push_back(Args.MakeArgString(P));
+
+  // Libfuzzer is written in C++ and requires libcxx.
+  AddCXXStdlibLibArgs(Args, CmdArgs);
+}
+
 StringRef Darwin::getPlatformFamily() const {
   switch (TargetPlatform) {
     case DarwinPlatformKind::MacOS:
@@ -1035,10 +1047,14 @@ void DarwinClang::AddLinkRuntimeLibArgs(const ArgList &Args,
   const SanitizerArgs &Sanitize = getSanitizerArgs();
   if (Sanitize.needsAsanRt())
     AddLinkSanitizerLibArgs(Args, CmdArgs, "asan");
+  if (Sanitize.needsLsanRt())
+    AddLinkSanitizerLibArgs(Args, CmdArgs, "lsan");
   if (Sanitize.needsUbsanRt())
     AddLinkSanitizerLibArgs(Args, CmdArgs, "ubsan");
   if (Sanitize.needsTsanRt())
     AddLinkSanitizerLibArgs(Args, CmdArgs, "tsan");
+  if (Sanitize.needsFuzzer())
+    AddFuzzerLinkArgs(Args, CmdArgs);
   if (Sanitize.needsStatsRt()) {
     StringRef OS = isTargetMacOS() ? "osx" : "iossim";
     AddLinkRuntimeLib(Args, CmdArgs,
@@ -1892,6 +1908,8 @@ SanitizerMask Darwin::getSupportedSanitizers() const {
   const bool IsX86_64 = getTriple().getArch() == llvm::Triple::x86_64;
   SanitizerMask Res = ToolChain::getSupportedSanitizers();
   Res |= SanitizerKind::Address;
+  Res |= SanitizerKind::Leak;
+  Res |= SanitizerKind::Fuzzer;
   if (isTargetMacOS()) {
     if (!isMacosxVersionLT(10, 9))
       Res |= SanitizerKind::Vptr;
diff --git a/lib/Driver/ToolChains/Darwin.h b/lib/Driver/ToolChains/Darwin.h
index 984f8ef0c41f..16ed04286ac0 100644
--- a/lib/Driver/ToolChains/Darwin.h
+++ b/lib/Driver/ToolChains/Darwin.h
@@ -154,6 +154,8 @@ public:
   /// Add the linker arguments to link the compiler runtime library.
   virtual void AddLinkRuntimeLibArgs(const llvm::opt::ArgList &Args,
                                      llvm::opt::ArgStringList &CmdArgs) const;
+  virtual void AddFuzzerLinkArgs(const llvm::opt::ArgList &Args,
+                               llvm::opt::ArgStringList &CmdArgs) const;
 
   virtual void addStartObjectFileArgs(const llvm::opt::ArgList &Args,
                                       llvm::opt::ArgStringList &CmdArgs) const {
diff --git a/lib/Driver/ToolChains/Gnu.cpp b/lib/Driver/ToolChains/Gnu.cpp
index 313bf38c2860..d29d826b5f44 100644
--- a/lib/Driver/ToolChains/Gnu.cpp
+++ b/lib/Driver/ToolChains/Gnu.cpp
@@ -1747,7 +1747,9 @@ bool Generic_GCC::GCCInstallationDetector::getBiarchSibling(Multilib &M) const {
   static const char *const ARMTriples[] = {"arm-linux-gnueabi",
                                            "arm-linux-androideabi"};
   static const char *const ARMHFTriples[] = {"arm-linux-gnueabihf",
-                                             "armv7hl-redhat-linux-gnueabi"};
+                                             "armv7hl-redhat-linux-gnueabi",
+                                             "armv6hl-suse-linux-gnueabi",
+                                             "armv7hl-suse-linux-gnueabi"};
   static const char *const ARMebLibDirs[] = {"/lib"};
   static const char *const ARMebTriples[] = {"armeb-linux-gnueabi",
                                              "armeb-linux-androideabi"};
diff --git a/lib/Driver/ToolChains/Hexagon.cpp b/lib/Driver/ToolChains/Hexagon.cpp
index c143b7f89868..1d7bcf8e4df0 100644
--- a/lib/Driver/ToolChains/Hexagon.cpp
+++ b/lib/Driver/ToolChains/Hexagon.cpp
@@ -402,6 +402,40 @@ Tool *HexagonToolChain::buildLinker() const {
   return new tools::hexagon::Linker(*this);
 }
 
+unsigned HexagonToolChain::getOptimizationLevel(
+    const llvm::opt::ArgList &DriverArgs) const {
+  // Copied in large part from lib/Frontend/CompilerInvocation.cpp.
+  Arg *A = DriverArgs.getLastArg(options::OPT_O_Group);
+  if (!A)
+    return 0;
+
+  if (A->getOption().matches(options::OPT_O0))
+    return 0;
+  if (A->getOption().matches(options::OPT_Ofast) ||
+      A->getOption().matches(options::OPT_O4))
+    return 3;
+  assert(A->getNumValues() != 0);
+  StringRef S(A->getValue());
+  if (S == "s" || S == "z" || S.empty())
+    return 2;
+  if (S == "g")
+    return 1;
+
+  unsigned OptLevel;
+  if (S.getAsInteger(10, OptLevel))
+    return 0;
+  return OptLevel;
+}
+
+void HexagonToolChain::addClangTargetOptions(const ArgList &DriverArgs,
+                                             ArgStringList &CC1Args) const {
+  if (DriverArgs.hasArg(options::OPT_ffp_contract))
+    return;
+  unsigned OptLevel = getOptimizationLevel(DriverArgs);
+  if (OptLevel >= 3)
+    CC1Args.push_back("-ffp-contract=fast");
+}
+
 void HexagonToolChain::AddClangSystemIncludeArgs(const ArgList &DriverArgs,
                                                  ArgStringList &CC1Args) const {
   if (DriverArgs.hasArg(options::OPT_nostdinc) ||
diff --git a/lib/Driver/ToolChains/Hexagon.h b/lib/Driver/ToolChains/Hexagon.h
index fb50ba3c72c3..78f97a3d59fd 100644
--- a/lib/Driver/ToolChains/Hexagon.h
+++ b/lib/Driver/ToolChains/Hexagon.h
@@ -61,11 +61,15 @@ protected:
   Tool *buildAssembler() const override;
   Tool *buildLinker() const override;
 
+  unsigned getOptimizationLevel(const llvm::opt::ArgList &DriverArgs) const;
+
 public:
   HexagonToolChain(const Driver &D, const llvm::Triple &Triple,
                    const llvm::opt::ArgList &Args);
   ~HexagonToolChain() override;
 
+  void addClangTargetOptions(const llvm::opt::ArgList &DriverArgs,
+                             llvm::opt::ArgStringList &CC1Args) const override;
   void
   AddClangSystemIncludeArgs(const llvm::opt::ArgList &DriverArgs,
                             llvm::opt::ArgStringList &CC1Args) const override;
diff --git a/lib/Driver/ToolChains/Linux.cpp b/lib/Driver/ToolChains/Linux.cpp
index 3ffb2f6e03e5..50443a125244 100644
--- a/lib/Driver/ToolChains/Linux.cpp
+++ b/lib/Driver/ToolChains/Linux.cpp
@@ -869,6 +869,7 @@ SanitizerMask Linux::getSupportedSanitizers() const {
                          llvm::Triple::thumbeb;
   SanitizerMask Res = ToolChain::getSupportedSanitizers();
   Res |= SanitizerKind::Address;
+  Res |= SanitizerKind::Fuzzer;
   Res |= SanitizerKind::KernelAddress;
   Res |= SanitizerKind::Vptr;
   Res |= SanitizerKind::SafeStack;
diff --git a/lib/Format/ContinuationIndenter.cpp b/lib/Format/ContinuationIndenter.cpp
index 73ae10a29f8f..3adb72c11da8 100644
--- a/lib/Format/ContinuationIndenter.cpp
+++ b/lib/Format/ContinuationIndenter.cpp
@@ -792,7 +792,8 @@ unsigned ContinuationIndenter::moveStateToNextToken(LineState &State,
     if (Previous && Previous->is(tok::question))
       State.Stack.back().QuestionColumn = State.Column;
   }
-  if (!Current.opensScope() && !Current.closesScope())
+  if (!Current.opensScope() && !Current.closesScope() &&
+      !Current.is(TT_PointerOrReference))
     State.LowestLevelOnLine =
         std::min(State.LowestLevelOnLine, Current.NestingLevel);
   if (Current.isMemberAccess())
diff --git a/lib/Format/Format.cpp b/lib/Format/Format.cpp
index 0e2da71343d5..f55a623a8d1f 100644
--- a/lib/Format/Format.cpp
+++ b/lib/Format/Format.cpp
@@ -908,8 +908,8 @@ private:
 class Formatter : public TokenAnalyzer {
 public:
   Formatter(const Environment &Env, const FormatStyle &Style,
-            bool *IncompleteFormat)
-      : TokenAnalyzer(Env, Style), IncompleteFormat(IncompleteFormat) {}
+            FormattingAttemptStatus *Status)
+      : TokenAnalyzer(Env, Style), Status(Status) {}
 
   tooling::Replacements
   analyze(TokenAnnotator &Annotator,
@@ -931,7 +931,7 @@ public:
                                   Env.getSourceManager(), Whitespaces, Encoding,
                                   BinPackInconclusiveFunctions);
     UnwrappedLineFormatter(&Indenter, &Whitespaces, Style, Tokens.getKeywords(),
-                           IncompleteFormat)
+                           Env.getSourceManager(), Status)
         .format(AnnotatedLines);
     for (const auto &R : Whitespaces.generateReplacements())
       if (Result.add(R))
@@ -1013,7 +1013,7 @@ private:
   }
 
   bool BinPackInconclusiveFunctions;
-  bool *IncompleteFormat;
+  FormattingAttemptStatus *Status;
 };
 
 // This class clean up the erroneous/redundant code around the given ranges in
@@ -1830,7 +1830,8 @@ cleanupAroundReplacements(StringRef Code, const tooling::Replacements &Replaces,
 
 tooling::Replacements reformat(const FormatStyle &Style, StringRef Code,
                                ArrayRef<tooling::Range> Ranges,
-                               StringRef FileName, bool *IncompleteFormat) {
+                               StringRef FileName,
+                               FormattingAttemptStatus *Status) {
   FormatStyle Expanded = expandPresets(Style);
   if (Expanded.DisableFormat)
     return tooling::Replacements();
@@ -1846,11 +1847,11 @@ tooling::Replacements reformat(const FormatStyle &Style, StringRef Code,
         auto NewEnv = Environment::CreateVirtualEnvironment(
             *NewCode, FileName,
             tooling::calculateRangesAfterReplacements(Fixes, Ranges));
-        Formatter Format(*NewEnv, Expanded, IncompleteFormat);
+        Formatter Format(*NewEnv, Expanded, Status);
         return Fixes.merge(Format.process());
       }
     }
-    Formatter Format(*Env, Expanded, IncompleteFormat);
+    Formatter Format(*Env, Expanded, Status);
     return Format.process();
   };
 
@@ -1866,7 +1867,7 @@ tooling::Replacements reformat(const FormatStyle &Style, StringRef Code,
     return reformatAfterApplying(Requoter);
   }
 
-  Formatter Format(*Env, Expanded, IncompleteFormat);
+  Formatter Format(*Env, Expanded, Status);
   return Format.process();
 }
 
@@ -1879,6 +1880,16 @@ tooling::Replacements cleanup(const FormatStyle &Style, StringRef Code,
   return Clean.process();
 }
 
+tooling::Replacements reformat(const FormatStyle &Style, StringRef Code,
+                               ArrayRef<tooling::Range> Ranges,
+                               StringRef FileName, bool *IncompleteFormat) {
+  FormattingAttemptStatus Status;
+  auto Result = reformat(Style, Code, Ranges, FileName, &Status);
+  if (!Status.FormatComplete)
+    *IncompleteFormat = true;
+  return Result;
+}
+
 tooling::Replacements fixNamespaceEndComments(const FormatStyle &Style,
                                               StringRef Code,
                                               ArrayRef<tooling::Range> Ranges,
diff --git a/lib/Format/FormatToken.h b/lib/Format/FormatToken.h
index c9649126d93f..3b3600fede97 100644
--- a/lib/Format/FormatToken.h
+++ b/lib/Format/FormatToken.h
@@ -617,10 +617,12 @@ struct AdditionalKeywords {
     kw_finally = &IdentTable.get("finally");
     kw_from = &IdentTable.get("from");
     kw_function = &IdentTable.get("function");
+    kw_get = &IdentTable.get("get");
     kw_import = &IdentTable.get("import");
     kw_is = &IdentTable.get("is");
     kw_let = &IdentTable.get("let");
     kw_module = &IdentTable.get("module");
+    kw_set = &IdentTable.get("set");
     kw_type = &IdentTable.get("type");
     kw_var = &IdentTable.get("var");
     kw_yield = &IdentTable.get("yield");
@@ -675,10 +677,12 @@ struct AdditionalKeywords {
   IdentifierInfo *kw_finally;
   IdentifierInfo *kw_from;
   IdentifierInfo *kw_function;
+  IdentifierInfo *kw_get;
   IdentifierInfo *kw_import;
   IdentifierInfo *kw_is;
   IdentifierInfo *kw_let;
   IdentifierInfo *kw_module;
+  IdentifierInfo *kw_set;
   IdentifierInfo *kw_type;
   IdentifierInfo *kw_var;
   IdentifierInfo *kw_yield;
diff --git a/lib/Format/TokenAnnotator.cpp b/lib/Format/TokenAnnotator.cpp
index bbc2d1e52b63..c274d7bf07f8 100644
--- a/lib/Format/TokenAnnotator.cpp
+++ b/lib/Format/TokenAnnotator.cpp
@@ -1120,7 +1120,11 @@ private:
                 Current.Type = TT_FunctionAnnotationRParen;
           }
         }
-    } else if (Current.is(tok::at) && Current.Next) {
+    } else if (Current.is(tok::at) && Current.Next &&
+               Style.Language != FormatStyle::LK_JavaScript &&
+               Style.Language != FormatStyle::LK_Java) {
+      // In Java & JavaScript, "@..." is a decorator or annotation. In ObjC, it
+      // marks declarations and properties that need special formatting.
       switch (Current.Next->Tok.getObjCKeywordID()) {
       case tok::objc_interface:
       case tok::objc_implementation:
@@ -2541,9 +2545,11 @@ bool TokenAnnotator::canBreakBefore(const AnnotatedLine &Line,
   } else if (Style.Language == FormatStyle::LK_JavaScript) {
     const FormatToken *NonComment = Right.getPreviousNonComment();
     if (NonComment &&
-        NonComment->isOneOf(tok::kw_return, tok::kw_continue, tok::kw_break,
-                            tok::kw_throw, Keywords.kw_interface,
-                            Keywords.kw_type))
+        NonComment->isOneOf(
+            tok::kw_return, tok::kw_continue, tok::kw_break, tok::kw_throw,
+            Keywords.kw_interface, Keywords.kw_type, tok::kw_static,
+            tok::kw_public, tok::kw_private, tok::kw_protected,
+            Keywords.kw_abstract, Keywords.kw_get, Keywords.kw_set))
       return false; // Otherwise a semicolon is inserted.
     if (Left.is(TT_JsFatArrow) && Right.is(tok::l_brace))
       return false;
diff --git a/lib/Format/UnwrappedLineFormatter.cpp b/lib/Format/UnwrappedLineFormatter.cpp
index c3c154afeb8a..8ff893426e25 100644
--- a/lib/Format/UnwrappedLineFormatter.cpp
+++ b/lib/Format/UnwrappedLineFormatter.cpp
@@ -835,8 +835,11 @@ UnwrappedLineFormatter::format(const SmallVectorImpl<AnnotatedLine *> &Lines,
     bool ShouldFormat = TheLine.Affected || FixIndentation;
     // We cannot format this line; if the reason is that the line had a
     // parsing error, remember that.
-    if (ShouldFormat && TheLine.Type == LT_Invalid && IncompleteFormat)
-      *IncompleteFormat = true;
+    if (ShouldFormat && TheLine.Type == LT_Invalid && Status) {
+      Status->FormatComplete = false;
+      Status->Line =
+          SourceMgr.getSpellingLineNumber(TheLine.First->Tok.getLocation());
+    }
 
     if (ShouldFormat && TheLine.Type != LT_Invalid) {
       if (!DryRun)
diff --git a/lib/Format/UnwrappedLineFormatter.h b/lib/Format/UnwrappedLineFormatter.h
index 93247f71d6e0..55f0d1cac689 100644
--- a/lib/Format/UnwrappedLineFormatter.h
+++ b/lib/Format/UnwrappedLineFormatter.h
@@ -32,9 +32,11 @@ public:
                          WhitespaceManager *Whitespaces,
                          const FormatStyle &Style,
                          const AdditionalKeywords &Keywords,
-                         bool *IncompleteFormat)
+                         const SourceManager &SourceMgr,
+                         FormattingAttemptStatus *Status)
       : Indenter(Indenter), Whitespaces(Whitespaces), Style(Style),
-        Keywords(Keywords), IncompleteFormat(IncompleteFormat) {}
+        Keywords(Keywords), SourceMgr(SourceMgr),
+        Status(Status) {}
 
   /// \brief Format the current block and return the penalty.
   unsigned format(const SmallVectorImpl<AnnotatedLine *> &Lines,
@@ -63,7 +65,8 @@ private:
   WhitespaceManager *Whitespaces;
   const FormatStyle &Style;
   const AdditionalKeywords &Keywords;
-  bool *IncompleteFormat;
+  const SourceManager &SourceMgr;
+  FormattingAttemptStatus *Status;
 };
 } // end namespace format
 } // end namespace clang
diff --git a/lib/Frontend/CompilerInvocation.cpp b/lib/Frontend/CompilerInvocation.cpp
index 0e0eb40eb334..8cdb8298ee9e 100644
--- a/lib/Frontend/CompilerInvocation.cpp
+++ b/lib/Frontend/CompilerInvocation.cpp
@@ -519,6 +519,7 @@ static bool ParseCodeGenArgs(CodeGenOptions &Opts, ArgList &Args, InputKind IK,
   Opts.MacroDebugInfo = Args.hasArg(OPT_debug_info_macro);
   Opts.WholeProgramVTables = Args.hasArg(OPT_fwhole_program_vtables);
   Opts.LTOVisibilityPublicStd = Args.hasArg(OPT_flto_visibility_public_std);
+  Opts.EnableSplitDwarf = Args.hasArg(OPT_enable_split_dwarf);
   Opts.SplitDwarfFile = Args.getLastArgValue(OPT_split_dwarf_file);
   Opts.SplitDwarfInlining = !Args.hasArg(OPT_fno_split_dwarf_inlining);
   Opts.DebugTypeExtRefs = Args.hasArg(OPT_dwarf_ext_refs);
diff --git a/lib/Frontend/InitPreprocessor.cpp b/lib/Frontend/InitPreprocessor.cpp
index 88be7732f519..0dd04e8a8fff 100644
--- a/lib/Frontend/InitPreprocessor.cpp
+++ b/lib/Frontend/InitPreprocessor.cpp
@@ -882,14 +882,16 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
     // The value written by __atomic_test_and_set.
     // FIXME: This is target-dependent.
     Builder.defineMacro("__GCC_ATOMIC_TEST_AND_SET_TRUEVAL", "1");
+  }
 
+  auto addLockFreeMacros = [&](const llvm::Twine &Prefix) {
     // Used by libc++ and libstdc++ to implement ATOMIC_<foo>_LOCK_FREE.
     unsigned InlineWidthBits = TI.getMaxAtomicInlineWidth();
-#define DEFINE_LOCK_FREE_MACRO(TYPE, Type) \
-    Builder.defineMacro("__GCC_ATOMIC_" #TYPE "_LOCK_FREE", \
-                        getLockFreeValue(TI.get##Type##Width(), \
-                                         TI.get##Type##Align(), \
-                                         InlineWidthBits));
+#define DEFINE_LOCK_FREE_MACRO(TYPE, Type)                                     \
+  Builder.defineMacro(Prefix + #TYPE "_LOCK_FREE",                             \
+                      getLockFreeValue(TI.get##Type##Width(),                  \
+                                       TI.get##Type##Align(),                  \
+                                       InlineWidthBits));
     DEFINE_LOCK_FREE_MACRO(BOOL, Bool);
     DEFINE_LOCK_FREE_MACRO(CHAR, Char);
     DEFINE_LOCK_FREE_MACRO(CHAR16_T, Char16);
@@ -899,12 +901,15 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
     DEFINE_LOCK_FREE_MACRO(INT, Int);
     DEFINE_LOCK_FREE_MACRO(LONG, Long);
     DEFINE_LOCK_FREE_MACRO(LLONG, LongLong);
-    Builder.defineMacro("__GCC_ATOMIC_POINTER_LOCK_FREE",
+    Builder.defineMacro(Prefix + "POINTER_LOCK_FREE",
                         getLockFreeValue(TI.getPointerWidth(0),
                                          TI.getPointerAlign(0),
                                          InlineWidthBits));
 #undef DEFINE_LOCK_FREE_MACRO
-  }
+  };
+  addLockFreeMacros("__CLANG_ATOMIC_");
+  if (!LangOpts.MSVCCompat)
+    addLockFreeMacros("__GCC_ATOMIC_");
 
   if (LangOpts.NoInlineDefine)
     Builder.defineMacro("__NO_INLINE__");
diff --git a/lib/Headers/stdatomic.h b/lib/Headers/stdatomic.h
index 23bb3a357768..b4845a74e494 100644
--- a/lib/Headers/stdatomic.h
+++ b/lib/Headers/stdatomic.h
@@ -40,16 +40,16 @@ extern "C" {
 
 /* 7.17.1 Introduction */
 
-#define ATOMIC_BOOL_LOCK_FREE       __GCC_ATOMIC_BOOL_LOCK_FREE
-#define ATOMIC_CHAR_LOCK_FREE       __GCC_ATOMIC_CHAR_LOCK_FREE
-#define ATOMIC_CHAR16_T_LOCK_FREE   __GCC_ATOMIC_CHAR16_T_LOCK_FREE
-#define ATOMIC_CHAR32_T_LOCK_FREE   __GCC_ATOMIC_CHAR32_T_LOCK_FREE
-#define ATOMIC_WCHAR_T_LOCK_FREE    __GCC_ATOMIC_WCHAR_T_LOCK_FREE
-#define ATOMIC_SHORT_LOCK_FREE      __GCC_ATOMIC_SHORT_LOCK_FREE
-#define ATOMIC_INT_LOCK_FREE        __GCC_ATOMIC_INT_LOCK_FREE
-#define ATOMIC_LONG_LOCK_FREE       __GCC_ATOMIC_LONG_LOCK_FREE
-#define ATOMIC_LLONG_LOCK_FREE      __GCC_ATOMIC_LLONG_LOCK_FREE
-#define ATOMIC_POINTER_LOCK_FREE    __GCC_ATOMIC_POINTER_LOCK_FREE
+#define ATOMIC_BOOL_LOCK_FREE       __CLANG_ATOMIC_BOOL_LOCK_FREE
+#define ATOMIC_CHAR_LOCK_FREE       __CLANG_ATOMIC_CHAR_LOCK_FREE
+#define ATOMIC_CHAR16_T_LOCK_FREE   __CLANG_ATOMIC_CHAR16_T_LOCK_FREE
+#define ATOMIC_CHAR32_T_LOCK_FREE   __CLANG_ATOMIC_CHAR32_T_LOCK_FREE
+#define ATOMIC_WCHAR_T_LOCK_FREE    __CLANG_ATOMIC_WCHAR_T_LOCK_FREE
+#define ATOMIC_SHORT_LOCK_FREE      __CLANG_ATOMIC_SHORT_LOCK_FREE
+#define ATOMIC_INT_LOCK_FREE        __CLANG_ATOMIC_INT_LOCK_FREE
+#define ATOMIC_LONG_LOCK_FREE       __CLANG_ATOMIC_LONG_LOCK_FREE
+#define ATOMIC_LLONG_LOCK_FREE      __CLANG_ATOMIC_LLONG_LOCK_FREE
+#define ATOMIC_POINTER_LOCK_FREE    __CLANG_ATOMIC_POINTER_LOCK_FREE
 
 /* 7.17.2 Initialization */
 
diff --git a/lib/Index/IndexDecl.cpp b/lib/Index/IndexDecl.cpp
index 0e893505516f..c1eed1684cbd 100644
--- a/lib/Index/IndexDecl.cpp
+++ b/lib/Index/IndexDecl.cpp
@@ -14,6 +14,13 @@
 using namespace clang;
 using namespace index;
 
+#define TRY_DECL(D,CALL_EXPR)                                                  \
+  do {                                                                         \
+    if (!IndexCtx.shouldIndex(D)) return true;                                 \
+    if (!CALL_EXPR)                                                            \
+      return false;                                                            \
+  } while (0)
+
 #define TRY_TO(CALL_EXPR)                                                      \
   do {                                                                         \
     if (!CALL_EXPR)                                                            \
@@ -120,8 +127,7 @@ public:
                                D->getDeclContext(), 0);
     }
 
-    if (!IndexCtx.handleDecl(D, MethodLoc, Roles, Relations))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D, MethodLoc, Roles, Relations));
     IndexCtx.indexTypeSourceInfo(D->getReturnTypeSourceInfo(), D);
     bool hasIBActionAndFirst = D->hasAttr<IBActionAttr>();
     for (const auto *I : D->parameters()) {
@@ -138,6 +144,53 @@ public:
     return true;
   }
 
+  /// Gather the declarations which the given declaration \D overrides in a
+  /// pseudo-override manner.
+  ///
+  /// Pseudo-overrides occur when a class template specialization declares
+  /// a declaration that has the same name as a similar declaration in the
+  /// non-specialized template.
+  void
+  gatherTemplatePseudoOverrides(const NamedDecl *D,
+                                SmallVectorImpl<SymbolRelation> &Relations) {
+    if (!IndexCtx.getLangOpts().CPlusPlus)
+      return;
+    const auto *CTSD =
+        dyn_cast<ClassTemplateSpecializationDecl>(D->getLexicalDeclContext());
+    if (!CTSD)
+      return;
+    llvm::PointerUnion<ClassTemplateDecl *,
+                       ClassTemplatePartialSpecializationDecl *>
+        Template = CTSD->getSpecializedTemplateOrPartial();
+    if (const auto *CTD = Template.dyn_cast<ClassTemplateDecl *>()) {
+      const CXXRecordDecl *Pattern = CTD->getTemplatedDecl();
+      bool TypeOverride = isa<TypeDecl>(D);
+      for (const NamedDecl *ND : Pattern->lookup(D->getDeclName())) {
+        if (const auto *CTD = dyn_cast<ClassTemplateDecl>(ND))
+          ND = CTD->getTemplatedDecl();
+        if (ND->isImplicit())
+          continue;
+        // Types can override other types.
+        if (!TypeOverride) {
+          if (ND->getKind() != D->getKind())
+            continue;
+        } else if (!isa<TypeDecl>(ND))
+          continue;
+        if (const auto *FD = dyn_cast<FunctionDecl>(ND)) {
+          const auto *DFD = cast<FunctionDecl>(D);
+          // Function overrides are approximated using the number of parameters.
+          if (FD->getStorageClass() != DFD->getStorageClass() ||
+              FD->getNumParams() != DFD->getNumParams())
+            continue;
+        }
+        Relations.emplace_back(
+            SymbolRoleSet(SymbolRole::RelationOverrideOf) |
+                SymbolRoleSet(SymbolRole::RelationSpecializationOf),
+            ND);
+      }
+    }
+  }
+
   bool VisitFunctionDecl(const FunctionDecl *D) {
     if (D->isDeleted())
       return true;
@@ -152,9 +205,13 @@ public:
         Relations.emplace_back((unsigned)SymbolRole::RelationOverrideOf, *I);
       }
     }
+    gatherTemplatePseudoOverrides(D, Relations);
+    if (const auto *Base = D->getPrimaryTemplate())
+      Relations.push_back(
+          SymbolRelation(SymbolRoleSet(SymbolRole::RelationSpecializationOf),
+                         Base->getTemplatedDecl()));
 
-    if (!IndexCtx.handleDecl(D, Roles, Relations))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D, Roles, Relations));
     handleDeclarator(D);
 
     if (const CXXConstructorDecl *Ctor = dyn_cast<CXXConstructorDecl>(D)) {
@@ -189,16 +246,18 @@ public:
   }
 
   bool VisitVarDecl(const VarDecl *D) {
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    SmallVector<SymbolRelation, 4> Relations;
+    gatherTemplatePseudoOverrides(D, Relations);
+    TRY_DECL(D, IndexCtx.handleDecl(D, SymbolRoleSet(), Relations));
     handleDeclarator(D);
     IndexCtx.indexBody(D->getInit(), D);
     return true;
   }
 
   bool VisitFieldDecl(const FieldDecl *D) {
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    SmallVector<SymbolRelation, 4> Relations;
+    gatherTemplatePseudoOverrides(D, Relations);
+    TRY_DECL(D, IndexCtx.handleDecl(D, SymbolRoleSet(), Relations));
     handleDeclarator(D);
     if (D->isBitField())
       IndexCtx.indexBody(D->getBitWidth(), D);
@@ -212,8 +271,7 @@ public:
       // handled in VisitObjCPropertyImplDecl
       return true;
     }
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D));
     handleDeclarator(D);
     return true;
   }
@@ -224,17 +282,18 @@ public:
   }
 
   bool VisitEnumConstantDecl(const EnumConstantDecl *D) {
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D));
     IndexCtx.indexBody(D->getInitExpr(), D);
     return true;
   }
 
   bool VisitTypedefNameDecl(const TypedefNameDecl *D) {
-    if (!D->isTransparentTag())
-      if (!IndexCtx.handleDecl(D))
-        return false;
-    IndexCtx.indexTypeSourceInfo(D->getTypeSourceInfo(), D);
+    if (!D->isTransparentTag()) {
+      SmallVector<SymbolRelation, 4> Relations;
+      gatherTemplatePseudoOverrides(D, Relations);
+      TRY_DECL(D, IndexCtx.handleDecl(D, SymbolRoleSet(), Relations));
+      IndexCtx.indexTypeSourceInfo(D->getTypeSourceInfo(), D);
+    }
     return true;
   }
 
@@ -242,7 +301,9 @@ public:
     // Non-free standing tags are handled in indexTypeSourceInfo.
     if (D->isFreeStanding()) {
       if (D->isThisDeclarationADefinition()) {
-        IndexCtx.indexTagDecl(D);
+        SmallVector<SymbolRelation, 4> Relations;
+        gatherTemplatePseudoOverrides(D, Relations);
+        IndexCtx.indexTagDecl(D, Relations);
       } else {
         auto *Parent = dyn_cast<NamedDecl>(D->getDeclContext());
         return IndexCtx.handleReference(D, D->getLocation(), Parent,
@@ -272,7 +333,7 @@ public:
 
   bool VisitObjCInterfaceDecl(const ObjCInterfaceDecl *D) {
     if (D->isThisDeclarationADefinition()) {
-      TRY_TO(IndexCtx.handleDecl(D));
+      TRY_DECL(D, IndexCtx.handleDecl(D));
       SourceLocation SuperLoc = D->getSuperClassLoc();
       if (auto *SuperD = D->getSuperClass()) {
         bool hasSuperTypedef = false;
@@ -303,7 +364,7 @@ public:
 
   bool VisitObjCProtocolDecl(const ObjCProtocolDecl *D) {
     if (D->isThisDeclarationADefinition()) {
-      TRY_TO(IndexCtx.handleDecl(D));
+      TRY_DECL(D, IndexCtx.handleDecl(D));
       TRY_TO(handleReferencedProtocols(D->getReferencedProtocols(), D,
                                        /*superLoc=*/SourceLocation()));
       TRY_TO(IndexCtx.indexDeclContext(D));
@@ -322,8 +383,7 @@ public:
     if (Class->isImplicitInterfaceDecl())
       IndexCtx.handleDecl(Class);
 
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D));
 
     // Visit implicit @synthesize property implementations first as their
     // location is reported at the name of the @implementation block. This
@@ -342,6 +402,8 @@ public:
   }
 
   bool VisitObjCCategoryDecl(const ObjCCategoryDecl *D) {
+    if (!IndexCtx.shouldIndex(D))
+      return true;
     const ObjCInterfaceDecl *C = D->getClassInterface();
     if (!C)
       return true;
@@ -370,8 +432,7 @@ public:
     SourceLocation CategoryLoc = D->getCategoryNameLoc();
     if (!CategoryLoc.isValid())
       CategoryLoc = D->getLocation();
-    if (!IndexCtx.handleDecl(D, CategoryLoc))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D, CategoryLoc));
     IndexCtx.indexDeclContext(D);
     return true;
   }
@@ -393,8 +454,7 @@ public:
     if (ObjCMethodDecl *MD = D->getSetterMethodDecl())
       if (MD->getLexicalDeclContext() == D->getLexicalDeclContext())
         handleObjCMethod(MD, D);
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D));
     if (IBOutletCollectionAttr *attr = D->getAttr<IBOutletCollectionAttr>())
       IndexCtx.indexTypeSourceInfo(attr->getInterfaceLoc(), D,
                                    D->getLexicalDeclContext(), false, true);
@@ -415,8 +475,7 @@ public:
       Loc = Container->getLocation();
       Roles |= (SymbolRoleSet)SymbolRole::Implicit;
     }
-    if (!IndexCtx.handleDecl(D, Loc, Roles, Relations))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D, Loc, Roles, Relations));
 
     if (D->getPropertyImplementation() == ObjCPropertyImplDecl::Dynamic)
       return true;
@@ -450,8 +509,7 @@ public:
         } else if (D->getLocation() == IvarLoc) {
           IvarRoles = (SymbolRoleSet)SymbolRole::Implicit;
         }
-        if(!IndexCtx.handleDecl(IvarD, IvarLoc, IvarRoles))
-          return false;
+        TRY_DECL(IvarD, IndexCtx.handleDecl(IvarD, IvarLoc, IvarRoles));
       } else {
         IndexCtx.handleReference(IvarD, D->getPropertyIvarDeclLoc(), nullptr,
                                  D->getDeclContext(), SymbolRoleSet());
@@ -461,8 +519,7 @@ public:
   }
 
   bool VisitNamespaceDecl(const NamespaceDecl *D) {
-    if (!IndexCtx.handleDecl(D))
-      return false;
+    TRY_DECL(D, IndexCtx.handleDecl(D));
     IndexCtx.indexDeclContext(D);
     return true;
   }
@@ -507,6 +564,9 @@ public:
           D, SymbolRelation(SymbolRoleSet(SymbolRole::RelationSpecializationOf),
                             SpecializationOf));
     }
+    if (TypeSourceInfo *TSI = D->getTypeAsWritten())
+      IndexCtx.indexTypeSourceInfo(TSI, /*Parent=*/nullptr,
+                                   D->getLexicalDeclContext());
     return true;
   }
 
diff --git a/lib/Index/IndexSymbol.cpp b/lib/Index/IndexSymbol.cpp
index ea66b7017951..0bfa19346b4e 100644
--- a/lib/Index/IndexSymbol.cpp
+++ b/lib/Index/IndexSymbol.cpp
@@ -318,6 +318,20 @@ SymbolInfo index::getSymbolInfo(const Decl *D) {
   if (Info.Properties & (unsigned)SymbolProperty::Generic)
     Info.Lang = SymbolLanguage::CXX;
 
+  auto getExternalSymAttr = [](const Decl *D) -> ExternalSourceSymbolAttr* {
+    if (auto *attr = D->getAttr<ExternalSourceSymbolAttr>())
+      return attr;
+    if (auto *dcd = dyn_cast<Decl>(D->getDeclContext())) {
+      if (auto *attr = dcd->getAttr<ExternalSourceSymbolAttr>())
+        return attr;
+    }
+    return nullptr;
+  };
+  if (auto *attr = getExternalSymAttr(D)) {
+    if (attr->getLanguage() == "Swift")
+      Info.Lang = SymbolLanguage::Swift;
+  }
+
   return Info;
 }
 
@@ -458,6 +472,7 @@ StringRef index::getSymbolLanguageString(SymbolLanguage K) {
   case SymbolLanguage::C: return "C";
   case SymbolLanguage::ObjC: return "ObjC";
   case SymbolLanguage::CXX: return "C++";
+  case SymbolLanguage::Swift: return "Swift";
   }
   llvm_unreachable("invalid symbol language kind");
 }
diff --git a/lib/Index/IndexTypeSourceInfo.cpp b/lib/Index/IndexTypeSourceInfo.cpp
index a3566a9f2ae8..44d1241fb930 100644
--- a/lib/Index/IndexTypeSourceInfo.cpp
+++ b/lib/Index/IndexTypeSourceInfo.cpp
@@ -210,6 +210,8 @@ void IndexingContext::indexNestedNameSpecifierLoc(NestedNameSpecifierLoc NNS,
 
 void IndexingContext::indexTagDecl(const TagDecl *D,
                                    ArrayRef<SymbolRelation> Relations) {
+  if (!shouldIndex(D))
+    return;
   if (!shouldIndexFunctionLocalSymbols() && isFunctionLocalSymbol(D))
     return;
 
diff --git a/lib/Index/IndexingContext.cpp b/lib/Index/IndexingContext.cpp
index 85574d0a314d..709a23657b07 100644
--- a/lib/Index/IndexingContext.cpp
+++ b/lib/Index/IndexingContext.cpp
@@ -17,6 +17,21 @@
 using namespace clang;
 using namespace index;
 
+static bool isGeneratedDecl(const Decl *D) {
+  if (auto *attr = D->getAttr<ExternalSourceSymbolAttr>()) {
+    return attr->getGeneratedDeclaration();
+  }
+  return false;
+}
+
+bool IndexingContext::shouldIndex(const Decl *D) {
+  return !isGeneratedDecl(D);
+}
+
+const LangOptions &IndexingContext::getLangOpts() const {
+  return Ctx->getLangOpts();
+}
+
 bool IndexingContext::shouldIndexFunctionLocalSymbols() const {
   return IndexOpts.IndexFunctionLocals;
 }
diff --git a/lib/Index/IndexingContext.h b/lib/Index/IndexingContext.h
index 1ebf6f9ce67a..566651c83a75 100644
--- a/lib/Index/IndexingContext.h
+++ b/lib/Index/IndexingContext.h
@@ -48,6 +48,10 @@ public:
 
   void setASTContext(ASTContext &ctx) { Ctx = &ctx; }
 
+  bool shouldIndex(const Decl *D);
+
+  const LangOptions &getLangOpts() const;
+
   bool shouldSuppressRefs() const {
     return false;
   }
diff --git a/lib/Index/USRGeneration.cpp b/lib/Index/USRGeneration.cpp
index 73dddd9a8b36..ed469f677a34 100644
--- a/lib/Index/USRGeneration.cpp
+++ b/lib/Index/USRGeneration.cpp
@@ -46,6 +46,15 @@ static bool printLoc(llvm::raw_ostream &OS, SourceLocation Loc,
   return false;
 }
 
+static StringRef GetExternalSourceContainer(const NamedDecl *D) {
+  if (!D)
+    return StringRef();
+  if (auto *attr = D->getAttr<ExternalSourceSymbolAttr>()) {
+    return attr->getDefinedIn();
+  }
+  return StringRef();
+}
+
 namespace {
 class USRGenerator : public ConstDeclVisitor<USRGenerator> {
   SmallVectorImpl<char> &Buf;
@@ -79,7 +88,8 @@ public:
   void VisitNamespaceAliasDecl(const NamespaceAliasDecl *D);
   void VisitFunctionTemplateDecl(const FunctionTemplateDecl *D);
   void VisitClassTemplateDecl(const ClassTemplateDecl *D);
-  void VisitObjCContainerDecl(const ObjCContainerDecl *CD);
+  void VisitObjCContainerDecl(const ObjCContainerDecl *CD,
+                              const ObjCCategoryDecl *CatD = nullptr);
   void VisitObjCMethodDecl(const ObjCMethodDecl *MD);
   void VisitObjCPropertyDecl(const ObjCPropertyDecl *D);
   void VisitObjCPropertyImplDecl(const ObjCPropertyImplDecl *D);
@@ -116,6 +126,8 @@ public:
     return D->getParentFunctionOrMethod() != nullptr;
   }
 
+  void GenExtSymbolContainer(const NamedDecl *D);
+
   /// Generate the string component containing the location of the
   ///  declaration.
   bool GenLoc(const Decl *D, bool IncludeOffset);
@@ -127,13 +139,16 @@ public:
   /// itself.
 
   /// Generate a USR for an Objective-C class.
-  void GenObjCClass(StringRef cls) {
-    generateUSRForObjCClass(cls, Out);
+  void GenObjCClass(StringRef cls, StringRef ExtSymDefinedIn,
+                    StringRef CategoryContextExtSymbolDefinedIn) {
+    generateUSRForObjCClass(cls, Out, ExtSymDefinedIn,
+                            CategoryContextExtSymbolDefinedIn);
   }
 
   /// Generate a USR for an Objective-C class category.
-  void GenObjCCategory(StringRef cls, StringRef cat) {
-    generateUSRForObjCCategory(cls, cat, Out);
+  void GenObjCCategory(StringRef cls, StringRef cat,
+                       StringRef clsExt, StringRef catExt) {
+    generateUSRForObjCCategory(cls, cat, Out, clsExt, catExt);
   }
 
   /// Generate a USR fragment for an Objective-C property.
@@ -142,8 +157,8 @@ public:
   }
 
   /// Generate a USR for an Objective-C protocol.
-  void GenObjCProtocol(StringRef prot) {
-    generateUSRForObjCProtocol(prot, Out);
+  void GenObjCProtocol(StringRef prot, StringRef ext) {
+    generateUSRForObjCProtocol(prot, Out, ext);
   }
 
   void VisitType(QualType T);
@@ -204,7 +219,11 @@ void USRGenerator::VisitFunctionDecl(const FunctionDecl *D) {
   if (ShouldGenerateLocation(D) && GenLoc(D, /*IncludeOffset=*/isLocal(D)))
     return;
 
+  const unsigned StartSize = Buf.size();
   VisitDeclContext(D->getDeclContext());
+  if (Buf.size() == StartSize)
+    GenExtSymbolContainer(D);
+
   bool IsTemplate = false;
   if (FunctionTemplateDecl *FunTmpl = D->getDescribedFunctionTemplate()) {
     IsTemplate = true;
@@ -367,7 +386,16 @@ void USRGenerator::VisitObjCMethodDecl(const ObjCMethodDecl *D) {
       IgnoreResults = true;
       return;
     }
-    Visit(ID);
+    auto getCategoryContext = [](const ObjCMethodDecl *D) ->
+                                    const ObjCCategoryDecl * {
+      if (auto *CD = dyn_cast<ObjCCategoryDecl>(D->getDeclContext()))
+        return CD;
+      if (auto *ICD = dyn_cast<ObjCCategoryImplDecl>(D->getDeclContext()))
+        return ICD->getCategoryDecl();
+      return nullptr;
+    };
+    auto *CD = getCategoryContext(D);
+    VisitObjCContainerDecl(ID, CD);
   }
   // Ideally we would use 'GenObjCMethod', but this is such a hot path
   // for Objective-C code that we don't want to use
@@ -376,13 +404,15 @@ void USRGenerator::VisitObjCMethodDecl(const ObjCMethodDecl *D) {
       << DeclarationName(D->getSelector());
 }
 
-void USRGenerator::VisitObjCContainerDecl(const ObjCContainerDecl *D) {
+void USRGenerator::VisitObjCContainerDecl(const ObjCContainerDecl *D,
+                                          const ObjCCategoryDecl *CatD) {
   switch (D->getKind()) {
     default:
       llvm_unreachable("Invalid ObjC container.");
     case Decl::ObjCInterface:
     case Decl::ObjCImplementation:
-      GenObjCClass(D->getName());
+      GenObjCClass(D->getName(), GetExternalSourceContainer(D),
+                   GetExternalSourceContainer(CatD));
       break;
     case Decl::ObjCCategory: {
       const ObjCCategoryDecl *CD = cast<ObjCCategoryDecl>(D);
@@ -402,7 +432,9 @@ void USRGenerator::VisitObjCContainerDecl(const ObjCContainerDecl *D) {
         GenLoc(CD, /*IncludeOffset=*/true);
       }
       else
-        GenObjCCategory(ID->getName(), CD->getName());
+        GenObjCCategory(ID->getName(), CD->getName(),
+                        GetExternalSourceContainer(ID),
+                        GetExternalSourceContainer(CD));
 
       break;
     }
@@ -417,12 +449,16 @@ void USRGenerator::VisitObjCContainerDecl(const ObjCContainerDecl *D) {
         IgnoreResults = true;
         return;
       }
-      GenObjCCategory(ID->getName(), CD->getName());
+      GenObjCCategory(ID->getName(), CD->getName(),
+                      GetExternalSourceContainer(ID),
+                      GetExternalSourceContainer(CD));
       break;
     }
-    case Decl::ObjCProtocol:
-      GenObjCProtocol(cast<ObjCProtocolDecl>(D)->getName());
+    case Decl::ObjCProtocol: {
+      const ObjCProtocolDecl *PD = cast<ObjCProtocolDecl>(D);
+      GenObjCProtocol(PD->getName(), GetExternalSourceContainer(PD));
       break;
+    }
   }
 }
 
@@ -452,6 +488,8 @@ void USRGenerator::VisitTagDecl(const TagDecl *D) {
       ShouldGenerateLocation(D) && GenLoc(D, /*IncludeOffset=*/isLocal(D)))
     return;
 
+  GenExtSymbolContainer(D);
+
   D = D->getCanonicalDecl();
   VisitDeclContext(D->getDeclContext());
 
@@ -544,6 +582,12 @@ void USRGenerator::VisitTemplateTypeParmDecl(const TemplateTypeParmDecl *D) {
   GenLoc(D, /*IncludeOffset=*/true);
 }
 
+void USRGenerator::GenExtSymbolContainer(const NamedDecl *D) {
+  StringRef Container = GetExternalSourceContainer(D);
+  if (!Container.empty())
+    Out << "@M@" << Container;
+}
+
 bool USRGenerator::GenLoc(const Decl *D, bool IncludeOffset) {
   if (generatedLoc)
     return IgnoreResults;
@@ -866,12 +910,34 @@ void USRGenerator::VisitTemplateArgument(const TemplateArgument &Arg) {
 // USR generation functions.
 //===----------------------------------------------------------------------===//
 
-void clang::index::generateUSRForObjCClass(StringRef Cls, raw_ostream &OS) {
+static void combineClassAndCategoryExtContainers(StringRef ClsSymDefinedIn,
+                                                 StringRef CatSymDefinedIn,
+                                                 raw_ostream &OS) {
+  if (ClsSymDefinedIn.empty() && CatSymDefinedIn.empty())
+    return;
+  if (CatSymDefinedIn.empty()) {
+    OS << "@M@" << ClsSymDefinedIn << '@';
+    return;
+  }
+  OS << "@CM@" << CatSymDefinedIn << '@';
+  if (ClsSymDefinedIn != CatSymDefinedIn) {
+    OS << ClsSymDefinedIn << '@';
+  }
+}
+
+void clang::index::generateUSRForObjCClass(StringRef Cls, raw_ostream &OS,
+                                           StringRef ExtSymDefinedIn,
+                                  StringRef CategoryContextExtSymbolDefinedIn) {
+  combineClassAndCategoryExtContainers(ExtSymDefinedIn,
+                                       CategoryContextExtSymbolDefinedIn, OS);
   OS << "objc(cs)" << Cls;
 }
 
 void clang::index::generateUSRForObjCCategory(StringRef Cls, StringRef Cat,
-                                              raw_ostream &OS) {
+                                              raw_ostream &OS,
+                                              StringRef ClsSymDefinedIn,
+                                              StringRef CatSymDefinedIn) {
+  combineClassAndCategoryExtContainers(ClsSymDefinedIn, CatSymDefinedIn, OS);
   OS << "objc(cy)" << Cls << '@' << Cat;
 }
 
@@ -890,10 +956,25 @@ void clang::index::generateUSRForObjCProperty(StringRef Prop, bool isClassProp,
   OS << (isClassProp ? "(cpy)" : "(py)") << Prop;
 }
 
-void clang::index::generateUSRForObjCProtocol(StringRef Prot, raw_ostream &OS) {
+void clang::index::generateUSRForObjCProtocol(StringRef Prot, raw_ostream &OS,
+                                              StringRef ExtSymDefinedIn) {
+  if (!ExtSymDefinedIn.empty())
+    OS << "@M@" << ExtSymDefinedIn << '@';
   OS << "objc(pl)" << Prot;
 }
 
+void clang::index::generateUSRForGlobalEnum(StringRef EnumName, raw_ostream &OS,
+                                            StringRef ExtSymDefinedIn) {
+  if (!ExtSymDefinedIn.empty())
+    OS << "@M@" << ExtSymDefinedIn;
+  OS << "@E@" << EnumName;
+}
+
+void clang::index::generateUSRForEnumConstant(StringRef EnumConstantName,
+                                              raw_ostream &OS) {
+  OS << '@' << EnumConstantName;
+}
+
 bool clang::index::generateUSRForDecl(const Decl *D,
                                       SmallVectorImpl<char> &Buf) {
   if (!D)
diff --git a/lib/Lex/ModuleMap.cpp b/lib/Lex/ModuleMap.cpp
index 5b60ed3f812a..512d7dc5de68 100644
--- a/lib/Lex/ModuleMap.cpp
+++ b/lib/Lex/ModuleMap.cpp
@@ -581,6 +581,7 @@ Module *ModuleMap::createModuleForInterfaceUnit(SourceLocation Loc,
   auto *Result =
       new Module(Name, Loc, nullptr, /*IsFramework*/ false,
                  /*IsExplicit*/ false, NumCreatedModules++);
+  Result->Kind = Module::ModuleInterfaceUnit;
   Modules[Name] = SourceModule = Result;
 
   // Mark the main source file as being within the newly-created module so that
diff --git a/lib/Lex/Pragma.cpp b/lib/Lex/Pragma.cpp
index f81eaa31e9e9..87e105d1d03d 100644
--- a/lib/Lex/Pragma.cpp
+++ b/lib/Lex/Pragma.cpp
@@ -989,9 +989,9 @@ struct PragmaDebugHandler : public PragmaHandler {
 #ifdef _MSC_VER
     #pragma warning(disable : 4717)
 #endif
-  static void DebugOverflowStack() {
-    void (*volatile Self)() = DebugOverflowStack;
-    Self();
+  static void DebugOverflowStack(void (*P)() = nullptr) {
+    void (*volatile Self)(void(*P)()) = DebugOverflowStack;
+    Self(reinterpret_cast<void(*)()>(Self));
   }
 #ifdef _MSC_VER
     #pragma warning(default : 4717)
diff --git a/lib/Parse/ParseExpr.cpp b/lib/Parse/ParseExpr.cpp
index ad45ab114fde..3e02e46ddc7d 100644
--- a/lib/Parse/ParseExpr.cpp
+++ b/lib/Parse/ParseExpr.cpp
@@ -339,7 +339,7 @@ Parser::ParseRHSOfBinaryExpression(ExprResult LHS, prec::Level MinPrec) {
         ColonLoc = Tok.getLocation();
       }
     }
-    
+
     // Code completion for the right-hand side of an assignment expression
     // goes through a special hook that takes the left-hand side into account.
     if (Tok.is(tok::code_completion) && NextTokPrec == prec::Assignment) {
@@ -347,7 +347,7 @@ Parser::ParseRHSOfBinaryExpression(ExprResult LHS, prec::Level MinPrec) {
       cutOffParsing();
       return ExprError();
     }
-    
+
     // Parse another leaf here for the RHS of the operator.
     // ParseCastExpression works here because all RHS expressions in C have it
     // as a prefix, at least. However, in C++, an assignment-expression could
@@ -456,6 +456,7 @@ Parser::ParseRHSOfBinaryExpression(ExprResult LHS, prec::Level MinPrec) {
       if (!getLangOpts().CPlusPlus)
         continue;
     }
+
     // Ensure potential typos aren't left undiagnosed.
     if (LHS.isInvalid()) {
       Actions.CorrectDelayedTyposInExpr(OrigLHS);
diff --git a/lib/Parse/Parser.cpp b/lib/Parse/Parser.cpp
index edbfc636bc46..38476ce7e15d 100644
--- a/lib/Parse/Parser.cpp
+++ b/lib/Parse/Parser.cpp
@@ -534,23 +534,6 @@ void Parser::LateTemplateParserCleanupCallback(void *P) {
 }
 
 bool Parser::ParseFirstTopLevelDecl(DeclGroupPtrTy &Result) {
-  // C++ Modules TS: module-declaration must be the first declaration in the
-  // file. (There can be no preceding preprocessor directives, but we expect
-  // the lexer to check that.)
-  if (Tok.is(tok::kw_module)) {
-    Result = ParseModuleDecl();
-    return false;
-  } else if (getLangOpts().getCompilingModule() ==
-             LangOptions::CMK_ModuleInterface) {
-    // FIXME: We avoid providing this diagnostic when generating an object file
-    // from an existing PCM file. This is not a good way to detect this
-    // condition; we should provide a mechanism to indicate whether we've
-    // already parsed a declaration in this translation unit and avoid calling
-    // ParseFirstTopLevelDecl in that case.
-    if (Actions.TUKind == TU_Module)
-      Diag(Tok, diag::err_expected_module_interface_decl);
-  }
-
   // C11 6.9p1 says translation units must have at least one top-level
   // declaration. C++ doesn't have this restriction. We also don't want to
   // complain if we have a precompiled header, although technically if the PCH
@@ -583,6 +566,14 @@ bool Parser::ParseTopLevelDecl(DeclGroupPtrTy &Result) {
     Result = ParseModuleImport(SourceLocation());
     return false;
 
+  case tok::kw_export:
+    if (NextToken().isNot(tok::kw_module))
+      break;
+    LLVM_FALLTHROUGH;
+  case tok::kw_module:
+    Result = ParseModuleDecl();
+    return false;
+
   case tok::annot_module_include:
     Actions.ActOnModuleInclude(Tok.getLocation(),
                                reinterpret_cast<Module *>(
@@ -2049,30 +2040,28 @@ void Parser::ParseMicrosoftIfExistsExternalDeclaration() {
 /// Parse a C++ Modules TS module declaration, which appears at the beginning
 /// of a module interface, module partition, or module implementation file.
 ///
-///   module-declaration:   [Modules TS + P0273R0]
-///     'module' module-kind[opt] module-name attribute-specifier-seq[opt] ';'
-///   module-kind:
-///     'implementation'
-///     'partition'
+///   module-declaration:   [Modules TS + P0273R0 + P0629R0]
+///     'export'[opt] 'module' 'partition'[opt]
+///            module-name attribute-specifier-seq[opt] ';'
 ///
-/// Note that the module-kind values are context-sensitive keywords.
+/// Note that 'partition' is a context-sensitive keyword.
 Parser::DeclGroupPtrTy Parser::ParseModuleDecl() {
-  assert(Tok.is(tok::kw_module) && getLangOpts().ModulesTS &&
-         "should not be parsing a module declaration");
+  SourceLocation StartLoc = Tok.getLocation();
+
+  Sema::ModuleDeclKind MDK = TryConsumeToken(tok::kw_export)
+                                 ? Sema::ModuleDeclKind::Module
+                                 : Sema::ModuleDeclKind::Implementation;
+
+  assert(Tok.is(tok::kw_module) && "not a module declaration");
   SourceLocation ModuleLoc = ConsumeToken();
 
-  // Check for a module-kind.
-  Sema::ModuleDeclKind MDK = Sema::ModuleDeclKind::Module;
-  if (Tok.is(tok::identifier) && NextToken().is(tok::identifier)) {
-    if (Tok.getIdentifierInfo()->isStr("implementation"))
-      MDK = Sema::ModuleDeclKind::Implementation;
-    else if (Tok.getIdentifierInfo()->isStr("partition"))
-      MDK = Sema::ModuleDeclKind::Partition;
-    else {
-      Diag(Tok, diag::err_unexpected_module_kind) << Tok.getIdentifierInfo();
-      SkipUntil(tok::semi);
-      return nullptr;
-    }
+  if (Tok.is(tok::identifier) && NextToken().is(tok::identifier) &&
+      Tok.getIdentifierInfo()->isStr("partition")) {
+    // If 'partition' is present, this must be a module interface unit.
+    if (MDK != Sema::ModuleDeclKind::Module)
+      Diag(Tok.getLocation(), diag::err_module_implementation_partition)
+        << FixItHint::CreateInsertion(ModuleLoc, "export ");
+    MDK = Sema::ModuleDeclKind::Partition;
     ConsumeToken();
   }
 
@@ -2080,14 +2069,14 @@ Parser::DeclGroupPtrTy Parser::ParseModuleDecl() {
   if (ParseModuleName(ModuleLoc, Path, /*IsImport*/false))
     return nullptr;
 
+  // We don't support any module attributes yet; just parse them and diagnose.
   ParsedAttributesWithRange Attrs(AttrFactory);
   MaybeParseCXX11Attributes(Attrs);
-  // We don't support any module attributes yet.
   ProhibitCXX11Attributes(Attrs, diag::err_attribute_not_module_attr);
 
   ExpectAndConsumeSemi(diag::err_module_expected_semi);
 
-  return Actions.ActOnModuleDecl(ModuleLoc, MDK, Path);
+  return Actions.ActOnModuleDecl(StartLoc, ModuleLoc, MDK, Path);
 }
 
 /// Parse a module import declaration. This is essentially the same for
diff --git a/lib/Sema/Sema.cpp b/lib/Sema/Sema.cpp
index 950f04088822..94e792c1d17d 100644
--- a/lib/Sema/Sema.cpp
+++ b/lib/Sema/Sema.cpp
@@ -1161,10 +1161,14 @@ void Sema::PushFunctionScope() {
     // memory for a new scope.
     FunctionScopes.back()->Clear();
     FunctionScopes.push_back(FunctionScopes.back());
+    if (LangOpts.OpenMP)
+      pushOpenMPFunctionRegion();
     return;
   }
 
   FunctionScopes.push_back(new FunctionScopeInfo(getDiagnostics()));
+  if (LangOpts.OpenMP)
+    pushOpenMPFunctionRegion();
 }
 
 void Sema::PushBlockScope(Scope *BlockScope, BlockDecl *Block) {
@@ -1192,6 +1196,9 @@ void Sema::PopFunctionScopeInfo(const AnalysisBasedWarnings::Policy *WP,
   FunctionScopeInfo *Scope = FunctionScopes.pop_back_val();
   assert(!FunctionScopes.empty() && "mismatched push/pop!");
 
+  if (LangOpts.OpenMP)
+    popOpenMPFunctionRegion(Scope);
+
   // Issue any analysis-based warnings.
   if (WP && D)
     AnalysisWarnings.IssueWarnings(*WP, Scope, D, blkExpr);
diff --git a/lib/Sema/SemaChecking.cpp b/lib/Sema/SemaChecking.cpp
index 45523b30ef22..044ec74679d5 100644
--- a/lib/Sema/SemaChecking.cpp
+++ b/lib/Sema/SemaChecking.cpp
@@ -408,7 +408,7 @@ static bool SemaOpenCLBuiltinEnqueueKernel(Sema &S, CallExpr *TheCall) {
   }
 
   // Third argument is always an ndrange_t type.
-  if (Arg2->getType().getAsString() != "ndrange_t") {
+  if (Arg2->getType().getUnqualifiedType().getAsString() != "ndrange_t") {
     S.Diag(TheCall->getArg(2)->getLocStart(),
            diag::err_opencl_enqueue_kernel_expected_type)
         << "'ndrange_t'";
diff --git a/lib/Sema/SemaDecl.cpp b/lib/Sema/SemaDecl.cpp
index f3ffcf5d696c..f838c9a4877d 100644
--- a/lib/Sema/SemaDecl.cpp
+++ b/lib/Sema/SemaDecl.cpp
@@ -615,7 +615,7 @@ bool Sema::isMicrosoftMissingTypename(const CXXScopeSpec *SS, Scope *S) {
 
     CXXRecordDecl *RD = cast<CXXRecordDecl>(CurContext);
     for (const auto &Base : RD->bases())
-      if (Context.hasSameUnqualifiedType(QualType(Ty, 1), Base.getType()))
+      if (Ty && Context.hasSameUnqualifiedType(QualType(Ty, 1), Base.getType()))
         return true;
     return S->isFunctionPrototypeScope();
   }
@@ -13523,7 +13523,8 @@ CreateNewDecl:
     // If this is an undefined enum, warn.
     if (TUK != TUK_Definition && !Invalid) {
       TagDecl *Def;
-      if ((getLangOpts().CPlusPlus11 || getLangOpts().ObjC2) &&
+      if (!EnumUnderlyingIsImplicit &&
+          (getLangOpts().CPlusPlus11 || getLangOpts().ObjC2) &&
           cast<EnumDecl>(New)->isFixed()) {
         // C++0x: 7.2p2: opaque-enum-declaration.
         // Conflicts are diagnosed above. Do nothing.
@@ -15687,30 +15688,41 @@ static void checkModuleImportContext(Sema &S, Module *M,
   }
 }
 
-Sema::DeclGroupPtrTy Sema::ActOnModuleDecl(SourceLocation ModuleLoc,
+Sema::DeclGroupPtrTy Sema::ActOnModuleDecl(SourceLocation StartLoc,
+                                           SourceLocation ModuleLoc,
                                            ModuleDeclKind MDK,
                                            ModuleIdPath Path) {
-  // 'module implementation' requires that we are not compiling a module of any
-  // kind. 'module' and 'module partition' require that we are compiling a
-  // module inteface (not a module map).
-  auto CMK = getLangOpts().getCompilingModule();
-  if (MDK == ModuleDeclKind::Implementation
-          ? CMK != LangOptions::CMK_None
-          : CMK != LangOptions::CMK_ModuleInterface) {
+  // A module implementation unit requires that we are not compiling a module
+  // of any kind. A module interface unit requires that we are not compiling a
+  // module map.
+  switch (getLangOpts().getCompilingModule()) {
+  case LangOptions::CMK_None:
+    // It's OK to compile a module interface as a normal translation unit.
+    break;
+
+  case LangOptions::CMK_ModuleInterface:
+    if (MDK != ModuleDeclKind::Implementation)
+      break;
+
+    // We were asked to compile a module interface unit but this is a module
+    // implementation unit. That indicates the 'export' is missing.
     Diag(ModuleLoc, diag::err_module_interface_implementation_mismatch)
-      << (unsigned)MDK;
+      << FixItHint::CreateInsertion(ModuleLoc, "export ");
+    break;
+
+  case LangOptions::CMK_ModuleMap:
+    Diag(ModuleLoc, diag::err_module_decl_in_module_map_module);
     return nullptr;
   }
 
   // FIXME: Create a ModuleDecl and return it.
 
   // FIXME: Most of this work should be done by the preprocessor rather than
-  // here, in case we look ahead across something where the current
-  // module matters (eg a #include).
+  // here, in order to support macro import.
 
-  // The dots in a module name in the Modules TS are a lie. Unlike Clang's
-  // hierarchical module map modules, the dots here are just another character
-  // that can appear in a module name. Flatten down to the actual module name.
+  // Flatten the dots in a module name. Unlike Clang's hierarchical module map
+  // modules, the dots here are just another character that can appear in a
+  // module name.
   std::string ModuleName;
   for (auto &Piece : Path) {
     if (!ModuleName.empty())
@@ -15735,8 +15747,8 @@ Sema::DeclGroupPtrTy Sema::ActOnModuleDecl(SourceLocation ModuleLoc,
   case ModuleDeclKind::Module: {
     // FIXME: Check we're not in a submodule.
 
-    // We can't have imported a definition of this module or parsed a module
-    // map defining it already.
+    // We can't have parsed or imported a definition of this module or parsed a
+    // module map defining it already.
     if (auto *M = Map.findModule(ModuleName)) {
       Diag(Path[0].second, diag::err_module_redefinition) << ModuleName;
       if (M->DefinitionLoc.isValid())
@@ -15910,6 +15922,12 @@ Decl *Sema::ActOnStartExportDecl(Scope *S, SourceLocation ExportLoc,
   ExportDecl *D = ExportDecl::Create(Context, CurContext, ExportLoc);
 
   // C++ Modules TS draft:
+  //   An export-declaration shall appear in the purview of a module other than
+  //   the global module.
+  if (ModuleScopes.empty() || !ModuleScopes.back().Module ||
+      ModuleScopes.back().Module->Kind != Module::ModuleInterfaceUnit)
+    Diag(ExportLoc, diag::err_export_not_in_module_interface);
+
   //   An export-declaration [...] shall not contain more than one
   //   export keyword.
   //
diff --git a/lib/Sema/SemaDeclAttr.cpp b/lib/Sema/SemaDeclAttr.cpp
index 027b3fe0e782..17c6975ca5e9 100644
--- a/lib/Sema/SemaDeclAttr.cpp
+++ b/lib/Sema/SemaDeclAttr.cpp
@@ -7220,6 +7220,8 @@ void Sema::DiagnoseUnguardedAvailabilityViolations(Decl *D) {
     Body = FD->getBody();
   } else if (auto *MD = dyn_cast<ObjCMethodDecl>(D))
     Body = MD->getBody();
+  else if (auto *BD = dyn_cast<BlockDecl>(D))
+    Body = BD->getBody();
 
   assert(Body && "Need a body here!");
 
diff --git a/lib/Sema/SemaExpr.cpp b/lib/Sema/SemaExpr.cpp
index 5a56f7093777..00480f821fc6 100644
--- a/lib/Sema/SemaExpr.cpp
+++ b/lib/Sema/SemaExpr.cpp
@@ -171,9 +171,14 @@ DiagnoseAvailabilityOfDecl(Sema &S, NamedDecl *D, SourceLocation Loc,
   if (AvailabilityResult Result =
           S.ShouldDiagnoseAvailabilityOfDecl(D, &Message)) {
 
-    if (Result == AR_NotYetIntroduced && S.getCurFunctionOrMethodDecl()) {
-      S.getEnclosingFunction()->HasPotentialAvailabilityViolations = true;
-      return;
+    if (Result == AR_NotYetIntroduced) {
+      if (S.getCurFunctionOrMethodDecl()) {
+        S.getEnclosingFunction()->HasPotentialAvailabilityViolations = true;
+        return;
+      } else if (S.getCurBlock() || S.getCurLambda()) {
+        S.getCurFunction()->HasPotentialAvailabilityViolations = true;
+        return;
+      }
     }
 
     const ObjCPropertyDecl *ObjCPDecl = nullptr;
@@ -12498,6 +12503,9 @@ ExprResult Sema::ActOnBlockStmtExpr(SourceLocation CaretLoc,
 
   BSI->TheDecl->setBody(cast<CompoundStmt>(Body));
 
+  if (Body && getCurFunction()->HasPotentialAvailabilityViolations)
+    DiagnoseUnguardedAvailabilityViolations(BSI->TheDecl);
+
   // Try to apply the named return value optimization. We have to check again
   // if we can do this, though, because blocks keep return statements around
   // to deduce an implicit return type.
diff --git a/lib/Sema/SemaLookup.cpp b/lib/Sema/SemaLookup.cpp
index e4d420f36832..ce76e14982db 100644
--- a/lib/Sema/SemaLookup.cpp
+++ b/lib/Sema/SemaLookup.cpp
@@ -1326,12 +1326,6 @@ bool Sema::CppLookupName(LookupResult &R, Scope *S) {
   return !R.empty();
 }
 
-/// \brief Find the declaration that a class temploid member specialization was
-/// instantiated from, or the member itself if it is an explicit specialization.
-static Decl *getInstantiatedFrom(Decl *D, MemberSpecializationInfo *MSInfo) {
-  return MSInfo->isExplicitSpecialization() ? D : MSInfo->getInstantiatedFrom();
-}
-
 Module *Sema::getOwningModule(Decl *Entity) {
   // If it's imported, grab its owning module.
   Module *M = Entity->getImportedOwningModule();
@@ -1413,12 +1407,11 @@ static Module *getDefiningModule(Sema &S, Decl *Entity) {
     if (CXXRecordDecl *Pattern = RD->getTemplateInstantiationPattern())
       Entity = Pattern;
   } else if (EnumDecl *ED = dyn_cast<EnumDecl>(Entity)) {
-    if (MemberSpecializationInfo *MSInfo = ED->getMemberSpecializationInfo())
-      Entity = getInstantiatedFrom(ED, MSInfo);
+    if (auto *Pattern = ED->getTemplateInstantiationPattern())
+      Entity = Pattern;
   } else if (VarDecl *VD = dyn_cast<VarDecl>(Entity)) {
-    // FIXME: Map from variable template specializations back to the template.
-    if (MemberSpecializationInfo *MSInfo = VD->getMemberSpecializationInfo())
-      Entity = getInstantiatedFrom(VD, MSInfo);
+    if (VarDecl *Pattern = VD->getTemplateInstantiationPattern())
+      Entity = Pattern;
   }
 
   // Walk up to the containing context. That might also have been instantiated
diff --git a/lib/Sema/SemaOpenMP.cpp b/lib/Sema/SemaOpenMP.cpp
index d7a6d2e5885e..fb13669407fc 100644
--- a/lib/Sema/SemaOpenMP.cpp
+++ b/lib/Sema/SemaOpenMP.cpp
@@ -118,7 +118,9 @@ private:
   typedef SmallVector<SharingMapTy, 4> StackTy;
 
   /// \brief Stack of used declaration and their data-sharing attributes.
-  StackTy Stack;
+  DeclSAMapTy Threadprivates;
+  const FunctionScopeInfo *CurrentNonCapturingFunctionScope = nullptr;
+  SmallVector<std::pair<StackTy, const FunctionScopeInfo *>, 4> Stack;
   /// \brief true, if check for DSA must be from parent directive, false, if
   /// from current directive.
   OpenMPClauseKind ClauseKindMode = OMPC_unknown;
@@ -133,8 +135,14 @@ private:
   /// \brief Checks if the variable is a local for OpenMP region.
   bool isOpenMPLocal(VarDecl *D, StackTy::reverse_iterator Iter);
 
+  bool isStackEmpty() const {
+    return Stack.empty() ||
+           Stack.back().second != CurrentNonCapturingFunctionScope ||
+           Stack.back().first.empty();
+  }
+
 public:
-  explicit DSAStackTy(Sema &S) : Stack(1), SemaRef(S) {}
+  explicit DSAStackTy(Sema &S) : SemaRef(S) {}
 
   bool isClauseParsingMode() const { return ClauseKindMode != OMPC_unknown; }
   void setClauseParsingMode(OpenMPClauseKind K) { ClauseKindMode = K; }
@@ -144,13 +152,38 @@ public:
 
   void push(OpenMPDirectiveKind DKind, const DeclarationNameInfo &DirName,
             Scope *CurScope, SourceLocation Loc) {
-    Stack.push_back(SharingMapTy(DKind, DirName, CurScope, Loc));
-    Stack.back().DefaultAttrLoc = Loc;
+    if (Stack.empty() ||
+        Stack.back().second != CurrentNonCapturingFunctionScope)
+      Stack.emplace_back(StackTy(), CurrentNonCapturingFunctionScope);
+    Stack.back().first.emplace_back(DKind, DirName, CurScope, Loc);
+    Stack.back().first.back().DefaultAttrLoc = Loc;
   }
 
   void pop() {
-    assert(Stack.size() > 1 && "Data-sharing attributes stack is empty!");
-    Stack.pop_back();
+    assert(!Stack.back().first.empty() &&
+           "Data-sharing attributes stack is empty!");
+    Stack.back().first.pop_back();
+  }
+
+  /// Start new OpenMP region stack in new non-capturing function.
+  void pushFunction() {
+    const FunctionScopeInfo *CurFnScope = SemaRef.getCurFunction();
+    assert(!isa<CapturingScopeInfo>(CurFnScope));
+    CurrentNonCapturingFunctionScope = CurFnScope;
+  }
+  /// Pop region stack for non-capturing function.
+  void popFunction(const FunctionScopeInfo *OldFSI) {
+    if (!Stack.empty() && Stack.back().second == OldFSI) {
+      assert(Stack.back().first.empty());
+      Stack.pop_back();
+    }
+    CurrentNonCapturingFunctionScope = nullptr;
+    for (const FunctionScopeInfo *FSI : llvm::reverse(SemaRef.FunctionScopes)) {
+      if (!isa<CapturingScopeInfo>(FSI)) {
+        CurrentNonCapturingFunctionScope = FSI;
+        break;
+      }
+    }
   }
 
   void addCriticalWithHint(OMPCriticalDirective *D, llvm::APSInt Hint) {
@@ -229,31 +262,35 @@ public:
 
   /// \brief Returns currently analyzed directive.
   OpenMPDirectiveKind getCurrentDirective() const {
-    return Stack.back().Directive;
+    return isStackEmpty() ? OMPD_unknown : Stack.back().first.back().Directive;
   }
   /// \brief Returns parent directive.
   OpenMPDirectiveKind getParentDirective() const {
-    if (Stack.size() > 2)
-      return Stack[Stack.size() - 2].Directive;
-    return OMPD_unknown;
+    if (isStackEmpty() || Stack.back().first.size() == 1)
+      return OMPD_unknown;
+    return std::next(Stack.back().first.rbegin())->Directive;
   }
 
   /// \brief Set default data sharing attribute to none.
   void setDefaultDSANone(SourceLocation Loc) {
-    Stack.back().DefaultAttr = DSA_none;
-    Stack.back().DefaultAttrLoc = Loc;
+    assert(!isStackEmpty());
+    Stack.back().first.back().DefaultAttr = DSA_none;
+    Stack.back().first.back().DefaultAttrLoc = Loc;
   }
   /// \brief Set default data sharing attribute to shared.
   void setDefaultDSAShared(SourceLocation Loc) {
-    Stack.back().DefaultAttr = DSA_shared;
-    Stack.back().DefaultAttrLoc = Loc;
+    assert(!isStackEmpty());
+    Stack.back().first.back().DefaultAttr = DSA_shared;
+    Stack.back().first.back().DefaultAttrLoc = Loc;
   }
 
   DefaultDataSharingAttributes getDefaultDSA() const {
-    return Stack.back().DefaultAttr;
+    return isStackEmpty() ? DSA_unspecified
+                          : Stack.back().first.back().DefaultAttr;
   }
   SourceLocation getDefaultDSALocation() const {
-    return Stack.back().DefaultAttrLoc;
+    return isStackEmpty() ? SourceLocation()
+                          : Stack.back().first.back().DefaultAttrLoc;
   }
 
   /// \brief Checks if the specified variable is a threadprivate.
@@ -264,52 +301,64 @@ public:
 
   /// \brief Marks current region as ordered (it has an 'ordered' clause).
   void setOrderedRegion(bool IsOrdered, Expr *Param) {
-    Stack.back().OrderedRegion.setInt(IsOrdered);
-    Stack.back().OrderedRegion.setPointer(Param);
+    assert(!isStackEmpty());
+    Stack.back().first.back().OrderedRegion.setInt(IsOrdered);
+    Stack.back().first.back().OrderedRegion.setPointer(Param);
   }
   /// \brief Returns true, if parent region is ordered (has associated
   /// 'ordered' clause), false - otherwise.
   bool isParentOrderedRegion() const {
-    if (Stack.size() > 2)
-      return Stack[Stack.size() - 2].OrderedRegion.getInt();
-    return false;
+    if (isStackEmpty() || Stack.back().first.size() == 1)
+      return false;
+    return std::next(Stack.back().first.rbegin())->OrderedRegion.getInt();
   }
   /// \brief Returns optional parameter for the ordered region.
   Expr *getParentOrderedRegionParam() const {
-    if (Stack.size() > 2)
-      return Stack[Stack.size() - 2].OrderedRegion.getPointer();
-    return nullptr;
+    if (isStackEmpty() || Stack.back().first.size() == 1)
+      return nullptr;
+    return std::next(Stack.back().first.rbegin())->OrderedRegion.getPointer();
   }
   /// \brief Marks current region as nowait (it has a 'nowait' clause).
   void setNowaitRegion(bool IsNowait = true) {
-    Stack.back().NowaitRegion = IsNowait;
+    assert(!isStackEmpty());
+    Stack.back().first.back().NowaitRegion = IsNowait;
   }
   /// \brief Returns true, if parent region is nowait (has associated
   /// 'nowait' clause), false - otherwise.
   bool isParentNowaitRegion() const {
-    if (Stack.size() > 2)
-      return Stack[Stack.size() - 2].NowaitRegion;
-    return false;
+    if (isStackEmpty() || Stack.back().first.size() == 1)
+      return false;
+    return std::next(Stack.back().first.rbegin())->NowaitRegion;
   }
   /// \brief Marks parent region as cancel region.
   void setParentCancelRegion(bool Cancel = true) {
-    if (Stack.size() > 2)
-      Stack[Stack.size() - 2].CancelRegion =
-          Stack[Stack.size() - 2].CancelRegion || Cancel;
+    if (!isStackEmpty() && Stack.back().first.size() > 1) {
+      auto &StackElemRef = *std::next(Stack.back().first.rbegin());
+      StackElemRef.CancelRegion |= StackElemRef.CancelRegion || Cancel;
+    }
   }
   /// \brief Return true if current region has inner cancel construct.
-  bool isCancelRegion() const { return Stack.back().CancelRegion; }
+  bool isCancelRegion() const {
+    return isStackEmpty() ? false : Stack.back().first.back().CancelRegion;
+  }
 
   /// \brief Set collapse value for the region.
-  void setAssociatedLoops(unsigned Val) { Stack.back().AssociatedLoops = Val; }
+  void setAssociatedLoops(unsigned Val) {
+    assert(!isStackEmpty());
+    Stack.back().first.back().AssociatedLoops = Val;
+  }
   /// \brief Return collapse value for region.
-  unsigned getAssociatedLoops() const { return Stack.back().AssociatedLoops; }
+  unsigned getAssociatedLoops() const {
+    return isStackEmpty() ? 0 : Stack.back().first.back().AssociatedLoops;
+  }
 
   /// \brief Marks current target region as one with closely nested teams
   /// region.
   void setParentTeamsRegionLoc(SourceLocation TeamsRegionLoc) {
-    if (Stack.size() > 2)
-      Stack[Stack.size() - 2].InnerTeamsRegionLoc = TeamsRegionLoc;
+    if (!isStackEmpty() && Stack.back().first.size() > 1) {
+      std::next(Stack.back().first.rbegin())->InnerTeamsRegionLoc =
+          TeamsRegionLoc;
+    }
   }
   /// \brief Returns true, if current region has closely nested teams region.
   bool hasInnerTeamsRegion() const {
@@ -317,14 +366,20 @@ public:
   }
   /// \brief Returns location of the nested teams region (if any).
   SourceLocation getInnerTeamsRegionLoc() const {
-    if (Stack.size() > 1)
-      return Stack.back().InnerTeamsRegionLoc;
-    return SourceLocation();
+    return isStackEmpty() ? SourceLocation()
+                          : Stack.back().first.back().InnerTeamsRegionLoc;
   }
 
-  Scope *getCurScope() const { return Stack.back().CurScope; }
-  Scope *getCurScope() { return Stack.back().CurScope; }
-  SourceLocation getConstructLoc() { return Stack.back().ConstructLoc; }
+  Scope *getCurScope() const {
+    return isStackEmpty() ? nullptr : Stack.back().first.back().CurScope;
+  }
+  Scope *getCurScope() {
+    return isStackEmpty() ? nullptr : Stack.back().first.back().CurScope;
+  }
+  SourceLocation getConstructLoc() {
+    return isStackEmpty() ? SourceLocation()
+                          : Stack.back().first.back().ConstructLoc;
+  }
 
   /// Do the check specified in \a Check to all component lists and return true
   /// if any issue is found.
@@ -333,8 +388,10 @@ public:
       const llvm::function_ref<
           bool(OMPClauseMappableExprCommon::MappableExprComponentListRef,
                OpenMPClauseKind)> &Check) {
-    auto SI = Stack.rbegin();
-    auto SE = Stack.rend();
+    if (isStackEmpty())
+      return false;
+    auto SI = Stack.back().first.rbegin();
+    auto SE = Stack.back().first.rend();
 
     if (SI == SE)
       return false;
@@ -361,9 +418,9 @@ public:
       ValueDecl *VD,
       OMPClauseMappableExprCommon::MappableExprComponentListRef Components,
       OpenMPClauseKind WhereFoundClauseKind) {
-    assert(Stack.size() > 1 &&
+    assert(!isStackEmpty() &&
            "Not expecting to retrieve components from a empty stack!");
-    auto &MEC = Stack.back().MappedExprComponents[VD];
+    auto &MEC = Stack.back().first.back().MappedExprComponents[VD];
     // Create new entry and append the new components there.
     MEC.Components.resize(MEC.Components.size() + 1);
     MEC.Components.back().append(Components.begin(), Components.end());
@@ -371,23 +428,25 @@ public:
   }
 
   unsigned getNestingLevel() const {
-    assert(Stack.size() > 1);
-    return Stack.size() - 2;
+    assert(!isStackEmpty());
+    return Stack.back().first.size() - 1;
   }
   void addDoacrossDependClause(OMPDependClause *C, OperatorOffsetTy &OpsOffs) {
-    assert(Stack.size() > 2);
-    assert(isOpenMPWorksharingDirective(Stack[Stack.size() - 2].Directive));
-    Stack[Stack.size() - 2].DoacrossDepends.insert({C, OpsOffs});
+    assert(!isStackEmpty() && Stack.back().first.size() > 1);
+    auto &StackElem = *std::next(Stack.back().first.rbegin());
+    assert(isOpenMPWorksharingDirective(StackElem.Directive));
+    StackElem.DoacrossDepends.insert({C, OpsOffs});
   }
   llvm::iterator_range<DoacrossDependMapTy::const_iterator>
   getDoacrossDependClauses() const {
-    assert(Stack.size() > 1);
-    if (isOpenMPWorksharingDirective(Stack[Stack.size() - 1].Directive)) {
-      auto &Ref = Stack[Stack.size() - 1].DoacrossDepends;
+    assert(!isStackEmpty());
+    auto &StackElem = Stack.back().first.back();
+    if (isOpenMPWorksharingDirective(StackElem.Directive)) {
+      auto &Ref = StackElem.DoacrossDepends;
       return llvm::make_range(Ref.begin(), Ref.end());
     }
-    return llvm::make_range(Stack[0].DoacrossDepends.end(),
-                            Stack[0].DoacrossDepends.end());
+    return llvm::make_range(StackElem.DoacrossDepends.end(),
+                            StackElem.DoacrossDepends.end());
   }
 };
 bool isParallelOrTaskRegion(OpenMPDirectiveKind DKind) {
@@ -416,7 +475,7 @@ DSAStackTy::DSAVarData DSAStackTy::getDSA(StackTy::reverse_iterator &Iter,
   auto *VD = dyn_cast<VarDecl>(D);
   auto *FD = dyn_cast<FieldDecl>(D);
   DSAVarData DVar;
-  if (Iter == std::prev(Stack.rend())) {
+  if (isStackEmpty() || Iter == Stack.back().first.rend()) {
     // OpenMP [2.9.1.1, Data-sharing Attribute Rules for Variables Referenced
     // in a region but not in construct]
     //  File-scope or namespace-scope variables referenced in called routines
@@ -490,8 +549,9 @@ DSAStackTy::DSAVarData DSAStackTy::getDSA(StackTy::reverse_iterator &Iter,
     //  bound to the current team is shared.
     if (isOpenMPTaskingDirective(DVar.DKind)) {
       DSAVarData DVarTemp;
-      for (StackTy::reverse_iterator I = std::next(Iter), EE = Stack.rend();
-           I != EE; ++I) {
+      auto I = Iter, E = Stack.back().first.rend();
+      do {
+        ++I;
         // OpenMP [2.9.1.1, Data-sharing Attribute Rules for Variables
         // Referenced in a Construct, implicitly determined, p.6]
         //  In a task construct, if no default clause is present, a variable
@@ -503,9 +563,7 @@ DSAStackTy::DSAVarData DSAStackTy::getDSA(StackTy::reverse_iterator &Iter,
           DVar.CKind = OMPC_firstprivate;
           return DVar;
         }
-        if (isParallelOrTaskRegion(I->Directive))
-          break;
-      }
+      } while (I != E && !isParallelOrTaskRegion(I->Directive));
       DVar.CKind =
           (DVarTemp.CKind == OMPC_unknown) ? OMPC_firstprivate : OMPC_shared;
       return DVar;
@@ -520,12 +578,13 @@ DSAStackTy::DSAVarData DSAStackTy::getDSA(StackTy::reverse_iterator &Iter,
 }
 
 Expr *DSAStackTy::addUniqueAligned(ValueDecl *D, Expr *NewDE) {
-  assert(Stack.size() > 1 && "Data sharing attributes stack is empty");
+  assert(!isStackEmpty() && "Data sharing attributes stack is empty");
   D = getCanonicalDecl(D);
-  auto It = Stack.back().AlignedMap.find(D);
-  if (It == Stack.back().AlignedMap.end()) {
+  auto &StackElem = Stack.back().first.back();
+  auto It = StackElem.AlignedMap.find(D);
+  if (It == StackElem.AlignedMap.end()) {
     assert(NewDE && "Unexpected nullptr expr to be added into aligned map");
-    Stack.back().AlignedMap[D] = NewDE;
+    StackElem.AlignedMap[D] = NewDE;
     return nullptr;
   } else {
     assert(It->second && "Unexpected nullptr expr in the aligned map");
@@ -535,35 +594,43 @@ Expr *DSAStackTy::addUniqueAligned(ValueDecl *D, Expr *NewDE) {
 }
 
 void DSAStackTy::addLoopControlVariable(ValueDecl *D, VarDecl *Capture) {
-  assert(Stack.size() > 1 && "Data-sharing attributes stack is empty");
+  assert(!isStackEmpty() && "Data-sharing attributes stack is empty");
   D = getCanonicalDecl(D);
-  Stack.back().LCVMap.insert(
-      std::make_pair(D, LCDeclInfo(Stack.back().LCVMap.size() + 1, Capture)));
+  auto &StackElem = Stack.back().first.back();
+  StackElem.LCVMap.insert(
+      {D, LCDeclInfo(StackElem.LCVMap.size() + 1, Capture)});
 }
 
 DSAStackTy::LCDeclInfo DSAStackTy::isLoopControlVariable(ValueDecl *D) {
-  assert(Stack.size() > 1 && "Data-sharing attributes stack is empty");
+  assert(!isStackEmpty() && "Data-sharing attributes stack is empty");
   D = getCanonicalDecl(D);
-  return Stack.back().LCVMap.count(D) > 0 ? Stack.back().LCVMap[D]
-                                          : LCDeclInfo(0, nullptr);
+  auto &StackElem = Stack.back().first.back();
+  auto It = StackElem.LCVMap.find(D);
+  if (It != StackElem.LCVMap.end())
+    return It->second;
+  return {0, nullptr};
 }
 
 DSAStackTy::LCDeclInfo DSAStackTy::isParentLoopControlVariable(ValueDecl *D) {
-  assert(Stack.size() > 2 && "Data-sharing attributes stack is empty");
+  assert(!isStackEmpty() && Stack.back().first.size() > 1 &&
+         "Data-sharing attributes stack is empty");
   D = getCanonicalDecl(D);
-  return Stack[Stack.size() - 2].LCVMap.count(D) > 0
-             ? Stack[Stack.size() - 2].LCVMap[D]
-             : LCDeclInfo(0, nullptr);
+  auto &StackElem = *std::next(Stack.back().first.rbegin());
+  auto It = StackElem.LCVMap.find(D);
+  if (It != StackElem.LCVMap.end())
+    return It->second;
+  return {0, nullptr};
 }
 
 ValueDecl *DSAStackTy::getParentLoopControlVariable(unsigned I) {
-  assert(Stack.size() > 2 && "Data-sharing attributes stack is empty");
-  if (Stack[Stack.size() - 2].LCVMap.size() < I)
+  assert(!isStackEmpty() && Stack.back().first.size() > 1 &&
+         "Data-sharing attributes stack is empty");
+  auto &StackElem = *std::next(Stack.back().first.rbegin());
+  if (StackElem.LCVMap.size() < I)
     return nullptr;
-  for (auto &Pair : Stack[Stack.size() - 2].LCVMap) {
+  for (auto &Pair : StackElem.LCVMap)
     if (Pair.second.first == I)
       return Pair.first;
-  }
   return nullptr;
 }
 
@@ -571,13 +638,13 @@ void DSAStackTy::addDSA(ValueDecl *D, Expr *E, OpenMPClauseKind A,
                         DeclRefExpr *PrivateCopy) {
   D = getCanonicalDecl(D);
   if (A == OMPC_threadprivate) {
-    auto &Data = Stack[0].SharingMap[D];
+    auto &Data = Threadprivates[D];
     Data.Attributes = A;
     Data.RefExpr.setPointer(E);
     Data.PrivateCopy = nullptr;
   } else {
-    assert(Stack.size() > 1 && "Data-sharing attributes stack is empty");
-    auto &Data = Stack.back().SharingMap[D];
+    assert(!isStackEmpty() && "Data-sharing attributes stack is empty");
+    auto &Data = Stack.back().first.back().SharingMap[D];
     assert(Data.Attributes == OMPC_unknown || (A == Data.Attributes) ||
            (A == OMPC_firstprivate && Data.Attributes == OMPC_lastprivate) ||
            (A == OMPC_lastprivate && Data.Attributes == OMPC_firstprivate) ||
@@ -592,7 +659,7 @@ void DSAStackTy::addDSA(ValueDecl *D, Expr *E, OpenMPClauseKind A,
     Data.RefExpr.setPointerAndInt(E, IsLastprivate);
     Data.PrivateCopy = PrivateCopy;
     if (PrivateCopy) {
-      auto &Data = Stack.back().SharingMap[PrivateCopy->getDecl()];
+      auto &Data = Stack.back().first.back().SharingMap[PrivateCopy->getDecl()];
       Data.Attributes = A;
       Data.RefExpr.setPointerAndInt(PrivateCopy, IsLastprivate);
       Data.PrivateCopy = nullptr;
@@ -602,19 +669,17 @@ void DSAStackTy::addDSA(ValueDecl *D, Expr *E, OpenMPClauseKind A,
 
 bool DSAStackTy::isOpenMPLocal(VarDecl *D, StackTy::reverse_iterator Iter) {
   D = D->getCanonicalDecl();
-  if (Stack.size() > 2) {
-    reverse_iterator I = Iter, E = std::prev(Stack.rend());
+  if (!isStackEmpty() && Stack.back().first.size() > 1) {
+    reverse_iterator I = Iter, E = Stack.back().first.rend();
     Scope *TopScope = nullptr;
-    while (I != E && !isParallelOrTaskRegion(I->Directive)) {
+    while (I != E && !isParallelOrTaskRegion(I->Directive))
       ++I;
-    }
     if (I == E)
       return false;
     TopScope = I->CurScope ? I->CurScope->getParent() : nullptr;
     Scope *CurScope = getCurScope();
-    while (CurScope != TopScope && !CurScope->isDeclScope(D)) {
+    while (CurScope != TopScope && !CurScope->isDeclScope(D))
       CurScope = CurScope->getParent();
-    }
     return CurScope != TopScope;
   }
   return false;
@@ -665,16 +730,16 @@ DSAStackTy::DSAVarData DSAStackTy::getTopDSA(ValueDecl *D, bool FromParent) {
                                D->getLocation()),
            OMPC_threadprivate);
   }
-  if (Stack[0].SharingMap.count(D)) {
-    DVar.RefExpr = Stack[0].SharingMap[D].RefExpr.getPointer();
+  auto TI = Threadprivates.find(D);
+  if (TI != Threadprivates.end()) {
+    DVar.RefExpr = TI->getSecond().RefExpr.getPointer();
     DVar.CKind = OMPC_threadprivate;
     return DVar;
   }
 
-  if (Stack.size() == 1) {
+  if (isStackEmpty())
     // Not in OpenMP execution region and top scope was already checked.
     return DVar;
-  }
 
   // OpenMP [2.9.1.1, Data-sharing Attribute Rules for Variables Referenced
   // in a Construct, C/C++, predetermined, p.4]
@@ -722,11 +787,10 @@ DSAStackTy::DSAVarData DSAStackTy::getTopDSA(ValueDecl *D, bool FromParent) {
 
   // Explicitly specified attributes and local variables with predetermined
   // attributes.
-  auto StartI = std::next(Stack.rbegin());
-  auto EndI = std::prev(Stack.rend());
-  if (FromParent && StartI != EndI) {
+  auto StartI = std::next(Stack.back().first.rbegin());
+  auto EndI = Stack.back().first.rend();
+  if (FromParent && StartI != EndI)
     StartI = std::next(StartI);
-  }
   auto I = std::prev(StartI);
   if (I->SharingMap.count(D)) {
     DVar.RefExpr = I->SharingMap[D].RefExpr.getPointer();
@@ -740,12 +804,15 @@ DSAStackTy::DSAVarData DSAStackTy::getTopDSA(ValueDecl *D, bool FromParent) {
 
 DSAStackTy::DSAVarData DSAStackTy::getImplicitDSA(ValueDecl *D,
                                                   bool FromParent) {
+  if (isStackEmpty()) {
+    StackTy::reverse_iterator I;
+    return getDSA(I, D);
+  }
   D = getCanonicalDecl(D);
-  auto StartI = Stack.rbegin();
-  auto EndI = std::prev(Stack.rend());
-  if (FromParent && StartI != EndI) {
+  auto StartI = Stack.back().first.rbegin();
+  auto EndI = Stack.back().first.rend();
+  if (FromParent && StartI != EndI)
     StartI = std::next(StartI);
-  }
   return getDSA(StartI, D);
 }
 
@@ -754,33 +821,40 @@ DSAStackTy::hasDSA(ValueDecl *D,
                    const llvm::function_ref<bool(OpenMPClauseKind)> &CPred,
                    const llvm::function_ref<bool(OpenMPDirectiveKind)> &DPred,
                    bool FromParent) {
+  if (isStackEmpty())
+    return {};
   D = getCanonicalDecl(D);
-  auto StartI = std::next(Stack.rbegin());
-  auto EndI = Stack.rend();
-  if (FromParent && StartI != EndI) {
+  auto StartI = std::next(Stack.back().first.rbegin());
+  auto EndI = Stack.back().first.rend();
+  if (FromParent && StartI != EndI)
     StartI = std::next(StartI);
-  }
-  for (auto I = StartI, EE = EndI; I != EE; ++I) {
+  if (StartI == EndI)
+    return {};
+  auto I = std::prev(StartI);
+  do {
+    ++I;
     if (!DPred(I->Directive) && !isParallelOrTaskRegion(I->Directive))
       continue;
     DSAVarData DVar = getDSA(I, D);
     if (CPred(DVar.CKind))
       return DVar;
-  }
-  return DSAVarData();
+  } while (I != EndI);
+  return {};
 }
 
 DSAStackTy::DSAVarData DSAStackTy::hasInnermostDSA(
     ValueDecl *D, const llvm::function_ref<bool(OpenMPClauseKind)> &CPred,
     const llvm::function_ref<bool(OpenMPDirectiveKind)> &DPred,
     bool FromParent) {
+  if (isStackEmpty())
+    return {};
   D = getCanonicalDecl(D);
-  auto StartI = std::next(Stack.rbegin());
-  auto EndI = Stack.rend();
+  auto StartI = std::next(Stack.back().first.rbegin());
+  auto EndI = Stack.back().first.rend();
   if (FromParent && StartI != EndI)
     StartI = std::next(StartI);
   if (StartI == EndI || !DPred(StartI->Directive))
-    return DSAVarData();
+    return {};
   DSAVarData DVar = getDSA(StartI, D);
   return CPred(DVar.CKind) ? DVar : DSAVarData();
 }
@@ -790,9 +864,11 @@ bool DSAStackTy::hasExplicitDSA(
     unsigned Level, bool NotLastprivate) {
   if (CPred(ClauseKindMode))
     return true;
+  if (isStackEmpty())
+    return false;
   D = getCanonicalDecl(D);
-  auto StartI = std::next(Stack.begin());
-  auto EndI = Stack.end();
+  auto StartI = Stack.back().first.begin();
+  auto EndI = Stack.back().first.end();
   if (std::distance(StartI, EndI) <= (int)Level)
     return false;
   std::advance(StartI, Level);
@@ -805,8 +881,10 @@ bool DSAStackTy::hasExplicitDSA(
 bool DSAStackTy::hasExplicitDirective(
     const llvm::function_ref<bool(OpenMPDirectiveKind)> &DPred,
     unsigned Level) {
-  auto StartI = std::next(Stack.begin());
-  auto EndI = Stack.end();
+  if (isStackEmpty())
+    return false;
+  auto StartI = Stack.back().first.begin();
+  auto EndI = Stack.back().first.end();
   if (std::distance(StartI, EndI) <= (int)Level)
     return false;
   std::advance(StartI, Level);
@@ -819,13 +897,12 @@ bool DSAStackTy::hasDirective(
         &DPred,
     bool FromParent) {
   // We look only in the enclosing region.
-  if (Stack.size() < 2)
+  if (isStackEmpty())
     return false;
-  auto StartI = std::next(Stack.rbegin());
-  auto EndI = std::prev(Stack.rend());
-  if (FromParent && StartI != EndI) {
+  auto StartI = std::next(Stack.back().first.rbegin());
+  auto EndI = Stack.back().first.rend();
+  if (FromParent && StartI != EndI)
     StartI = std::next(StartI);
-  }
   for (auto I = StartI, EE = EndI; I != EE; ++I) {
     if (DPred(I->Directive, I->DirectiveName, I->ConstructLoc))
       return true;
@@ -839,6 +916,14 @@ void Sema::InitDataSharingAttributesStack() {
 
 #define DSAStack static_cast<DSAStackTy *>(VarDataSharingAttributesStack)
 
+void Sema::pushOpenMPFunctionRegion() {
+  DSAStack->pushFunction();
+}
+
+void Sema::popOpenMPFunctionRegion(const FunctionScopeInfo *OldFSI) {
+  DSAStack->popFunction(OldFSI);
+}
+
 bool Sema::IsOpenMPCapturedByRef(ValueDecl *D, unsigned Level) {
   assert(LangOpts.OpenMP && "OpenMP is not allowed");
 
diff --git a/lib/Sema/TreeTransform.h b/lib/Sema/TreeTransform.h
index 693b5c088acc..f4a97f55789e 100644
--- a/lib/Sema/TreeTransform.h
+++ b/lib/Sema/TreeTransform.h
@@ -2220,6 +2220,9 @@ public:
     Base = BaseResult.get();
     QualType BaseType = Base->getType();
 
+    if (isArrow && !BaseType->isPointerType())
+      return ExprError();
+
     // FIXME: this involves duplicating earlier analysis in a lot of
     // cases; we should avoid this when possible.
     LookupResult R(getSema(), MemberNameInfo, Sema::LookupMemberName);
diff --git a/lib/Serialization/ASTReader.cpp b/lib/Serialization/ASTReader.cpp
index 406c4b50d92b..7b1edc062d1e 100644
--- a/lib/Serialization/ASTReader.cpp
+++ b/lib/Serialization/ASTReader.cpp
@@ -4779,6 +4779,7 @@ ASTReader::ReadSubmoduleBlock(ModuleFile &F, unsigned ClientLoadCapabilities) {
   ModuleMap &ModMap = PP.getHeaderSearchInfo().getModuleMap();
   bool First = true;
   Module *CurrentModule = nullptr;
+  Module::ModuleKind ModuleKind = Module::ModuleMapModule;
   RecordData Record;
   while (true) {
     llvm::BitstreamEntry Entry = F.Stream.advanceSkippingSubblocks();
@@ -4871,6 +4872,7 @@ ASTReader::ReadSubmoduleBlock(ModuleFile &F, unsigned ClientLoadCapabilities) {
         CurrentModule->setASTFile(F.File);
       }
 
+      CurrentModule->Kind = ModuleKind;
       CurrentModule->Signature = F.Signature;
       CurrentModule->IsFromModuleFile = true;
       CurrentModule->IsSystem = IsSystem || CurrentModule->IsSystem;
@@ -4969,6 +4971,7 @@ ASTReader::ReadSubmoduleBlock(ModuleFile &F, unsigned ClientLoadCapabilities) {
 
         SubmodulesLoaded.resize(SubmodulesLoaded.size() + F.LocalNumSubmodules);
       }
+      ModuleKind = (Module::ModuleKind)Record[2];
       break;
     }
 
diff --git a/lib/Serialization/ASTWriter.cpp b/lib/Serialization/ASTWriter.cpp
index bb869077f4cd..84d2420f4ba9 100644
--- a/lib/Serialization/ASTWriter.cpp
+++ b/lib/Serialization/ASTWriter.cpp
@@ -2694,9 +2694,10 @@ void ASTWriter::WriteSubmodules(Module *WritingModule) {
   unsigned ConflictAbbrev = Stream.EmitAbbrev(std::move(Abbrev));
 
   // Write the submodule metadata block.
-  RecordData::value_type Record[] = {getNumberOfModules(WritingModule),
-                                     FirstSubmoduleID -
-                                         NUM_PREDEF_SUBMODULE_IDS};
+  RecordData::value_type Record[] = {
+      getNumberOfModules(WritingModule),
+      FirstSubmoduleID - NUM_PREDEF_SUBMODULE_IDS,
+      (unsigned)WritingModule->Kind};
   Stream.EmitRecord(SUBMODULE_METADATA, Record);
   
   // Write all of the submodules.
diff --git a/lib/StaticAnalyzer/Checkers/MallocChecker.cpp b/lib/StaticAnalyzer/Checkers/MallocChecker.cpp
index 6e9b7fefa3d0..5730517289bb 100644
--- a/lib/StaticAnalyzer/Checkers/MallocChecker.cpp
+++ b/lib/StaticAnalyzer/Checkers/MallocChecker.cpp
@@ -177,7 +177,10 @@ public:
         II_wcsdup(nullptr), II_win_wcsdup(nullptr), II_g_malloc(nullptr),
         II_g_malloc0(nullptr), II_g_realloc(nullptr), II_g_try_malloc(nullptr), 
         II_g_try_malloc0(nullptr), II_g_try_realloc(nullptr), 
-        II_g_free(nullptr), II_g_memdup(nullptr) {}
+        II_g_free(nullptr), II_g_memdup(nullptr), II_g_malloc_n(nullptr), 
+        II_g_malloc0_n(nullptr), II_g_realloc_n(nullptr), 
+        II_g_try_malloc_n(nullptr), II_g_try_malloc0_n(nullptr), 
+        II_g_try_realloc_n(nullptr) {}
 
   /// In pessimistic mode, the checker assumes that it does not know which
   /// functions might free the memory.
@@ -241,7 +244,10 @@ private:
                          *II_if_nameindex, *II_if_freenameindex, *II_wcsdup,
                          *II_win_wcsdup, *II_g_malloc, *II_g_malloc0, 
                          *II_g_realloc, *II_g_try_malloc, *II_g_try_malloc0, 
-                         *II_g_try_realloc, *II_g_free, *II_g_memdup;
+                         *II_g_try_realloc, *II_g_free, *II_g_memdup, 
+                         *II_g_malloc_n, *II_g_malloc0_n, *II_g_realloc_n, 
+                         *II_g_try_malloc_n, *II_g_try_malloc0_n, 
+                         *II_g_try_realloc_n;
   mutable Optional<uint64_t> KernelZeroFlagVal;
 
   void initIdentifierInfo(ASTContext &C) const;
@@ -321,9 +327,12 @@ private:
                              bool &ReleasedAllocated,
                              bool ReturnsNullOnFailure = false) const;
 
-  ProgramStateRef ReallocMem(CheckerContext &C, const CallExpr *CE,
-                             bool FreesMemOnFailure,
-                             ProgramStateRef State) const;
+  ProgramStateRef ReallocMemAux(CheckerContext &C, const CallExpr *CE,
+                                bool FreesMemOnFailure,
+                                ProgramStateRef State, 
+                                bool SuffixWithN = false) const;
+  static SVal evalMulForBufferSize(CheckerContext &C, const Expr *Blocks,
+                                   const Expr *BlockBytes);
   static ProgramStateRef CallocMem(CheckerContext &C, const CallExpr *CE,
                                    ProgramStateRef State);
 
@@ -569,6 +578,12 @@ void MallocChecker::initIdentifierInfo(ASTContext &Ctx) const {
   II_g_try_realloc = &Ctx.Idents.get("g_try_realloc");
   II_g_free = &Ctx.Idents.get("g_free");
   II_g_memdup = &Ctx.Idents.get("g_memdup");
+  II_g_malloc_n = &Ctx.Idents.get("g_malloc_n");
+  II_g_malloc0_n = &Ctx.Idents.get("g_malloc0_n");
+  II_g_realloc_n = &Ctx.Idents.get("g_realloc_n");
+  II_g_try_malloc_n = &Ctx.Idents.get("g_try_malloc_n");
+  II_g_try_malloc0_n = &Ctx.Idents.get("g_try_malloc0_n");
+  II_g_try_realloc_n = &Ctx.Idents.get("g_try_realloc_n");
 }
 
 bool MallocChecker::isMemFunction(const FunctionDecl *FD, ASTContext &C) const {
@@ -617,7 +632,10 @@ bool MallocChecker::isCMemFunction(const FunctionDecl *FD,
           FunI == II_g_malloc || FunI == II_g_malloc0 || 
           FunI == II_g_realloc || FunI == II_g_try_malloc || 
           FunI == II_g_try_malloc0 || FunI == II_g_try_realloc ||
-          FunI == II_g_memdup)
+          FunI == II_g_memdup || FunI == II_g_malloc_n || 
+          FunI == II_g_malloc0_n || FunI == II_g_realloc_n || 
+          FunI == II_g_try_malloc_n || FunI == II_g_try_malloc0_n || 
+          FunI == II_g_try_realloc_n)
         return true;
     }
 
@@ -767,6 +785,17 @@ llvm::Optional<ProgramStateRef> MallocChecker::performKernelMalloc(
   return None;
 }
 
+SVal MallocChecker::evalMulForBufferSize(CheckerContext &C, const Expr *Blocks,
+                                         const Expr *BlockBytes) {
+  SValBuilder &SB = C.getSValBuilder();
+  SVal BlocksVal = C.getSVal(Blocks);
+  SVal BlockBytesVal = C.getSVal(BlockBytes);
+  ProgramStateRef State = C.getState();
+  SVal TotalSize = SB.evalBinOp(State, BO_Mul, BlocksVal, BlockBytesVal,
+                                SB.getContext().getSizeType());
+  return TotalSize;
+}
+
 void MallocChecker::checkPostStmt(const CallExpr *CE, CheckerContext &C) const {
   if (C.wasInlined)
     return;
@@ -813,10 +842,10 @@ void MallocChecker::checkPostStmt(const CallExpr *CE, CheckerContext &C) const {
       State = ProcessZeroAllocation(C, CE, 0, State);
     } else if (FunI == II_realloc || FunI == II_g_realloc || 
                FunI == II_g_try_realloc) {
-      State = ReallocMem(C, CE, false, State);
+      State = ReallocMemAux(C, CE, false, State);
       State = ProcessZeroAllocation(C, CE, 1, State);
     } else if (FunI == II_reallocf) {
-      State = ReallocMem(C, CE, true, State);
+      State = ReallocMemAux(C, CE, true, State);
       State = ProcessZeroAllocation(C, CE, 1, State);
     } else if (FunI == II_calloc) {
       State = CallocMem(C, CE, State);
@@ -874,6 +903,25 @@ void MallocChecker::checkPostStmt(const CallExpr *CE, CheckerContext &C) const {
         return;
       State = MallocMemAux(C, CE, CE->getArg(1), UndefinedVal(), State);
       State = ProcessZeroAllocation(C, CE, 1, State);
+    } else if (FunI == II_g_malloc_n || FunI == II_g_try_malloc_n || 
+               FunI == II_g_malloc0_n || FunI == II_g_try_malloc0_n) {
+      if (CE->getNumArgs() < 2)
+        return;
+      SVal Init = UndefinedVal();
+      if (FunI == II_g_malloc0_n || FunI == II_g_try_malloc0_n) {
+        SValBuilder &SB = C.getSValBuilder();
+        Init = SB.makeZeroVal(SB.getContext().CharTy);
+      }
+      SVal TotalSize = evalMulForBufferSize(C, CE->getArg(0), CE->getArg(1));
+      State = MallocMemAux(C, CE, TotalSize, Init, State);
+      State = ProcessZeroAllocation(C, CE, 0, State);
+      State = ProcessZeroAllocation(C, CE, 1, State);
+    } else if (FunI == II_g_realloc_n || FunI == II_g_try_realloc_n) {
+      if (CE->getNumArgs() < 3)
+        return;
+      State = ReallocMemAux(C, CE, false, State, true);
+      State = ProcessZeroAllocation(C, CE, 1, State);
+      State = ProcessZeroAllocation(C, CE, 2, State);
     }
   }
 
@@ -1976,14 +2024,17 @@ void MallocChecker::ReportUseZeroAllocated(CheckerContext &C,
   }
 }
 
-ProgramStateRef MallocChecker::ReallocMem(CheckerContext &C,
-                                          const CallExpr *CE,
-                                          bool FreesOnFail,
-                                          ProgramStateRef State) const {
+ProgramStateRef MallocChecker::ReallocMemAux(CheckerContext &C,
+                                             const CallExpr *CE,
+                                             bool FreesOnFail,
+                                             ProgramStateRef State, 
+                                             bool SuffixWithN) const {
   if (!State)
     return nullptr;
 
-  if (CE->getNumArgs() < 2)
+  if (SuffixWithN && CE->getNumArgs() < 3)
+    return nullptr;
+  else if (CE->getNumArgs() < 2)
     return nullptr;
 
   const Expr *arg0Expr = CE->getArg(0);
@@ -1998,20 +2049,19 @@ ProgramStateRef MallocChecker::ReallocMem(CheckerContext &C,
   DefinedOrUnknownSVal PtrEQ =
     svalBuilder.evalEQ(State, arg0Val, svalBuilder.makeNull());
 
-  // Get the size argument. If there is no size arg then give up.
+  // Get the size argument.
   const Expr *Arg1 = CE->getArg(1);
-  if (!Arg1)
-    return nullptr;
 
   // Get the value of the size argument.
-  SVal Arg1ValG = State->getSVal(Arg1, LCtx);
-  if (!Arg1ValG.getAs<DefinedOrUnknownSVal>())
+  SVal TotalSize = State->getSVal(Arg1, LCtx);
+  if (SuffixWithN)
+    TotalSize = evalMulForBufferSize(C, Arg1, CE->getArg(2));
+  if (!TotalSize.getAs<DefinedOrUnknownSVal>())
     return nullptr;
-  DefinedOrUnknownSVal Arg1Val = Arg1ValG.castAs<DefinedOrUnknownSVal>();
 
   // Compare the size argument to 0.
   DefinedOrUnknownSVal SizeZero =
-    svalBuilder.evalEQ(State, Arg1Val,
+    svalBuilder.evalEQ(State, TotalSize.castAs<DefinedOrUnknownSVal>(),
                        svalBuilder.makeIntValWithPtrWidth(0, false));
 
   ProgramStateRef StatePtrIsNull, StatePtrNotNull;
@@ -2025,8 +2075,8 @@ ProgramStateRef MallocChecker::ReallocMem(CheckerContext &C,
 
   // If the ptr is NULL and the size is not 0, the call is equivalent to
   // malloc(size).
-  if ( PrtIsNull && !SizeIsZero) {
-    ProgramStateRef stateMalloc = MallocMemAux(C, CE, CE->getArg(1),
+  if (PrtIsNull && !SizeIsZero) {
+    ProgramStateRef stateMalloc = MallocMemAux(C, CE, TotalSize,
                                                UndefinedVal(), StatePtrIsNull);
     return stateMalloc;
   }
@@ -2059,7 +2109,7 @@ ProgramStateRef MallocChecker::ReallocMem(CheckerContext &C,
   if (ProgramStateRef stateFree =
         FreeMemAux(C, CE, State, 0, false, ReleasedAllocated)) {
 
-    ProgramStateRef stateRealloc = MallocMemAux(C, CE, CE->getArg(1),
+    ProgramStateRef stateRealloc = MallocMemAux(C, CE, TotalSize,
                                                 UnknownVal(), stateFree);
     if (!stateRealloc)
       return nullptr;
@@ -2090,12 +2140,8 @@ ProgramStateRef MallocChecker::CallocMem(CheckerContext &C, const CallExpr *CE,
     return nullptr;
 
   SValBuilder &svalBuilder = C.getSValBuilder();
-  const LocationContext *LCtx = C.getLocationContext();
-  SVal count = State->getSVal(CE->getArg(0), LCtx);
-  SVal elementSize = State->getSVal(CE->getArg(1), LCtx);
-  SVal TotalSize = svalBuilder.evalBinOp(State, BO_Mul, count, elementSize,
-                                        svalBuilder.getContext().getSizeType());
   SVal zeroVal = svalBuilder.makeZeroVal(svalBuilder.getContext().CharTy);
+  SVal TotalSize = evalMulForBufferSize(C, CE->getArg(0), CE->getArg(1));
 
   return MallocMemAux(C, CE, TotalSize, zeroVal, State);
 }
diff --git a/lib/StaticAnalyzer/Core/BugReporterVisitors.cpp b/lib/StaticAnalyzer/Core/BugReporterVisitors.cpp
index 7309741d22e3..d00182a871c1 100644
--- a/lib/StaticAnalyzer/Core/BugReporterVisitors.cpp
+++ b/lib/StaticAnalyzer/Core/BugReporterVisitors.cpp
@@ -61,7 +61,9 @@ const Expr *bugreporter::getDerefExpr(const Stmt *S) {
         return U->getSubExpr()->IgnoreParenCasts();
     }
     else if (const MemberExpr *ME = dyn_cast<MemberExpr>(E)) {
-      if (ME->isArrow() || isDeclRefExprToReference(ME->getBase())) {
+      if (ME->isImplicitAccess()) {
+        return ME;
+      } else if (ME->isArrow() || isDeclRefExprToReference(ME->getBase())) {
         return ME->getBase()->IgnoreParenCasts();
       } else {
         // If we have a member expr with a dot, the base must have been
@@ -73,9 +75,9 @@ const Expr *bugreporter::getDerefExpr(const Stmt *S) {
       return IvarRef->getBase()->IgnoreParenCasts();
     }
     else if (const ArraySubscriptExpr *AE = dyn_cast<ArraySubscriptExpr>(E)) {
-      return AE->getBase();
+      return getDerefExpr(AE->getBase());
     }
-    else if (isDeclRefExprToReference(E)) {
+    else if (isa<DeclRefExpr>(E)) {
       return E;
     }
     break;
@@ -961,7 +963,24 @@ bool bugreporter::trackNullOrUndefValue(const ExplodedNode *N,
   const Expr *Inner = nullptr;
   if (const Expr *Ex = dyn_cast<Expr>(S)) {
     Ex = Ex->IgnoreParenCasts();
-    if (ExplodedGraph::isInterestingLValueExpr(Ex) || CallEvent::isCallStmt(Ex))
+
+    // Performing operator `&' on an lvalue expression is essentially a no-op.
+    // Then, if we are taking addresses of fields or elements, these are also
+    // unlikely to matter.
+    // FIXME: There's a hack in our Store implementation that always computes
+    // field offsets around null pointers as if they are always equal to 0.
+    // The idea here is to report accesses to fields as null dereferences
+    // even though the pointer value that's being dereferenced is actually
+    // the offset of the field rather than exactly 0.
+    // See the FIXME in StoreManager's getLValueFieldOrIvar() method.
+    // This code interacts heavily with this hack; otherwise the value
+    // would not be null at all for most fields, so we'd be unable to track it.
+    if (const auto *Op = dyn_cast<UnaryOperator>(Ex))
+      if (Op->getOpcode() == UO_AddrOf && Op->getSubExpr()->isLValue())
+        if (const Expr *DerefEx = getDerefExpr(Op->getSubExpr()))
+          Ex = DerefEx;
+
+    if (Ex && (ExplodedGraph::isInterestingLValueExpr(Ex) || CallEvent::isCallStmt(Ex)))
       Inner = Ex;
   }
 
diff --git a/lib/StaticAnalyzer/Core/ExprEngine.cpp b/lib/StaticAnalyzer/Core/ExprEngine.cpp
index 9e6ec09010e9..8ee34190891a 100644
--- a/lib/StaticAnalyzer/Core/ExprEngine.cpp
+++ b/lib/StaticAnalyzer/Core/ExprEngine.cpp
@@ -1904,8 +1904,8 @@ void ExprEngine::processSwitch(SwitchNodeBuilder& builder) {
 
     // Evaluate the LHS of the case value.
     llvm::APSInt V1 = Case->getLHS()->EvaluateKnownConstInt(getContext());
-    assert(V1.getBitWidth() == getContext().getTypeSize(CondE->getType()));
-
+    assert(V1.getBitWidth() == getContext().getIntWidth(CondE->getType()));
+    
     // Get the RHS of the case, if it exists.
     llvm::APSInt V2;
     if (const Expr *E = Case->getRHS())
diff --git a/lib/StaticAnalyzer/Core/RegionStore.cpp b/lib/StaticAnalyzer/Core/RegionStore.cpp
index dd7e9dd11781..3000e13d32c6 100644
--- a/lib/StaticAnalyzer/Core/RegionStore.cpp
+++ b/lib/StaticAnalyzer/Core/RegionStore.cpp
@@ -1338,6 +1338,9 @@ RegionStoreManager::getSizeInElements(ProgramStateRef state,
 ///  the array).  This is called by ExprEngine when evaluating casts
 ///  from arrays to pointers.
 SVal RegionStoreManager::ArrayToPointer(Loc Array, QualType T) {
+  if (Array.getAs<loc::ConcreteInt>())
+    return Array;
+
   if (!Array.getAs<loc::MemRegionVal>())
     return UnknownVal();
 
diff --git a/lib/StaticAnalyzer/Core/Store.cpp b/lib/StaticAnalyzer/Core/Store.cpp
index ba48a60d5a1c..1af49f68cc05 100644
--- a/lib/StaticAnalyzer/Core/Store.cpp
+++ b/lib/StaticAnalyzer/Core/Store.cpp
@@ -404,9 +404,15 @@ SVal StoreManager::getLValueFieldOrIvar(const Decl *D, SVal Base) {
 
   case loc::ConcreteIntKind:
     // While these seem funny, this can happen through casts.
-    // FIXME: What we should return is the field offset.  For example,
-    //  add the field offset to the integer value.  That way funny things
+    // FIXME: What we should return is the field offset, not base. For example,
+    //  add the field offset to the integer value.  That way things
     //  like this work properly:  &(((struct foo *) 0xa)->f)
+    //  However, that's not easy to fix without reducing our abilities
+    //  to catch null pointer dereference. Eg., ((struct foo *)0x0)->f = 7
+    //  is a null dereference even though we're dereferencing offset of f
+    //  rather than null. Coming up with an approach that computes offsets
+    //  over null pointers properly while still being able to catch null
+    //  dereferences might be worth it.
     return Base;
 
   default:
@@ -431,7 +437,7 @@ SVal StoreManager::getLValueElement(QualType elementType, NonLoc Offset,
   // If the base is an unknown or undefined value, just return it back.
   // FIXME: For absolute pointer addresses, we just return that value back as
   //  well, although in reality we should return the offset added to that
-  //  value.
+  //  value. See also the similar FIXME in getLValueFieldOrIvar().
   if (Base.isUnknownOrUndef() || Base.getAs<loc::ConcreteInt>())
     return Base;
 
diff --git a/test/Analysis/enum.cpp b/test/Analysis/enum.cpp
index e26b8f00d98c..b561e65ac86e 100644
--- a/test/Analysis/enum.cpp
+++ b/test/Analysis/enum.cpp
@@ -24,3 +24,16 @@ void testCasting(int i) {
     clang_analyzer_eval(j == 0); // expected-warning{{FALSE}}
   }
 }
+
+enum class EnumBool : bool {
+  F = false,
+  T = true
+};
+
+bool testNoCrashOnSwitchEnumBool(EnumBool E) {
+  switch (E) {
+  case EnumBool::F:
+    return false;
+  }
+  return true;
+}
diff --git a/test/Analysis/gmalloc.c b/test/Analysis/gmalloc.c
index 10c4fe284054..50413e2e9b13 100644
--- a/test/Analysis/gmalloc.c
+++ b/test/Analysis/gmalloc.c
@@ -13,6 +13,12 @@ gpointer g_realloc(gpointer mem, gsize n_bytes);
 gpointer g_try_malloc(gsize n_bytes);
 gpointer g_try_malloc0(gsize n_bytes);
 gpointer g_try_realloc(gpointer mem, gsize n_bytes);
+gpointer g_malloc_n(gsize n_blocks, gsize n_block_bytes);
+gpointer g_malloc0_n(gsize n_blocks, gsize n_block_bytes);
+gpointer g_realloc_n(gpointer mem, gsize n_blocks, gsize n_block_bytes);
+gpointer g_try_malloc_n(gsize n_blocks, gsize n_block_bytes);
+gpointer g_try_malloc0_n(gsize n_blocks, gsize n_block_bytes);
+gpointer g_try_realloc_n(gpointer mem, gsize n_blocks, gsize n_block_bytes);
 void g_free(gpointer mem);
 gpointer g_memdup(gconstpointer mem, guint byte_size);
 
@@ -25,6 +31,12 @@ void f1() {
   gpointer g3 = g_try_malloc(n_bytes);
   gpointer g4 = g_try_malloc0(n_bytes);
   g3 = g_try_realloc(g3, n_bytes * 2);
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char));
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char));
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char));
 
   g_free(g1);
   g_free(g2);
@@ -38,6 +50,12 @@ void f2() {
   gpointer g3 = g_try_malloc(n_bytes);
   gpointer g4 = g_try_malloc0(n_bytes);
   g3 = g_try_realloc(g3, n_bytes * 2);
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char));
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char));
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char));
 
   g_free(g1);
   g_free(g2);
@@ -52,8 +70,100 @@ void f3() {
   gpointer g3 = g_try_malloc(n_bytes);
   gpointer g4 = g_try_malloc0(n_bytes);
   g3 = g_try_realloc(g3, n_bytes * 2); // expected-warning{{Potential leak of memory pointed to by 'g4'}}
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g6'}}
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g5'}}
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}}
+
+  g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}}
+  g_free(g2);
+  g_free(g3);
+}
+
+void f4() {
+  gpointer g1 = g_malloc(n_bytes);
+  gpointer g2 = g_malloc0(n_bytes);
+  g1 = g_realloc(g1, n_bytes * 2);
+  gpointer g3 = g_try_malloc(n_bytes);
+  gpointer g4 = g_try_malloc0(n_bytes);
+  g3 = g_try_realloc(g3, n_bytes * 2);
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g6'}}
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g5'}}
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}}
+
+  g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}}
+  g_free(g2);
+  g_free(g3);
+  g_free(g4);
+}
+
+void f5() {
+  gpointer g1 = g_malloc(n_bytes);
+  gpointer g2 = g_malloc0(n_bytes);
+  g1 = g_realloc(g1, n_bytes * 2);
+  gpointer g3 = g_try_malloc(n_bytes);
+  gpointer g4 = g_try_malloc0(n_bytes);
+  g3 = g_try_realloc(g3, n_bytes * 2);
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g6'}}
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char));
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}}
+
+  g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}}
+  g_free(g2);
+  g_free(g3);
+  g_free(g4);
+  g_free(g5);
+}
+
+void f6() {
+  gpointer g1 = g_malloc(n_bytes);
+  gpointer g2 = g_malloc0(n_bytes);
+  g1 = g_realloc(g1, n_bytes * 2);
+  gpointer g3 = g_try_malloc(n_bytes);
+  gpointer g4 = g_try_malloc0(n_bytes);
+  g3 = g_try_realloc(g3, n_bytes * 2);
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char));
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char));
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}}
+
+  g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}}
+  g_free(g2);
+  g_free(g3);
+  g_free(g4);
+  g_free(g5);
+  g_free(g6);
+}
+
+void f7() {
+  gpointer g1 = g_malloc(n_bytes);
+  gpointer g2 = g_malloc0(n_bytes);
+  g1 = g_realloc(g1, n_bytes * 2);
+  gpointer g3 = g_try_malloc(n_bytes);
+  gpointer g4 = g_try_malloc0(n_bytes);
+  g3 = g_try_realloc(g3, n_bytes * 2);
+  gpointer g5 = g_malloc_n(n_bytes, sizeof(char));
+  gpointer g6 = g_malloc0_n(n_bytes, sizeof(char));
+  g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char));
+  gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char));
+  gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char));
+  g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}}
 
   g_free(g1);
   g_free(g2);
   g_free(g3);
+  g_free(g4);
+  g_free(g5);
+  g_free(g6);
+  g_free(g7);
 }
diff --git a/test/Analysis/inlining/inline-defensive-checks.c b/test/Analysis/inlining/inline-defensive-checks.c
index 4029da651b65..010d3a774756 100644
--- a/test/Analysis/inlining/inline-defensive-checks.c
+++ b/test/Analysis/inlining/inline-defensive-checks.c
@@ -1,7 +1,7 @@
 // RUN: %clang_analyze_cc1 -analyzer-checker=core -analyzer-config suppress-inlined-defensive-checks=true -verify %s
 
 // Perform inline defensive checks.
-void idc(int *p) {
+void idc(void *p) {
 	if (p)
 		;
 }
@@ -139,3 +139,42 @@ void idcTrackZeroThroughDoubleAssignemnt(int x) {
   int z = y;
   idcTriggerZeroValueThroughCall(z);
 }
+
+struct S {
+  int f1;
+  int f2;
+};
+
+void idcTrackZeroValueThroughUnaryPointerOperators(struct S *s) {
+  idc(s);
+  *(&(s->f1)) = 7; // no-warning
+}
+
+void idcTrackZeroValueThroughUnaryPointerOperatorsWithOffset1(struct S *s) {
+  idc(s);
+  int *x = &(s->f2);
+  *x = 7; // no-warning
+}
+
+void idcTrackZeroValueThroughUnaryPointerOperatorsWithOffset2(struct S *s) {
+  idc(s);
+  int *x = &(s->f2) - 1;
+  // FIXME: Should not warn.
+  *x = 7; // expected-warning{{Dereference of null pointer}}
+}
+
+void idcTrackZeroValueThroughUnaryPointerOperatorsWithAssignment(struct S *s) {
+  idc(s);
+  int *x = &(s->f1);
+  *x = 7; // no-warning
+}
+
+
+struct S2 {
+  int a[1];
+};
+
+void idcTrackZeroValueThroughUnaryPointerOperatorsWithArrayField(struct S2 *s) {
+  idc(s);
+  *(&(s->a[0])) = 7; // no-warning
+}
diff --git a/test/Analysis/inlining/inline-defensive-checks.cpp b/test/Analysis/inlining/inline-defensive-checks.cpp
index 6a803fa695c6..eaae8d2ae28f 100644
--- a/test/Analysis/inlining/inline-defensive-checks.cpp
+++ b/test/Analysis/inlining/inline-defensive-checks.cpp
@@ -70,4 +70,17 @@ int *retNull() {
 void test(int *p1, int *p2) {
   idc(p1);
 	Foo f(p1);
-}
\ No newline at end of file
+}
+
+struct Bar {
+  int x;
+};
+void idcBar(Bar *b) {
+  if (b)
+    ;
+}
+void testRefToField(Bar *b) {
+  idcBar(b);
+  int &x = b->x; // no-warning
+  x = 5;
+}
diff --git a/test/Analysis/null-deref-offsets.c b/test/Analysis/null-deref-offsets.c
new file mode 100644
index 000000000000..988cec4985da
--- /dev/null
+++ b/test/Analysis/null-deref-offsets.c
@@ -0,0 +1,37 @@
+// RUN: %clang_analyze_cc1 -w -triple i386-apple-darwin10 -analyzer-checker=core,debug.ExprInspection -verify %s
+
+void clang_analyzer_eval(int);
+
+struct S {
+  int x, y;
+  int z[2];
+};
+
+void testOffsets(struct S *s, int coin) {
+  if (s != 0)
+    return;
+
+  // FIXME: Here we are testing the hack that computes offsets to null pointers
+  // as 0 in order to find null dereferences of not-exactly-null pointers,
+  // such as &(s->y) below, which is equal to 4 rather than 0 in run-time.
+
+  // These are indeed null.
+  clang_analyzer_eval(s == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(&(s->x) == 0); // expected-warning{{TRUE}}
+
+  // FIXME: These should ideally be true.
+  clang_analyzer_eval(&(s->y) == 4); // expected-warning{{FALSE}}
+  clang_analyzer_eval(&(s->z[0]) == 8); // expected-warning{{FALSE}}
+  clang_analyzer_eval(&(s->z[1]) == 12); // expected-warning{{FALSE}}
+
+  // FIXME: These should ideally be false.
+  clang_analyzer_eval(&(s->y) == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(&(s->z[0]) == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(&(s->z[1]) == 0); // expected-warning{{TRUE}}
+
+  // But these should still be reported as null dereferences.
+  if (coin)
+    s->y = 5; // expected-warning{{Access to field 'y' results in a dereference of a null pointer (loaded from variable 's')}}
+  else
+    s->z[1] = 6; // expected-warning{{Array access (via field 'z') results in a null pointer dereference}}
+}
diff --git a/test/Analysis/uninit-const.cpp b/test/Analysis/uninit-const.cpp
index 75e932a77ced..db969bfb67d6 100644
--- a/test/Analysis/uninit-const.cpp
+++ b/test/Analysis/uninit-const.cpp
@@ -122,7 +122,7 @@ void f1(void) {
 }
 
 void f_uninit(void) {
-      int x;
+      int x;               // expected-note {{'x' declared without an initial value}}
       doStuff_uninit(&x);  // expected-warning {{1st function call argument is a pointer to uninitialized value}}
                            // expected-note@-1 {{1st function call argument is a pointer to uninitialized value}}
 }
diff --git a/test/CXX/modules-ts/basic/basic.link/module-declaration.cpp b/test/CXX/modules-ts/basic/basic.link/module-declaration.cpp
new file mode 100644
index 000000000000..ee696d14585c
--- /dev/null
+++ b/test/CXX/modules-ts/basic/basic.link/module-declaration.cpp
@@ -0,0 +1,55 @@
+// Tests for module-declaration syntax.
+//
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: echo 'export module x; int a, b;' > %t/x.cppm
+// RUN: echo 'export module x.y; int c;' > %t/x.y.cppm
+//
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %t/x.cppm -o %t/x.pcm
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface -fmodule-file=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm
+//
+// Module implementation for unknown and known module. (The former is ill-formed.)
+// FIXME: TEST=1 should fail because we don't have an interface for module z.
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=1 -DEXPORT= -DPARTITION= -DMODULE_NAME=z
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=2 -DEXPORT= -DPARTITION= -DMODULE_NAME=x
+//
+// Module interface for unknown and known module. (The latter is ill-formed due to
+// redefinition.)
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=3 -DEXPORT=export -DPARTITION= -DMODULE_NAME=z
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=4 -DEXPORT=export -DPARTITION= -DMODULE_NAME=x
+//
+// Defining a module partition.
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=5 -DEXPORT=export -DPARTITION=partition -DMODULE_NAME=z
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=6 -DEXPORT= -DPARTITION=partition -DMODULE_NAME=z
+//
+// Miscellaneous syntax.
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=7 -DEXPORT= -DPARTITION=elderberry -DMODULE_NAME=z
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=8 -DEXPORT= -DPARTITION= -DMODULE_NAME='z [[]]'
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=9 -DEXPORT= -DPARTITION= -DMODULE_NAME='z [[fancy]]'
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DTEST=10 -DEXPORT= -DPARTITION= -DMODULE_NAME='z [[maybe_unused]]'
+
+EXPORT module PARTITION MODULE_NAME;
+#if TEST == 4
+// expected-error@-2 {{redefinition of module 'x'}}
+// expected-note-re@module-declaration.cpp:* {{loaded from '{{.*}}/x.pcm'}}
+#elif TEST == 6
+// expected-error@-5 {{module partition must be declared 'export'}}
+#elif TEST == 7
+// expected-error@-7 {{expected ';'}} expected-error@-7 {{requires a type specifier}}
+#elif TEST == 9
+// expected-warning@-9 {{unknown attribute 'fancy' ignored}}
+#elif TEST == 10
+// expected-error-re@-11 {{'maybe_unused' attribute cannot be applied to a module{{$}}}}
+#else
+// expected-no-diagnostics
+#endif
diff --git a/test/CodeGenCXX/modules-ts.cppm b/test/CXX/modules-ts/codegen-basics.cppm
index 90f6b5319eb6..d1552d9edd26 100644
--- a/test/CodeGenCXX/modules-ts.cppm
+++ b/test/CXX/modules-ts/codegen-basics.cppm
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -fmodules-ts -std=c++1z -triple=x86_64-linux-gnu -fmodules-codegen -emit-module-interface %s -o %t.pcm
 // RUN: %clang_cc1 -fmodules-ts -std=c++1z -triple=x86_64-linux-gnu %t.pcm -emit-llvm -o - | FileCheck %s
 
-module FooBar;
+export module FooBar;
 
 export {
   // CHECK-LABEL: define i32 @_Z1fv(
diff --git a/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.import/p1.cpp b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.import/p1.cpp
new file mode 100644
index 000000000000..aaf43d6584a4
--- /dev/null
+++ b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.import/p1.cpp
@@ -0,0 +1,41 @@
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: echo 'export module x; int a, b;' > %t/x.cppm
+// RUN: echo 'export module x.y; int c;' > %t/x.y.cppm
+//
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %t/x.cppm -o %t/x.pcm
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface -fmodule-file=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm
+//
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DMODULE_NAME=z
+// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
+// RUN:            -DMODULE_X -DMODULE_NAME=x
+
+module MODULE_NAME;
+
+int use_1 = a;
+#if !MODULE_X
+// expected-error@-2 {{declaration of 'a' must be imported from module 'x' before it is required}}
+// expected-note@x.cppm:1 {{here}}
+#endif
+
+import x;
+
+int use_2 = b; // ok
+
+// There is no relation between module x and module x.y.
+int use_3 = c; // expected-error {{declaration of 'c' must be imported from module 'x.y'}}
+               // expected-note@x.y.cppm:1 {{here}}
+
+import x [[]];
+import x [[foo]]; // expected-warning {{unknown attribute 'foo' ignored}}
+import x [[noreturn]]; // expected-error {{'noreturn' attribute cannot be applied to a module import}}
+import x [[blarg::noreturn]]; // expected-warning {{unknown attribute 'noreturn' ignored}}
+
+import x.y;
+import x.; // expected-error {{expected a module name after 'import'}}
+import .x; // expected-error {{expected a module name after 'import'}}
+
+int use_4 = c; // ok
+
+import blarg; // expected-error {{module 'blarg' not found}}
diff --git a/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.interface/p1.cpp b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.interface/p1.cpp
new file mode 100644
index 000000000000..afe3a7a48589
--- /dev/null
+++ b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.interface/p1.cpp
@@ -0,0 +1,27 @@
+// RUN: %clang_cc1 -fmodules-ts %s -verify -o /dev/null
+// RUN: %clang_cc1 -fmodules-ts %s -DINTERFACE -verify -o /dev/null
+// RUN: %clang_cc1 -fmodules-ts %s -DIMPLEMENTATION -verify -o /dev/null
+//
+// RUN: %clang_cc1 -fmodules-ts %s -DBUILT_AS_INTERFACE -emit-module-interface -verify -o /dev/null
+// RUN: %clang_cc1 -fmodules-ts %s -DINTERFACE -DBUILT_AS_INTERFACE -emit-module-interface -verify -o /dev/null
+// RUN: %clang_cc1 -fmodules-ts %s -DIMPLEMENTATION -DBUILT_AS_INTERFACE -emit-module-interface -verify -o /dev/null
+
+#if INTERFACE
+export module A;
+#elif IMPLEMENTATION
+module A;
+ #ifdef BUILT_AS_INTERFACE
+  // expected-error@-2 {{missing 'export' specifier in module declaration while building module interface}}
+ #endif
+#else
+ #ifdef BUILT_AS_INTERFACE
+  // FIXME: Diagnose missing module declaration (at end of TU)
+ #endif
+#endif
+
+export int a;
+#ifndef INTERFACE
+// expected-error@-2 {{export declaration can only be used within a module interface unit}}
+#else
+// expected-no-diagnostics
+#endif
diff --git a/test/CodeGen/asan-globals-gc.cpp b/test/CodeGen/asan-globals-gc.cpp
new file mode 100644
index 000000000000..6d64f41dda8d
--- /dev/null
+++ b/test/CodeGen/asan-globals-gc.cpp
@@ -0,0 +1,7 @@
+// RUN: %clang_cc1 -fsanitize=address -emit-llvm -o - -triple x86_64-windows-msvc %s | FileCheck %s --check-prefix=WITH-GC
+// RUN: %clang_cc1 -fsanitize=address -emit-llvm -o - -triple x86_64-windows-msvc -fdata-sections %s | FileCheck %s --check-prefix=WITH-GC
+
+int global;
+
+// WITH-GC-NOT: call void @__asan_register_globals
+// WITHOUT-GC: call void @__asan_register_globals
diff --git a/test/CodeGen/catch-undef-behavior.c b/test/CodeGen/catch-undef-behavior.c
index c5f3a79429a0..e67f0a1f8eec 100644
--- a/test/CodeGen/catch-undef-behavior.c
+++ b/test/CodeGen/catch-undef-behavior.c
@@ -36,13 +36,7 @@ void foo() {
 
   // CHECK-COMMON:      %[[I8PTR:.*]] = bitcast i32* %[[PTR:.*]] to i8*
   // CHECK-COMMON-NEXT: %[[SIZE:.*]] = call i64 @llvm.objectsize.i64.p0i8(i8* %[[I8PTR]], i1 false, i1 false)
-  // CHECK-COMMON-NEXT: %[[CHECK0:.*]] = icmp uge i64 %[[SIZE]], 4
-
-  // CHECK-COMMON:      %[[PTRTOINT:.*]] = ptrtoint {{.*}}* %[[PTR]] to i64
-  // CHECK-COMMON-NEXT: %[[MISALIGN:.*]] = and i64 %[[PTRTOINT]], 3
-  // CHECK-COMMON-NEXT: %[[CHECK1:.*]] = icmp eq i64 %[[MISALIGN]], 0
-
-  // CHECK-COMMON:       %[[OK:.*]] = and i1 %[[CHECK0]], %[[CHECK1]]
+  // CHECK-COMMON-NEXT: %[[OK:.*]] = icmp uge i64 %[[SIZE]], 4
 
   // CHECK-UBSAN: br i1 %[[OK]], {{.*}} !prof ![[WEIGHT_MD:.*]], !nosanitize
   // CHECK-TRAP:  br i1 %[[OK]], {{.*}}
diff --git a/test/CodeGen/pgo-sample-thinlto-summary.c b/test/CodeGen/pgo-sample-thinlto-summary.c
index a284af3c8087..7782aeb76616 100644
--- a/test/CodeGen/pgo-sample-thinlto-summary.c
+++ b/test/CodeGen/pgo-sample-thinlto-summary.c
@@ -32,10 +32,10 @@ void unroll() {
     baz(i);
 }
 
-// Checks if icp is invoked by normal compile, but not thinlto compile.
+// Check that icp is not invoked (both -O2 and ThinLTO).
 // O2-LABEL: define void @icp
 // THINLTO-LABEL: define void @icp
-// O2: if.true.direct_targ
+// O2-NOT: if.true.direct_targ
 // ThinLTO-NOT: if.true.direct_targ
 void icp(void (*p)()) {
   p();
diff --git a/test/CodeGen/sanitize-recover.c b/test/CodeGen/sanitize-recover.c
index 99eff8518ce8..d714d58c7f26 100644
--- a/test/CodeGen/sanitize-recover.c
+++ b/test/CodeGen/sanitize-recover.c
@@ -22,15 +22,7 @@ void foo() {
   // PARTIAL:      %[[SIZE:.*]] = call i64 @llvm.objectsize.i64.p0i8(i8* {{.*}}, i1 false, i1 false)
   // PARTIAL-NEXT: %[[CHECK0:.*]] = icmp uge i64 %[[SIZE]], 4
 
-  // PARTIAL:      %[[MISALIGN:.*]] = and i64 {{.*}}, 3
-  // PARTIAL-NEXT: %[[CHECK1:.*]] = icmp eq i64 %[[MISALIGN]], 0
+  // PARTIAL:      br i1 %[[CHECK0]], {{.*}} !nosanitize
 
-  // PARTIAL:      %[[CHECK01:.*]] = and i1 %[[CHECK1]], %[[CHECK0]]
-
-  // PARTIAL:      br i1 %[[CHECK01]], {{.*}} !nosanitize
-  // PARTIAL:      br i1 %[[CHECK1]], {{.*}} !nosanitize
-
-  // PARTIAL:      call void @__ubsan_handle_type_mismatch_v1_abort(
-  // PARTIAL-NEXT: unreachable
   // PARTIAL:      call void @__ubsan_handle_type_mismatch_v1(
 }
diff --git a/test/CodeGen/split-debug-filename.c b/test/CodeGen/split-debug-filename.c
index 43daeada96ab..9ca7b0f3287a 100644
--- a/test/CodeGen/split-debug-filename.c
+++ b/test/CodeGen/split-debug-filename.c
@@ -1,7 +1,12 @@
 // RUN: %clang_cc1 -debug-info-kind=limited -split-dwarf-file foo.dwo -S -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -debug-info-kind=limited -enable-split-dwarf -split-dwarf-file foo.dwo -S -emit-llvm -o - %s | FileCheck --check-prefix=VANILLA %s
 int main (void) {
   return 0;
 }
 
 // Testing to ensure that the dwo name gets output into the compile unit.
 // CHECK: !DICompileUnit({{.*}}, splitDebugFilename: "foo.dwo"
+
+// Testing to ensure that the dwo name is not output into the compile unit if
+// it's for vanilla split-dwarf rather than split-dwarf for implicit modules.
+// VANILLA-NOT: splitDebugFilename
diff --git a/test/CodeGenCXX/invariant.group-for-vptrs.cpp b/test/CodeGenCXX/invariant.group-for-vptrs.cpp
index 5c763538088b..4e5d9b49d985 100644
--- a/test/CodeGenCXX/invariant.group-for-vptrs.cpp
+++ b/test/CodeGenCXX/invariant.group-for-vptrs.cpp
@@ -12,15 +12,15 @@ struct D : A {
 void testExternallyVisible() {
   A *a = new A;
 
-  // CHECK: load {{.*}} !invariant.group ![[A_MD:[0-9]+]]
+  // CHECK: load {{.*}} !invariant.group ![[MD:[0-9]+]]
   a->foo();
 
   D *d = new D;
   // CHECK: call void @_ZN1DC1Ev(
-  // CHECK: load {{.*}} !invariant.group ![[D_MD:[0-9]+]]
+  // CHECK: load {{.*}} !invariant.group ![[MD]]
   d->foo();
   A *a2 = d;
-  // CHECK: load {{.*}} !invariant.group ![[A_MD]]
+  // CHECK: load {{.*}} !invariant.group ![[MD]]
   a2->foo();
 }
 // CHECK-LABEL: {{^}}}
@@ -40,35 +40,32 @@ struct C : B {
 // CHECK-LABEL: define void @_Z21testInternallyVisibleb(
 void testInternallyVisible(bool p) {
   B *b = new B;
-  // CHECK: = load {{.*}}, !invariant.group ![[B_MD:[0-9]+]]
+  // CHECK: = load {{.*}}, !invariant.group ![[MD]]
   b->bar();
 
   // CHECK: call void @_ZN12_GLOBAL__N_11CC1Ev(
   C *c = new C;
-  // CHECK: = load {{.*}}, !invariant.group ![[C_MD:[0-9]+]]
+  // CHECK: = load {{.*}}, !invariant.group ![[MD]]
   c->bar();
 }
 
 // Checking A::A()
 // CHECK-LABEL: define linkonce_odr void @_ZN1AC2Ev(
-// CHECK: store {{.*}}, !invariant.group ![[A_MD]]
+// CHECK: store {{.*}}, !invariant.group ![[MD]]
 // CHECK-LABEL: {{^}}}
 
 // Checking D::D()
 // CHECK-LABEL: define linkonce_odr void @_ZN1DC2Ev(
 // CHECK:  = call i8* @llvm.invariant.group.barrier(i8*
 // CHECK:  call void @_ZN1AC2Ev(%struct.A*
-// CHECK: store {{.*}} !invariant.group ![[D_MD]]
+// CHECK: store {{.*}} !invariant.group ![[MD]]
 
 // Checking B::B()
 // CHECK-LABEL: define internal void @_ZN12_GLOBAL__N_11BC2Ev(
-// CHECK:  store {{.*}}, !invariant.group ![[B_MD]]
+// CHECK:  store {{.*}}, !invariant.group ![[MD]]
 
 // Checking C::C()
 // CHECK-LABEL: define internal void @_ZN12_GLOBAL__N_11CC2Ev(
-// CHECK:  store {{.*}}, !invariant.group ![[C_MD]]
+// CHECK:  store {{.*}}, !invariant.group ![[MD]]
 
-// CHECK: ![[A_MD]] = !{!"_ZTS1A"}
-// CHECK: ![[D_MD]] = !{!"_ZTS1D"}
-// CHECK: ![[B_MD]] = distinct !{}
-// CHECK: ![[C_MD]] = distinct !{}
+// CHECK: ![[MD]] = !{}
diff --git a/test/CodeGenCXX/ubsan-suppress-checks.cpp b/test/CodeGenCXX/ubsan-suppress-checks.cpp
index d0e7b32ef96c..ae7c94b34f29 100644
--- a/test/CodeGenCXX/ubsan-suppress-checks.cpp
+++ b/test/CodeGenCXX/ubsan-suppress-checks.cpp
@@ -12,6 +12,7 @@ void load_non_null_pointers() {
 
   char c = "foo"[0];
 
+  // CHECK-NOT: and i64 {{.*}}, !nosanitize
   // CHECK-NOT: icmp ne {{.*}}, null, !nosanitize
   // CHECK: ret void
 }
@@ -43,10 +44,6 @@ struct A {
     };
     f();
 
-    // LAMBDA: %[[LAMBDAINT:[0-9]+]] = ptrtoint %class.anon* %[[FUNCVAR:.*]] to i64, !nosanitize
-    // LAMBDA: and i64 %[[LAMBDAINT]], 7, !nosanitize
-    // LAMBDA: call void @__ubsan_handle_type_mismatch
-
     // LAMBDA-NOT: call void @__ubsan_handle_type_mismatch
     // LAMBDA: ret void
   }
diff --git a/test/CoverageMapping/empty-destructor.cpp b/test/CoverageMapping/empty-destructor.cpp
new file mode 100644
index 000000000000..cfc29a75c6d7
--- /dev/null
+++ b/test/CoverageMapping/empty-destructor.cpp
@@ -0,0 +1,11 @@
+// RUN: %clang_cc1 -triple i686-windows -emit-llvm-only -fcoverage-mapping -dump-coverage-mapping -fprofile-instrument=clang %s | FileCheck %s
+
+struct A {
+  virtual ~A();
+};
+
+// CHECK: ?PR32761@@YAXXZ:
+// CHECK-NEXT: File 0, [[@LINE+1]]:16 -> [[@LINE+3]]:2 = #0
+void PR32761() {
+  A a;
+}
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crt1.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crt1.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crt1.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crti.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crti.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crti.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crtn.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crtn.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crtn.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtbegin.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtbegin.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtbegin.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtend.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtend.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtend.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crt1.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crt1.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crt1.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crti.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crti.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crti.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crtn.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crtn.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crtn.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtbegin.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtbegin.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtbegin.o
diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtend.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtend.o
new file mode 100644
index 000000000000..e69de29bb2d1
--- /dev/null
+++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtend.o
diff --git a/test/Driver/fsanitize.c b/test/Driver/fsanitize.c
index b6304b515b73..05e239c74243 100644
--- a/test/Driver/fsanitize.c
+++ b/test/Driver/fsanitize.c
@@ -358,6 +358,27 @@
 // RUN: %clang -target i386-pc-openbsd -fsanitize=address %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ASAN-OPENBSD
 // CHECK-ASAN-OPENBSD: unsupported option '-fsanitize=address' for target 'i386-pc-openbsd'
 
+// RUN: %clang -target x86_64-apple-darwin -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-X86-64-DARWIN
+// CHECK-LSAN-X86-64-DARWIN-NOT: unsupported option
+
+// RUN: %clang -target x86_64-apple-iossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-X86-64-IOSSIMULATOR
+// CHECK-LSAN-X86-64-IOSSIMULATOR-NOT: unsupported option
+
+// RUN: %clang -target x86_64-apple-tvossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-X86-64-TVOSSIMULATOR
+// CHECK-LSAN-X86-64-TVOSSIMULATOR-NOT: unsupported option
+
+// RUN: %clang -target i386-apple-darwin -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-I386-DARWIN
+// CHECK-LSAN-I386-DARWIN-NOT: unsupported option
+
+// RUN: %clang -target arm-apple-ios -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-ARM-IOS
+// CHECK-LSAN-ARM-IOS-NOT: unsupported option
+
+// RUN: %clang -target i386-apple-iossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-I386-IOSSIMULATOR
+// CHECK-LSAN-I386-IOSSIMULATOR-NOT: unsupported option
+
+// RUN: %clang -target i386-apple-tvossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-I386-TVOSSIMULATOR
+// CHECK-LSAN-I386-TVOSSIMULATOR-NOT: unsupported option
+
 // RUN: %clang -target i686-linux-gnu -fsanitize=efficiency-cache-frag %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ESAN-X86
 // RUN: %clang -target i686-linux-gnu -fsanitize=efficiency-working-set %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ESAN-X86
 // CHECK-ESAN-X86: error: unsupported option '-fsanitize=efficiency-{{.*}}' for target 'i686--linux-gnu'
diff --git a/test/Driver/fuzzer.c b/test/Driver/fuzzer.c
new file mode 100644
index 000000000000..e534a73a647f
--- /dev/null
+++ b/test/Driver/fuzzer.c
@@ -0,0 +1,20 @@
+// Test flags inserted by -fsanitize=fuzzer.
+
+// RUN: %clang -fsanitize=fuzzer %s -target x86_64-apple-darwin14 -### 2>&1 | FileCheck --check-prefixes=CHECK-FUZZER-LIB,CHECK-COVERAGE-FLAGS %s
+//
+// CHECK-FUZZER-LIB: libLLVMFuzzer.a
+// CHECK-COVERAGE: -fsanitize-coverage-trace-pc-guard
+// CHECK-COVERAGE-SAME: -fsanitize-coverage-indirect-calls
+// CHECK-COVERAGE-SAME: -fsanitize-coverage-trace-cmp
+
+// RUN: %clang -fsanitize=fuzzer -target i386-unknown-linux -stdlib=platform %s -### 2>&1 | FileCheck --check-prefixes=CHECK-LIBCXX-LINUX %s
+//
+// CHECK-LIBCXX-LINUX: -lstdc++
+
+// RUN: %clang -target x86_64-apple-darwin14 -fsanitize=fuzzer %s -### 2>&1 | FileCheck --check-prefixes=CHECK-LIBCXX-DARWIN %s
+//
+// CHECK-LIBCXX-DARWIN: -lc++
+
+int LLVMFuzzerTestOneInput(const char *Data, long Size) {
+  return 0;
+}
diff --git a/test/Driver/hexagon-toolchain-elf.c b/test/Driver/hexagon-toolchain-elf.c
index f6f2191b3eb6..98582450e3fd 100644
--- a/test/Driver/hexagon-toolchain-elf.c
+++ b/test/Driver/hexagon-toolchain-elf.c
@@ -97,6 +97,22 @@
 // CHECK024: "-cc1" {{.*}} "-target-cpu" "hexagonv62"
 // CHECK024: hexagon-link{{.*}}/Inputs/hexagon_tree/Tools/bin/../target/hexagon/lib/v62/crt0
 
+// RUN: %clang -### -target hexagon-unknown-elf \
+// RUN:   -ccc-install-dir %S/Inputs/hexagon_tree/Tools/bin \
+// RUN:   -O3 \
+// RUN:   %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK025 %s
+// CHECK025: "-ffp-contract=fast"
+// CHECK025: hexagon-link
+
+// RUN: %clang -### -target hexagon-unknown-elf \
+// RUN:   -ccc-install-dir %S/Inputs/hexagon_tree/Tools/bin \
+// RUN:   -O3 -ffp-contract=off \
+// RUN:   %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK026 %s
+// CHECK026-NOT: "-ffp-contract=fast"
+// CHECK026: hexagon-link
+
 // -----------------------------------------------------------------------------
 // Test Linker related args
 // -----------------------------------------------------------------------------
diff --git a/test/Driver/linux-ld.c b/test/Driver/linux-ld.c
index e5aa870866d2..1c5f1a4556b1 100644
--- a/test/Driver/linux-ld.c
+++ b/test/Driver/linux-ld.c
@@ -638,6 +638,46 @@
 // CHECK-OPENSUSE-42-2-AARCH64: "{{.*}}/usr/lib64/gcc/aarch64-suse-linux/4.8{{/|\\\\}}crtend.o"
 // CHECK-OPENSUSE-42-2-AARCH64: "{{.*}}/usr/lib64/gcc/aarch64-suse-linux/4.8/../../../../lib64{{/|\\\\}}crtn.o"
 //
+// Check openSUSE Tumbleweed on armv6hl
+// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \
+// RUN:     --target=armv6hl-suse-linux-gnueabi \
+// RUN:     --gcc-toolchain="" \
+// RUN:     --sysroot=%S/Inputs/opensuse_tumbleweed_armv6hl_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV6HL %s
+// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \
+// RUN:     --target=armv6hl-suse-linux-gnueabi \
+// RUN:     --gcc-toolchain="" \
+// RUN:     --sysroot=%S/Inputs/opensuse_tumbleweed_armv6hl_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV6HL %s
+// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
+// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crt1.o"
+// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crti.o"
+// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5{{/|\\\\}}crtbegin.o"
+// CHECK-OPENSUSE-TW-ARMV6HL: "-L[[SYSROOT]]/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5"
+// CHECK-OPENSUSE-TW-ARMV6HL: "-L[[SYSROOT]]/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib"
+// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5{{/|\\\\}}crtend.o"
+// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crtn.o"
+//
+// Check openSUSE Tumbleweed on armv7hl
+// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \
+// RUN:     --target=armv7hl-suse-linux-gnueabi \
+// RUN:     --gcc-toolchain="" \
+// RUN:     --sysroot=%S/Inputs/opensuse_tumbleweed_armv7hl_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV7HL %s
+// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \
+// RUN:     --target=armv7hl-suse-linux-gnueabi \
+// RUN:     --gcc-toolchain="" \
+// RUN:     --sysroot=%S/Inputs/opensuse_tumbleweed_armv7hl_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV7HL %s
+// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
+// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crt1.o"
+// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crti.o"
+// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5{{/|\\\\}}crtbegin.o"
+// CHECK-OPENSUSE-TW-ARMV7HL: "-L[[SYSROOT]]/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5"
+// CHECK-OPENSUSE-TW-ARMV7HL: "-L[[SYSROOT]]/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib"
+// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5{{/|\\\\}}crtend.o"
+// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crtn.o"
+//
 // Check dynamic-linker for different archs
 // RUN: %clang %s -### -o %t.o 2>&1 \
 // RUN:     --target=arm-linux-gnueabi \
diff --git a/test/Driver/modules-ts.cpp b/test/Driver/modules-ts.cpp
index fd3391479eea..3847b71f7b74 100644
--- a/test/Driver/modules-ts.cpp
+++ b/test/Driver/modules-ts.cpp
@@ -1,6 +1,6 @@
 // Check compiling a module interface to a .pcm file.
 //
-// RUN: %clang -fmodules-ts -x c++-module --precompile %s -Dimplementation= -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE
+// RUN: %clang -fmodules-ts -x c++-module --precompile %s -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE
 //
 // CHECK-PRECOMPILE: -cc1 {{.*}} -emit-module-interface
 // CHECK-PRECOMPILE-SAME: -o {{.*}}.pcm
@@ -18,7 +18,7 @@
 
 // Check use of a .pcm file in another compilation.
 //
-// RUN: %clang -fmodules-ts -fmodule-file=%t.pcm %s -c -o %t.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-USE
+// RUN: %clang -fmodules-ts -fmodule-file=%t.pcm -Dexport= %s -c -o %t.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-USE
 //
 // CHECK-USE: -cc1
 // CHECK-USE-SAME: -emit-obj
@@ -28,11 +28,11 @@
 
 // Check combining precompile and compile steps works.
 //
-// RUN: %clang -fmodules-ts -x c++-module %s -Dimplementation= -c -o %t.pcm.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE --check-prefix=CHECK-COMPILE
+// RUN: %clang -fmodules-ts -x c++-module %s -c -o %t.pcm.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE --check-prefix=CHECK-COMPILE
 
 // Check that .cppm is treated as a module implicitly.
 // RUN: cp %s %t.cppm
-// RUN: %clang -fmodules-ts --precompile %t.cppm -Dimplementation= -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE
+// RUN: %clang -fmodules-ts --precompile %t.cppm -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE
 
-// Note, we use -Dimplementation= to make this a valid module interface unit when building the interface.
-module implementation foo;
+// Note, we use -Dexport= to make this a module implementation unit when building the implementation.
+export module foo;
diff --git a/test/Driver/sanitizer-ld.c b/test/Driver/sanitizer-ld.c
index c4a6f4377b89..efdd1245778d 100644
--- a/test/Driver/sanitizer-ld.c
+++ b/test/Driver/sanitizer-ld.c
@@ -409,6 +409,15 @@
 // CHECK-ASAN-DARWIN106-CXX: libclang_rt.asan_osx_dynamic.dylib
 // CHECK-ASAN-DARWIN106-CXX-NOT: -lc++abi
 
+// RUN: %clangxx -fsanitize=leak %s -### -o %t.o 2>&1 \
+// RUN:     -mmacosx-version-min=10.6 \
+// RUN:     -target x86_64-apple-darwin13.4.0 -fuse-ld=ld -stdlib=platform \
+// RUN:     --sysroot=%S/Inputs/basic_linux_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-LSAN-DARWIN106-CXX %s
+// CHECK-LSAN-DARWIN106-CXX: "{{.*}}ld{{(.exe)?}}"
+// CHECK-LSAN-DARWIN106-CXX: libclang_rt.lsan_osx_dynamic.dylib
+// CHECK-LSAN-DARWIN106-CXX-NOT: -lc++abi
+
 // RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \
 // RUN:     -target x86_64-unknown-linux -fuse-ld=ld -fsanitize=safe-stack \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
diff --git a/test/Driver/split-debug.c b/test/Driver/split-debug.c
index eaa1e990c04c..c12164b4e2f1 100644
--- a/test/Driver/split-debug.c
+++ b/test/Driver/split-debug.c
@@ -36,53 +36,53 @@
 // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -gmlt -fno-split-dwarf-inlining -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-GMLT-WITH-SPLIT < %t %s
 //
-// CHECK-GMLT-WITH-SPLIT: "-split-dwarf=Enable"
+// CHECK-GMLT-WITH-SPLIT: "-enable-split-dwarf"
 // CHECK-GMLT-WITH-SPLIT: "-debug-info-kind=line-tables-only"
 // CHECK-GMLT-WITH-SPLIT: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -gmlt -gsplit-dwarf -fno-split-dwarf-inlining -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-SPLIT-WITH-GMLT < %t %s
 //
-// CHECK-SPLIT-WITH-GMLT: "-split-dwarf=Enable"
+// CHECK-SPLIT-WITH-GMLT: "-enable-split-dwarf"
 // CHECK-SPLIT-WITH-GMLT: "-debug-info-kind=line-tables-only"
 // CHECK-SPLIT-WITH-GMLT: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -fno-split-dwarf-inlining -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-SPLIT-WITH-NOINL < %t %s
 //
-// CHECK-SPLIT-WITH-NOINL: "-split-dwarf=Enable"
+// CHECK-SPLIT-WITH-NOINL: "-enable-split-dwarf"
 // CHECK-SPLIT-WITH-NOINL: "-debug-info-kind=limited"
 // CHECK-SPLIT-WITH-NOINL: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -gmlt -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-GMLT-OVER-SPLIT < %t %s
 //
-// CHECK-GMLT-OVER-SPLIT-NOT: "-split-dwarf=Enable"
+// CHECK-GMLT-OVER-SPLIT-NOT: "-enable-split-dwarf"
 // CHECK-GMLT-OVER-SPLIT: "-debug-info-kind=line-tables-only"
 // CHECK-GMLT-OVER-SPLIT-NOT: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -gmlt -gsplit-dwarf -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-SPLIT-OVER-GMLT < %t %s
 //
-// CHECK-SPLIT-OVER-GMLT: "-split-dwarf=Enable" "-debug-info-kind=limited"
+// CHECK-SPLIT-OVER-GMLT: "-enable-split-dwarf" "-debug-info-kind=limited"
 // CHECK-SPLIT-OVER-GMLT: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -g0 -fno-split-dwarf-inlining -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-G0-OVER-SPLIT < %t %s
 //
-// CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf=Enable"
+// CHECK-G0-OVER-SPLIT-NOT: "-enable-split-dwarf"
 // CHECK-G0-OVER-SPLIT-NOT: "-debug-info-kind
 // CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -g0 -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-G0-OVER-SPLIT < %t %s
 //
-// CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf=Enable"
+// CHECK-G0-OVER-SPLIT-NOT: "-enable-split-dwarf"
 // CHECK-G0-OVER-SPLIT-NOT: "-debug-info-kind
 // CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf-file"
 
 // RUN: %clang -target x86_64-unknown-linux-gnu -g0 -gsplit-dwarf -S -### %s 2> %t
 // RUN: FileCheck -check-prefix=CHECK-SPLIT-OVER-G0 < %t %s
 //
-// CHECK-SPLIT-OVER-G0: "-split-dwarf=Enable" "-debug-info-kind=limited"
+// CHECK-SPLIT-OVER-G0: "-enable-split-dwarf" "-debug-info-kind=limited"
 // CHECK-SPLIT-OVER-G0: "-split-dwarf-file"
diff --git a/test/Format/incomplete.cpp b/test/Format/incomplete.cpp
index 8a65fad657cd..f92d57619373 100644
--- a/test/Format/incomplete.cpp
+++ b/test/Format/incomplete.cpp
@@ -1,6 +1,6 @@
 // RUN: grep -Ev "// *[A-Z-]+:" %s | clang-format -style=LLVM -cursor=0 \
 // RUN:   | FileCheck -strict-whitespace %s
-// CHECK: {{"IncompleteFormat": true}}
+// CHECK: {{"IncompleteFormat": true, "Line": 2}}
 // CHECK: {{^int\ \i;$}}
  int    i;
 // CHECK: {{^f  \( g  \(;$}}
diff --git a/test/Headers/stdatomic.c b/test/Headers/stdatomic.c
index 76112cfe40ea..3643fd4245b3 100644
--- a/test/Headers/stdatomic.c
+++ b/test/Headers/stdatomic.c
@@ -1,4 +1,5 @@
 // RUN: %clang_cc1 -std=c11 -E %s | FileCheck %s
+// RUN: %clang_cc1 -std=c11 -fms-compatibility -E %s | FileCheck %s
 #include <stdatomic.h>
 
 int bool_lock_free = ATOMIC_BOOL_LOCK_FREE;
diff --git a/test/Index/Core/external-source-symbol-attr.m b/test/Index/Core/external-source-symbol-attr.m
new file mode 100644
index 000000000000..49a075fc74e0
--- /dev/null
+++ b/test/Index/Core/external-source-symbol-attr.m
@@ -0,0 +1,100 @@
+// RUN: c-index-test core -print-source-symbols -- %s -target x86_64-apple-macosx10.7 | FileCheck %s
+
+#define EXT_DECL(mod_name) __attribute__((external_source_symbol(language="Swift", defined_in=mod_name)))
+#define GEN_DECL(mod_name) __attribute__((external_source_symbol(language="Swift", defined_in=mod_name, generated_declaration)))
+#define PUSH_GEN_DECL(mod_name) push(GEN_DECL(mod_name), apply_to=any(enum, objc_interface, objc_category, objc_protocol))
+
+// This should not be indexed.
+GEN_DECL("some_module")
+@interface I1
+// CHECK-NOT: [[@LINE-1]]:12 |
+-(void)method;
+// CHECK-NOT: [[@LINE-1]]:8 |
+@end
+
+EXT_DECL("some_module")
+@interface I2
+// CHECK: [[@LINE-1]]:12 | class/Swift | I2 | c:@M@some_module@objc(cs)I2 | {{.*}} | Decl | rel: 0
+-(void)method;
+// CHECK: [[@LINE-1]]:8 | instance-method/Swift | method | c:@M@some_module@objc(cs)I2(im)method | -[I2 method] | Decl,Dyn,RelChild | rel: 1
+@end
+
+void test1(I1 *o) {
+// CHECK: [[@LINE-1]]:12 | class/Swift | I1 | c:@M@some_module@objc(cs)I1 |
+  [o method];
+  // CHECK: [[@LINE-1]]:6 | instance-method/Swift | method | c:@M@some_module@objc(cs)I1(im)method |
+}
+
+EXT_DECL("some_module")
+@protocol ExtProt
+// CHECK: [[@LINE-1]]:11 | protocol/Swift | ExtProt | c:@M@some_module@objc(pl)ExtProt |
+@end
+
+@interface I1(cat)
+// CHECK: [[@LINE-1]]:15 | extension/ObjC | cat | c:@M@some_module@objc(cy)I1@cat |
+-(void)cat_method;
+// CHECK: [[@LINE-1]]:8 | instance-method/ObjC | cat_method | c:@M@some_module@objc(cs)I1(im)cat_method
+@end
+
+EXT_DECL("cat_module")
+@interface I1(cat2)
+// CHECK: [[@LINE-1]]:15 | extension/Swift | cat2 | c:@CM@cat_module@some_module@objc(cy)I1@cat2 |
+-(void)cat_method2;
+// CHECK: [[@LINE-1]]:8 | instance-method/Swift | cat_method2 | c:@CM@cat_module@some_module@objc(cs)I1(im)cat_method2
+@end
+
+#define NS_ENUM(_name, _type) enum _name:_type _name; enum _name : _type
+
+#pragma clang attribute PUSH_GEN_DECL("modname")
+
+@interface I3
+// CHECK-NOT: [[@LINE-1]]:12 |
+-(void)meth;
+// CHECK-NOT: [[@LINE-1]]:8 |
+@end
+
+@interface I3(cat)
+// CHECK-NOT: [[@LINE-1]]:12 |
+// CHECK-NOT: [[@LINE-2]]:15 |
+-(void)meth2;
+// CHECK-NOT: [[@LINE-1]]:8 |
+@end
+
+@protocol ExtProt2
+// CHECK-NOT: [[@LINE-1]]:11 |
+-(void)meth;
+// CHECK-NOT: [[@LINE-1]]:8 |
+@end
+
+typedef NS_ENUM(SomeEnum, int) {
+// CHECK-NOT: [[@LINE-1]]:17 |
+  SomeEnumFirst = 0,
+  // CHECK-NOT: [[@LINE-1]]:3 |
+};
+
+#pragma clang attribute pop
+
+void test2(I3 *i3, id<ExtProt2> prot2, SomeEnum some) {
+  // CHECK: [[@LINE-1]]:12 | class/Swift | I3 | c:@M@modname@objc(cs)I3 |
+  // CHECK: [[@LINE-2]]:23 | protocol/Swift | ExtProt2 | c:@M@modname@objc(pl)ExtProt2 |
+  // CHECK: [[@LINE-3]]:40 | enum/Swift | SomeEnum | c:@M@modname@E@SomeEnum |
+  [i3 meth];
+  // CHECK: [[@LINE-1]]:7 | instance-method/Swift | meth | c:@M@modname@objc(cs)I3(im)meth |
+  [i3 meth2];
+  // CHECK: [[@LINE-1]]:7 | instance-method/Swift | meth2 | c:@CM@modname@objc(cs)I3(im)meth2 |
+  [prot2 meth];
+  // CHECK: [[@LINE-1]]:10 | instance-method/Swift | meth | c:@M@modname@objc(pl)ExtProt2(im)meth |
+  some = SomeEnumFirst;
+  // CHECK: [[@LINE-1]]:10 | enumerator/Swift | SomeEnumFirst | c:@M@modname@E@SomeEnum@SomeEnumFirst |
+}
+
+#pragma clang attribute PUSH_GEN_DECL("other_mod_for_cat")
+@interface I3(cat_other_mod)
+-(void)meth_other_mod;
+@end
+#pragma clang attribute pop
+
+void test3(I3 *i3) {
+  [i3 meth_other_mod];
+  // CHECK: [[@LINE-1]]:7 | instance-method/Swift | meth_other_mod | c:@CM@other_mod_for_cat@modname@objc(cs)I3(im)meth_other_mod |
+}
diff --git a/test/Index/Core/index-source.cpp b/test/Index/Core/index-source.cpp
index 5d4a1be45683..7bb366c53611 100644
--- a/test/Index/Core/index-source.cpp
+++ b/test/Index/Core/index-source.cpp
@@ -111,3 +111,180 @@ template<>
 class PartialSpecilizationClass<int, int> { };
 // CHECK: [[@LINE-1]]:7 | class(Gen,TS)/C++ | PartialSpecilizationClass | c:@S@PartialSpecilizationClass>#I#I | <no-cgname> | Def,RelSpecialization | rel: 1
 // CHECK-NEXT: RelSpecialization | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass
+
+template<typename T, typename S>
+class PseudoOverridesInSpecializations {
+  void function() { }
+  void function(int) { }
+
+  static void staticFunction() { }
+
+  int field;
+  static int variable;
+
+  typedef T TypeDef;
+  using TypeAlias = T;
+
+  enum anEnum { };
+
+  struct Struct { };
+  union Union { };
+
+  using TypealiasOrRecord = void;
+
+  template<typename U> struct InnerTemplate { };
+  template<typename U> struct InnerTemplate <U*> { };
+};
+
+template<>
+class PseudoOverridesInSpecializations<double, int> {
+  void function() { }
+// CHECK: [[@LINE-1]]:8 | instance-method/C++ | function | c:@S@PseudoOverridesInSpecializations>#d#I@F@function# | __ZN32PseudoOverridesInSpecializationsIdiE8functionEv | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | function | c:@ST>2#T#T@PseudoOverridesInSpecializations@F@function#
+
+  void staticFunction() { }
+// CHECK: [[@LINE-1]]:8 | instance-method/C++ | staticFunction | c:@S@PseudoOverridesInSpecializations>#d#I@F@staticFunction# | __ZN32PseudoOverridesInSpecializationsIdiE14staticFunctionEv | Def,RelChild | rel: 1
+// CHECK-NOT: RelOver
+
+  int notOverridingField = 0;
+
+// CHECK-LABEL: checLabelBreak
+  int checLabelBreak = 0;
+
+  int field = 0;
+// CHECK: [[@LINE-1]]:7 | field/C++ | field | c:@S@PseudoOverridesInSpecializations>#d#I@FI@field | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | field | c:@ST>2#T#T@PseudoOverridesInSpecializations@FI@field
+
+  static double variable;
+// CHECK: [[@LINE-1]]:17 | static-property/C++ | variable | c:@S@PseudoOverridesInSpecializations>#d#I@variable | __ZN32PseudoOverridesInSpecializationsIdiE8variableE | Decl,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | variable | c:@ST>2#T#T@PseudoOverridesInSpecializations@variable
+
+  typedef double TypeDef;
+// CHECK: [[@LINE-1]]:18 | type-alias/C | TypeDef | c:index-source.cpp@S@PseudoOverridesInSpecializations>#d#I@T@TypeDef | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | TypeDef | c:index-source.cpp@ST>2#T#T@PseudoOverridesInSpecializations@T@TypeDef
+
+  using TypeAlias = int;
+// CHECK: [[@LINE-1]]:9 | type-alias/C++ | TypeAlias | c:@S@PseudoOverridesInSpecializations>#d#I@TypeAlias | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | TypeAlias | c:@ST>2#T#T@PseudoOverridesInSpecializations@TypeAlias
+
+  enum anEnum { };
+// CHECK: [[@LINE-1]]:8 | enum/C | anEnum | c:@S@PseudoOverridesInSpecializations>#d#I@E@anEnum | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | anEnum | c:@ST>2#T#T@PseudoOverridesInSpecializations@E@anEnum
+  class Struct { };
+// CHECK: [[@LINE-1]]:9 | class/C++ | Struct | c:@S@PseudoOverridesInSpecializations>#d#I@S@Struct | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | Struct | c:@ST>2#T#T@PseudoOverridesInSpecializations@S@Struct
+  union Union { };
+// CHECK: [[@LINE-1]]:9 | union/C | Union | c:@S@PseudoOverridesInSpecializations>#d#I@U@Union | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | Union | c:@ST>2#T#T@PseudoOverridesInSpecializations@U@Union
+
+  struct TypealiasOrRecord { };
+// CHECK: [[@LINE-1]]:10 | struct/C | TypealiasOrRecord | c:@S@PseudoOverridesInSpecializations>#d#I@S@TypealiasOrRecord | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | TypealiasOrRecord | c:@ST>2#T#T@PseudoOverridesInSpecializations@TypealiasOrRecord
+
+  template<typename U> struct InnerTemplate { };
+// CHECK: [[@LINE-1]]:31 | struct(Gen)/C++ | InnerTemplate | c:@S@PseudoOverridesInSpecializations>#d#I@ST>1#T@InnerTemplate | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | InnerTemplate | c:@ST>2#T#T@PseudoOverridesInSpecializations@ST>1#T@InnerTemplate
+  template<typename U> struct InnerTemplate <U*> { };
+// CHECK-NOT: RelOver
+};
+
+template<typename S>
+class PseudoOverridesInSpecializations<float, S> {
+  typedef float TypealiasOrRecord;
+// CHECK: [[@LINE-1]]:17 | type-alias/C | TypealiasOrRecord | c:index-source.cpp@SP>1#T@PseudoOverridesInSpecializations>#f#t0.0@T@TypealiasOrRecord | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | TypealiasOrRecord | c:@ST>2#T#T@PseudoOverridesInSpecializations@TypealiasOrRecord
+};
+
+template<typename T, typename U>
+class ConflictingPseudoOverridesInSpecialization {
+  void foo(T x);
+  void foo(U x);
+};
+
+template<typename T>
+class ConflictingPseudoOverridesInSpecialization<int, T> {
+  void foo(T x);
+// CHECK: [[@LINE-1]]:8 | instance-method/C++ | foo | c:@SP>1#T@ConflictingPseudoOverridesInSpecialization>#I#t0.0@F@foo#S0_# | <no-cgname> | Decl,RelChild,RelOver,RelSpecialization | rel: 3
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelOver,RelSpecialization | foo | c:@ST>2#T#T@ConflictingPseudoOverridesInSpecialization@F@foo#t0.0#
+// CHECK-NEXT: RelOver,RelSpecialization | foo | c:@ST>2#T#T@ConflictingPseudoOverridesInSpecialization@F@foo#t0.1#
+};
+
+template<typename T, typename U, int x>
+void functionSpecialization();
+
+template<typename T, typename U, int x>
+void functionSpecialization() { }
+// CHECK: [[@LINE-1]]:6 | function/C | functionSpecialization | c:@FT@>3#T#T#NIfunctionSpecialization#v# | <no-cgname> | Def | rel: 0
+
+template<>
+void functionSpecialization<int, int, 0>();
+// CHECK: [[@LINE-1]]:6 | function(Gen,TS)/C++ | functionSpecialization | c:@F@functionSpecialization<#I#I#VI0># | __Z22functionSpecializationIiiLi0EEvv | Decl,RelSpecialization | rel: 1
+// CHECK-NEXT: RelSpecialization | functionSpecialization | c:@FT@>3#T#T#NIfunctionSpecialization#v#
+
+template<>
+void functionSpecialization<int, int, 0>() { }
+// CHECK: [[@LINE-1]]:6 | function(Gen,TS)/C++ | functionSpecialization | c:@F@functionSpecialization<#I#I#VI0># | __Z22functionSpecializationIiiLi0EEvv | Def,RelSpecialization | rel: 1
+// CHECK-NEXT: RelSpecialization | functionSpecialization | c:@FT@>3#T#T#NIfunctionSpecialization#v#
+
+struct ContainsSpecializedMemberFunction {
+  template<typename T>
+  void memberSpecialization();
+};
+
+template<typename T>
+void ContainsSpecializedMemberFunction::memberSpecialization() {
+// CHECK: [[@LINE-1]]:41 | instance-method/C++ | memberSpecialization | c:@S@ContainsSpecializedMemberFunction@FT@>1#TmemberSpecialization#v# | <no-cgname> | Def,RelChild | rel: 1
+// CHECK-NEXT: RelChild
+}
+
+template<>
+void ContainsSpecializedMemberFunction::memberSpecialization<int>() {
+// CHECK: [[@LINE-1]]:41 | instance-method(Gen,TS)/C++ | memberSpecialization | c:@S@ContainsSpecializedMemberFunction@F@memberSpecialization<#I># | __ZN33ContainsSpecializedMemberFunction20memberSpecializationIiEEvv | Def,RelChild,RelSpecialization | rel: 2
+// CHECK-NEXT: RelChild
+// CHECK-NEXT: RelSpecialization | memberSpecialization | c:@S@ContainsSpecializedMemberFunction@FT@>1#TmemberSpecialization#v#
+}
+
+template<typename T>
+class SpecializationDecl;
+// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Ref | rel: 0
+
+template<typename T>
+class SpecializationDecl { };
+// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Def | rel: 0
+
+template<>
+class SpecializationDecl<int>;
+// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Ref | rel: 0
+
+template<>
+class SpecializationDecl<int> { };
+// CHECK: [[@LINE-1]]:7 | class(Gen,TS)/C++ | SpecializationDecl | c:@S@SpecializationDecl>#I | <no-cgname> | Def,RelSpecialization | rel: 1
+// CHECK-NEXT: RelSpecialization | SpecializationDecl | c:@ST>1#T@SpecializationDecl
+// CHECK-NEXT: [[@LINE-3]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Ref | rel: 0
+
+template<typename T>
+class PartialSpecilizationClass<Cls, T>;
+// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass | <no-cgname> | Ref | rel: 0
+// CHECK-NEXT: [[@LINE-2]]:33 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref | rel: 0
+
+template<>
+class PartialSpecilizationClass<Cls, Cls> : Cls { };
+// CHECK: [[@LINE-1]]:7 | class(Gen,TS)/C++ | PartialSpecilizationClass | c:@S@PartialSpecilizationClass>#$@S@Cls#S0_ | <no-cgname> | Def,RelSpecialization | rel: 1
+// CHECK-NEXT: RelSpecialization | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass
+// CHECK-NEXT: [[@LINE-3]]:45 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref,RelBase,RelCont | rel: 1
+// CHECK-NEXT: RelBase,RelCont | PartialSpecilizationClass | c:@S@PartialSpecilizationClass>#$@S@Cls#S0_
+// CHECK-NEXT: [[@LINE-5]]:7 | class(Gen)/C++ | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass | <no-cgname> | Ref | rel: 0
+// CHECK-NEXT: [[@LINE-6]]:33 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref | rel: 0
+// CHECK-NEXT: [[@LINE-7]]:38 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref | rel: 0
diff --git a/test/Index/Core/index-source.m b/test/Index/Core/index-source.m
index f48891417f5f..a2eef89deea7 100644
--- a/test/Index/Core/index-source.m
+++ b/test/Index/Core/index-source.m
@@ -353,7 +353,8 @@ typedef MyGenCls<Base *><MyEnumerating> MyEnumerator;
 
 typedef NS_ENUM(AnotherEnum, int) {
 // CHECK-NOT: [[@LINE-1]]:17 | type-alias/C | AnotherEnum |
-// CHECK: [[@LINE-2]]:17 | enum/C | AnotherEnum | [[AnotherEnum_USR:.*]] | {{.*}} | Ref,RelCont | rel: 1
+// CHECK-NOT: [[@LINE-2]]:17 | {{.*}} | Ref
+// CHECK: [[@LINE-3]]:17 | enum/C | AnotherEnum | [[AnotherEnum_USR:.*]] | {{.*}} | Def | rel: 0
   AnotherEnumFirst = 0,
   AnotherEnumSecond = 1,
   AnotherEnumThird = 2,
diff --git a/test/Index/index-refs.cpp b/test/Index/index-refs.cpp
index 5a1ba1e8b43a..d8bede0b4761 100644
--- a/test/Index/index-refs.cpp
+++ b/test/Index/index-refs.cpp
@@ -105,6 +105,7 @@ int ginitlist[] = {EnumVal};
 // CHECK:      [indexDeclaration]: kind: c++-class-template | name: TS | {{.*}} | loc: 47:8
 // CHECK-NEXT: [indexDeclaration]: kind: struct-template-partial-spec | name: TS | USR: c:@SP>1#T@TS>#t0.0#I | {{.*}} | loc: 50:8
 // CHECK-NEXT: [indexDeclaration]: kind: typedef | name: MyInt | USR: c:index-refs.cpp@SP>1#T@TS>#t0.0#I@T@MyInt | {{.*}} | loc: 51:15 | semantic-container: [TS:50:8] | lexical-container: [TS:50:8]
+// CHECK-NEXT: [indexEntityReference]: kind: c++-class-template | name: TS | USR: c:@ST>2#T#T@TS | lang: C++ | cursor: TemplateRef=TS:47:8 | loc: 50:8 | <parent>:: <<NULL>> | container: [TU] | refkind: direct
 /* when indexing implicit instantiations
   [indexDeclaration]: kind: struct-template-spec | name: TS | USR: c:@S@TS>#I | {{.*}} | loc: 50:8
   [indexDeclaration]: kind: typedef | name: MyInt | USR: c:index-refs.cpp@593@S@TS>#I@T@MyInt | {{.*}} | loc: 51:15 | semantic-container: [TS:50:8] | lexical-container: [TS:50:8]
diff --git a/test/Index/print-type.cpp b/test/Index/print-type.cpp
index 108ba53c80b0..ff150f7dd613 100644
--- a/test/Index/print-type.cpp
+++ b/test/Index/print-type.cpp
@@ -71,6 +71,11 @@ struct Specialization<int>;
 Specialization<Specialization<bool>& > templRefParam;
 auto autoTemplRefParam = templRefParam;
 
+template<typename T, typename U = int>
+struct DefaultedTypeExample {};
+
+typedef DefaultedTypeExample<int> DefaultedTypeAlias;
+
 // RUN: c-index-test -test-print-type %s -std=c++14 | FileCheck %s
 // CHECK: Namespace=outer:1:11 (Definition) [type=] [typekind=Invalid] [isPOD=0]
 // CHECK: ClassTemplate=Foo:4:8 (Definition) [type=] [typekind=Invalid] [isPOD=0]
@@ -115,7 +120,7 @@ auto autoTemplRefParam = templRefParam;
 // CHECK: TemplateRef=Baz:9:8 [type=] [typekind=Invalid] [isPOD=0]
 // CHECK: IntegerLiteral= [type=int] [typekind=Int] [isPOD=1]
 // CHECK: TemplateRef=Foo:4:8 [type=] [typekind=Invalid] [isPOD=0]
-// CHECK: FieldDecl=qux:29:38 (Definition) [type=Qux<int, char *, Foo<int>, outer::inner::Bar::FooType>] [typekind=Unexposed] [templateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=Foo<int>] [typekind=Unexposed] [type=outer::inner::Bar::FooType] [typekind=Typedef]] [canonicaltype=outer::Qux<int, char *, outer::Foo<int>, int>] [canonicaltypekind=Record] [canonicaltemplateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=outer::Foo<int>] [typekind=Record] [type=int] [typekind=Int]] [isPOD=1]
+// CHECK: FieldDecl=qux:29:38 (Definition) [type=Qux<int, char *, Foo<int>, outer::inner::Bar::FooType>] [typekind=Unexposed] [templateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=outer::Foo<int>] [typekind=Record] [type=int] [typekind=Int]] [canonicaltype=outer::Qux<int, char *, outer::Foo<int>, int>] [canonicaltypekind=Record] [canonicaltemplateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=outer::Foo<int>] [typekind=Record] [type=int] [typekind=Int]] [isPOD=1]
 // CHECK: TemplateRef=Qux:12:8 [type=] [typekind=Invalid] [isPOD=0]
 // CHECK: TemplateRef=Foo:4:8 [type=] [typekind=Invalid] [isPOD=0]
 // CHECK: FunctionTemplate=tbar:36:3 [type=T (int)] [typekind=FunctionProto] [canonicaltype=type-parameter-0-0 (int)] [canonicaltypekind=FunctionProto] [resulttype=T] [resulttypekind=Unexposed] [isPOD=0]
@@ -177,3 +182,4 @@ auto autoTemplRefParam = templRefParam;
 // CHECK: VarDecl=autoTemplRefParam:72:6 (Definition) [type=Specialization<Specialization<bool> &>] [typekind=Auto] [templateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [canonicaltype=Specialization<Specialization<bool> &>] [canonicaltypekind=Record] [canonicaltemplateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [isPOD=1]
 // CHECK: UnexposedExpr=templRefParam:71:40 [type=const Specialization<Specialization<bool> &>] [typekind=Unexposed] const [templateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [canonicaltype=const Specialization<Specialization<bool> &>] [canonicaltypekind=Record] [canonicaltemplateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [isPOD=1]
 // CHECK: DeclRefExpr=templRefParam:71:40 [type=Specialization<Specialization<bool> &>] [typekind=Unexposed] [templateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [canonicaltype=Specialization<Specialization<bool> &>] [canonicaltypekind=Record] [canonicaltemplateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [isPOD=1]
+// CHECK: TypedefDecl=DefaultedTypeAlias:77:35 (Definition) [type=DefaultedTypeAlias] [typekind=Typedef] [templateargs/2= [type=int] [typekind=Int] [type=int] [typekind=Int]] [canonicaltype=DefaultedTypeExample<int, int>] [canonicaltypekind=Record] [canonicaltemplateargs/2= [type=int] [typekind=Int] [type=int] [typekind=Int]] [isPOD=0]
diff --git a/test/Modules/Inputs/objc-desig-init/A.h b/test/Modules/Inputs/objc-desig-init/A.h
new file mode 100644
index 000000000000..02e0105f3add
--- /dev/null
+++ b/test/Modules/Inputs/objc-desig-init/A.h
@@ -0,0 +1 @@
+#import "A2.h"
diff --git a/test/Modules/Inputs/objc-desig-init/A2.h b/test/Modules/Inputs/objc-desig-init/A2.h
new file mode 100644
index 000000000000..5a4ceb495b32
--- /dev/null
+++ b/test/Modules/Inputs/objc-desig-init/A2.h
@@ -0,0 +1,4 @@
+#import "Base.h"
+
+@interface A2 : Base
+@end
diff --git a/test/Modules/Inputs/objc-desig-init/Base.h b/test/Modules/Inputs/objc-desig-init/Base.h
new file mode 100644
index 000000000000..47184256b139
--- /dev/null
+++ b/test/Modules/Inputs/objc-desig-init/Base.h
@@ -0,0 +1,4 @@
+@class NSString;
+@interface Base
+- (id)initWithNibName:(NSString *)nibNameOrNil __attribute__((objc_designated_initializer));
+@end
diff --git a/test/Modules/Inputs/objc-desig-init/X.h b/test/Modules/Inputs/objc-desig-init/X.h
new file mode 100644
index 000000000000..cb46ba70a374
--- /dev/null
+++ b/test/Modules/Inputs/objc-desig-init/X.h
@@ -0,0 +1,4 @@
+#import "A2.h"
+
+@interface X : A2
+@end
diff --git a/test/Modules/Inputs/objc-desig-init/module.modulemap b/test/Modules/Inputs/objc-desig-init/module.modulemap
new file mode 100644
index 000000000000..0150efb8eb18
--- /dev/null
+++ b/test/Modules/Inputs/objc-desig-init/module.modulemap
@@ -0,0 +1,9 @@
+module Base {
+  header "Base.h"
+  export *
+}
+
+module A {
+  header "A.h"
+  export *
+}
diff --git a/test/Modules/Inputs/template-default-args/a.h b/test/Modules/Inputs/template-default-args/a.h
index 532cbc8be035..1541d5f6a1b5 100644
--- a/test/Modules/Inputs/template-default-args/a.h
+++ b/test/Modules/Inputs/template-default-args/a.h
@@ -14,3 +14,16 @@ struct FriendL {
   template<typename T> friend struct L;
 };
 END
+
+namespace DeferredLookup {
+  template<typename T, typename U = T> using X = U;
+  template<typename T> void f() { (void) X<T>(); }
+  template<typename T> int n = X<T>();
+  template<typename T> struct S { X<T> xt; enum E : int; };
+  template<typename T> enum S<T>::E : int { a = X<T>() };
+
+  namespace Indirect {
+    template<typename, bool = true> struct A {};
+    template<typename> struct B { template<typename T> using C = A<T>; };
+  }
+}
diff --git a/test/Modules/Inputs/template-default-args/d.h b/test/Modules/Inputs/template-default-args/d.h
index 5961c91a82ec..07ececc1f41d 100644
--- a/test/Modules/Inputs/template-default-args/d.h
+++ b/test/Modules/Inputs/template-default-args/d.h
@@ -4,3 +4,10 @@ struct FriendL {
   template<typename T> friend struct L;
 };
 END
+
+namespace DeferredLookup {
+  namespace Indirect {
+    template<typename, bool = true> struct A {};
+    template<typename> struct B { template<typename T> using C = A<T>; };
+  }
+}
diff --git a/test/Modules/localsubmodulevis.m b/test/Modules/localsubmodulevis.m
index f8c4b3c755eb..bf317651df96 100644
--- a/test/Modules/localsubmodulevis.m
+++ b/test/Modules/localsubmodulevis.m
@@ -1,4 +1,4 @@
-// RUN: rm -rf %t
+// RUN: rm -rf %t.a %t.b
 // RUN: %clang_cc1 -fmodules -fmodules-local-submodule-visibility -fimplicit-module-maps -fmodules-cache-path=%t.a -I %S/Inputs/preprocess -verify %s
 // RUN: %clang_cc1 -fmodules -fmodules-local-submodule-visibility -fimplicit-module-maps -fmodules-cache-path=%t.b -I %S/Inputs/preprocess -x c -verify -x c %s
 
diff --git a/test/Modules/objc-designated-init-mod.m b/test/Modules/objc-designated-init-mod.m
new file mode 100644
index 000000000000..15c25a3ec2ef
--- /dev/null
+++ b/test/Modules/objc-designated-init-mod.m
@@ -0,0 +1,17 @@
+// RUN: rm -rf %t
+// RUN: %clang_cc1 -fmodules-cache-path=%t -fmodules -fimplicit-module-maps -I %S/Inputs/objc-desig-init %s -verify
+// expected-no-diagnostics
+
+#import "X.h"
+#import "Base.h"
+#import "A.h"
+
+@implementation X
+
+- (instancetype)initWithNibName:(NSString *)nibName {
+  if ((self = [super initWithNibName:nibName])) {
+		return self;
+  }
+  return self;
+}
+@end
diff --git a/test/Modules/template-default-args.cpp b/test/Modules/template-default-args.cpp
index 9d16cbf43420..c51cb2840884 100644
--- a/test/Modules/template-default-args.cpp
+++ b/test/Modules/template-default-args.cpp
@@ -44,3 +44,20 @@ H<> h; // expected-error {{default argument of 'H' must be imported from module
 I<> i;
 L<> *l;
 END
+
+namespace DeferredLookup {
+  template<typename T, typename U = T> using X = U;
+  template<typename T> void f() { (void) X<T>(); }
+  template<typename T> int n = X<T>(); // expected-warning {{extension}}
+  template<typename T> struct S { X<T> xt; enum E : int; };
+  template<typename T> enum S<T>::E : int { a = X<T>() };
+
+  void test() {
+    f<int>();
+    n<int> = 1;
+    S<int> s;
+    S<int>::E e = S<int>::E::a;
+
+    Indirect::B<int>::C<int> indirect;
+  }
+}
diff --git a/test/OpenMP/capturing_in_templates.cpp b/test/OpenMP/capturing_in_templates.cpp
new file mode 100644
index 000000000000..a4dcbee8fab9
--- /dev/null
+++ b/test/OpenMP/capturing_in_templates.cpp
@@ -0,0 +1,30 @@
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-ibm-linux-gnu -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s
+// expected-no-diagnostics
+
+template <typename T1, typename T2>
+struct pair {
+  T1 t1;
+  T2 t2;
+  pair(T1 t1, T2 t2) : t1(t1), t2(t2) {}
+};
+
+template <typename T1, typename T2>
+pair<T1, T2> make_pair(T1 &&t1, T2 &&t2) {
+  return {t1, t2};
+}
+
+// CHECK-LABEL: @main
+int main(int argc, char **argv) {
+// CHECK: call i32 @__tgt_target(i32 -1, i8* @{{.+}}.region_id, i32 0, i8** null, i8** null, i64* null, i32* null)
+#pragma omp target
+ {
+    for (int i = 0; i < 64; ++i) {
+      for (int j = 0; j < 64; ++j) {
+        auto foo = make_pair(i * i, j * j);
+      }
+    }
+  }
+  return 0;
+}
+
+// CHECK: call {{.+}} @{{.*}}make_pair
diff --git a/test/OpenMP/distribute_parallel_for_codegen.cpp b/test/OpenMP/distribute_parallel_for_codegen.cpp
new file mode 100644
index 000000000000..62c38f188851
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_codegen.cpp
@@ -0,0 +1,2260 @@
+// Test host code gen
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+
+// RUN: %clang_cc1  -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// RUN: %clang_cc1  -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+
+template <typename T>
+T tmain() {
+  T *a, *b, *c;
+  int n = 10000;
+  int ch = 100;
+
+  // no schedule clauses
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  // dist_schedule: static no chunk
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for dist_schedule(static)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  // dist_schedule: static chunk
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for dist_schedule(static, ch)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  // schedule: static no chunk
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for schedule(static)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  // schedule: static chunk
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for schedule(static, ch)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  // schedule: dynamic no chunk
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for schedule(dynamic)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  // schedule: dynamic chunk
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for schedule(dynamic, ch)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+  }
+
+  return T();
+}
+
+int main() {
+  double *a, *b, *c;
+  int n = 10000;
+  int ch = 100;
+
+#ifdef LAMBDA
+  // LAMBDA-LABEL: @main
+  // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
+  [&]() {
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_1:@.+]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_2:@.+]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_3:@.+]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_4:@.+]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_5:@.+]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_6:@.+]](
+
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN_7:@.+]](
+
+    // no schedule clauses
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_1]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_1]](
+      // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca
+      // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+      // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+      // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+
+      // check EUB for distribute
+      // LAMBDA-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+      // LAMBDA: [[NUM_IT_1:%.+]] = load{{.+}},
+      // LAMBDA-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+      // LAMBDA: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+      // LAMBDA-DAG: [[EUB_TRUE]]:
+      // LAMBDA: [[NUM_IT_2:%.+]] = load{{.+}},
+      // LAMBDA: br label %[[EUB_END:.+]]
+      // LAMBDA-DAG: [[EUB_FALSE]]:
+      // LAMBDA: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+      // LAMBDA: br label %[[EUB_END]]
+      // LAMBDA-DAG: [[EUB_END]]:
+      // LAMBDA-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+      // LAMBDA: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+      // initialize omp.iv
+      // LAMBDA: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+      // LAMBDA: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_JUMP_BACK:.+]]
+
+      // check exit condition
+      // LAMBDA: [[OMP_JUMP_BACK]]:
+      // LAMBDA-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+      // LAMBDA: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+      // LAMBDA: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]]
+
+      // check that PrevLB and PrevUB are passed to the 'for'
+      // LAMBDA: [[DIST_BODY]]:
+      // LAMBDA-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to
+      // LAMBDA-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to
+      // check that distlb and distub are properly passed to fork_call
+      // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+      // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+      // LAMBDA: br label %[[DIST_INC:.+]]
+
+      // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch
+      // LAMBDA: [[DIST_INC]]:
+      // LAMBDA-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+      // LAMBDA: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]]
+      // LAMBDA: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_JUMP_BACK]]
+
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+
+      // implementation of 'parallel for'
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+      // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+      // initialize lb and ub to PrevLB and PrevUB
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+      // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+      // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+      // In this case we use EUB
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+      // LAMBDA-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+      // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+      // LAMBDA: [[PF_EUB_TRUE]]:
+      // LAMBDA: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+      // LAMBDA: br label %[[PF_EUB_END:.+]]
+      // LAMBDA-DAG: [[PF_EUB_FALSE]]:
+      // LAMBDA: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA: br label %[[PF_EUB_END]]
+      // LAMBDA-DAG: [[PF_EUB_END]]:
+      // LAMBDA-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+      // LAMBDA: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+      // initialize omp.iv
+      // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+      // check exit condition
+      // LAMBDA: [[OMP_PF_JUMP_BACK]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+      // LAMBDA: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+      // check that PrevLB and PrevUB are passed to the 'for'
+      // LAMBDA: [[PF_BODY]]:
+      // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label {{.+}}
+
+      // check stride 1 for 'for' in 'distribute parallel for'
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+      // LAMBDA: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_JUMP_BACK]]
+
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+
+    // dist_schedule: static no chunk (same sa default - no dist_schedule)
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_2]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_2:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for dist_schedule(static)
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_2]](
+      // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca
+      // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+      // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+      // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+
+      // check EUB for distribute
+      // LAMBDA-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+      // LAMBDA: [[NUM_IT_1:%.+]] = load{{.+}},
+      // LAMBDA-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+      // LAMBDA: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+      // LAMBDA-DAG: [[EUB_TRUE]]:
+      // LAMBDA: [[NUM_IT_2:%.+]] = load{{.+}},
+      // LAMBDA: br label %[[EUB_END:.+]]
+      // LAMBDA-DAG: [[EUB_FALSE]]:
+      // LAMBDA: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+      // LAMBDA: br label %[[EUB_END]]
+      // LAMBDA-DAG: [[EUB_END]]:
+      // LAMBDA-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+      // LAMBDA: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+      // initialize omp.iv
+      // LAMBDA: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+      // LAMBDA: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_JUMP_BACK:.+]]
+
+      // check exit condition
+      // LAMBDA: [[OMP_JUMP_BACK]]:
+      // LAMBDA-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+      // LAMBDA: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+      // LAMBDA: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]]
+
+      // check that PrevLB and PrevUB are passed to the 'for'
+      // LAMBDA: [[DIST_BODY]]:
+      // LAMBDA-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to
+      // LAMBDA-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to
+      // check that distlb and distub are properly passed to fork_call
+      // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+      // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+      // LAMBDA: br label %[[DIST_INC:.+]]
+
+      // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch
+      // LAMBDA: [[DIST_INC]]:
+      // LAMBDA-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+      // LAMBDA: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]]
+      // LAMBDA: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_JUMP_BACK]]
+
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+
+      // implementation of 'parallel for'
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_2]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+      // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+      // initialize lb and ub to PrevLB and PrevUB
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+      // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+      // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+      // In this case we use EUB
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+      // LAMBDA-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+      // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+      // LAMBDA: [[PF_EUB_TRUE]]:
+      // LAMBDA: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+      // LAMBDA: br label %[[PF_EUB_END:.+]]
+      // LAMBDA-DAG: [[PF_EUB_FALSE]]:
+      // LAMBDA: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA: br label %[[PF_EUB_END]]
+      // LAMBDA-DAG: [[PF_EUB_END]]:
+      // LAMBDA-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+      // LAMBDA: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+      // initialize omp.iv
+      // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+      // check exit condition
+      // LAMBDA: [[OMP_PF_JUMP_BACK]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+      // LAMBDA: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+      // check that PrevLB and PrevUB are passed to the 'for'
+      // LAMBDA: [[PF_BODY]]:
+      // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label {{.+}}
+
+      // check stride 1 for 'for' in 'distribute parallel for'
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+      // LAMBDA: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_JUMP_BACK]]
+
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+
+    // dist_schedule: static chunk
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_3]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for dist_schedule(static, ch)
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_3]](
+      // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca
+      // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+      // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+      // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+      // unlike the previous tests, in this one we have a outer and inner loop for 'distribute'
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 91,
+      // LAMBDA: br label %[[DIST_OUTER_LOOP_HEADER:.+]]
+
+      // LAMBDA: [[DIST_OUTER_LOOP_HEADER]]:
+      // check EUB for distribute
+      // LAMBDA-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+      // LAMBDA: [[NUM_IT_1:%.+]] = load{{.+}},
+      // LAMBDA-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+      // LAMBDA: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+      // LAMBDA-DAG: [[EUB_TRUE]]:
+      // LAMBDA: [[NUM_IT_2:%.+]] = load{{.+}},
+      // LAMBDA: br label %[[EUB_END:.+]]
+      // LAMBDA-DAG: [[EUB_FALSE]]:
+      // LAMBDA: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+      // LAMBDA: br label %[[EUB_END]]
+      // LAMBDA-DAG: [[EUB_END]]:
+      // LAMBDA-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+      // LAMBDA: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+      // initialize omp.iv
+      // LAMBDA: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+      // LAMBDA: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+
+      // check exit condition
+      // LAMBDA-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+      // LAMBDA: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+      // LAMBDA: br {{.+}} [[CMP_IV_UB]], label %[[DIST_OUTER_LOOP_BODY:.+]], label %[[DIST_OUTER_LOOP_END:.+]]
+
+      // LAMBDA: [[DIST_OUTER_LOOP_BODY]]:
+      // LAMBDA: br label %[[DIST_INNER_LOOP_HEADER:.+]]
+
+      // LAMBDA: [[DIST_INNER_LOOP_HEADER]]:
+      // LAMBDA-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}} [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_UB_VAL_4:%.+]] = load {{.+}} [[OMP_UB]],
+      // LAMBDA: [[CMP_IV_UB_2:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_2]], [[OMP_UB_VAL_4]]
+      // LAMBDA: br{{.+}} [[CMP_IV_UB_2]], label %[[DIST_INNER_LOOP_BODY:.+]], label %[[DIST_INNER_LOOP_END:.+]]
+
+      // check that PrevLB and PrevUB are passed to the 'for'
+      // LAMBDA: [[DIST_INNER_LOOP_BODY]]:
+      // LAMBDA-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+      // check that distlb and distub are properly passed to fork_call
+      // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+      // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+      // LAMBDA: br label %[[DIST_INNER_LOOP_INC:.+]]
+
+      // check DistInc
+      // LAMBDA: [[DIST_INNER_LOOP_INC]]:
+      // LAMBDA-DAG: [[OMP_IV_VAL_3:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+      // LAMBDA: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_3]], [[OMP_ST_VAL_1]]
+      // LAMBDA: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[DIST_INNER_LOOP_HEADER]]
+
+      // LAMBDA: [[DIST_INNER_LOOP_END]]:
+      // LAMBDA: br label %[[DIST_OUTER_LOOP_INC:.+]]
+
+      // LAMBDA: [[DIST_OUTER_LOOP_INC]]:
+      // check NextLB and NextUB
+      // LAMBDA-DAG: [[OMP_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_LB]],
+      // LAMBDA-DAG: [[OMP_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]],
+      // LAMBDA-DAG: [[OMP_LB_NEXT:%.+]] = add{{.+}} [[OMP_LB_VAL_2]], [[OMP_ST_VAL_2]]
+      // LAMBDA: store{{.+}} [[OMP_LB_NEXT]], {{.+}}* [[OMP_LB]],
+      // LAMBDA-DAG: [[OMP_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_UB]],
+      // LAMBDA-DAG: [[OMP_ST_VAL_3:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]],
+      // LAMBDA-DAG: [[OMP_UB_NEXT:%.+]] = add{{.+}} [[OMP_UB_VAL_5]], [[OMP_ST_VAL_3]]
+      // LAMBDA: store{{.+}} [[OMP_UB_NEXT]], {{.+}}* [[OMP_UB]],
+      // LAMBDA: br label %[[DIST_OUTER_LOOP_HEADER]]
+
+      // outer loop exit
+      // LAMBDA: [[DIST_OUTER_LOOP_END]]:
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+
+      // skip implementation of 'parallel for': using default scheduling and was tested above
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+
+    // schedule: static no chunk
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_4]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_4:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for schedule(static)
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_4]](
+      // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca
+      // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+      // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+      // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+      // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_4:@.+]] to {{.+}},
+      // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+      // LAMBDA: ret
+
+      // 'parallel for' implementation is the same as the case without schedule clase (static no chunk is the default)
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_4]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+      // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+      // initialize lb and ub to PrevLB and PrevUB
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+      // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+      // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+      // In this case we use EUB
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+      // LAMBDA-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+      // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+      // LAMBDA: [[PF_EUB_TRUE]]:
+      // LAMBDA: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+      // LAMBDA: br label %[[PF_EUB_END:.+]]
+      // LAMBDA-DAG: [[PF_EUB_FALSE]]:
+      // LAMBDA: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA: br label %[[PF_EUB_END]]
+      // LAMBDA-DAG: [[PF_EUB_END]]:
+      // LAMBDA-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+      // LAMBDA: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+      // initialize omp.iv
+      // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+      // check exit condition
+      // LAMBDA: [[OMP_PF_JUMP_BACK]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+      // LAMBDA: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+      // check that PrevLB and PrevUB are passed to the 'for'
+      // LAMBDA: [[PF_BODY]]:
+      // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label {{.+}}
+
+      // check stride 1 for 'for' in 'distribute parallel for'
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+      // LAMBDA: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_JUMP_BACK]]
+
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+
+    // schedule: static chunk
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_5]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_5:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for schedule(static, ch)
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_5]](
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+      // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_5:@.+]] to {{.+}},
+      // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+      // LAMBDA: ret
+
+      // 'parallel for' implementation using outer and inner loops and PrevEUB
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_5]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}})
+      // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+      // initialize lb and ub to PrevLB and PrevUB
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+      // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 33, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+      // check PrevEUB (using PrevUB instead of NumIt as upper bound)
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_HEADER]]:
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA-64-DAG: [[OMP_PF_UB_VAL_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_1]] to
+      // LAMBDA: [[PF_PREV_UB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_CONV]], [[PF_PREV_UB_VAL_1]]
+      // LAMBDA-32-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_PREV_UB_VAL_1]]
+      // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+      // LAMBDA: [[PF_EUB_TRUE]]:
+      // LAMBDA: [[PF_PREV_UB_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA: br label %[[PF_EUB_END:.+]]
+      // LAMBDA-DAG: [[PF_EUB_FALSE]]:
+      // LAMBDA: [[OMP_PF_UB_VAL_2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+      // LAMBDA-64: [[OMP_PF_UB_VAL_2_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_2]] to
+      // LAMBDA: br label %[[PF_EUB_END]]
+      // LAMBDA-DAG: [[PF_EUB_END]]:
+      // LAMBDA-64-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2_CONV]], %[[PF_EUB_FALSE]] ]
+      // LAMBDA-32-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2]], %[[PF_EUB_FALSE]] ]
+      // LAMBDA-64-DAG: [[PF_EUB_RES_CONV:%.+]] = trunc{{.+}} [[PF_EUB_RES]] to
+      // LAMBDA-64: store{{.+}} [[PF_EUB_RES_CONV]],{{.+}}  [[OMP_PF_UB]],
+      // LAMBDA-32: store{{.+}} [[PF_EUB_RES]], {{.+}} [[OMP_PF_UB]],
+
+      // initialize omp.iv (IV = LB)
+      // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+
+      // outer loop: while (IV < UB) {
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB_1:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+      // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_1]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_BODY]]:
+      // LAMBDA: br label %[[OMP_PF_INNER_FOR_HEADER:.+]]
+
+      // LAMBDA: [[OMP_PF_INNER_FOR_HEADER]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+      // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+      // LAMBDA: [[OMP_PF_INNER_LOOP_BODY]]:
+      // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // skip body branch
+      // LAMBDA: br{{.+}}
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+      // IV = IV + 1 and inner loop latch
+      // LAMBDA: [[OMP_PF_INNER_LOOP_INC]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+      // LAMBDA-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_PF_INNER_FOR_HEADER]]
+
+      // check NextLB and NextUB
+      // LAMBDA: [[OMP_PF_INNER_LOOP_END]]:
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_INC]]:
+      // LAMBDA-DAG: [[OMP_PF_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA-DAG: [[OMP_PF_ST_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]],
+      // LAMBDA-DAG: [[OMP_PF_LB_NEXT:%.+]] = add{{.+}} [[OMP_PF_LB_VAL_2]], [[OMP_PF_ST_VAL_1]]
+      // LAMBDA: store{{.+}} [[OMP_PF_LB_NEXT]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+      // LAMBDA-DAG: [[OMP_PF_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]],
+      // LAMBDA-DAG: [[OMP_PF_UB_NEXT:%.+]] = add{{.+}} [[OMP_PF_UB_VAL_5]], [[OMP_PF_ST_VAL_2]]
+      // LAMBDA: store{{.+}} [[OMP_PF_UB_NEXT]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_END]]:
+      // LAMBDA-DAG: call void @__kmpc_for_static_fini(
+      // LAMBDA: ret
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+
+    // schedule: dynamic no chunk
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_6]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_6:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for schedule(dynamic)
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_6]](
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+      // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_6:@.+]] to {{.+}},
+      // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+      // LAMBDA: ret
+
+      // 'parallel for' implementation using outer and inner loops and PrevEUB
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_6]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+      // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+      // initialize lb and ub to PrevLB and PrevUB
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+      // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+      // LAMBDA: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}})
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_HEADER]]:
+      // LAMBDA: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]])
+      // LAMBDA: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0
+      // LAMBDA: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+      // initialize omp.iv (IV = LB)
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_BODY]]:
+      // LAMBDA-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]]
+
+      // LAMBDA: [[OMP_PF_INNER_LOOP_HEADER]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+      // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+      // LAMBDA: [[OMP_PF_INNER_LOOP_BODY]]:
+      // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // skip body branch
+      // LAMBDA: br{{.+}}
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+      // IV = IV + 1 and inner loop latch
+      // LAMBDA: [[OMP_PF_INNER_LOOP_INC]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+      // LAMBDA-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER]]
+
+      // check NextLB and NextUB
+      // LAMBDA: [[OMP_PF_INNER_LOOP_END]]:
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_INC]]:
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_END]]:
+      // LAMBDA: ret
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+
+    // schedule: dynamic chunk
+    #pragma omp target
+    #pragma omp teams
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_7]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_7:@.+]] to {{.+}})
+
+    #pragma omp distribute parallel for schedule(dynamic, ch)
+    for (int i = 0; i < n; ++i) {
+      a[i] = b[i] + c[i];
+      // LAMBDA: define{{.+}} void [[OMP_OUTLINED_7]](
+      // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+      // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_7:@.+]] to {{.+}},
+      // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+      // LAMBDA: ret
+
+      // 'parallel for' implementation using outer and inner loops and PrevEUB
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_7]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}})
+      // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+      // LAMBDA-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+      // initialize lb and ub to PrevLB and PrevUB
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+      // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+      // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+      // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+      // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+      // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+      // LAMBDA: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}})
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_HEADER]]:
+      // LAMBDA: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]])
+      // LAMBDA: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0
+      // LAMBDA: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+      // initialize omp.iv (IV = LB)
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_BODY]]:
+      // LAMBDA-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+      // LAMBDA-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]]
+
+      // LAMBDA: [[OMP_PF_INNER_LOOP_HEADER]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // LAMBDA-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+      // LAMBDA: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+      // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+      // LAMBDA: [[OMP_PF_INNER_LOOP_BODY]]:
+      // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+      // skip body branch
+      // LAMBDA: br{{.+}}
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+      // IV = IV + 1 and inner loop latch
+      // LAMBDA: [[OMP_PF_INNER_LOOP_INC]]:
+      // LAMBDA-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+      // LAMBDA-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+      // LAMBDA-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+      // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER]]
+
+      // check NextLB and NextUB
+      // LAMBDA: [[OMP_PF_INNER_LOOP_END]]:
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_INC]]:
+      // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+      // LAMBDA: [[OMP_PF_OUTER_LOOP_END]]:
+      // LAMBDA: ret
+      [&]() {
+	a[i] = b[i] + c[i];
+      }();
+    }
+  }();
+  return 0;
+#else
+  // CHECK-LABEL: @main
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_1:@.+]](
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_2:@.+]](
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_3:@.+]](
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_4:@.+]](
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_5:@.+]](
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_6:@.+]](
+
+  // CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+  // CHECK: call void [[OFFLOADING_FUN_7:@.+]](
+
+  // CHECK: call{{.+}} [[TMAIN:@.+]]()
+
+  // no schedule clauses
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define internal void [[OFFLOADING_FUN_1]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_1]](
+    // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+    // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+    // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+    // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+
+    // check EUB for distribute
+    // CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+    // CHECK: [[NUM_IT_1:%.+]] = load{{.+}},
+    // CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+    // CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+    // CHECK-DAG: [[EUB_TRUE]]:
+    // CHECK: [[NUM_IT_2:%.+]] = load{{.+}},
+    // CHECK: br label %[[EUB_END:.+]]
+    // CHECK-DAG: [[EUB_FALSE]]:
+    // CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+    // CHECK: br label %[[EUB_END]]
+    // CHECK-DAG: [[EUB_END]]:
+    // CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+    // CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+    // initialize omp.iv
+    // CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+    // CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_JUMP_BACK:.+]]
+
+    // check exit condition
+    // CHECK: [[OMP_JUMP_BACK]]:
+    // CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+    // CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+    // CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+    // CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]]
+
+    // check that PrevLB and PrevUB are passed to the 'for'
+    // CHECK: [[DIST_BODY]]:
+    // CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+    // check that distlb and distub are properly passed to fork_call
+    // CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+    // CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+    // CHECK: br label %[[DIST_INC:.+]]
+
+    // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch
+    // CHECK: [[DIST_INC]]:
+    // CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+    // CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+    // CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]]
+    // CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_JUMP_BACK]]
+
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+
+    // implementation of 'parallel for'
+    // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+    // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+    // initialize lb and ub to PrevLB and PrevUB
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+    // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+    // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+    // In this case we use EUB
+    // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+    // CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+    // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+    // CHECK: [[PF_EUB_TRUE]]:
+    // CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+    // CHECK: br label %[[PF_EUB_END:.+]]
+    // CHECK-DAG: [[PF_EUB_FALSE]]:
+    // CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK: br label %[[PF_EUB_END]]
+    // CHECK-DAG: [[PF_EUB_END]]:
+    // CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+    // CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+    // initialize omp.iv
+    // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+    // check exit condition
+    // CHECK: [[OMP_PF_JUMP_BACK]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+    // CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+    // check that PrevLB and PrevUB are passed to the 'for'
+    // CHECK: [[PF_BODY]]:
+    // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label {{.+}}
+
+    // check stride 1 for 'for' in 'distribute parallel for'
+    // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+    // CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_JUMP_BACK]]
+
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+  }
+
+  // dist_schedule: static no chunk
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define{{.+}} void [[OFFLOADING_FUN_2]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_2:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for dist_schedule(static)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_2]](
+    // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+    // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+    // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+    // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+
+    // check EUB for distribute
+    // CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+    // CHECK: [[NUM_IT_1:%.+]] = load{{.+}},
+    // CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+    // CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+    // CHECK-DAG: [[EUB_TRUE]]:
+    // CHECK: [[NUM_IT_2:%.+]] = load{{.+}},
+    // CHECK: br label %[[EUB_END:.+]]
+    // CHECK-DAG: [[EUB_FALSE]]:
+    // CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+    // CHECK: br label %[[EUB_END]]
+    // CHECK-DAG: [[EUB_END]]:
+    // CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+    // CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+    // initialize omp.iv
+    // CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+    // CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_JUMP_BACK:.+]]
+
+    // check exit condition
+    // CHECK: [[OMP_JUMP_BACK]]:
+    // CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+    // CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+    // CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+    // CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]]
+
+    // check that PrevLB and PrevUB are passed to the 'for'
+    // CHECK: [[DIST_BODY]]:
+    // CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+    // check that distlb and distub are properly passed to fork_call
+    // CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+    // CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+    // CHECK: br label %[[DIST_INC:.+]]
+
+    // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch
+    // CHECK: [[DIST_INC]]:
+    // CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+    // CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+    // CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]]
+    // CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_JUMP_BACK]]
+
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+
+    // implementation of 'parallel for'
+    // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_2]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+    // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+    // initialize lb and ub to PrevLB and PrevUB
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+    // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+    // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+    // In this case we use EUB
+    // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+    // CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+    // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+    // CHECK: [[PF_EUB_TRUE]]:
+    // CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+    // CHECK: br label %[[PF_EUB_END:.+]]
+    // CHECK-DAG: [[PF_EUB_FALSE]]:
+    // CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK: br label %[[PF_EUB_END]]
+    // CHECK-DAG: [[PF_EUB_END]]:
+    // CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+    // CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+    // initialize omp.iv
+    // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+    // check exit condition
+    // CHECK: [[OMP_PF_JUMP_BACK]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+    // CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+    // check that PrevLB and PrevUB are passed to the 'for'
+    // CHECK: [[PF_BODY]]:
+    // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label {{.+}}
+
+    // check stride 1 for 'for' in 'distribute parallel for'
+    // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+    // CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_JUMP_BACK]]
+
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+  }
+
+  // dist_schedule: static chunk
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define{{.+}} void [[OFFLOADING_FUN_3]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for dist_schedule(static, ch)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_3]](
+    // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+    // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+    // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+    // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+    // unlike the previous tests, in this one we have a outer and inner loop for 'distribute'
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 91,
+    // CHECK: br label %[[DIST_OUTER_LOOP_HEADER:.+]]
+
+    // CHECK: [[DIST_OUTER_LOOP_HEADER]]:
+    // check EUB for distribute
+    // CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+    // CHECK: [[NUM_IT_1:%.+]] = load{{.+}},
+    // CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+    // CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+    // CHECK-DAG: [[EUB_TRUE]]:
+    // CHECK: [[NUM_IT_2:%.+]] = load{{.+}},
+    // CHECK: br label %[[EUB_END:.+]]
+    // CHECK-DAG: [[EUB_FALSE]]:
+    // CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+    // CHECK: br label %[[EUB_END]]
+    // CHECK-DAG: [[EUB_END]]:
+    // CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+    // CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+    // initialize omp.iv
+    // CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+    // CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+
+    // check exit condition
+    // CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+    // CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+    // CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+    // CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_OUTER_LOOP_BODY:.+]], label %[[DIST_OUTER_LOOP_END:.+]]
+
+    // CHECK: [[DIST_OUTER_LOOP_BODY]]:
+    // CHECK: br label %[[DIST_INNER_LOOP_HEADER:.+]]
+
+    // CHECK: [[DIST_INNER_LOOP_HEADER]]:
+    // CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}} [[OMP_IV]],
+    // CHECK-DAG: [[OMP_UB_VAL_4:%.+]] = load {{.+}} [[OMP_UB]],
+    // CHECK: [[CMP_IV_UB_2:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_2]], [[OMP_UB_VAL_4]]
+    // CHECK: br{{.+}} [[CMP_IV_UB_2]], label %[[DIST_INNER_LOOP_BODY:.+]], label %[[DIST_INNER_LOOP_END:.+]]
+
+    // check that PrevLB and PrevUB are passed to the 'for'
+    // CHECK: [[DIST_INNER_LOOP_BODY]]:
+    // CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+    // check that distlb and distub are properly passed to fork_call
+    // CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+    // CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+    // CHECK: br label %[[DIST_INNER_LOOP_INC:.+]]
+
+    // check DistInc
+    // CHECK: [[DIST_INNER_LOOP_INC]]:
+    // CHECK-DAG: [[OMP_IV_VAL_3:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+    // CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+    // CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_3]], [[OMP_ST_VAL_1]]
+    // CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[DIST_INNER_LOOP_HEADER]]
+
+    // CHECK: [[DIST_INNER_LOOP_END]]:
+    // CHECK: br label %[[DIST_OUTER_LOOP_INC:.+]]
+
+    // CHECK: [[DIST_OUTER_LOOP_INC]]:
+    // check NextLB and NextUB
+    // CHECK-DAG: [[OMP_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_LB]],
+    // CHECK-DAG: [[OMP_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]],
+    // CHECK-DAG: [[OMP_LB_NEXT:%.+]] = add{{.+}} [[OMP_LB_VAL_2]], [[OMP_ST_VAL_2]]
+    // CHECK: store{{.+}} [[OMP_LB_NEXT]], {{.+}}* [[OMP_LB]],
+    // CHECK-DAG: [[OMP_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_UB]],
+    // CHECK-DAG: [[OMP_ST_VAL_3:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]],
+    // CHECK-DAG: [[OMP_UB_NEXT:%.+]] = add{{.+}} [[OMP_UB_VAL_5]], [[OMP_ST_VAL_3]]
+    // CHECK: store{{.+}} [[OMP_UB_NEXT]], {{.+}}* [[OMP_UB]],
+    // CHECK: br label %[[DIST_OUTER_LOOP_HEADER]]
+
+    // outer loop exit
+    // CHECK: [[DIST_OUTER_LOOP_END]]:
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+
+    // skip implementation of 'parallel for': using default scheduling and was tested above
+  }
+
+  // schedule: static no chunk
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define{{.+}} void [[OFFLOADING_FUN_4]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_4:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for schedule(static)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_4]](
+    // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+    // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+    // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+    // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+    // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_4:@.+]] to {{.+}},
+    // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+    // CHECK: ret
+
+    // 'parallel for' implementation is the same as the case without schedule clase (static no chunk is the default)
+    // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_4]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+    // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+    // initialize lb and ub to PrevLB and PrevUB
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+    // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+    // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+    // In this case we use EUB
+    // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+    // CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+    // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+    // CHECK: [[PF_EUB_TRUE]]:
+    // CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+    // CHECK: br label %[[PF_EUB_END:.+]]
+    // CHECK-DAG: [[PF_EUB_FALSE]]:
+    // CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK: br label %[[PF_EUB_END]]
+    // CHECK-DAG: [[PF_EUB_END]]:
+    // CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+    // CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+    // initialize omp.iv
+    // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+    // check exit condition
+    // CHECK: [[OMP_PF_JUMP_BACK]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+    // CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+    // check that PrevLB and PrevUB are passed to the 'for'
+    // CHECK: [[PF_BODY]]:
+    // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label {{.+}}
+
+    // check stride 1 for 'for' in 'distribute parallel for'
+    // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+    // CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_JUMP_BACK]]
+
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+  }
+
+  // schedule: static chunk
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define{{.+}} void [[OFFLOADING_FUN_5]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_5:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for schedule(static, ch)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_5]](
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+    // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_5:@.+]] to {{.+}},
+    // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+    // CHECK: ret
+
+    // 'parallel for' implementation using outer and inner loops and PrevEUB
+    // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_5]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}})
+    // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+    // initialize lb and ub to PrevLB and PrevUB
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+    // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 33, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+    // check PrevEUB (using PrevUB instead of NumIt as upper bound)
+    // CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]:
+    // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK-64-DAG: [[OMP_PF_UB_VAL_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_1]] to
+    // CHECK: [[PF_PREV_UB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_CONV]], [[PF_PREV_UB_VAL_1]]
+    // CHECK-32-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_PREV_UB_VAL_1]]
+    // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+    // CHECK: [[PF_EUB_TRUE]]:
+    // CHECK: [[PF_PREV_UB_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK: br label %[[PF_EUB_END:.+]]
+    // CHECK-DAG: [[PF_EUB_FALSE]]:
+    // CHECK: [[OMP_PF_UB_VAL_2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+    // CHECK-64: [[OMP_PF_UB_VAL_2_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_2]] to
+    // CHECK: br label %[[PF_EUB_END]]
+    // CHECK-DAG: [[PF_EUB_END]]:
+    // CHECK-64-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2_CONV]], %[[PF_EUB_FALSE]] ]
+    // CHECK-32-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2]], %[[PF_EUB_FALSE]] ]
+    // CHECK-64-DAG: [[PF_EUB_RES_CONV:%.+]] = trunc{{.+}} [[PF_EUB_RES]] to
+    // CHECK-64: store{{.+}} [[PF_EUB_RES_CONV]],{{.+}}  [[OMP_PF_UB]],
+    // CHECK-32: store{{.+}} [[PF_EUB_RES]], {{.+}} [[OMP_PF_UB]],
+
+    // initialize omp.iv (IV = LB)
+    // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+
+    // outer loop: while (IV < UB) {
+    // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB_1:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+    // CHECK: br{{.+}} [[PF_CMP_IV_UB_1]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_BODY]]:
+    // CHECK: br label %[[OMP_PF_INNER_FOR_HEADER:.+]]
+
+    // CHECK: [[OMP_PF_INNER_FOR_HEADER]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+    // CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+    // CHECK: [[OMP_PF_INNER_LOOP_BODY]]:
+    // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // skip body branch
+    // CHECK: br{{.+}}
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+    // IV = IV + 1 and inner loop latch
+    // CHECK: [[OMP_PF_INNER_LOOP_INC]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+    // CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+    // CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_PF_INNER_FOR_HEADER]]
+
+    // check NextLB and NextUB
+    // CHECK: [[OMP_PF_INNER_LOOP_END]]:
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_INC]]:
+    // CHECK-DAG: [[OMP_PF_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK-DAG: [[OMP_PF_ST_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]],
+    // CHECK-DAG: [[OMP_PF_LB_NEXT:%.+]] = add{{.+}} [[OMP_PF_LB_VAL_2]], [[OMP_PF_ST_VAL_1]]
+    // CHECK: store{{.+}} [[OMP_PF_LB_NEXT]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+    // CHECK-DAG: [[OMP_PF_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]],
+    // CHECK-DAG: [[OMP_PF_UB_NEXT:%.+]] = add{{.+}} [[OMP_PF_UB_VAL_5]], [[OMP_PF_ST_VAL_2]]
+    // CHECK: store{{.+}} [[OMP_PF_UB_NEXT]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_END]]:
+    // CHECK-DAG: call void @__kmpc_for_static_fini(
+    // CHECK: ret
+  }
+
+  // schedule: dynamic no chunk
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define{{.+}} void [[OFFLOADING_FUN_6]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_6:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for schedule(dynamic)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_6]](
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+    // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_6:@.+]] to {{.+}},
+    // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+    // CHECK: ret
+
+    // 'parallel for' implementation using outer and inner loops and PrevEUB
+    // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_6]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+    // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+    // initialize lb and ub to PrevLB and PrevUB
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+    // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+    // CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}})
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]:
+    // CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]])
+    // CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0
+    // CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+    // initialize omp.iv (IV = LB)
+    // CHECK: [[OMP_PF_OUTER_LOOP_BODY]]:
+    // CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]]
+
+    // CHECK: [[OMP_PF_INNER_LOOP_HEADER]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+    // CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+    // CHECK: [[OMP_PF_INNER_LOOP_BODY]]:
+    // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // skip body branch
+    // CHECK: br{{.+}}
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+    // IV = IV + 1 and inner loop latch
+    // CHECK: [[OMP_PF_INNER_LOOP_INC]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+    // CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+    // CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]]
+
+    // check NextLB and NextUB
+    // CHECK: [[OMP_PF_INNER_LOOP_END]]:
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_INC]]:
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_END]]:
+    // CHECK: ret
+  }
+
+  // schedule: dynamic chunk
+  #pragma omp target
+  #pragma omp teams
+  // CHECK: define{{.+}} void [[OFFLOADING_FUN_7]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_7:@.+]] to {{.+}})
+
+  #pragma omp distribute parallel for schedule(dynamic, ch)
+  for (int i = 0; i < n; ++i) {
+    a[i] = b[i] + c[i];
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_7]](
+    // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+    // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_7:@.+]] to {{.+}},
+    // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+    // CHECK: ret
+
+    // 'parallel for' implementation using outer and inner loops and PrevEUB
+    // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_7]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}})
+    // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+    // CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+    // initialize lb and ub to PrevLB and PrevUB
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+    // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+    // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+    // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+    // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+    // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+    // CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+    // CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}})
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]:
+    // CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]])
+    // CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0
+    // CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+    // initialize omp.iv (IV = LB)
+    // CHECK: [[OMP_PF_OUTER_LOOP_BODY]]:
+    // CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+    // CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]]
+
+    // CHECK: [[OMP_PF_INNER_LOOP_HEADER]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+    // CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+    // CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+    // CHECK: [[OMP_PF_INNER_LOOP_BODY]]:
+    // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+    // skip body branch
+    // CHECK: br{{.+}}
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+    // IV = IV + 1 and inner loop latch
+    // CHECK: [[OMP_PF_INNER_LOOP_INC]]:
+    // CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+    // CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+    // CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+    // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]]
+
+    // check NextLB and NextUB
+    // CHECK: [[OMP_PF_INNER_LOOP_END]]:
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_INC]]:
+    // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+    // CHECK: [[OMP_PF_OUTER_LOOP_END]]:
+    // CHECK: ret
+  }
+
+  return tmain<int>();
+#endif
+}
+
+// check code
+// CHECK: define{{.+}} [[TMAIN]]()
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_1:@.+]](
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_2:@.+]](
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_3:@.+]](
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_4:@.+]](
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_5:@.+]](
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_6:@.+]](
+
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_7:@.+]](
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_1]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_1]](
+// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+
+// check EUB for distribute
+// CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+// CHECK: [[NUM_IT_1:%.+]] = load{{.+}},
+// CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+// CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+// CHECK-DAG: [[EUB_TRUE]]:
+// CHECK: [[NUM_IT_2:%.+]] = load{{.+}},
+// CHECK: br label %[[EUB_END:.+]]
+// CHECK-DAG: [[EUB_FALSE]]:
+// CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+// CHECK: br label %[[EUB_END]]
+// CHECK-DAG: [[EUB_END]]:
+// CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+// CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+// initialize omp.iv
+// CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+// CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_JUMP_BACK:.+]]
+
+// check exit condition
+// CHECK: [[OMP_JUMP_BACK]]:
+// CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+// CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+// CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+// CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]]
+
+// check that PrevLB and PrevUB are passed to the 'for'
+// CHECK: [[DIST_BODY]]:
+// CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+// CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+// CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+// check that distlb and distub are properly passed to fork_call
+// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+// CHECK: br label %[[DIST_INC:.+]]
+
+// increment by stride (distInc - 'parallel for' executes the whole chunk) and latch
+// CHECK: [[DIST_INC]]:
+// CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+// CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+// CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]]
+// CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_JUMP_BACK]]
+
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// implementation of 'parallel for'
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+// initialize lb and ub to PrevLB and PrevUB
+// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+// PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+// In this case we use EUB
+// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+// CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+// CHECK: [[PF_EUB_TRUE]]:
+// CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+// CHECK: br label %[[PF_EUB_END:.+]]
+// CHECK-DAG: [[PF_EUB_FALSE]]:
+// CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK: br label %[[PF_EUB_END]]
+// CHECK-DAG: [[PF_EUB_END]]:
+// CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+// CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+// initialize omp.iv
+// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+// check exit condition
+// CHECK: [[OMP_PF_JUMP_BACK]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+// CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+// check that PrevLB and PrevUB are passed to the 'for'
+// CHECK: [[PF_BODY]]:
+// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label {{.+}}
+
+// check stride 1 for 'for' in 'distribute parallel for'
+// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+// CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_JUMP_BACK]]
+
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_2]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_2:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_2]](
+// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+
+// check EUB for distribute
+// CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+// CHECK: [[NUM_IT_1:%.+]] = load{{.+}},
+// CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+// CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+// CHECK-DAG: [[EUB_TRUE]]:
+// CHECK: [[NUM_IT_2:%.+]] = load{{.+}},
+// CHECK: br label %[[EUB_END:.+]]
+// CHECK-DAG: [[EUB_FALSE]]:
+// CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+// CHECK: br label %[[EUB_END]]
+// CHECK-DAG: [[EUB_END]]:
+// CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+// CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+// initialize omp.iv
+// CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+// CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_JUMP_BACK:.+]]
+
+// check exit condition
+// CHECK: [[OMP_JUMP_BACK]]:
+// CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+// CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+// CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+// CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]]
+
+// check that PrevLB and PrevUB are passed to the 'for'
+// CHECK: [[DIST_BODY]]:
+// CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+// CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+// CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+// check that distlb and distub are properly passed to fork_call
+// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+// CHECK: br label %[[DIST_INC:.+]]
+
+// increment by stride (distInc - 'parallel for' executes the whole chunk) and latch
+// CHECK: [[DIST_INC]]:
+// CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+// CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+// CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]]
+// CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_JUMP_BACK]]
+
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// implementation of 'parallel for'
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_2]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+// initialize lb and ub to PrevLB and PrevUB
+// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+// PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+// In this case we use EUB
+// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+// CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+// CHECK: [[PF_EUB_TRUE]]:
+// CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+// CHECK: br label %[[PF_EUB_END:.+]]
+// CHECK-DAG: [[PF_EUB_FALSE]]:
+// CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK: br label %[[PF_EUB_END]]
+// CHECK-DAG: [[PF_EUB_END]]:
+// CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+// CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+// initialize omp.iv
+// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+// check exit condition
+// CHECK: [[OMP_PF_JUMP_BACK]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+// CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+// check that PrevLB and PrevUB are passed to the 'for'
+// CHECK: [[PF_BODY]]:
+// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label {{.+}}
+
+// check stride 1 for 'for' in 'distribute parallel for'
+// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+// CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_JUMP_BACK]]
+
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_3]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_3]](
+// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+// unlike the previous tests, in this one we have a outer and inner loop for 'distribute'
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 91,
+// CHECK: br label %[[DIST_OUTER_LOOP_HEADER:.+]]
+
+// CHECK: [[DIST_OUTER_LOOP_HEADER]]:
+// check EUB for distribute
+// CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]],
+// CHECK: [[NUM_IT_1:%.+]] = load{{.+}},
+// CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}}  [[OMP_UB_VAL_1]], [[NUM_IT_1]]
+// CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]]
+// CHECK-DAG: [[EUB_TRUE]]:
+// CHECK: [[NUM_IT_2:%.+]] = load{{.+}},
+// CHECK: br label %[[EUB_END:.+]]
+// CHECK-DAG: [[EUB_FALSE]]:
+// CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]],
+// CHECK: br label %[[EUB_END]]
+// CHECK-DAG: [[EUB_END]]:
+// CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ]
+// CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]],
+
+// initialize omp.iv
+// CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]],
+// CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]],
+
+// check exit condition
+// CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]],
+// CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]],
+// CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]]
+// CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_OUTER_LOOP_BODY:.+]], label %[[DIST_OUTER_LOOP_END:.+]]
+
+// CHECK: [[DIST_OUTER_LOOP_BODY]]:
+// CHECK: br label %[[DIST_INNER_LOOP_HEADER:.+]]
+
+// CHECK: [[DIST_INNER_LOOP_HEADER]]:
+// CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}} [[OMP_IV]],
+// CHECK-DAG: [[OMP_UB_VAL_4:%.+]] = load {{.+}} [[OMP_UB]],
+// CHECK: [[CMP_IV_UB_2:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_2]], [[OMP_UB_VAL_4]]
+// CHECK: br{{.+}} [[CMP_IV_UB_2]], label %[[DIST_INNER_LOOP_BODY:.+]], label %[[DIST_INNER_LOOP_END:.+]]
+
+// check that PrevLB and PrevUB are passed to the 'for'
+// CHECK: [[DIST_INNER_LOOP_BODY]]:
+// CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]],
+// CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]],
+// CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}}
+// check that distlb and distub are properly passed to fork_call
+// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}})
+// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}})
+// CHECK: br label %[[DIST_INNER_LOOP_INC:.+]]
+
+// check DistInc
+// CHECK: [[DIST_INNER_LOOP_INC]]:
+// CHECK-DAG: [[OMP_IV_VAL_3:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]],
+// CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]],
+// CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_3]], [[OMP_ST_VAL_1]]
+// CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[DIST_INNER_LOOP_HEADER]]
+
+// CHECK: [[DIST_INNER_LOOP_END]]:
+// CHECK: br label %[[DIST_OUTER_LOOP_INC:.+]]
+
+// CHECK: [[DIST_OUTER_LOOP_INC]]:
+// check NextLB and NextUB
+// CHECK-DAG: [[OMP_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_LB]],
+// CHECK-DAG: [[OMP_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]],
+// CHECK-DAG: [[OMP_LB_NEXT:%.+]] = add{{.+}} [[OMP_LB_VAL_2]], [[OMP_ST_VAL_2]]
+// CHECK: store{{.+}} [[OMP_LB_NEXT]], {{.+}}* [[OMP_LB]],
+// CHECK-DAG: [[OMP_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_UB]],
+// CHECK-DAG: [[OMP_ST_VAL_3:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]],
+// CHECK-DAG: [[OMP_UB_NEXT:%.+]] = add{{.+}} [[OMP_UB_VAL_5]], [[OMP_ST_VAL_3]]
+// CHECK: store{{.+}} [[OMP_UB_NEXT]], {{.+}}* [[OMP_UB]],
+// CHECK: br label %[[DIST_OUTER_LOOP_HEADER]]
+
+// outer loop exit
+// CHECK: [[DIST_OUTER_LOOP_END]]:
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// skip implementation of 'parallel for': using default scheduling and was tested above
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_4]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_4:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_4]](
+// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca
+// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca
+// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca
+// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca
+
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_4:@.+]] to {{.+}},
+// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+// CHECK: ret
+
+// 'parallel for' implementation is the same as the case without schedule clase (static no chunk is the default)
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_4]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+
+// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+
+// initialize lb and ub to PrevLB and PrevUB
+// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+
+// PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used
+// In this case we use EUB
+// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}},
+// CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]]
+// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+// CHECK: [[PF_EUB_TRUE]]:
+// CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}},
+// CHECK: br label %[[PF_EUB_END:.+]]
+// CHECK-DAG: [[PF_EUB_FALSE]]:
+// CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK: br label %[[PF_EUB_END]]
+// CHECK-DAG: [[PF_EUB_END]]:
+// CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ]
+// CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}}  [[OMP_PF_UB]],
+
+// initialize omp.iv
+// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_JUMP_BACK:.+]]
+
+// check exit condition
+// CHECK: [[OMP_PF_JUMP_BACK]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+// CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]]
+
+// check that PrevLB and PrevUB are passed to the 'for'
+// CHECK: [[PF_BODY]]:
+// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label {{.+}}
+
+// check stride 1 for 'for' in 'distribute parallel for'
+// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1
+// CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_JUMP_BACK]]
+
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_5]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_5:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_5]](
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_5:@.+]] to {{.+}},
+// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+// CHECK: ret
+
+// 'parallel for' implementation using outer and inner loops and PrevEUB
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_5]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}})
+// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+// initialize lb and ub to PrevLB and PrevUB
+// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 33, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}})
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+// check PrevEUB (using PrevUB instead of NumIt as upper bound)
+// CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]:
+// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK-64-DAG: [[OMP_PF_UB_VAL_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_1]] to
+// CHECK: [[PF_PREV_UB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_CONV]], [[PF_PREV_UB_VAL_1]]
+// CHECK-32-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_PREV_UB_VAL_1]]
+// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]]
+// CHECK: [[PF_EUB_TRUE]]:
+// CHECK: [[PF_PREV_UB_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK: br label %[[PF_EUB_END:.+]]
+// CHECK-DAG: [[PF_EUB_FALSE]]:
+// CHECK: [[OMP_PF_UB_VAL_2:%.+]] = load{{.+}} [[OMP_PF_UB]],
+// CHECK-64: [[OMP_PF_UB_VAL_2_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_2]] to
+// CHECK: br label %[[PF_EUB_END]]
+// CHECK-DAG: [[PF_EUB_END]]:
+// CHECK-64-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2_CONV]], %[[PF_EUB_FALSE]] ]
+// CHECK-32-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2]], %[[PF_EUB_FALSE]] ]
+// CHECK-64-DAG: [[PF_EUB_RES_CONV:%.+]] = trunc{{.+}} [[PF_EUB_RES]] to
+// CHECK-64: store{{.+}} [[PF_EUB_RES_CONV]],{{.+}}  [[OMP_PF_UB]],
+// CHECK-32: store{{.+}} [[PF_EUB_RES]], {{.+}} [[OMP_PF_UB]],
+
+// initialize omp.iv (IV = LB)
+// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+
+// outer loop: while (IV < UB) {
+// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB_1:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]]
+// CHECK: br{{.+}} [[PF_CMP_IV_UB_1]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_BODY]]:
+// CHECK: br label %[[OMP_PF_INNER_FOR_HEADER:.+]]
+
+// CHECK: [[OMP_PF_INNER_FOR_HEADER]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+// CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+// CHECK: [[OMP_PF_INNER_LOOP_BODY]]:
+// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// skip body branch
+// CHECK: br{{.+}}
+// CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+// IV = IV + 1 and inner loop latch
+// CHECK: [[OMP_PF_INNER_LOOP_INC]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+// CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+// CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_PF_INNER_FOR_HEADER]]
+
+// check NextLB and NextUB
+// CHECK: [[OMP_PF_INNER_LOOP_END]]:
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_INC]]:
+// CHECK-DAG: [[OMP_PF_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK-DAG: [[OMP_PF_ST_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]],
+// CHECK-DAG: [[OMP_PF_LB_NEXT:%.+]] = add{{.+}} [[OMP_PF_LB_VAL_2]], [[OMP_PF_ST_VAL_1]]
+// CHECK: store{{.+}} [[OMP_PF_LB_NEXT]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+// CHECK-DAG: [[OMP_PF_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]],
+// CHECK-DAG: [[OMP_PF_UB_NEXT:%.+]] = add{{.+}} [[OMP_PF_UB_VAL_5]], [[OMP_PF_ST_VAL_2]]
+// CHECK: store{{.+}} [[OMP_PF_UB_NEXT]], {{.+}}* [[OMP_PF_UB]],
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_END]]:
+// CHECK-DAG: call void @__kmpc_for_static_fini(
+// CHECK: ret
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_6]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_6:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_6]](
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_6:@.+]] to {{.+}},
+// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+// CHECK: ret
+
+// 'parallel for' implementation using outer and inner loops and PrevEUB
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_6]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}})
+// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+// initialize lb and ub to PrevLB and PrevUB
+// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+// CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}})
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]:
+// CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]])
+// CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0
+// CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+// initialize omp.iv (IV = LB)
+// CHECK: [[OMP_PF_OUTER_LOOP_BODY]]:
+// CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]]
+
+// CHECK: [[OMP_PF_INNER_LOOP_HEADER]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+// CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+// CHECK: [[OMP_PF_INNER_LOOP_BODY]]:
+// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// skip body branch
+// CHECK: br{{.+}}
+// CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+// IV = IV + 1 and inner loop latch
+// CHECK: [[OMP_PF_INNER_LOOP_INC]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+// CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+// CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]]
+
+// check NextLB and NextUB
+// CHECK: [[OMP_PF_INNER_LOOP_END]]:
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_INC]]:
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_END]]:
+// CHECK: ret
+
+// CHECK: define{{.+}} void [[OFFLOADING_FUN_7]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_7:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[OMP_OUTLINED_7]](
+// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92,
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_7:@.+]] to {{.+}},
+// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case
+// CHECK: ret
+
+// 'parallel for' implementation using outer and inner loops and PrevEUB
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_7]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}})
+// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}},
+// CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}},
+
+// initialize lb and ub to PrevLB and PrevUB
+// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]],
+// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]],
+// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}}
+// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]],
+// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}}
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]],
+// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]],
+// CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]],
+// CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}})
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]:
+// CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]])
+// CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0
+// CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]]
+
+// initialize omp.iv (IV = LB)
+// CHECK: [[OMP_PF_OUTER_LOOP_BODY]]:
+// CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]],
+// CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]],
+// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]]
+
+// CHECK: [[OMP_PF_INNER_LOOP_HEADER]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]],
+// CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]]
+// CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]]
+
+// CHECK: [[OMP_PF_INNER_LOOP_BODY]]:
+// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]],
+// skip body branch
+// CHECK: br{{.+}}
+// CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]]
+
+// IV = IV + 1 and inner loop latch
+// CHECK: [[OMP_PF_INNER_LOOP_INC]]:
+// CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]],
+// CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1
+// CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]],
+// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]]
+
+// check NextLB and NextUB
+// CHECK: [[OMP_PF_INNER_LOOP_END]]:
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_INC]]:
+// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]]
+
+// CHECK: [[OMP_PF_OUTER_LOOP_END]]:
+// CHECK: ret
+#endif
diff --git a/test/OpenMP/distribute_parallel_for_firstprivate_codegen.cpp b/test/OpenMP/distribute_parallel_for_firstprivate_codegen.cpp
new file mode 100644
index 000000000000..d12c0aa11cf2
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_firstprivate_codegen.cpp
@@ -0,0 +1,619 @@
+// RxUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+
+// RUN: %clang_cc1  -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+template <class T>
+struct S {
+  T f;
+  S(T a) : f(a) {}
+  S() : f() {}
+  operator T() { return T(); }
+  ~S() {}
+};
+
+// CHECK: [[S_FLOAT_TY:%.+]] = type { float }
+// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
+template <typename T>
+T tmain() {
+  S<T> test;
+  T t_var = T();
+  T vec[] = {1, 2};
+  S<T> s_arr[] = {1, 2};
+  S<T> &var = test;
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for firstprivate(t_var, vec, s_arr, s_arr, var, var)
+  for (int i = 0; i < 2; ++i) {
+    vec[i] = t_var;
+    s_arr[i] = var;
+  }
+  return T();
+}
+
+int main() {
+  static int svar;
+  volatile double g;
+  volatile double &g1 = g;
+
+  #ifdef LAMBDA
+  // LAMBDA-LABEL: @main
+  // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
+  [&]() {
+    static float sfvar;
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN:@.+]](
+
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED:@.+]] to {{.+}})
+    #pragma omp target
+    #pragma omp teams
+    #pragma omp distribute parallel for firstprivate(g, g1, svar, sfvar)
+    for (int i = 0; i < 2; ++i) {
+      // LAMBDA-64: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i{{[0-9]+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]])
+      // LAMBDA-32: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, double* {{.+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]])
+
+      // addr alloca's
+      // LAMBDA-64: [[G_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA-32: [[G_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[G1_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[G1_REF:%.+]] = alloca double*,
+      // LAMBDA: [[TMP:%.+]] = alloca double*,
+
+      // private alloca's
+      // LAMBDA: [[G_PRIV:%.+]] = alloca double,
+      // LAMBDA: [[G1_PRIV:%.+]] = alloca double,
+      // LAMBDA: [[TMP_PRIV:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_PRIV:%.+]] = alloca float,
+
+      // transfer input parameters into addr alloca's
+      // LAMBDA-DAG: store {{.+}} [[G_IN]], {{.+}} [[G_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[G1_IN]], {{.+}} [[G1_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[SFVAR_IN]], {{.+}} [[SFVAR_ADDR]],
+
+      // init private alloca's with addr alloca's
+      // g
+      // LAMBDA-64-DAG: [[G_CONV:%.+]] = bitcast {{.+}}* [[G_ADDR]] to
+      // LAMBDA-32-DAG: [[G_CONV:%.+]] = load {{.+}}*, {{.+}}** [[G_ADDR]]
+      // LAMBDA-DAG: [[G_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[G_CONV]],
+      // LAMBDA-DAG: store {{.+}} [[G_ADDR_VAL]], {{.+}}* [[G_PRIV]],
+
+      // g1
+      // LAMBDA-DAG: [[G1_CONV:%.+]] = bitcast {{.+}}* [[G1_ADDR]] to
+      // LAMBDA-DAG: store {{.+}}* [[G1_CONV]], {{.+}}** [[G1_REF]],
+      // LAMBDA-DAG: [[G1_REF_VAL:%.+]] = load {{.+}}*, {{.+}}** [[G1_REF]],
+      // LAMBDA-DAG: store {{.+}}* [[G1_REF_VAL]], {{.+}}** [[TMP]],
+      // LAMBDA-DAG: [[TMP_REF:%.+]] = load {{.+}}*, {{.+}}** [[TMP]],
+      // LAMBDA-DAG: [[TMP_VAL:%.+]] = load {{.+}}, {{.+}}* [[TMP_REF]],
+      // LAMBDA-DAG: store {{.+}} [[TMP_VAL]], {{.+}}* [[G1_PRIV]]
+      // LAMBDA-DAG: store {{.+}}* [[G1_PRIV]], {{.+}}** [[TMP_PRIV]],
+
+      // svar
+      // LAMBDA-64-DAG: [[SVAR_CONV:%.+]] = bitcast {{.+}}* [[SVAR_ADDR]] to
+      // LAMBDA-64-DAG: [[SVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CONV]],
+      // LAMBDA-32-DAG: [[SVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[SVAR_VAL]], {{.+}}* [[SVAR_PRIV]],
+
+      // sfvar
+      // LAMBDA-DAG: [[SFVAR_CONV:%.+]] = bitcast {{.+}}* [[SFVAR_ADDR]] to
+      // LAMBDA-DAG: [[SFVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SFVAR_CONV]],
+      // LAMBDA-DAG: store {{.+}} [[SFVAR_VAL]], {{.+}}* [[SFVAR_PRIV]],
+
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
+      // pass firstprivate parameters to parallel outlined function
+      // g
+      // LAMBDA-64-DAG: [[G_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[G_PRIV]],
+      // LAMBDA-64: [[G_CAST_CONV:%.+]] = bitcast {{.+}}* [[G_CAST:%.+]] to
+      // LAMBDA-64-DAG: store {{.+}} [[G_PRIV_VAL]], {{.+}}* [[G_CAST_CONV]],
+      // LAMBDA-64-DAG: [[G_PAR:%.+]] = load {{.+}}, {{.+}}* [[G_CAST]],
+
+      // g1
+      // LAMBDA-DAG: [[TMP_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[TMP_PRIV]],
+      // LAMBDA-DAG: [[G1_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[TMP_PRIV_VAL]],
+      // LAMBDA: [[G1_CAST_CONV:%.+]] = bitcast {{.+}}* [[G1_CAST:%.+]] to
+      // LAMBDA-DAG: store {{.+}} [[G1_PRIV_VAL]], {{.+}}* [[G1_CAST_CONV]],
+      // LAMBDA-DAG: [[G1_PAR:%.+]] = load {{.+}}, {{.+}}* [[G1_CAST]],
+
+      // svar
+      // LAMBDA: [[SVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_PRIV]],
+      // LAMBDA-64-DAG: [[SVAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[SVAR_CAST:%.+]] to
+      // LAMBDA-64-DAG: store {{.+}} [[SVAR_VAL]], {{.+}}* [[SVAR_CAST_CONV]],
+      // LAMBDA-32-DAG: store {{.+}} [[SVAR_VAL]], {{.+}}* [[SVAR_CAST:%.+]],
+      // LAMBDA-DAG: [[SVAR_PAR:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CAST]],
+
+      // sfvar
+      // LAMBDA: [[SFVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SFVAR_PRIV]],
+      // LAMBDA-DAG: [[SFVAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[SFVAR_CAST:%.+]] to
+      // LAMBDA-DAG: store {{.+}} [[SFVAR_VAL]], {{.+}}* [[SFVAR_CAST_CONV]],
+      // LAMBDA-DAG: [[SFVAR_PAR:%.+]] = load {{.+}}, {{.+}}* [[SFVAR_CAST]],
+
+      // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PAR]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PAR]], {{.+}} [[SFVAR_PAR]])
+      // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PRIV]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PAR]], {{.+}} [[SFVAR_PAR]])
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_fini(
+      // LAMBDA: ret void
+
+
+      // LAMBDA-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, i{{[0-9]+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]])
+      // LAMBDA-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, double* {{.+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]])
+      // skip initial params
+      // LAMBDA: {{.+}} = alloca{{.+}},
+      // LAMBDA: {{.+}} = alloca{{.+}},
+      // LAMBDA: {{.+}} = alloca{{.+}},
+      // LAMBDA: {{.+}} = alloca{{.+}},
+
+      // addr alloca's
+      // LAMBDA-64: [[G_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA-32: [[G_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[G1_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[G1_REF:%.+]] = alloca double*,
+
+      // private alloca's (only for 32-bit)
+      // LAMBDA-32: [[G_PRIV:%.+]] = alloca double,
+
+      // transfer input parameters into addr alloca's
+      // LAMBDA-DAG: store {{.+}} [[G_IN]], {{.+}} [[G_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[G1_IN]], {{.+}} [[G1_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]],
+      // LAMBDA-DAG: store {{.+}} [[SFVAR_IN]], {{.+}} [[SFVAR_ADDR]],
+
+      // prepare parameters for lambda
+      // g
+      // LAMBDA-64-DAG: [[G_CONV:%.+]] = bitcast {{.+}}* [[G_ADDR]] to
+      // LAMBDA-32-DAG: [[G_ADDR_REF:%.+]] = load {{.+}}*, {{.+}}** [[G_ADDR]]
+      // LAMBDA-32-DAG: [[G_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[G_ADDR_REF]],
+      // LAMBDA-32-DAG: store {{.+}} [[G_ADDR_VAL]], {{.+}}* [[G_PRIV]],
+
+      // g1
+      // LAMBDA-DAG: [[G1_CONV:%.+]] = bitcast {{.+}}* [[G1_ADDR]] to
+      // LAMBDA-DAG: store {{.+}}* [[G1_CONV]], {{.+}}* [[G1_REF]],
+
+      // svar
+      // LAMBDA-64-DAG: [[SVAR_CONV:%.+]] = bitcast {{.+}}* [[SVAR_ADDR]] to
+
+      // sfvar
+      // LAMBDA-DAG: [[SFVAR_CONV:%.+]] = bitcast {{.+}}* [[SFVAR_ADDR]] to
+
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
+      g = 1;
+      g1 = 1;
+      svar = 3;
+      sfvar = 4.0;
+      // LAMBDA-64: store double 1.0{{.+}}, double* [[G_CONV]],
+      // LAMBDA-32: store double 1.0{{.+}}, double* [[G_PRIV]],
+      // LAMBDA: [[G1_REF_REF:%.+]] = load {{.+}}*, {{.+}}** [[G1_REF]],
+      // LAMBDA: store {{.+}} 1.0{{.+}}, {{.+}}* [[G1_REF_REF]],
+      // LAMBDA-64: store {{.+}} 3, {{.+}}* [[SVAR_CONV]],
+      // LAMBDA-32: store {{.+}} 3, {{.+}}* [[SVAR_ADDR]],
+      // LAMBDA: store {{.+}} 4.0{{.+}}, {{.+}}* [[SFVAR_CONV]],
+
+      // pass params to inner lambda
+      // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+      // LAMBDA-64: store double* [[G_CONV]], double** [[G_PRIVATE_ADDR_REF]],
+      // LAMBDA-32: store double* [[G_PRIV]], double** [[G_PRIVATE_ADDR_REF]],
+      // LAMBDA: [[G1_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+      // LAMBDA: [[G1_REF_REF:%.+]] = load double*, double** [[G1_REF]],
+      // LAMBDA: store double* [[G1_REF_REF]], double** [[G1_PRIVATE_ADDR_REF]],
+      // LAMBDA: [[SVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 2
+      // LAMBDA-64: store i{{[0-9]+}}* [[SVAR_CONV]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA-32: store i{{[0-9]+}}* [[SVAR_ADDR]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
+      // LAMBDA: store float* [[SFVAR_CONV]], float** [[SFVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]])
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_fini(
+      // LAMBDA: ret void
+      [&]() {
+	// LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]])
+	// LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]],
+	g = 2;
+	g1 = 2;
+	svar = 4;
+	sfvar = 8.0;
+	// LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]]
+	// LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+	// LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]]
+	// LAMBDA: store double 2.0{{.+}}, double* [[G_REF]]
+
+	// LAMBDA: [[TMP_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+	// LAMBDA: [[G1_REF:%.+]] = load double*, double** [[TMP_PTR_REF]]
+	// LAMBDA: store double 2.0{{.+}}, double* [[G1_REF]],
+	// LAMBDA: [[SVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 2
+	// LAMBDA: [[SVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PTR_REF]]
+	// LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SVAR_REF]]
+	// LAMBDA: [[SFVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
+	// LAMBDA: [[SFVAR_REF:%.+]] = load float*, float** [[SFVAR_PTR_REF]]
+	// LAMBDA: store float 8.0{{.+}}, float* [[SFVAR_REF]]
+      }();
+    }
+  }();
+  return 0;
+  #else
+  S<float> test;
+  int t_var = 0;
+  int vec[] = {1, 2};
+  S<float> s_arr[] = {1, 2};
+  S<float> &var = test;
+
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for firstprivate(t_var, vec, s_arr, s_arr, var, var, svar)
+  for (int i = 0; i < 2; ++i) {
+    vec[i] = t_var;
+    s_arr[i] = var;
+  }
+  return tmain<int>();
+  #endif
+}
+
+// CHECK-LABEL: define{{.*}} i{{[0-9]+}} @main()
+// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOAD_FUN_0:@.+]](
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
+
+// CHECK: define{{.+}} [[OFFLOAD_FUN_0]](i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_FLOAT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]* {{.+}} [[VAR_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]])
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 5, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i{{[0-9]+}}, [2 x i{{[0-9]+}}]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*, i{{[0-9]+}})* [[OMP_OUTLINED_0:@.+]] to void
+// CHECK: ret
+
+// CHECK: define internal void [[OMP_OUTLINED_0]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_FLOAT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]* {{.+}} [[VAR_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]])
+
+// addr alloca's
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[TMP:%.+]] = alloca [[S_FLOAT_TY]]*,
+
+// skip loop alloca's
+// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_LB:.omp.comb.lb+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_UB:.omp.comb.ub+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}},
+
+// private alloca's
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[SVAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+
+// init addr alloca's with input values
+// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]],
+// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]],
+// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]],
+
+// init private alloca's with addr alloca's
+// t-var
+// CHECK-64-DAG: [[T_VAR_CONV:%.+]] = bitcast {{.+}} [[T_VAR_ADDR]] to
+// CHECK-64-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CONV]],
+// CHECK-32-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[T_VAR_ADDR_VAL]], {{.+}} [[T_VAR_PRIV]],
+
+// vec
+// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]],
+// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to
+// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to
+// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]],
+
+// s_arr
+// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]],
+// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]],
+// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to
+// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]],
+// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]]
+// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]]
+// CHECK-DAG: [[CPY_BODY]]:
+// CHECK-DAG: call void @llvm.memcpy{{.+}}(
+// CHECK-DAG: [[CPY_DONE]]:
+
+// var
+// CHECK-DAG: [[TMP_REF:%.+]] = load {{.+}}*, {{.+}}* [[TMP]],
+// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to
+// CHECK-DAG: [[TMP_REF_BCAST:%.+]] = bitcast {{.+}}* [[TMP_REF]] to
+// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[TMP_REF_BCAST]],
+// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]],
+
+// svar
+// CHECK-64-DAG: [[SVAR_CONV:%.+]] = bitcast {{.+}}* [[SVAR_ADDR]] to
+// CHECK-64-DAG: [[SVAR_CONV_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CONV]],
+// CHECK-32-DAG: [[SVAR_CONV_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[SVAR_CONV_VAL]], {{.+}}* [[SVAR_PRIV]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// pass private alloca's to fork
+// CHECK-DAG: [[T_VAR_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_PRIV]],
+// not dag to distinguish with S_VAR_CAST
+// CHECK-64: [[T_VAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[T_VAR_CAST:%.+]] to
+// CHECK-64-DAG: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST_CONV]],
+// CHECK-32: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST:%.+]],
+// CHECK-DAG: [[T_VAR_CAST_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CAST]],
+// CHECK-DAG: [[TMP_PRIV_VAL:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]],
+// CHECK-DAG: [[SVAR_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_PRIV]],
+// CHECK-64-DAG: [[SVAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[SVAR_CAST:%.+]] to
+// CHECK-64-DAG: store {{.+}} [[SVAR_PRIV_VAL]], {{.+}}* [[SVAR_CAST_CONV]],
+// CHECK-32-DAG: store {{.+}} [[SVAR_PRIV_VAL]], {{.+}}* [[SVAR_CAST:%.+]],
+// CHECK-DAG: [[SVAR_CAST_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CAST]],
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_0:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} [[T_VAR_CAST_VAL]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], [[S_FLOAT_TY]]* [[TMP_PRIV_VAL]], i{{[0-9]+}} [[SVAR_CAST_VAL]])
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+// By OpenMP specifications, 'firstprivate' applies to both distribute and parallel for.
+// However, the support for 'firstprivate' of 'parallel' is only used when 'parallel'
+// is found alone. Therefore we only have one 'firstprivate' support for 'parallel for'
+// in combination
+// CHECK: define internal void [[OMP_PARFOR_OUTLINED_0]]({{.+}}, {{.+}}, {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x [[S_FLOAT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]* {{.+}} [[VAR_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]])
+
+// addr alloca's
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+
+// skip loop alloca's
+// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_LB:.omp.lb+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_UB:.omp.ub+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}},
+
+// private alloca's
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*,
+
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+
+// init addr alloca's with input values
+// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]],
+// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]],
+// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]],
+
+// init private alloca's with addr alloca's
+// vec
+// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]],
+// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to
+// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to
+// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]],
+
+// s_arr
+// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]],
+// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]],
+// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to
+// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]],
+// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]]
+// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]]
+// CHECK-DAG: [[CPY_BODY]]:
+// CHECK-DAG: call void @llvm.memcpy{{.+}}(
+// CHECK-DAG: [[CPY_DONE]]:
+
+// var
+// CHECK-DAG: [[VAR_ADDR_REF:%.+]] = load {{.+}}*, {{.+}}* [[VAR_ADDR]],
+// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to
+// CHECK-DAG: [[VAR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[VAR_ADDR_REF]] to
+// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[VAR_ADDR_BCAST]],
+// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+// template tmain with S_INT_TY
+// CHECK-LABEL: define{{.*}} i{{[0-9]+}} @{{.+}}tmain{{.+}}()
+// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOAD_FUN_0:@.+]](
+// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR:@.+]]([[S_INT_TY]]* [[TEST]])
+
+// CHECK: define{{.+}} [[OFFLOAD_FUN_0]](i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_INT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]* {{.+}} [[VAR_IN:%.+]])
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i{{[0-9]+}}, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[OMP_OUTLINED_0:@.+]] to void
+// CHECK: ret
+
+// CHECK: define internal void [[OMP_OUTLINED_0]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_INT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]* {{.+}} [[VAR_IN:%.+]])
+
+// addr alloca's
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*,
+// CHECK: [[TMP:%.+]] = alloca [[S_INT_TY]]*,
+
+// skip loop alloca's
+// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_LB:.omp.comb.lb+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_UB:.omp.comb.ub+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}},
+
+// private alloca's
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*,
+
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+
+// init addr alloca's with input values
+// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]],
+// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]],
+// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]],
+
+// init private alloca's with addr alloca's
+// t-var
+// CHECK-64-DAG: [[T_VAR_CONV:%.+]] = bitcast {{.+}} [[T_VAR_ADDR]] to
+// CHECK-64-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CONV]],
+// CHECK-32-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[T_VAR_ADDR_VAL]], {{.+}} [[T_VAR_PRIV]],
+
+// vec
+// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]],
+// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to
+// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to
+// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]],
+
+// s_arr
+// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]],
+// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]],
+// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to
+// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]],
+// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]]
+// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]]
+// CHECK-DAG: [[CPY_BODY]]:
+// CHECK-DAG: call void @llvm.memcpy{{.+}}(
+// CHECK-DAG: [[CPY_DONE]]:
+
+// var
+// CHECK-DAG: [[TMP_REF:%.+]] = load {{.+}}*, {{.+}}* [[TMP]],
+// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to
+// CHECK-DAG: [[TMP_REF_BCAST:%.+]] = bitcast {{.+}}* [[TMP_REF]] to
+// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[TMP_REF_BCAST]],
+// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// pass private alloca's to fork
+// CHECK-DAG: [[T_VAR_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_PRIV]],
+// not dag to distinguish with S_VAR_CAST
+// CHECK-64: [[T_VAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[T_VAR_CAST:%.+]] to
+// CHECK-64-DAG: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST_CONV]],
+// CHECK-32: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST:%.+]],
+// CHECK-DAG: [[T_VAR_CAST_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CAST]],
+// CHECK-DAG: [[TMP_PRIV_VAL:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]],
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_0:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} [[T_VAR_CAST_VAL]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], [[S_INT_TY]]* [[TMP_PRIV_VAL]])
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_INT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+// By OpenMP specifications, 'firstprivate' applies to both distribute and parallel for.
+// However, the support for 'firstprivate' of 'parallel' is only used when 'parallel'
+// is found alone. Therefore we only have one 'firstprivate' support for 'parallel for'
+// in combination
+// CHECK: define internal void [[OMP_PARFOR_OUTLINED_0]]({{.+}}, {{.+}}, {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x [[S_INT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]* {{.+}} [[VAR_IN:%.+]])
+
+// addr alloca's
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*,
+
+// skip loop alloca's
+// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_LB:.omp.lb+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_UB:.omp.ub+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}},
+
+// private alloca's
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*,
+
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+
+// init addr alloca's with input values
+// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]],
+// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]],
+// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]],
+// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]],
+
+// init private alloca's with addr alloca's
+// vec
+// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]],
+// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to
+// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to
+// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]],
+
+// s_arr
+// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]],
+// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]],
+// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to
+// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]],
+// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]]
+// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]]
+// CHECK-DAG: [[CPY_BODY]]:
+// CHECK-DAG: call void @llvm.memcpy{{.+}}(
+// CHECK-DAG: [[CPY_DONE]]:
+
+// var
+// CHECK-DAG: [[VAR_ADDR_REF:%.+]] = load {{.+}}*, {{.+}}* [[VAR_ADDR]],
+// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to
+// CHECK-DAG: [[VAR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[VAR_ADDR_REF]] to
+// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[VAR_ADDR_BCAST]],
+// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_INT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+#endif
diff --git a/test/OpenMP/distribute_parallel_for_if_codegen.cpp b/test/OpenMP/distribute_parallel_for_if_codegen.cpp
new file mode 100644
index 000000000000..e6cd7215f585
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_if_codegen.cpp
@@ -0,0 +1,192 @@
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -std=c++11 -triple %itanium_abi_triple -emit-pch -o %t %s
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck --check-prefix=CHECK %s
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+void fn1();
+void fn2();
+void fn3();
+void fn4();
+void fn5();
+void fn6();
+
+int Arg;
+
+// CHECK-LABEL: define {{.*}}void @{{.+}}gtid_test
+void gtid_test() {
+#pragma omp target
+#pragma omp teams
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_0:@.+]](
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_1:@.+]](
+#pragma omp distribute parallel for
+  for(int i = 0 ; i < 100; i++) {}
+  // CHECK: define internal void [[OFFLOADING_FUN_0]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}})
+  // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_0]](
+  // CHECK: call void @__kmpc_for_static_init_4(
+  // CHECK:  call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, {{.+}}* [[OMP_OUTLINED_0:@.+]] to void
+  // CHECK: call void @__kmpc_for_static_fini(
+
+  // CHECK: define{{.+}} void [[OMP_OUTLINED_0]](
+  // CHECK: call void @__kmpc_for_static_init_4(
+  // CHECK: call void @__kmpc_for_static_fini(
+  // CHECK: ret
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (parallel: false)
+  for(int i = 0 ; i < 100; i++) {
+  // CHECK: define internal void [[OFFLOADING_FUN_1]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}})
+  // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_1]](
+  // CHECK: call void @__kmpc_for_static_init_4(
+  // CHECK: call void @__kmpc_serialized_parallel(
+  // CHECK: call void [[OMP_OUTLINED_1:@.+]](
+  // CHECK: call void @__kmpc_end_serialized_parallel(
+  // CHECK: call void @__kmpc_for_static_fini(
+  // CHECK: define{{.+}} void [[OMP_OUTLINED_1]](
+  // CHECK: call void @__kmpc_for_static_init_4(
+  // CHECK: call void @{{.+}}gtid_test
+  // CHECK: call void @__kmpc_for_static_fini(
+  // CHECK: ret
+    gtid_test();
+  }
+}
+
+
+template <typename T>
+int tmain(T Arg) {
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (true)
+  for(int i = 0 ; i < 100; i++) {
+    fn1();
+  }
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (false)
+  for(int i = 0 ; i < 100; i++) {
+    fn2();
+  }
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (parallel: Arg)
+  for(int i = 0 ; i < 100; i++) {
+    fn3();
+  }
+  return 0;
+}
+
+// CHECK-LABEL: define {{.*}}i{{[0-9]+}} @main()
+int main() {
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_0:@.+]](
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_1:@.+]](
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_2:@.+]](
+// CHECK: = call {{.*}}i{{.+}} @{{.+}}tmain
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (true)
+  for(int i = 0 ; i < 100; i++) {
+    // CHECK: define internal void [[OFFLOADING_FUN_0]](
+    // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}})
+    // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_0]](
+
+    // CHECK: call void @__kmpc_for_static_init_4(
+    // CHECK: call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, {{.+}}* [[OMP_OUTLINED_2:@.+]] to void
+    // CHECK: call void @__kmpc_for_static_fini(
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_2]](
+    // CHECK: call void @__kmpc_for_static_init_4(
+    // CHECK: call {{.*}}void @{{.+}}fn4
+    // CHECK: call void @__kmpc_for_static_fini(
+
+    fn4();
+  }
+
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (false)
+  for(int i = 0 ; i < 100; i++) {
+    // CHECK: define internal void [[OFFLOADING_FUN_1]](
+    // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}})
+    // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_1]](
+
+    // CHECK: call void @__kmpc_for_static_init_4(
+    // CHECK: call void @__kmpc_serialized_parallel(
+    // CHECK: call void [[OMP_OUTLINED_3:@.+]](
+    // CHECK: call void @__kmpc_end_serialized_parallel(
+    // CHECK: call void @__kmpc_for_static_fini(
+
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_3]](
+    // CHECK: call void @__kmpc_for_static_init_4(
+    // CHECK: call {{.*}}void @{{.+}}fn5
+    // CHECK: call void @__kmpc_for_static_fini(
+    fn5();
+  }
+
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for if (Arg)
+  for(int i = 0 ; i < 100; i++) {
+    // CHECK: define internal void [[OFFLOADING_FUN_2]](
+    // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 1, {{.+}}* [[OMP_TEAMS_OUTLINED_2:@.+]] to {{.+}})
+    // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_2]](
+
+    // CHECK: call void @__kmpc_for_static_init_4(
+    // CHECK: call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, {{.+}}* [[OMP_OUTLINED_4:@.+]] to void
+    // CHECK: call void @__kmpc_serialized_parallel(
+    // CHECK: call void [[OMP_OUTLINED_4:@.+]](
+    // CHECK: call void @__kmpc_end_serialized_parallel(
+    // CHECK: call void @__kmpc_for_static_fini(
+
+    // CHECK: define{{.+}} void [[OMP_OUTLINED_4]](
+    // CHECK: call void @__kmpc_for_static_init_4(
+    // CHECK: call {{.*}}void @{{.+}}fn6
+    // CHECK: call void @__kmpc_for_static_fini(
+    fn6();
+  }
+
+  return tmain(Arg);
+}
+
+// CHECK-LABEL: define {{.+}} @{{.+}}tmain
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, void {{.+}}* [[T_OUTLINE_FUN_1:@.+]] to void
+// CHECK: call void @__kmpc_for_static_fini(
+
+// CHECK: define internal {{.*}}void [[T_OUTLINE_FUN_1]]
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call {{.*}}void @{{.+}}fn1
+// CHECK: call void @__kmpc_for_static_fini(
+// CHECK: ret void
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call {{.*}}void @__kmpc_serialized_parallel(
+// CHECK: call void [[T_OUTLINE_FUN_2:@.+]](
+// CHECK: call {{.*}}void @__kmpc_end_serialized_parallel(
+// CHECK: call void @__kmpc_for_static_fini(
+
+// CHECK: define internal {{.*}}void [[T_OUTLINE_FUN_2]]
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call {{.*}}void @{{.+}}fn2
+// CHECK: call void @__kmpc_for_static_fini(
+// CHECK: ret void
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, void {{.+}}* [[T_OUTLINE_FUN_3:@.+]] to void
+// CHECK: call {{.*}}void @__kmpc_serialized_parallel(
+// call void [[T_OUTLINE_FUN_3:@.+]](
+// CHECK: call {{.*}}void @__kmpc_end_serialized_parallel(
+
+// CHECK: define internal {{.*}}void [[T_OUTLINE_FUN_3]]
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call {{.*}}void @{{.+}}fn3
+// CHECK: call void @__kmpc_for_static_fini(
+// CHECK: ret void
+#endif
diff --git a/test/OpenMP/distribute_parallel_for_lastprivate_codegen.cpp b/test/OpenMP/distribute_parallel_for_lastprivate_codegen.cpp
new file mode 100644
index 000000000000..2e8da79c03bd
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_lastprivate_codegen.cpp
@@ -0,0 +1,653 @@
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+
+// RUN: %clang_cc1  -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+template <class T>
+struct S {
+  T f;
+  S(T a) : f(a) {}
+  S() : f() {}
+  operator T() { return T(); }
+  ~S() {}
+};
+
+// CHECK: [[S_FLOAT_TY:%.+]] = type { float }
+// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
+template <typename T>
+T tmain() {
+  S<T> test;
+  T t_var = T();
+  T vec[] = {1, 2};
+  S<T> s_arr[] = {1, 2};
+  S<T> &var = test;
+  #pragma omp target
+  #pragma omp teams
+#pragma omp distribute parallel for lastprivate(t_var, vec, s_arr, s_arr, var, var)
+  for (int i = 0; i < 2; ++i) {
+    vec[i] = t_var;
+    s_arr[i] = var;
+  }
+  return T();
+}
+
+int main() {
+  static int svar;
+  volatile double g;
+  volatile double &g1 = g;
+
+  #ifdef LAMBDA
+  // LAMBDA-LABEL: @main
+  // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
+  [&]() {
+    static float sfvar;
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN:@.+]](
+
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN]](
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED:@.+]] to {{.+}})
+    #pragma omp target
+    #pragma omp teams
+#pragma omp distribute parallel for lastprivate(g, g1, svar, sfvar)
+    for (int i = 0; i < 2; ++i) {
+      // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, double*{{.+}} [[G_IN:%.+]], double*{{.+}} [[G1_IN:%.+]], i{{[0-9]+}}*{{.+}} [[SVAR_IN:%.+]], float*{{.+}} [[SFVAR_IN:%.+]])
+      // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float*,
+      // LAMBDA: [[TMP_G1:%.+]] = alloca double*,
+      // loop variables
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[G_PRIVATE:%.+]] = alloca double,
+      // LAMBDA: [[G1_PRIVATE:%.+]] = alloca double,
+      // LAMBDA: [[TMP_G1_PRIVATE:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIVATE:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_PRIVATE:%.+]] = alloca float,
+
+      // init addr alloca's
+      // LAMBDA: store double* [[G_IN]], double** [[G_PRIVATE_ADDR]],
+      // LAMBDA: store double* [[G1_IN]], double** [[G1_PRIVATE_ADDR]],
+      // LAMBDA: store i{{[0-9]+}}* [[SVAR_IN]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]],
+      // LAMBDA: store float* [[SFVAR_IN]], float** [[SFVAR_PRIVATE_ADDR]],
+
+      // init private variables
+      // LAMBDA: [[G_IN_REF:%.+]] = load double*, double** [[G_PRIVATE_ADDR]],
+      // LAMBDA: [[SVAR_IN_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]],
+      // LAMBDA: [[SFVAR_IN_REF:%.+]] = load float*, float** [[SFVAR_PRIVATE_ADDR]],
+      // LAMBDA: [[G1_IN_REF:%.+]] = load double*, double** [[G1_PRIVATE_ADDR]],
+      // LAMBDA: store double* [[G1_IN_REF]], double** [[TMP_G1]],
+      // LAMBDA: [[TMP_G1_VAL:%.+]] = load double*, double** [[TMP_G1]],
+      // LAMBDA: store double* [[G1_PRIVATE]], double** [[TMP_G1_PRIVATE]],
+
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
+      // LAMBDA: [[G1_PAR:%.+]] = load{{.+}}, {{.+}} [[TMP_G1_PRIVATE]],
+      // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PRIVATE]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PRIVATE]], {{.+}} [[SFVAR_PRIVATE]])
+      // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PRIVATE]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PRIVATE]], {{.+}} [[SFVAR_PRIVATE]])
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_fini(
+
+      // lastprivate
+      // LAMBDA: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]],
+      // LAMBDA: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0
+      // LAMBDA: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]]
+
+      // LAMBDA: [[OMP_LASTPRIV_BLOCK]]:
+      // LAMBDA: [[G_PRIV_VAL:%.+]] = load double, double* [[G_PRIVATE]],
+      // LAMBDA: store{{.*}} double [[G_PRIV_VAL]], double* [[G_IN_REF]],
+      // LAMBDA: [[TMP_G1_PRIV_REF:%.+]] = load double*, double** [[TMP_G1_PRIVATE]],
+      // LAMBDA: [[TMP_G1_PRIV_VAL:%.+]] = load double, double* [[TMP_G1_PRIV_REF]],
+      // LAMBDA: store{{.*}} double [[TMP_G1_PRIV_VAL]], double* [[TMP_G1_VAL]],
+
+      // LAMBDA: [[SVAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[SVAR_PRIVATE]],
+      // LAMBDA: store i{{[0-9]+}} [[SVAR_PRIV_VAL]], i{{[0-9]+}}* [[SVAR_IN_REF]],
+      // LAMBDA: [[SFVAR_PRIV_VAL:%.+]] = load float, float* [[SFVAR_PRIVATE]],
+      // LAMBDA: store float [[SFVAR_PRIV_VAL]], float* [[SFVAR_IN_REF]],
+      // LAMBDA: br label %[[OMP_LASTPRIV_DONE]]
+      // LAMBDA: [[OMP_LASTPRIV_DONE]]:
+      // LAMBDA: ret
+
+      g = 1;
+      g1 = 1;
+      svar = 3;
+      sfvar = 4.0;
+      // outlined function for 'parallel for'
+      // LAMBDA-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[G_IN:%.+]], {{.+}} [[G1_IN:%.+]], {{.+}} [[SVAR_IN:%.+]], {{.+}} [[SFVAR_IN:%.+]])
+      // LAMBDA-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[G_IN:%.+]], {{.+}} [[G1_IN:%.+]], {{.+}} [[SVAR_IN:%.+]], {{.+}} [[SFVAR_IN:%.+]])
+
+      // addr alloca's
+      // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float*,
+
+      // loop variables
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+      // LAMBDA: {{.+}} = alloca i{{[0-9]+}},
+
+      // private alloca's
+      // LAMBDA: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[G_PRIVATE:%.+]] = alloca double,
+      // LAMBDA: [[G1_PRIVATE:%.+]] = alloca double,
+      // LAMBDA: [[TMP_G1_PRIVATE:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIVATE:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_PRIVATE:%.+]] = alloca float,
+
+            // init addr alloca's
+      // LAMBDA: store double* [[G_IN]], double** [[G_PRIVATE_ADDR]],
+      // LAMBDA: store double* [[G1_IN]], double** [[G1_PRIVATE_ADDR]],
+      // LAMBDA: store i{{[0-9]+}}* [[SVAR_IN]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]],
+      // LAMBDA: store float* [[SFVAR_IN]], float** [[SFVAR_PRIVATE_ADDR]],
+
+      // init private variables
+      // LAMBDA: [[G_IN_REF:%.+]] = load double*, double** [[G_PRIVATE_ADDR]],
+      // LAMBDA: [[SVAR_IN_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]],
+      // LAMBDA: [[SFVAR_IN_REF:%.+]] = load float*, float** [[SFVAR_PRIVATE_ADDR]],
+
+      // LAMBDA: [[G1_IN_REF:%.+]] = load double*, double** [[G1_PRIVATE_ADDR]],
+      // LAMBDA: store double* [[G1_PRIVATE]], double** [[TMP_G1]],
+
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
+
+      // loop body
+      // LAMBDA: store double 1.0{{.+}}, double* [[G_PRIVATE]],
+      // LAMBDA: [[TMP_G1_REF:%.+]] = load double*, double** [[TMP_G1_PRIVATE]],
+      // LAMBDA: store{{.+}} double 1.0{{.+}}, double* [[TMP_G1_REF]],
+      // LAMBDA: store i{{[0-9]+}} 3, i{{[0-9]+}}* [[SVAR_PRIVATE]],
+      // LAMBDA: store float 4.0{{.+}}, float* [[SFVAR_PRIVATE]],
+      // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+      // LAMBDA: store double* [[G_PRIVATE]], double** [[G_PRIVATE_ADDR_REF]],
+      // LAMBDA: [[TMP_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+      // LAMBDA: [[G1_PRIVATE_ADDR_FROM_TMP:%.+]] = load double*, double** [[TMP_G1_PRIVATE]],
+      // LAMBDA: store double* [[G1_PRIVATE_ADDR_FROM_TMP]], double** [[TMP_PRIVATE_ADDR_REF]],
+      // LAMBDA: [[SVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 2
+      // LAMBDA: store i{{[0-9]+}}* [[SVAR_PRIVATE]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
+      // LAMBDA: store float* [[SFVAR_PRIVATE]], float** [[SFVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]])
+
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_fini(
+
+      // lastprivate
+      // LAMBDA: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]],
+      // LAMBDA: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0
+      // LAMBDA: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]]
+      // LAMBDA: [[OMP_LASTPRIV_BLOCK]]:
+      // LAMBDA: [[G_PRIV_VAL:%.+]] = load double, double* [[G_PRIVATE]],
+      // LAMBDA: store{{.*}} double [[G_PRIV_VAL]], double* [[G_IN_REF]],
+      // LAMBDA: [[TMP_G1_PRIV_REF:%.+]] = load double*, double** [[TMP_G1_PRIVATE]],
+      // LAMBDA: [[TMP_G1_PRIV_VAL:%.+]] = load double, double* [[TMP_G1_PRIV_REF]],
+      // LAMBDA: store{{.*}} double [[TMP_G1_PRIV_VAL]], double* [[G1_IN_REF]],
+      // LAMBDA: [[SVAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[SVAR_PRIVATE]],
+      // LAMBDA: store i{{[0-9]+}} [[SVAR_PRIV_VAL]], i{{[0-9]+}}* [[SVAR_IN_REF]],
+      // LAMBDA: [[SFVAR_PRIV_VAL:%.+]] = load float, float* [[SFVAR_PRIVATE]],
+      // LAMBDA: store float [[SFVAR_PRIV_VAL]], float* [[SFVAR_IN_REF]],
+      // LAMBDA: br label %[[OMP_LASTPRIV_DONE]]
+      // LAMBDA: [[OMP_LASTPRIV_DONE]]:
+      // LAMBDA: ret
+
+      [&]() {
+	// LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]])
+	// LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]],
+	g = 2;
+	g1 = 2;
+	svar = 4;
+	sfvar = 8.0;
+	// LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]]
+	// LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+	// LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]]
+	// LAMBDA: store double 2.0{{.+}}, double* [[G_REF]]
+
+	// LAMBDA: [[TMP_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+	// LAMBDA: [[G1_REF:%.+]] = load double*, double** [[TMP_PTR_REF]]
+	// LAMBDA: store double 2.0{{.+}}, double* [[G1_REF]],
+	// LAMBDA: [[SVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 2
+	// LAMBDA: [[SVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PTR_REF]]
+	// LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SVAR_REF]]
+	// LAMBDA: [[SFVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
+	// LAMBDA: [[SFVAR_REF:%.+]] = load float*, float** [[SFVAR_PTR_REF]]
+	// LAMBDA: store float 8.0{{.+}}, float* [[SFVAR_REF]]
+      }();
+    }
+  }();
+  return 0;
+  #else
+  S<float> test;
+  int t_var = 0;
+  int vec[] = {1, 2};
+  S<float> s_arr[] = {1, 2};
+  S<float> &var = test;
+
+  #pragma omp target
+  #pragma omp teams
+#pragma omp distribute parallel for lastprivate(t_var, vec, s_arr, s_arr, var, var, svar)
+  for (int i = 0; i < 2; ++i) {
+    vec[i] = t_var;
+    s_arr[i] = var;
+  }
+  int i;
+
+  return tmain<int>();
+  #endif
+}
+
+// CHECK: define{{.*}} i{{[0-9]+}} @main()
+// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOAD_FUN:@.+]](i{{[0-9]+}} {{.+}}, [2 x i{{[0-9]+}}]* {{.+}}, [2 x [[S_FLOAT_TY]]]* {{.+}}, [[S_FLOAT_TY]]* {{.+}}, i{{[0-9]+}} {{.+}})
+// CHECK: ret
+
+// CHECK: define{{.+}} [[OFFLOAD_FUN]](i{{[0-9]+}} {{.+}}, [2 x i{{[0-9]+}}]*{{.+}} {{.+}}, [2 x [[S_FLOAT_TY]]]*{{.+}} {{.+}}, [[S_FLOAT_TY]]*{{.+}} {{.+}}, i{{[0-9]+}} {{.+}})
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(
+// CHECK: ret
+//
+// CHECK: define internal void [[OMP_OUTLINED:@.+]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}}*{{.+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]*{{.+}} [[VEC_IN:%.+]], [2 x [[S_FLOAT_TY]]]*{{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]*{{.+}} [[VAR_IN:%.+]], i{{[0-9]+}}*{{.*}} [[S_VAR_IN:%.+]])
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+// CHECK: [[TMP:%.*]] = alloca [[S_FLOAT_TY]]*,
+// skip loop variables
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[S_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+
+// copy from parameters to local address variables
+// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: store [2 x [[S_FLOAT_TY]]]* [[S_ARR_IN]], [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: store [[S_FLOAT_TY]]* [[VAR_IN]], [[S_FLOAT_TY]]** [[VAR_ADDR]],
+// CHECK: store i{{[0-9]+}}* [[S_VAR_IN]], i{{[0-9]+}}** [[SVAR_ADDR]],
+
+// load content of local address variables
+// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: [[SVAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_ADDR]],
+// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_ADDR]],
+// CHECK: store [[S_FLOAT_TY]]* [[VAR_ADDR_REF]], [[S_FLOAT_TY]]** [[TMP]],
+// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]],
+
+// call constructor for s_arr
+// CHECK: [[S_ARR_BGN:%.+]] = getelementptr{{.+}} [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]],
+// CHECK: [[S_ARR_END:%.+]] = getelementptr {{.+}} [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BGN]],
+// CHECK: br label %[[S_ARR_CST_LOOP:.+]]
+// CHECK: [[S_ARR_CST_LOOP]]:
+// CHECK: [[S_ARR_CTOR:%.+]] = phi {{.+}}
+// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_CTOR]])
+// CHECK: [[S_ARR_NEXT:%.+]] = getelementptr {{.+}} [[S_ARR_CTOR]],
+// CHECK: [[S_ARR_DONE:%.+]] = icmp {{.+}} [[S_ARR_NEXT]], [[S_ARR_END]]
+// CHECK: br i1 [[S_ARR_DONE]], label %[[S_ARR_CST_END:.+]], label %[[S_ARR_CST_LOOP]]
+// CHECK: [[S_ARR_CST_END]]:
+// CHECK: [[TMP_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP]],
+// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+// CHECK: store [[S_FLOAT_TY]]* [[VAR_PRIV]], [[S_FLOAT_TY]]** [[TMP_PRIV]],
+
+// the distribute loop
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: [[TMP_PRIV_VAL:%.+]] = load {{.+}}, {{.+}} [[TMP_PRIV]],
+// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]], {{.+}} [[S_VAR_PRIV]])
+// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]], {{.+}} [[S_VAR_PRIV]])
+
+// CHECK: call void @__kmpc_for_static_fini(
+
+// lastprivates
+// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]],
+// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0
+// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]]
+
+// CHECK: [[OMP_LASTPRIV_BLOCK]]:
+// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]],
+// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]],
+// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8*
+// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]],
+// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]] to [[S_FLOAT_TY]]*
+// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2
+// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]]
+// CHECK: [[S_ARR_COPY_BLOCK]]:
+// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_DST_EL]] to i8*
+// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_SRC_EL]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}})
+// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1
+// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}}
+// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]]
+// CHECK: [[S_ARR_COPY_DONE]]:
+// CHECK: [[TMP_VAL1:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]],
+// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_REF]] to i8*
+// CHECK: [[TMP_VAL1_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_VAL1]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL1_BCAST]],{{.+}})
+// CHECK: [[SVAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[S_VAR_PRIV]],
+// CHECK: store i{{[0-9]+}} [[SVAR_VAL]], i{{[0-9]+}}* [[SVAR_ADDR_REF]],
+// CHECK: ret void
+
+// outlined function for 'parallel for'
+// CHECK-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]], {{.+}} [[SVAR_IN:%.+]])
+// CHECK-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]], {{.+}} [[SVAR_IN:%.+]])
+
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+// skip loop variables
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*,
+// CHECK: [[S_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+
+// copy from parameters to local address variables
+// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: store [2 x [[S_FLOAT_TY]]]* [[S_ARR_IN]], [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: store [[S_FLOAT_TY]]* [[VAR_IN]], [[S_FLOAT_TY]]** [[VAR_ADDR]],
+// CHECK: store i{{[0-9]+}}* [[S_VAR_IN]], i{{[0-9]+}}** [[SVAR_ADDR]],
+
+// load content of local address variables
+// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: [[SVAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_ADDR]],
+// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]],
+
+// call constructor for s_arr
+// CHECK: [[S_ARR_BGN:%.+]] = getelementptr{{.+}} [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]],
+// CHECK: [[S_ARR_END:%.+]] = getelementptr {{.+}} [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BGN]],
+// CHECK: br label %[[S_ARR_CST_LOOP:.+]]
+// CHECK: [[S_ARR_CST_LOOP]]:
+// CHECK: [[S_ARR_CTOR:%.+]] = phi {{.+}}
+// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_CTOR]])
+// CHECK: [[S_ARR_NEXT:%.+]] = getelementptr {{.+}} [[S_ARR_CTOR]],
+// CHECK: [[S_ARR_DONE:%.+]] = icmp {{.+}} [[S_ARR_NEXT]], [[S_ARR_END]]
+// CHECK: br i1 [[S_ARR_DONE]], label %[[S_ARR_CST_END:.+]], label %[[S_ARR_CST_LOOP]]
+// CHECK: [[S_ARR_CST_END]]:
+// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_ADDR]],
+// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+// CHECK: store [[S_FLOAT_TY]]* [[VAR_PRIV]], [[S_FLOAT_TY]]** [[TMP_PRIV]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+
+// loop body
+// assignment: vec[i] = t_var;
+// CHECK: [[T_VAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]],
+// CHECK: [[VEC_PTR:%.+]] = getelementptr inbounds [2 x i{{[0-9]+}}], [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} {{.+}}
+// CHECK:  store i{{[0-9]+}} [[T_VAR_PRIV_VAL]], i{{[0-9]+}}* [[VEC_PTR]],
+
+// assignment: s_arr[i] = var;
+// CHECK-DAG: [[S_ARR_PTR:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]],
+// CHECK-DAG: [[TMP_VAL:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]],
+// CHECK-DAG: [[S_ARR_PTR_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_PTR]] to i8*
+// CHECK-DAG: [[TMP_VAL_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_VAL]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_PTR_BCAST]], i8* [[TMP_VAL_BCAST]],
+
+// CHECK: call void @__kmpc_for_static_fini(
+
+// lastprivates
+// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]],
+// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0
+// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]]
+
+// CHECK: [[OMP_LASTPRIV_BLOCK]]:
+// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]],
+// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]],
+// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8*
+// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]],
+// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]] to [[S_FLOAT_TY]]*
+// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2
+// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]]
+// CHECK: [[S_ARR_COPY_BLOCK]]:
+// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_DST_EL]] to i8*
+// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_SRC_EL]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}})
+// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1
+// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}}
+// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]]
+// CHECK: [[S_ARR_COPY_DONE]]:
+// CHECK: [[TMP_VAL1:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]],
+// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_ADDR_REF]] to i8*
+// CHECK: [[TMP_VAL1_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_VAL1]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL1_BCAST]],{{.+}})
+// CHECK: [[SVAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[S_VAR_PRIV]],
+// CHECK: store i{{[0-9]+}} [[SVAR_VAL]], i{{[0-9]+}}* [[SVAR_ADDR_REF]],
+// CHECK: ret void
+
+// template tmain
+// CHECK: define{{.*}} i{{[0-9]+}} [[TMAIN_INT:@.+]]()
+// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOAD_FUN_1:@.+]](i{{[0-9]+}} {{.+}}, [2 x i{{[0-9]+}}]* {{.+}}, [2 x [[S_INT_TY]]]* {{.+}}, [[S_INT_TY]]* {{.+}})
+// CHECK: ret
+
+// CHECK: define internal void [[OFFLOAD_FUN_1]](
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4,
+// CHECK: ret
+
+// CHECK: define internal void [[OMP_OUTLINED_1:@.+]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}}*{{.+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]*{{.+}} [[VEC_IN:%.+]],  [2 x [[S_INT_TY]]]*{{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]*{{.+}} [[VAR_IN:%.+]])
+// skip alloca of global_tid and bound_tid
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*,
+// CHECK: [[TMP:%.+]] = alloca [[S_INT_TY]]*,
+// skip loop variables
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*,
+
+// skip init of bound and global tid
+// CHECK: store i{{[0-9]+}}* {{.*}},
+// CHECK: store i{{[0-9]+}}* {{.*}},
+// copy from parameters to local address variables
+// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: store [2 x [[S_INT_TY]]]* [[S_ARR_IN]], [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: store [[S_INT_TY]]* [[VAR_IN]], [[S_INT_TY]]** [[VAR_ADDR]],
+
+// load content of local address variables
+// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_ADDR]],
+// CHECK-DAG: store [[S_INT_TY]]* [[VAR_ADDR_REF]], [[S_INT_TY]]** [[TMP]],
+// CHECK-DAG: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]],
+// CHECK-DAG: [[TMP_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: [[TMP_PRIV_VAL:%.+]] = load {{.+}}, {{.+}} [[TMP_PRIV]],
+// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]])
+// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]])
+
+// CHECK: call void @__kmpc_for_static_fini(
+
+// lastprivates
+// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]],
+// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0
+// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]]
+
+// CHECK: [[OMP_LASTPRIV_BLOCK]]:
+// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]],
+// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]],
+// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8*
+// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]],
+// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]] to [[S_INT_TY]]*
+// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2
+// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]]
+// CHECK: [[S_ARR_COPY_BLOCK]]:
+// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_INT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_INT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_DST_EL]] to i8*
+// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_SRC_EL]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}})
+// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1
+// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}}
+// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]]
+// CHECK: [[S_ARR_COPY_DONE]]:
+// CHECK: [[TMP_VAL:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]],
+// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_REF]] to i8*
+// CHECK: [[TMP_VAL_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_VAL]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL_BCAST]],{{.+}})
+// CHECK: ret void
+
+// outlined function for 'parallel for'
+// CHECK-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]])
+// CHECK-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]])
+
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: {{.+}} = alloca i{{[0-9]+}}*,
+// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*,
+// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*,
+// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*,
+// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*,
+// skip loop variables
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: {{.+}} = alloca i{{[0-9]+}},
+// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
+// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*,
+
+// copy from parameters to local address variables
+// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: store [2 x [[S_INT_TY]]]* [[S_ARR_IN]], [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: store [[S_INT_TY]]* [[VAR_IN]], [[S_INT_TY]]** [[VAR_ADDR]],
+
+// load content of local address variables
+// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]],
+// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]],
+// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]],
+// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]],
+
+// call constructor for s_arr
+// CHECK: [[S_ARR_BGN:%.+]] = getelementptr{{.+}} [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]],
+// CHECK: [[S_ARR_END:%.+]] = getelementptr {{.+}} [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_BGN]],
+// CHECK: br label %[[S_ARR_CST_LOOP:.+]]
+// CHECK: [[S_ARR_CST_LOOP]]:
+// CHECK: [[S_ARR_CTOR:%.+]] = phi {{.+}}
+// CHECK: call void [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_CTOR]])
+// CHECK: [[S_ARR_NEXT:%.+]] = getelementptr {{.+}} [[S_ARR_CTOR]],
+// CHECK: [[S_ARR_DONE:%.+]] = icmp {{.+}} [[S_ARR_NEXT]], [[S_ARR_END]]
+// CHECK: br i1 [[S_ARR_DONE]], label %[[S_ARR_CST_END:.+]], label %[[S_ARR_CST_LOOP]]
+// CHECK: [[S_ARR_CST_END]]:
+// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_ADDR]],
+// CHECK: call void [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
+// CHECK: store [[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]** [[TMP_PRIV]],
+
+// CHECK: call void @__kmpc_for_static_init_4(
+
+// assignment: vec[i] = t_var;
+// CHECK: [[IV_VAL:%.+]] =
+// CHECK: [[T_VAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]],
+// CHECK: [[VEC_PTR:%.+]] = getelementptr inbounds [2 x i{{[0-9]+}}], [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} {{.+}}
+// CHECK:  store i{{[0-9]+}} [[T_VAR_PRIV_VAL]], i{{[0-9]+}}* [[VEC_PTR]],
+
+// assignment: s_arr[i] = var;
+// CHECK-DAG: [[S_ARR_PTR:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]],
+// CHECK-DAG: [[TMP_VAL:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]],
+// CHECK-DAG: [[S_ARR_PTR_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_PTR]] to i8*
+// CHECK-DAG: [[TMP_VAL_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_VAL]] to i8*
+// CHECK-DAG: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_PTR_BCAST]], i8* [[TMP_VAL_BCAST]],
+
+// CHECK: call void @__kmpc_for_static_fini(
+
+// lastprivates
+// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]],
+// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0
+// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]]
+
+// CHECK: [[OMP_LASTPRIV_BLOCK]]:
+// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]],
+// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]],
+// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8*
+// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]],
+// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]] to [[S_INT_TY]]*
+// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2
+// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]]
+// CHECK: [[S_ARR_COPY_BLOCK]]:
+// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_INT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_INT_TY]]*{{.+}}
+// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_DST_EL]] to i8*
+// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_SRC_EL]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}})
+// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1
+// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}}
+// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]]
+// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]]
+// CHECK: [[S_ARR_COPY_DONE]]:
+// CHECK: [[TMP_VAL1:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]],
+// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[VAR_ADDR_REF]] to i8*
+// CHECK: [[TMP_VAL1_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_VAL1]] to i8*
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL1_BCAST]],{{.+}})
+// CHECK: ret void
+
+#endif
diff --git a/test/OpenMP/distribute_parallel_for_num_threads_codegen.cpp b/test/OpenMP/distribute_parallel_for_num_threads_codegen.cpp
new file mode 100644
index 000000000000..89e990504439
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_num_threads_codegen.cpp
@@ -0,0 +1,121 @@
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -emit-llvm %s -fexceptions -fcxx-exceptions -o - | FileCheck %s
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -std=c++11 -triple %itanium_abi_triple -fexceptions -fcxx-exceptions -emit-pch -o %t %s
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -fexceptions -fcxx-exceptions -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+typedef __INTPTR_TYPE__ intptr_t;
+
+// CHECK-DAG: [[IDENT_T_TY:%.+]] = type { i32, i32, i32, i32, i8* }
+// CHECK-DAG: [[S_TY:%.+]] = type { [[INTPTR_T_TY:i[0-9]+]], [[INTPTR_T_TY]], [[INTPTR_T_TY]] }
+// CHECK-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00"
+// CHECK-DAG: [[DEF_LOC_2:@.+]] = private unnamed_addr constant [[IDENT_T_TY]] { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) }
+
+void foo();
+
+struct S {
+  intptr_t a, b, c;
+  S(intptr_t a) : a(a) {}
+  operator char() { return a; }
+  ~S() {}
+};
+
+template <typename T, int C>
+int tmain() {
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for num_threads(C)
+  for (int i = 0; i < 100; i++)
+    foo();
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for num_threads(T(23))
+  for (int i = 0; i < 100; i++)
+    foo();
+  return 0;
+}
+
+int main() {
+  S s(0);
+  char a = s;
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_0:@.+]](
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOADING_FUN_1:@.+]](
+// CHECK: invoke{{.+}} [[TMAIN_5:@.+]]()
+// CHECK: invoke{{.+}} [[TMAIN_1:@.+]]()
+#pragma omp target
+#pragma omp teams
+  // CHECK: define internal void [[OFFLOADING_FUN_0]](
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}})
+#pragma omp distribute parallel for num_threads(2)
+  for (int i = 0; i < 100; i++) {
+    // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_0]](
+    // CHECK:       call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 2)
+    // CHECK:       call {{.*}}void {{.*}} @__kmpc_fork_call(
+    foo();
+  }
+#pragma omp target
+#pragma omp teams
+  // CHECK: define internal void [[OFFLOADING_FUN_1]](
+
+  // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 1, {{.+}}* [[OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}})
+#pragma omp distribute parallel for num_threads(a)
+  for (int i = 0; i < 100; i++) {
+    // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_1]](
+    // CHECK-DAG: [[A_ADDR:%.+]] = alloca i8*,
+    // CHECK-DAG: [[A_REF:%.+]] = load i8*, i8** [[A_ADDR]],
+    // CHECK-DAG: [[A_VAL:%.+]] = load i8, i8* [[A_REF]],
+    // CHECK-DAG: [[A_EXT:%.+]] = sext i8 [[A_VAL]] to {{.+}}
+    // CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 [[A_EXT]])
+    // CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call(
+    foo();
+  }
+  return a + tmain<char, 5>() + tmain<S, 1>();
+}
+
+// tmain 5
+// CHECK-DAG: define {{.*}}i{{[0-9]+}} [[TMAIN_5]]()
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[T_OFFLOADING_FUN_0:@.+]](
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[T_OFFLOADING_FUN_1:@.+]](
+
+// tmain 1
+// CHECK-DAG: define {{.*}}i{{[0-9]+}} [[TMAIN_1]]()
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[T_OFFLOADING_FUN_2:@.+]](
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[T_OFFLOADING_FUN_3:@.+]](
+
+// CHECK: define internal void [[T_OFFLOADING_FUN_0]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_0]](
+// CHECK:       call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 5)
+// CHECK:       call {{.*}}void {{.*}} @__kmpc_fork_call(
+
+// CHECK: define internal void [[T_OFFLOADING_FUN_1]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_1]](
+// CHECK:       call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 23)
+// CHECK:       call {{.*}}void {{.*}} @__kmpc_fork_call(
+
+// CHECK: define internal void [[T_OFFLOADING_FUN_2]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_2:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_2]](
+// CHECK:       call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 1)
+// CHECK:       call {{.*}}void {{.*}} @__kmpc_fork_call(
+
+// CHECK: define internal void [[T_OFFLOADING_FUN_3]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_3:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_3]](
+// CHECK-DAG:   [[CALL_RES:%.+]] = invoke{{.*}} i8 [[S_TY_CHAR_OP:@.+]]([[S_TY]]* {{.+}})
+// CHECK-DAG:   [[CALL_RES_SEXT:%.+]] = sext i8 [[CALL_RES]] to {{.+}}
+// CHECK:       call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 [[CALL_RES_SEXT]])
+// CHECK:       call {{.*}}void {{.*}} @__kmpc_fork_call(
+#endif
diff --git a/test/OpenMP/distribute_parallel_for_private_codegen.cpp b/test/OpenMP/distribute_parallel_for_private_codegen.cpp
new file mode 100644
index 000000000000..d8d8a9a1edec
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_private_codegen.cpp
@@ -0,0 +1,297 @@
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64
+// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32
+
+// RUN: %clang_cc1  -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1  -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1  -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+template <class T>
+struct S {
+  T f;
+  S(T a) : f(a) {}
+  S() : f() {}
+  operator T() { return T(); }
+  ~S() {}
+};
+
+// CHECK: [[S_FLOAT_TY:%.+]] = type { float }
+// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
+template <typename T>
+T tmain() {
+  S<T> test;
+  T t_var = T();
+  T vec[] = {1, 2};
+  S<T> s_arr[] = {1, 2};
+  S<T> &var = test;
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for private(t_var, vec, s_arr, s_arr, var, var)
+  for (int i = 0; i < 2; ++i) {
+    vec[i] = t_var;
+    s_arr[i] = var;
+  }
+  return T();
+}
+
+int main() {
+  static int svar;
+  volatile double g;
+  volatile double &g1 = g;
+
+  #ifdef LAMBDA
+  // LAMBDA-LABEL: @main
+  // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
+  [&]() {
+    static float sfvar;
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
+    // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams(
+    // LAMBDA: call void [[OFFLOADING_FUN:@.+]](
+
+    // LAMBDA: define{{.+}} void [[OFFLOADING_FUN]]()
+    // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_OUTLINED:@.+]] to {{.+}})
+    #pragma omp target
+    #pragma omp teams
+    #pragma omp distribute parallel for private(g, g1, svar, sfvar)
+    for (int i = 0; i < 2; ++i) {
+      // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
+      // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
+      // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double,
+      // LAMBDA: [[TMP_PRIVATE_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float,
+      // LAMBDA: store double* [[G1_PRIVATE_ADDR]], double** [[TMP_PRIVATE_ADDR]],
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
+      // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to {{.+}},
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_fini(
+      // LAMBDA: ret void
+
+      // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED]](
+      // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
+      // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double,
+      // LAMBDA: [[TMP_PRIVATE_ADDR:%.+]] = alloca double*,
+      // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float,
+
+      g = 1;
+      g1 = 1;
+      svar = 3;
+      sfvar = 4.0;
+      // LAMBDA: store double* [[G1_PRIVATE_ADDR]], double** [[TMP_PRIVATE_ADDR]],
+      // LAMBDA: store double 1.0{{.+}}, double* [[G_PRIVATE_ADDR]],
+      // LAMBDA: store i{{[0-9]+}} 3, i{{[0-9]+}}* [[SVAR_PRIVATE_ADDR]],
+      // LAMBDA: store float 4.0{{.+}}, float* [[SFVAR_PRIVATE_ADDR]],
+      // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+      // LAMBDA: store double* [[G_PRIVATE_ADDR]], double** [[G_PRIVATE_ADDR_REF]],
+      // LAMBDA: [[TMP_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+      // LAMBDA: [[G1_PRIVATE_ADDR_FROM_TMP:%.+]] = load double*, double** [[TMP_PRIVATE_ADDR]],
+      // LAMBDA: store double* [[G1_PRIVATE_ADDR_FROM_TMP]], double** [[TMP_PRIVATE_ADDR_REF]],
+      // LAMBDA: [[SVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 2
+      // LAMBDA: store i{{[0-9]+}}* [[SVAR_PRIVATE_ADDR]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA: [[SFVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
+      // LAMBDA: store float* [[SFVAR_PRIVATE_ADDR]], float** [[SFVAR_PRIVATE_ADDR_REF]]
+      // LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]])
+      // LAMBDA: call {{.*}}void @__kmpc_for_static_fini(
+      // LAMBDA: ret void
+      [&]() {
+	// LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]])
+	// LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]],
+	g = 2;
+	g1 = 2;
+	svar = 4;
+	sfvar = 8.0;
+	// LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]]
+	// LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+	// LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]]
+	// LAMBDA: store double 2.0{{.+}}, double* [[G_REF]]
+
+	// LAMBDA: [[TMP_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+	// LAMBDA: [[G1_REF:%.+]] = load double*, double** [[TMP_PTR_REF]]
+	// LAMBDA: store double 2.0{{.+}}, double* [[G1_REF]],
+	// LAMBDA: [[SVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 2
+	// LAMBDA: [[SVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PTR_REF]]
+	// LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SVAR_REF]]
+	// LAMBDA: [[SFVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
+	// LAMBDA: [[SFVAR_REF:%.+]] = load float*, float** [[SFVAR_PTR_REF]]
+	// LAMBDA: store float 8.0{{.+}}, float* [[SFVAR_REF]]
+      }();
+    }
+  }();
+  return 0;
+  #else
+  S<float> test;
+  int t_var = 0;
+  int vec[] = {1, 2};
+  S<float> s_arr[] = {1, 2};
+  S<float> &var = test;
+
+  #pragma omp target
+  #pragma omp teams
+  #pragma omp distribute parallel for private(t_var, vec, s_arr, s_arr, var, var, svar)
+  for (int i = 0; i < 2; ++i) {
+    vec[i] = t_var;
+    s_arr[i] = var;
+  }
+  return tmain<int>();
+  #endif
+}
+
+// CHECK: define{{.*}} i{{[0-9]+}} @main()
+// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOAD_FUN_0:@.+]](
+
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
+// CHECK: ret
+
+// CHECK: define{{.+}} [[OFFLOAD_FUN_0]]()
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[OMP_OUTLINED_0:@.+]] to void
+// CHECK: ret
+//
+// CHECK: define internal void [[OMP_OUTLINED_0]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
+// CHECK-NOT: alloca [2 x [[S_FLOAT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
+// CHECK-NOT: alloca [[S_FLOAT_TY]],
+// CHECK: [[S_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// this is the ctor loop
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_FLOAT_TY]]*
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_0:@.+]] to {{.+}},
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+// By OpenMP specifications, private applies to both distribute and parallel for.
+// However, the support for 'private' of 'parallel' is only used when 'parallel'
+// is found alone. Therefore we only have one 'private' support for 'parallel for'
+// in combination
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_0]](
+// CHECK: [[T_VAR_PRIV:%t_var+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%vec+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%s_arr+]] = alloca [2 x [[S_FLOAT_TY]]],
+// CHECK-NOT: alloca [2 x [[S_FLOAT_TY]]],
+// CHECK: [[VAR_PRIV:%var+]] = alloca [[S_FLOAT_TY]],
+// CHECK-NOT: alloca [[S_FLOAT_TY]],
+// CHECK: [[S_VAR_PRIV:%svar+]] = alloca i{{[0-9]+}},
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// this is the ctor loop
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_FLOAT_TY]]*
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+// template tmain with S_INT_TY
+// CHECK: define{{.*}} i{{[0-9]+}} [[TMAIN_INT:@.+]]()
+// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
+// CHECK: call i{{[0-9]+}} @__tgt_target_teams(
+// CHECK: call void [[OFFLOAD_FUN_1:@.+]](
+// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR:@.+]]([[S_INT_TY]]* [[TEST]])
+// CHECK: ret
+
+// CHECK: ret
+
+// CHECK: define internal void [[OFFLOAD_FUN_1]]()
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[OMP_OUTLINED_1:@.+]] to void
+// CHECK: ret
+//
+// CHECK: define internal void [[OMP_OUTLINED_1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
+// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
+// CHECK-NOT: alloca [2 x [[S_INT_TY]]],
+// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
+// CHECK-NOT: alloca [[S_INT_TY]],
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]*
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}},
+// CHECK: call void @__kmpc_for_static_fini(
+// CHECK: ret void
+
+// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]](
+// CHECK: [[T_VAR_PRIV:%t_var+]] = alloca i{{[0-9]+}},
+// CHECK: [[VEC_PRIV:%vec+]] = alloca [2 x i{{[0-9]+}}],
+// CHECK: [[S_ARR_PRIV:%s_arr+]] = alloca [2 x [[S_INT_TY]]],
+// CHECK-NOT: alloca [2 x [[S_INT_TY]]],
+// CHECK: [[VAR_PRIV:%var+]] = alloca [[S_INT_TY]],
+// CHECK-NOT: alloca [[S_INT_TY]],
+// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// this is the ctor loop
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]*
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
+// CHECK-NOT: [[T_VAR_PRIV]]
+// CHECK-NOT: [[VEC_PRIV]]
+// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
+// CHECK: call void @__kmpc_for_static_init_4(
+// CHECK: call void @__kmpc_for_static_fini(
+
+// call destructors: var..
+// CHECK-DAG: call {{.+}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]])
+
+// ..and s_arr
+// CHECK: {{.+}}:
+// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_INT_TY]]*
+// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]],
+// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
+
+// CHECK: ret void
+
+#endif
diff --git a/test/OpenMP/distribute_parallel_for_proc_bind_codegen.cpp b/test/OpenMP/distribute_parallel_for_proc_bind_codegen.cpp
new file mode 100644
index 000000000000..958b4e782f10
--- /dev/null
+++ b/test/OpenMP/distribute_parallel_for_proc_bind_codegen.cpp
@@ -0,0 +1,93 @@
+// add -fopenmp-targets
+
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+typedef __INTPTR_TYPE__ intptr_t;
+
+// CHECK-DAG: [[IDENT_T_TY:%.+]] = type { i32, i32, i32, i32, i8* }
+// CHECK-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00"
+// CHECK-DAG: [[DEF_LOC_2:@.+]] = private unnamed_addr constant [[IDENT_T_TY]] { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) }
+
+void foo();
+
+struct S {
+  intptr_t a, b, c;
+  S(intptr_t a) : a(a) {}
+  operator char() { return a; }
+  ~S() {}
+};
+
+template <typename T>
+T tmain() {
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for proc_bind(master)
+  for(int i = 0; i < 1000; i++) {}
+  return T();
+}
+
+int main() {
+  // CHECK-LABEL: @main
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for proc_bind(spread)
+  for(int i = 0; i < 1000; i++) {}
+#pragma omp target
+#pragma omp teams
+#pragma omp distribute parallel for proc_bind(close)
+  for(int i = 0; i < 1000; i++) {}
+  return tmain<int>();
+}
+
+// CHECK: call {{.*}}@__tgt_target_teams({{.+}})
+// CHECK: call void [[OFFL1:@.+]]()
+// CHECK: call {{.*}}@__tgt_target_teams({{.+}})
+// CHECK: call void [[OFFL2:@.+]]()
+// CHECK: [[CALL_RET:%.+]] = call{{.+}} i32 [[TMAIN:@.+]]()
+// CHECK: ret i32 [[CALL_RET]]
+
+// CHECK: define{{.+}} void [[OFFL1]](
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, {{.+}}, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} [[OMP_OUTLINED_1]](i32* {{.+}} [[GTID_IN:%.+]],
+// CHECK: [[GTID_ADDR:%.+]] = alloca i32*,
+// CHECK: store i32* [[GTID_IN]], i32** [[GTID_ADDR]],
+// CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_ADDR]],
+// CHECK: [[GTID_VAL:%.+]] = load i32, i32* [[GTID_REF]],
+// CHECK: call {{.*}}void @__kmpc_push_proc_bind([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 [[GTID_VAL]], i32 4)
+// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(
+// CHECK: ret void
+
+// CHECK: define{{.+}} [[OFFL2]]()
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, {{.+}}, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} [[OMP_OUTLINED_1]](i32* {{.+}} [[GTID_IN:%.+]],
+// CHECK: [[GTID_ADDR:%.+]] = alloca i32*,
+// CHECK: store i32* [[GTID_IN]], i32** [[GTID_ADDR]],
+// CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_ADDR]],
+// CHECK: [[GTID_VAL:%.+]] = load i32, i32* [[GTID_REF]],
+// CHECK: call {{.*}}void @__kmpc_push_proc_bind([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 [[GTID_VAL]], i32 3)
+// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(
+// CHECK: ret void
+
+// CHECK: define{{.+}} [[TMAIN]]()
+// CHECK: call {{.*}}@__tgt_target_teams({{.+}})
+// CHECK: call void [[OFFL3:@.+]]()
+
+// CHECK: define{{.+}} [[OFFL3]]()
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, {{.+}}, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}})
+
+// CHECK: define{{.+}} [[OMP_OUTLINED_3]](i32* {{.+}} [[GTID_IN:%.+]],
+// CHECK: [[GTID_ADDR:%.+]] = alloca i32*,
+// CHECK: store i32* [[GTID_IN]], i32** [[GTID_ADDR]],
+// CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_ADDR]],
+// CHECK: [[GTID_VAL:%.+]] = load i32, i32* [[GTID_REF]],
+// CHECK: call {{.*}}void @__kmpc_push_proc_bind([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 [[GTID_VAL]], i32 2)
+// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(
+// CHECK: ret void
+#endif
diff --git a/test/Parser/cxx-modules-import.cpp b/test/Parser/cxx-modules-import.cpp
deleted file mode 100644
index 0600eddecb83..000000000000
--- a/test/Parser/cxx-modules-import.cpp
+++ /dev/null
@@ -1,58 +0,0 @@
-// RUN: rm -rf %t
-// RUN: mkdir -p %t
-// RUN: echo 'module x; int a, b;' > %t/x.cppm
-// RUN: echo 'module x.y; int c;' > %t/x.y.cppm
-//
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %t/x.cppm -o %t/x.pcm
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface -fmodule-file=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm
-//
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=1 -DMODULE_KIND=implementation -DMODULE_NAME=z
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=2 -DMODULE_KIND=implementation -DMODULE_NAME=x
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=3 -DMODULE_KIND= -DMODULE_NAME=z
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=4 -DMODULE_KIND=partition -DMODULE_NAME=z
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=5 -DMODULE_KIND=elderberry -DMODULE_NAME=z
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=1 -DMODULE_KIND=implementation -DMODULE_NAME='z [[]]'
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=6 -DMODULE_KIND=implementation -DMODULE_NAME='z [[fancy]]'
-// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \
-// RUN:            -DTEST=7 -DMODULE_KIND=implementation -DMODULE_NAME='z [[maybe_unused]]'
-
-module MODULE_KIND MODULE_NAME;
-#if TEST == 3
-// expected-error@-2 {{'module' declaration found while not building module interface}}
-#elif TEST == 4
-// expected-error@-4 {{'module partition' declaration found while not building module interface}}
-#elif TEST == 5
-// expected-error@-6 {{unexpected module kind 'elderberry'}}
-#elif TEST == 6
-// expected-warning@-8 {{unknown attribute 'fancy' ignored}}
-#elif TEST == 7
-// expected-error-re@-10 {{'maybe_unused' attribute cannot be applied to a module{{$}}}}
-#endif
-
-int use_1 = a;
-#if TEST != 2
-// expected-error@-2 {{declaration of 'a' must be imported from module 'x' before it is required}}
-// expected-note@x.cppm:1 {{here}}
-#endif
-
-import x;
-
-int use_2 = b; // ok
-
-import x [[]];
-import x [[foo]]; // expected-warning {{unknown attribute 'foo' ignored}}
-import x [[noreturn]]; // expected-error {{'noreturn' attribute cannot be applied to a module import}}
-import x [[blarg::noreturn]]; // expected-warning {{unknown attribute 'noreturn' ignored}}
-
-import x.y;
-import x.; // expected-error {{expected a module name after module import}}
-import .x; // expected-error {{expected a module name after module import}}
-
-import blarg; // expected-error {{module 'blarg' not found}}
diff --git a/test/Parser/cxx-modules-interface.cppm b/test/Parser/cxx-modules-interface.cppm
index f7835bd167b2..0e7c74683353 100644
--- a/test/Parser/cxx-modules-interface.cppm
+++ b/test/Parser/cxx-modules-interface.cppm
@@ -1,13 +1,12 @@
 // RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify
-// RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -Dmodule=int -DERRORS
+// RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -DERRORS
 
-module foo;
+export module foo;
 #ifndef ERRORS
 // expected-no-diagnostics
 #else
-// expected-error@-4 {{expected module declaration at start of module interface}}
+// FIXME: diagnose missing module-declaration when building module interface
 
-// FIXME: support 'export module X;' and 'export { int n; module X; }'
 // FIXME: proclaimed-ownership-declarations?
 
 export {
diff --git a/test/Preprocessor/init.c b/test/Preprocessor/init.c
index 7548fbaae2eb..5fb2011cd025 100644
--- a/test/Preprocessor/init.c
+++ b/test/Preprocessor/init.c
@@ -8846,6 +8846,16 @@
 // WEBASSEMBLY32-NEXT:#define __CHAR32_TYPE__ unsigned int
 // WEBASSEMBLY32-NEXT:#define __CHAR_BIT__ 8
 // WEBASSEMBLY32-NOT:#define __CHAR_UNSIGNED__
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 1
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2
+// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2
 // WEBASSEMBLY32-NEXT:#define __CONSTANT_CFSTRINGS__ 1
 // WEBASSEMBLY32-NEXT:#define __DBL_DECIMAL_DIG__ 17
 // WEBASSEMBLY32-NEXT:#define __DBL_DENORM_MIN__ 4.9406564584124654e-324
@@ -9162,6 +9172,16 @@
 // WEBASSEMBLY64-NEXT:#define __CHAR32_TYPE__ unsigned int
 // WEBASSEMBLY64-NEXT:#define __CHAR_BIT__ 8
 // WEBASSEMBLY64-NOT:#define __CHAR_UNSIGNED__
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2
+// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2
 // WEBASSEMBLY64-NEXT:#define __CONSTANT_CFSTRINGS__ 1
 // WEBASSEMBLY64-NEXT:#define __DBL_DECIMAL_DIG__ 17
 // WEBASSEMBLY64-NEXT:#define __DBL_DENORM_MIN__ 4.9406564584124654e-324
@@ -9637,3 +9657,36 @@
 // AVR:#define __WCHAR_MAX__ 32767
 // AVR:#define __WCHAR_TYPE__ int
 // AVR:#define __WINT_TYPE__ int
+
+
+// RUN: %clang_cc1 -E -dM -ffreestanding \
+// RUN:    -triple i686-windows-msvc -fms-compatibility < /dev/null \
+// RUN:  | FileCheck -match-full-lines -check-prefix MSVC-X32 %s
+
+// RUN: %clang_cc1 -E -dM -ffreestanding \
+// RUN:    -triple x86_64-windows-msvc -fms-compatibility < /dev/null \
+// RUN:  | FileCheck -match-full-lines -check-prefix MSVC-X64 %s
+
+// MSVC-X32:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2
+// MSVC-X32-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2
+// MSVC-X32-NOT:#define __GCC_ATOMIC{{.*}}
+
+// MSVC-X64:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2
+// MSVC-X64-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2
+// MSVC-X86-NOT:#define __GCC_ATOMIC{{.*}}
diff --git a/test/Sema/atomic-ops.c b/test/Sema/atomic-ops.c
index 8ebf3eaed4af..eee1cda8635e 100644
--- a/test/Sema/atomic-ops.c
+++ b/test/Sema/atomic-ops.c
@@ -7,19 +7,29 @@
 struct S { char c[3]; };
 
 _Static_assert(__GCC_ATOMIC_BOOL_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_BOOL_LOCK_FREE == __CLANG_ATOMIC_BOOL_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_CHAR_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_CHAR_LOCK_FREE == __CLANG_ATOMIC_CHAR_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_CHAR16_T_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_CHAR16_T_LOCK_FREE == __CLANG_ATOMIC_CHAR16_T_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_CHAR32_T_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_CHAR32_T_LOCK_FREE == __CLANG_ATOMIC_CHAR32_T_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_WCHAR_T_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_WCHAR_T_LOCK_FREE == __CLANG_ATOMIC_WCHAR_T_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_SHORT_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_SHORT_LOCK_FREE == __CLANG_ATOMIC_SHORT_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_INT_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_INT_LOCK_FREE == __CLANG_ATOMIC_INT_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_LONG_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_LONG_LOCK_FREE == __CLANG_ATOMIC_LONG_LOCK_FREE, "");
 #ifdef __i386__
 _Static_assert(__GCC_ATOMIC_LLONG_LOCK_FREE == 1, "");
 #else
 _Static_assert(__GCC_ATOMIC_LLONG_LOCK_FREE == 2, "");
 #endif
+_Static_assert(__GCC_ATOMIC_LLONG_LOCK_FREE == __CLANG_ATOMIC_LLONG_LOCK_FREE, "");
 _Static_assert(__GCC_ATOMIC_POINTER_LOCK_FREE == 2, "");
+_Static_assert(__GCC_ATOMIC_POINTER_LOCK_FREE == __CLANG_ATOMIC_POINTER_LOCK_FREE, "");
 
 _Static_assert(__c11_atomic_is_lock_free(1), "");
 _Static_assert(__c11_atomic_is_lock_free(2), "");
diff --git a/test/Sema/warn-documentation.cpp b/test/Sema/warn-documentation.cpp
index 34d8f5fd2da2..0c92b2aa029f 100644
--- a/test/Sema/warn-documentation.cpp
+++ b/test/Sema/warn-documentation.cpp
@@ -1210,3 +1210,75 @@ template <class T> T test_function (T arg);
 /*!     @function test_function<int>
 */
 template <> int test_function<int> (int arg);
+
+namespace AllowParamAndReturnsOnFunctionPointerVars {
+
+/**
+ * functionPointerVariable
+ *
+ * @param i is integer.
+ * @returns integer.
+ */
+int (*functionPointerVariable)(int i);
+
+struct HasFields {
+  /**
+   * functionPointerField
+   *
+   * @param i is integer.
+   * @returns integer.
+   */
+  int (*functionPointerField)(int i);
+};
+
+// expected-warning@+5 {{'\returns' command used in a comment that is attached to a function returning void}}
+/**
+ * functionPointerVariable
+ *
+ * \param p not here.
+ * \returns integer.
+ */
+void (*functionPointerVariableThatLeadsNowhere)();
+
+// Still warn about param/returns commands for variables that don't specify
+// the type directly:
+
+/**
+ * FunctionPointerTypedef
+ *
+ * \param i is integer.
+ * \returns integer.
+ */
+typedef int (*FunctionPointerTypedef)(int i);
+
+/**
+ * FunctionPointerTypealias
+ *
+ * \param i is integer.
+ * \returns integer.
+ */
+using FunctionPointerTypealias = int (*)(int i);
+
+// expected-warning@+5 {{'@param' command used in a comment that is not attached to a function declaration}}
+// expected-warning@+5 {{'@returns' command used in a comment that is not attached to a function or method declaration}}
+/**
+ * functionPointerVariable
+ *
+ * @param i is integer.
+ * @returns integer.
+ */
+FunctionPointerTypedef functionPointerTypedefVariable;
+
+struct HasMoreFields {
+  // expected-warning@+5 {{'\param' command used in a comment that is not attached to a function declaration}}
+  // expected-warning@+5 {{'\returns' command used in a comment that is not attached to a function or method declaration}}
+  /**
+   * functionPointerTypealiasField
+   *
+   * \param i is integer.
+   * \returns integer.
+   */
+  FunctionPointerTypealias functionPointerTypealiasField;
+};
+
+}
diff --git a/test/Sema/warn-documentation.m b/test/Sema/warn-documentation.m
index 5e95e2a1e8a2..3c1a369c4b14 100644
--- a/test/Sema/warn-documentation.m
+++ b/test/Sema/warn-documentation.m
@@ -229,3 +229,73 @@ int FooBar();
 - (void) VarArgMeth : (id)arg, ... {}
 @end
 
+/**
+ * blockPointerVariable
+ *
+ * @param i is integer.
+ * @returns integer.
+ */
+int (^blockPointerVariable)(int i);
+
+struct HasFields {
+  /**
+   * blockPointerField
+   *
+   * \param i is integer.
+   * \returns integer.
+   */
+  int (^blockPointerFields)(int i);
+};
+
+// expected-warning@+5 {{'\returns' command used in a comment that is attached to a function returning void}}
+/**
+ * functionPointerVariable
+ *
+ * \param p not here.
+ * \returns integer.
+ */
+void (^_Nullable blockPointerVariableThatLeadsNowhere)();
+
+@interface CheckFunctionBlockPointerVars {
+  /**
+   * functionPointerIVar
+   *
+   * @param i is integer.
+   * @returns integer.
+   */
+  int (*functionPointerIVar)(int i);
+
+  /**
+   * blockPointerIVar
+   *
+   * \param i is integer.
+   * \returns integer.
+   */
+  int (^blockPointerIVar)(int i);
+}
+
+/**
+ * functionPointerProperty
+ *
+ * @param i is integer.
+ * @returns integer.
+ */
+@property int (*functionPointerProperty)(int i);
+
+/**
+ * blockPointerProperty
+ *
+ * \param i is integer.
+ * \returns integer.
+ */
+@property int (^blockPointerProperty)(int i);
+
+/**
+ * blockReturnsNothing
+ *
+ * \returns Nothing, but can allow this as this pattern is used to document the
+ * value that the property getter returns.
+ */
+@property void (^blockReturnsNothing)();
+
+@end
diff --git a/test/SemaCXX/MicrosoftCompatibility-cxx98.cpp b/test/SemaCXX/MicrosoftCompatibility-cxx98.cpp
deleted file mode 100644
index 3b80d0937e23..000000000000
--- a/test/SemaCXX/MicrosoftCompatibility-cxx98.cpp
+++ /dev/null
@@ -1,23 +0,0 @@
-// RUN: %clang_cc1 %s -triple i686-pc-win32 -fsyntax-only -std=c++98 -Wmicrosoft -verify -fms-compatibility -fexceptions -fcxx-exceptions
-
-
-//MSVC allows forward enum declaration
-enum ENUM; // expected-warning {{forward references to 'enum' types are a Microsoft extension}}
-ENUM *var = 0;     
-ENUM var2 = (ENUM)3;
-enum ENUM1* var3 = 0;// expected-warning {{forward references to 'enum' types are a Microsoft extension}}
-
-typedef void (*FnPtrTy)();
-void (*PR23733_1)() = static_cast<FnPtrTy>((void *)0); // expected-warning {{static_cast between pointer-to-function and pointer-to-object is a Microsoft extension}}
-void (*PR23733_2)() = FnPtrTy((void *)0);
-void (*PR23733_3)() = (FnPtrTy)((void *)0);
-void (*PR23733_4)() = reinterpret_cast<FnPtrTy>((void *)0);
-
-long function_prototype(int a);
-long (*function_ptr)(int a);
-
-void function_to_voidptr_conv() {
-  void *a1 = function_prototype;  // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}}
-  void *a2 = &function_prototype; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}}
-  void *a3 = function_ptr;        // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}}
-}
diff --git a/test/SemaCXX/MicrosoftCompatibility.cpp b/test/SemaCXX/MicrosoftCompatibility.cpp
index 26cd7825ee87..d095f6edc617 100644
--- a/test/SemaCXX/MicrosoftCompatibility.cpp
+++ b/test/SemaCXX/MicrosoftCompatibility.cpp
@@ -224,6 +224,15 @@ template void function_missing_typename<D>(const D::Type param);
 
 }
 
+//MSVC allows forward enum declaration
+enum ENUM; // expected-warning {{forward references to 'enum' types are a Microsoft extension}}
+ENUM *var = 0;     
+ENUM var2 = (ENUM)3;
+enum ENUM1* var3 = 0;// expected-warning {{forward references to 'enum' types are a Microsoft extension}}
+
+enum ENUM1 { kA };
+enum ENUM1;  // This way round is fine.
+
 enum ENUM2 {
 	ENUM2_a = (enum ENUM2) 4,
 	ENUM2_b = 0x9FFFFFFF, // expected-warning {{enumerator value is not representable in the underlying type 'int'}}
@@ -269,3 +278,18 @@ void g() {
   f(0xffffffffffffffffi64);
 }
 }
+
+typedef void (*FnPtrTy)();
+void (*PR23733_1)() = static_cast<FnPtrTy>((void *)0); // expected-warning {{static_cast between pointer-to-function and pointer-to-object is a Microsoft extension}}
+void (*PR23733_2)() = FnPtrTy((void *)0);
+void (*PR23733_3)() = (FnPtrTy)((void *)0);
+void (*PR23733_4)() = reinterpret_cast<FnPtrTy>((void *)0);
+
+long function_prototype(int a);
+long (*function_ptr)(int a);
+
+void function_to_voidptr_conv() {
+  void *a1 = function_prototype;  // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}}
+  void *a2 = &function_prototype; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}}
+  void *a3 = function_ptr;        // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}}
+}
diff --git a/test/SemaCXX/MicrosoftExtensions.cpp b/test/SemaCXX/MicrosoftExtensions.cpp
index 96088a084558..38949e11376d 100644
--- a/test/SemaCXX/MicrosoftExtensions.cpp
+++ b/test/SemaCXX/MicrosoftExtensions.cpp
@@ -2,6 +2,7 @@
 // RUN: %clang_cc1 -std=c++98 %s -triple i686-pc-win32 -fsyntax-only -Wmicrosoft -Wc++11-extensions -Wno-long-long -verify -fms-extensions -fexceptions -fcxx-exceptions -DTEST1
 // RUN: %clang_cc1 -std=c++11 %s -triple i686-pc-win32 -fsyntax-only -Wmicrosoft -Wc++11-extensions -Wno-long-long -verify -fms-extensions -fexceptions -fcxx-exceptions -DTEST1
 // RUN: %clang_cc1 %s -triple i686-pc-win32 -fsyntax-only -Wmicrosoft -Wc++11-extensions -Wno-long-long -verify -fexceptions -fcxx-exceptions -DTEST2
+// RUN: %clang_cc1 %s -triple i686-pc-win32 -fsyntax-only -std=c++11 -fms-compatibility -verify -DTEST3
 
 #if TEST1
 
@@ -508,6 +509,13 @@ int S::fn() { return 0; } // expected-warning {{is missing exception specificati
 // Check that __unaligned is not recognized if MS extensions are not enabled
 typedef char __unaligned *aligned_type; // expected-error {{expected ';' after top level declarator}}
 
+#elif TEST3
+
+namespace PR32750 {
+template<typename T> struct A {};
+template<typename T> struct B : A<A<T>> { A<T>::C::D d; }; // expected-error {{missing 'typename' prior to dependent type name 'A<T>::C::D'}}
+}
+
 #else
 
 #error Unknown test mode
diff --git a/test/SemaCXX/modules-ts.cppm b/test/SemaCXX/modules-ts.cppm
index 71c09d3bfe6c..16695f6463a8 100644
--- a/test/SemaCXX/modules-ts.cppm
+++ b/test/SemaCXX/modules-ts.cppm
@@ -1,22 +1,21 @@
 // RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -DTEST=0
-// RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -Dmodule=int -DTEST=1
-// RUN: not %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -fmodule-file=%t.pcm -o %t.pcm -DTEST=2 2>&1 | FileCheck %s --check-prefix=CHECK-2
+// RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -DTEST=1
+// RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -fmodule-file=%t.pcm -o %t.pcm -verify -DTEST=2
 // RUN:     %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -fmodule-file=%t.pcm -o %t.pcm -verify -Dfoo=bar -DTEST=3
 
 #if TEST == 0
 // expected-no-diagnostics
 #endif
 
-module foo;
-#if TEST == 1
-// expected-error@-2 {{expected module declaration at start of module interface}}
-#elif TEST == 2
-// CHECK-2: error: redefinition of module 'foo'
+export module foo;
+#if TEST == 2
+// expected-error@-2 {{redefinition of module 'foo'}}
+// expected-note@modules-ts.cppm:* {{loaded from}}
 #endif
 
 static int m; // ok, internal linkage, so no redefinition error
 int n;
-#if TEST == 3
+#if TEST >= 2
 // expected-error@-2 {{redefinition of '}}
 // expected-note@-3 {{previous}}
 #endif
diff --git a/test/SemaObjC/unguarded-availability.m b/test/SemaObjC/unguarded-availability.m
index 4369a0e329d2..ae921f4a27b3 100644
--- a/test/SemaObjC/unguarded-availability.m
+++ b/test/SemaObjC/unguarded-availability.m
@@ -1,5 +1,5 @@
 // RUN: %clang_cc1 -triple x86_64-apple-macosx-10.9 -Wunguarded-availability -fblocks -fsyntax-only -verify %s
-// RUN: %clang_cc1 -xobjective-c++ -DOBJCPP -triple x86_64-apple-macosx-10.9 -Wunguarded-availability -fblocks -fsyntax-only -verify %s
+// RUN: %clang_cc1 -xobjective-c++ -std=c++11 -DOBJCPP -triple x86_64-apple-macosx-10.9 -Wunguarded-availability -fblocks -fsyntax-only -verify %s
 
 #define AVAILABLE_10_0  __attribute__((availability(macos, introduced = 10.0)))
 #define AVAILABLE_10_11 __attribute__((availability(macos, introduced = 10.11)))
@@ -8,9 +8,9 @@
 int func_10_11() AVAILABLE_10_11; // expected-note 4 {{'func_10_11' has been explicitly marked partial here}}
 
 #ifdef OBJCPP
-// expected-note@+2 {{marked partial here}}
+// expected-note@+2 2 {{marked partial here}}
 #endif
-int func_10_12() AVAILABLE_10_12; // expected-note 5 {{'func_10_12' has been explicitly marked partial here}}
+int func_10_12() AVAILABLE_10_12; // expected-note 6 {{'func_10_12' has been explicitly marked partial here}}
 
 int func_10_0() AVAILABLE_10_0;
 
@@ -129,6 +129,12 @@ void test_params(int_10_12 x); // expected-warning {{'int_10_12' is partial: int
 
 void test_params2(int_10_12 x) AVAILABLE_10_12; // no warn
 
+void (^topLevelBlockDecl)() = ^ {
+  func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}}
+  if (@available(macos 10.12, *))
+    func_10_12();
+};
+
 #ifdef OBJCPP
 
 int f(char) AVAILABLE_10_12;
@@ -176,4 +182,10 @@ int instantiate_availability() {
     with_availability_attr<int_10_12>(); // expected-warning{{'with_availability_attr<int>' is only available on macOS 10.11 or newer}} expected-warning{{'int_10_12' is only available on macOS 10.12 or newer}} expected-note 2 {{enclose}}
 }
 
+auto topLevelLambda = [] () {
+  func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}}
+  if (@available(macos 10.12, *))
+    func_10_12();
+};
+
 #endif
diff --git a/test/SemaObjCXX/pr32725.mm b/test/SemaObjCXX/pr32725.mm
new file mode 100644
index 000000000000..8d7d1163250f
--- /dev/null
+++ b/test/SemaObjCXX/pr32725.mm
@@ -0,0 +1,9 @@
+// RUN: %clang_cc1 -triple x86_64-apple-macosx -fsyntax-only -verify -x objective-c++ %s -o /dev/null
+// REQUIRES: asserts
+
+struct objc_class {
+  unsigned long long bits;
+};
+typedef struct objc_class *Class;
+static void f(Class c) { (void)(c->bits & RW_HAS_OVERFLOW_REFCOUNT); }
+// expected-error@-1{{use of undeclared identifier 'RW_HAS_OVERFLOW_REFCOUNT}}
diff --git a/test/SemaOpenCL/cl20-device-side-enqueue.cl b/test/SemaOpenCL/cl20-device-side-enqueue.cl
index 3e87cfcddf66..bafbe447ee28 100644
--- a/test/SemaOpenCL/cl20-device-side-enqueue.cl
+++ b/test/SemaOpenCL/cl20-device-side-enqueue.cl
@@ -1,12 +1,14 @@
-// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir-unknown-unknown" -verify -pedantic -fsyntax-only -DB32
-// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir64-unknown-unknown" -verify -pedantic -fsyntax-only -Wconversion -DWCONV
+// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir-unknown-unknown" -verify -pedantic -fsyntax-only -DB32 -DQUALS=
+// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir-unknown-unknown" -verify -pedantic -fsyntax-only -DB32 -DQUALS="const volatile"
+// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir64-unknown-unknown" -verify -pedantic -fsyntax-only -Wconversion -DWCONV -DQUALS=
+// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir64-unknown-unknown" -verify -pedantic -fsyntax-only -Wconversion -DWCONV -DQUALS="const volatile"
 
 typedef struct {int a;} ndrange_t;
 // Diagnostic tests for different overloads of enqueue_kernel from Table 6.13.17.1 of OpenCL 2.0 Spec.
 kernel void enqueue_kernel_tests() {
   queue_t default_queue;
   unsigned flags = 0;
-  ndrange_t ndrange;
+  QUALS ndrange_t ndrange;
   clk_event_t evt;
   clk_event_t event_wait_list;
   clk_event_t event_wait_list2[] = {evt, evt};
diff --git a/tools/c-index-test/c-index-test.c b/tools/c-index-test/c-index-test.c
index 0dde4c7cc764..e8763ffc00ae 100644
--- a/tools/c-index-test/c-index-test.c
+++ b/tools/c-index-test/c-index-test.c
@@ -3014,6 +3014,7 @@ static const char *getEntityLanguageString(CXIdxEntityLanguage kind) {
   case CXIdxEntityLang_C: return "C";
   case CXIdxEntityLang_ObjC: return "ObjC";
   case CXIdxEntityLang_CXX: return "C++";
+  case CXIdxEntityLang_Swift: return "Swift";
   }
   assert(0 && "Garbage language kind");
   return 0;
diff --git a/tools/clang-format/ClangFormat.cpp b/tools/clang-format/ClangFormat.cpp
index ac0d0a8512f4..14bff19a1a0c 100644
--- a/tools/clang-format/ClangFormat.cpp
+++ b/tools/clang-format/ClangFormat.cpp
@@ -276,14 +276,17 @@ static bool format(StringRef FileName) {
   }
   // Get new affected ranges after sorting `#includes`.
   Ranges = tooling::calculateRangesAfterReplacements(Replaces, Ranges);
-  bool IncompleteFormat = false;
+  FormattingAttemptStatus Status;
   Replacements FormatChanges = reformat(*FormatStyle, *ChangedCode, Ranges,
-                                        AssumedFileName, &IncompleteFormat);
+                                        AssumedFileName, &Status);
   Replaces = Replaces.merge(FormatChanges);
   if (OutputXML) {
     outs() << "<?xml version='1.0'?>\n<replacements "
               "xml:space='preserve' incomplete_format='"
-           << (IncompleteFormat ? "true" : "false") << "'>\n";
+           << (Status.FormatComplete ? "false" : "true") << "'";
+    if (!Status.FormatComplete)
+      outs() << " line=" << Status.Line;
+    outs() << ">\n";
     if (Cursor.getNumOccurrences() != 0)
       outs() << "<cursor>"
              << FormatChanges.getShiftedCodePosition(CursorPosition)
@@ -307,11 +310,15 @@ static bool format(StringRef FileName) {
       if (Rewrite.overwriteChangedFiles())
         return true;
     } else {
-      if (Cursor.getNumOccurrences() != 0)
+      if (Cursor.getNumOccurrences() != 0) {
         outs() << "{ \"Cursor\": "
                << FormatChanges.getShiftedCodePosition(CursorPosition)
                << ", \"IncompleteFormat\": "
-               << (IncompleteFormat ? "true" : "false") << " }\n";
+               << (Status.FormatComplete ? "false" : "true");
+        if (!Status.FormatComplete)
+          outs() << ", \"Line\": " << Status.Line;
+        outs() << " }\n";
+      }
       Rewrite.getEditBuffer(ID).write(outs());
     }
   }
diff --git a/tools/clang-format/git-clang-format b/tools/clang-format/git-clang-format
index 74fd451a847b..3d1ba8a3c107 100755
--- a/tools/clang-format/git-clang-format
+++ b/tools/clang-format/git-clang-format
@@ -23,6 +23,7 @@ git clang-format -h
 Requires Python 2.7                                                              
 """               
 
+from __future__ import print_function
 import argparse
 import collections
 import contextlib
@@ -138,15 +139,15 @@ def main():
   if opts.verbose >= 1:
     ignored_files.difference_update(changed_lines)
     if ignored_files:
-      print 'Ignoring changes in the following files (wrong extension):'
+      print('Ignoring changes in the following files (wrong extension):')
       for filename in ignored_files:
-        print '   ', filename
+        print('    %s' % filename)
     if changed_lines:
-      print 'Running clang-format on the following files:'
+      print('Running clang-format on the following files:')
       for filename in changed_lines:
-        print '   ', filename
+        print('    %s' % filename)
   if not changed_lines:
-    print 'no modified files to format'
+    print('no modified files to format')
     return
   # The computed diff outputs absolute paths, so we must cd before accessing
   # those files.
@@ -163,20 +164,20 @@ def main():
                                                  binary=opts.binary,
                                                  style=opts.style)
   if opts.verbose >= 1:
-    print 'old tree:', old_tree
-    print 'new tree:', new_tree
+    print('old tree: %s' % old_tree)
+    print('new tree: %s' % new_tree)
   if old_tree == new_tree:
     if opts.verbose >= 0:
-      print 'clang-format did not modify any files'
+      print('clang-format did not modify any files')
   elif opts.diff:
     print_diff(old_tree, new_tree)
   else:
     changed_files = apply_changes(old_tree, new_tree, force=opts.force,
                                   patch_mode=opts.patch)
     if (opts.verbose >= 0 and not opts.patch) or opts.verbose >= 1:
-      print 'changed files:'
+      print('changed files:')
       for filename in changed_files:
-        print '   ', filename
+        print('    %s' % filename)
 
 
 def load_git_config(non_string_options=None):
@@ -320,7 +321,7 @@ def filter_by_extension(dictionary, allowed_extensions):
   `allowed_extensions` must be a collection of lowercase file extensions,
   excluding the period."""
   allowed_extensions = frozenset(allowed_extensions)
-  for filename in dictionary.keys():
+  for filename in list(dictionary.keys()):
     base_ext = filename.rsplit('.', 1)
     if len(base_ext) == 1 or base_ext[1].lower() not in allowed_extensions:
       del dictionary[filename]
@@ -344,8 +345,13 @@ def run_clang_format_and_save_to_tree(changed_lines, revision=None,
   """Run clang-format on each file and save the result to a git tree.
 
   Returns the object ID (SHA-1) of the created tree."""
+  def iteritems(container):
+      try:
+          return container.iteritems() # Python 2
+      except AttributeError:
+          return container.items() # Python 3
   def index_info_generator():
-    for filename, line_ranges in changed_lines.iteritems():
+    for filename, line_ranges in iteritems(changed_lines):
       if revision:
         git_metadata_cmd = ['git', 'ls-tree',
                             '%s:%s' % (revision, os.path.dirname(filename)),
@@ -356,6 +362,9 @@ def run_clang_format_and_save_to_tree(changed_lines, revision=None,
         mode = oct(int(stdout.split()[0], 8))
       else:
         mode = oct(os.stat(filename).st_mode)
+      # Adjust python3 octal format so that it matches what git expects
+      if mode.startswith('0o'):
+          mode = '0' + mode[2:]
       blob_id = clang_format_to_blob(filename, line_ranges,
                                      revision=revision,
                                      binary=binary,
@@ -488,10 +497,10 @@ def apply_changes(old_tree, new_tree, force=False, patch_mode=False):
   if not force:
     unstaged_files = run('git', 'diff-files', '--name-status', *changed_files)
     if unstaged_files:
-      print >>sys.stderr, ('The following files would be modified but '
-                           'have unstaged changes:')
-      print >>sys.stderr, unstaged_files
-      print >>sys.stderr, 'Please commit, stage, or stash them first.'
+      print('The following files would be modified but '
+                'have unstaged changes:', file=sys.stderr)
+      print(unstaged_files, file=sys.stderr)
+      print('Please commit, stage, or stash them first.', file=sys.stderr)
       sys.exit(2)
   if patch_mode:
     # In patch mode, we could just as well create an index from the new tree
@@ -521,20 +530,20 @@ def run(*args, **kwargs):
   if p.returncode == 0:
     if stderr:
       if verbose:
-        print >>sys.stderr, '`%s` printed to stderr:' % ' '.join(args)
-      print >>sys.stderr, stderr.rstrip()
+        print('`%s` printed to stderr:' % ' '.join(args), file=sys.stderr)
+      print(stderr.rstrip(), file=sys.stderr)
     if strip:
       stdout = stdout.rstrip('\r\n')
     return stdout
   if verbose:
-    print >>sys.stderr, '`%s` returned %s' % (' '.join(args), p.returncode)
+    print('`%s` returned %s' % (' '.join(args), p.returncode), file=sys.stderr)
   if stderr:
-    print >>sys.stderr, stderr.rstrip()
+    print(stderr.rstrip(), file=sys.stderr)
   sys.exit(2)
 
 
 def die(message):
-  print >>sys.stderr, 'error:', message
+  print('error:', message, file=sys.stderr)
   sys.exit(2)
 
 
diff --git a/tools/libclang/CXIndexDataConsumer.cpp b/tools/libclang/CXIndexDataConsumer.cpp
index 5d9776be3b6c..9cd5ff4f505f 100644
--- a/tools/libclang/CXIndexDataConsumer.cpp
+++ b/tools/libclang/CXIndexDataConsumer.cpp
@@ -1315,6 +1315,7 @@ static CXIdxEntityLanguage getEntityLangFromSymbolLang(SymbolLanguage L) {
   case SymbolLanguage::C: return CXIdxEntityLang_C;
   case SymbolLanguage::ObjC: return CXIdxEntityLang_ObjC;
   case SymbolLanguage::CXX: return CXIdxEntityLang_CXX;
+  case SymbolLanguage::Swift: return CXIdxEntityLang_Swift;
   }
   llvm_unreachable("invalid symbol language");
 }
diff --git a/tools/libclang/CXType.cpp b/tools/libclang/CXType.cpp
index 16e993e2ac01..fce7ef2c0d86 100644
--- a/tools/libclang/CXType.cpp
+++ b/tools/libclang/CXType.cpp
@@ -147,9 +147,6 @@ static inline CXTranslationUnit GetTU(CXType CT) {
 static Optional<ArrayRef<TemplateArgument>>
 GetTemplateArguments(QualType Type) {
   assert(!Type.isNull());
-  if (const auto *Specialization = Type->getAs<TemplateSpecializationType>())
-    return Specialization->template_arguments();
-
   if (const auto *RecordDecl = Type->getAsCXXRecordDecl()) {
     const auto *TemplateDecl =
       dyn_cast<ClassTemplateSpecializationDecl>(RecordDecl);
@@ -157,6 +154,9 @@ GetTemplateArguments(QualType Type) {
       return TemplateDecl->getTemplateArgs().asArray();
   }
 
+  if (const auto *Specialization = Type->getAs<TemplateSpecializationType>())
+    return Specialization->template_arguments();
+
   return None;
 }
 
diff --git a/unittests/Format/CleanupTest.cpp b/unittests/Format/CleanupTest.cpp
index 6ac5f695d437..e554a8191277 100644
--- a/unittests/Format/CleanupTest.cpp
+++ b/unittests/Format/CleanupTest.cpp
@@ -292,7 +292,7 @@ protected:
   }
 
   inline std::string apply(StringRef Code,
-                           const tooling::Replacements Replaces) {
+                           const tooling::Replacements &Replaces) {
     auto CleanReplaces = cleanupAroundReplacements(Code, Replaces, Style);
     EXPECT_TRUE(static_cast<bool>(CleanReplaces))
         << llvm::toString(CleanReplaces.takeError()) << "\n";
@@ -302,8 +302,7 @@ protected:
   }
 
   inline std::string formatAndApply(StringRef Code,
-                                    const tooling::Replacements Replaces) {
-
+                                    const tooling::Replacements &Replaces) {
     auto CleanReplaces = cleanupAroundReplacements(Code, Replaces, Style);
     EXPECT_TRUE(static_cast<bool>(CleanReplaces))
         << llvm::toString(CleanReplaces.takeError()) << "\n";
diff --git a/unittests/Format/FormatTest.cpp b/unittests/Format/FormatTest.cpp
index e6ba2230ac6d..9b833a96f16a 100644
--- a/unittests/Format/FormatTest.cpp
+++ b/unittests/Format/FormatTest.cpp
@@ -30,24 +30,25 @@ FormatStyle getGoogleStyle() { return getGoogleStyle(FormatStyle::LK_Cpp); }
 
 class FormatTest : public ::testing::Test {
 protected:
-  enum IncompleteCheck {
-    IC_ExpectComplete,
-    IC_ExpectIncomplete,
-    IC_DoNotCheck
+  enum StatusCheck {
+    SC_ExpectComplete,
+    SC_ExpectIncomplete,
+    SC_DoNotCheck
   };
 
   std::string format(llvm::StringRef Code,
                      const FormatStyle &Style = getLLVMStyle(),
-                     IncompleteCheck CheckIncomplete = IC_ExpectComplete) {
+                     StatusCheck CheckComplete = SC_ExpectComplete) {
     DEBUG(llvm::errs() << "---\n");
     DEBUG(llvm::errs() << Code << "\n\n");
     std::vector<tooling::Range> Ranges(1, tooling::Range(0, Code.size()));
-    bool IncompleteFormat = false;
+    FormattingAttemptStatus Status;
     tooling::Replacements Replaces =
-        reformat(Style, Code, Ranges, "<stdin>", &IncompleteFormat);
-    if (CheckIncomplete != IC_DoNotCheck) {
-      bool ExpectedIncompleteFormat = CheckIncomplete == IC_ExpectIncomplete;
-      EXPECT_EQ(ExpectedIncompleteFormat, IncompleteFormat) << Code << "\n\n";
+        reformat(Style, Code, Ranges, "<stdin>", &Status);
+    if (CheckComplete != SC_DoNotCheck) {
+      bool ExpectedCompleteFormat = CheckComplete == SC_ExpectComplete;
+      EXPECT_EQ(ExpectedCompleteFormat, Status.FormatComplete)
+          << Code << "\n\n";
     }
     ReplacementCount = Replaces.size();
     auto Result = applyAllReplacements(Code, Replaces);
@@ -83,7 +84,7 @@ protected:
   void verifyIncompleteFormat(llvm::StringRef Code,
                               const FormatStyle &Style = getLLVMStyle()) {
     EXPECT_EQ(Code.str(),
-              format(test::messUp(Code), Style, IC_ExpectIncomplete));
+              format(test::messUp(Code), Style, SC_ExpectIncomplete));
   }
 
   void verifyGoogleFormat(llvm::StringRef Code) {
@@ -98,7 +99,7 @@ protected:
   /// \brief Verify that clang-format does not crash on the given input.
   void verifyNoCrash(llvm::StringRef Code,
                      const FormatStyle &Style = getLLVMStyle()) {
-    format(Code, Style, IC_DoNotCheck);
+    format(Code, Style, SC_DoNotCheck);
   }
 
   int ReplacementCount;
@@ -5220,6 +5221,12 @@ TEST_F(FormatTest, FormatsFunctionTypes) {
   verifyFormat("void f() { function(*some_pointer_var)[0] = 10; }");
   verifyFormat("int x = f(&h)();");
   verifyFormat("returnsFunction(&param1, &param2)(param);");
+  verifyFormat("std::function<\n"
+               "    LooooooooooongTemplatedType<\n"
+               "        SomeType>*(\n"
+               "        LooooooooooooooooongType type)>\n"
+               "    function;",
+               getGoogleStyleWithColumns(40));
 }
 
 TEST_F(FormatTest, FormatsPointersToArrayTypes) {
@@ -6189,7 +6196,7 @@ TEST_F(FormatTest, SkipsDeeplyNestedLines) {
   // Deeply nested part is untouched, rest is formatted.
   EXPECT_EQ(std::string("int i;\n") + Code + "int j;\n",
             format(std::string("int    i;\n") + Code + "int    j;\n",
-                   getLLVMStyle(), IC_ExpectIncomplete));
+                   getLLVMStyle(), SC_ExpectIncomplete));
 }
 
 //===----------------------------------------------------------------------===//
diff --git a/unittests/Format/FormatTestComments.cpp b/unittests/Format/FormatTestComments.cpp
index e6310158e91e..4238efe54f06 100644
--- a/unittests/Format/FormatTestComments.cpp
+++ b/unittests/Format/FormatTestComments.cpp
@@ -29,24 +29,25 @@ FormatStyle getGoogleStyle() { return getGoogleStyle(FormatStyle::LK_Cpp); }
 
 class FormatTestComments : public ::testing::Test {
 protected:
-  enum IncompleteCheck {
-    IC_ExpectComplete,
-    IC_ExpectIncomplete,
-    IC_DoNotCheck
+  enum StatusCheck {
+    SC_ExpectComplete,
+    SC_ExpectIncomplete,
+    SC_DoNotCheck
   };
 
   std::string format(llvm::StringRef Code,
                      const FormatStyle &Style = getLLVMStyle(),
-                     IncompleteCheck CheckIncomplete = IC_ExpectComplete) {
+                     StatusCheck CheckComplete = SC_ExpectComplete) {
     DEBUG(llvm::errs() << "---\n");
     DEBUG(llvm::errs() << Code << "\n\n");
     std::vector<tooling::Range> Ranges(1, tooling::Range(0, Code.size()));
-    bool IncompleteFormat = false;
+    FormattingAttemptStatus Status;
     tooling::Replacements Replaces =
-        reformat(Style, Code, Ranges, "<stdin>", &IncompleteFormat);
-    if (CheckIncomplete != IC_DoNotCheck) {
-      bool ExpectedIncompleteFormat = CheckIncomplete == IC_ExpectIncomplete;
-      EXPECT_EQ(ExpectedIncompleteFormat, IncompleteFormat) << Code << "\n\n";
+        reformat(Style, Code, Ranges, "<stdin>", &Status);
+    if (CheckComplete != SC_DoNotCheck) {
+      bool ExpectedCompleteFormat = CheckComplete == SC_ExpectComplete;
+      EXPECT_EQ(ExpectedCompleteFormat, Status.FormatComplete)
+          << Code << "\n\n";
     }
     ReplacementCount = Replaces.size();
     auto Result = applyAllReplacements(Code, Replaces);
@@ -73,7 +74,7 @@ protected:
   /// \brief Verify that clang-format does not crash on the given input.
   void verifyNoCrash(llvm::StringRef Code,
                      const FormatStyle &Style = getLLVMStyle()) {
-    format(Code, Style, IC_DoNotCheck);
+    format(Code, Style, SC_DoNotCheck);
   }
 
   int ReplacementCount;
diff --git a/unittests/Format/FormatTestJS.cpp b/unittests/Format/FormatTestJS.cpp
index d8fa8e4b942c..7886c4fe27ad 100644
--- a/unittests/Format/FormatTestJS.cpp
+++ b/unittests/Format/FormatTestJS.cpp
@@ -24,10 +24,10 @@ protected:
     DEBUG(llvm::errs() << "---\n");
     DEBUG(llvm::errs() << Code << "\n\n");
     std::vector<tooling::Range> Ranges(1, tooling::Range(Offset, Length));
-    bool IncompleteFormat = false;
+    FormattingAttemptStatus Status;
     tooling::Replacements Replaces =
-        reformat(Style, Code, Ranges, "<stdin>", &IncompleteFormat);
-    EXPECT_FALSE(IncompleteFormat);
+        reformat(Style, Code, Ranges, "<stdin>", &Status);
+    EXPECT_TRUE(Status.FormatComplete);
     auto Result = applyAllReplacements(Code, Replaces);
     EXPECT_TRUE(static_cast<bool>(Result));
     DEBUG(llvm::errs() << "\n" << *Result << "\n\n");
@@ -318,6 +318,25 @@ TEST_F(FormatTestJS, MethodsInObjectLiterals) {
                "};");
 }
 
+TEST_F(FormatTestJS, GettersSettersVisibilityKeywords) {
+  // Don't break after "protected"
+  verifyFormat("class X {\n"
+               "  protected get getter():\n"
+               "      number {\n"
+               "    return 1;\n"
+               "  }\n"
+               "}",
+               getGoogleJSStyleWithColumns(12));
+  // Don't break after "get"
+  verifyFormat("class X {\n"
+               "  protected get someReallyLongGetterName():\n"
+               "      number {\n"
+               "    return 1;\n"
+               "  }\n"
+               "}",
+               getGoogleJSStyleWithColumns(40));
+}
+
 TEST_F(FormatTestJS, SpacesInContainerLiterals) {
   verifyFormat("var arr = [1, 2, 3];");
   verifyFormat("f({a: 1, b: 2, c: 3});");
@@ -1220,6 +1239,9 @@ TEST_F(FormatTestJS, MetadataAnnotations) {
                "}");
   verifyFormat("class X {}\n"
                "class Y {}");
+  verifyFormat("class X {\n"
+               "  @property() private isReply = false;\n"
+               "}\n");
 }
 
 TEST_F(FormatTestJS, TypeAliases) {
diff --git a/unittests/Format/FormatTestObjC.cpp b/unittests/Format/FormatTestObjC.cpp
index 680ff92f75e8..d2dc8bca00c5 100644
--- a/unittests/Format/FormatTestObjC.cpp
+++ b/unittests/Format/FormatTestObjC.cpp
@@ -33,23 +33,24 @@ protected:
     Style.Language = FormatStyle::LK_ObjC;
   }
 
-  enum IncompleteCheck {
-    IC_ExpectComplete,
-    IC_ExpectIncomplete,
-    IC_DoNotCheck
+  enum StatusCheck {
+    SC_ExpectComplete,
+    SC_ExpectIncomplete,
+    SC_DoNotCheck
   };
 
   std::string format(llvm::StringRef Code,
-                     IncompleteCheck CheckIncomplete = IC_ExpectComplete) {
+                     StatusCheck CheckComplete = SC_ExpectComplete) {
     DEBUG(llvm::errs() << "---\n");
     DEBUG(llvm::errs() << Code << "\n\n");
     std::vector<tooling::Range> Ranges(1, tooling::Range(0, Code.size()));
-    bool IncompleteFormat = false;
+    FormattingAttemptStatus Status;
     tooling::Replacements Replaces =
-        reformat(Style, Code, Ranges, "<stdin>", &IncompleteFormat);
-    if (CheckIncomplete != IC_DoNotCheck) {
-      bool ExpectedIncompleteFormat = CheckIncomplete == IC_ExpectIncomplete;
-      EXPECT_EQ(ExpectedIncompleteFormat, IncompleteFormat) << Code << "\n\n";
+        reformat(Style, Code, Ranges, "<stdin>", &Status);
+    if (CheckComplete != SC_DoNotCheck) {
+      bool ExpectedCompleteFormat = CheckComplete == SC_ExpectComplete;
+      EXPECT_EQ(ExpectedCompleteFormat, Status.FormatComplete)
+          << Code << "\n\n";
     }
     auto Result = applyAllReplacements(Code, Replaces);
     EXPECT_TRUE(static_cast<bool>(Result));
@@ -62,7 +63,7 @@ protected:
   }
 
   void verifyIncompleteFormat(StringRef Code) {
-    EXPECT_EQ(Code.str(), format(test::messUp(Code), IC_ExpectIncomplete));
+    EXPECT_EQ(Code.str(), format(test::messUp(Code), SC_ExpectIncomplete));
   }
 
   FormatStyle Style;
diff --git a/unittests/Format/FormatTestSelective.cpp b/unittests/Format/FormatTestSelective.cpp
index 367dbafcaf4c..8046d7fab2dd 100644
--- a/unittests/Format/FormatTestSelective.cpp
+++ b/unittests/Format/FormatTestSelective.cpp
@@ -24,10 +24,10 @@ protected:
     DEBUG(llvm::errs() << "---\n");
     DEBUG(llvm::errs() << Code << "\n\n");
     std::vector<tooling::Range> Ranges(1, tooling::Range(Offset, Length));
-    bool IncompleteFormat = false;
+    FormattingAttemptStatus Status;
     tooling::Replacements Replaces =
-        reformat(Style, Code, Ranges, "<stdin>", &IncompleteFormat);
-    EXPECT_FALSE(IncompleteFormat) << Code << "\n\n";
+        reformat(Style, Code, Ranges, "<stdin>", &Status);
+    EXPECT_TRUE(Status.FormatComplete) << Code << "\n\n";
     auto Result = applyAllReplacements(Code, Replaces);
     EXPECT_TRUE(static_cast<bool>(Result));
     DEBUG(llvm::errs() << "\n" << *Result << "\n\n");
diff --git a/unittests/Format/NamespaceEndCommentsFixerTest.cpp b/unittests/Format/NamespaceEndCommentsFixerTest.cpp
index 912638f45652..6c2d36985290 100644
--- a/unittests/Format/NamespaceEndCommentsFixerTest.cpp
+++ b/unittests/Format/NamespaceEndCommentsFixerTest.cpp
@@ -23,7 +23,7 @@ class NamespaceEndCommentsFixerTest : public ::testing::Test {
 protected:
   std::string
   fixNamespaceEndComments(llvm::StringRef Code,
-                          std::vector<tooling::Range> Ranges,
+                          const std::vector<tooling::Range> &Ranges,
                           const FormatStyle &Style = getLLVMStyle()) {
     DEBUG(llvm::errs() << "---\n");
     DEBUG(llvm::errs() << Code << "\n\n");