Index: include/llvm/Analysis/ProfileSummaryInfo.h
===================================================================
--- include/llvm/Analysis/ProfileSummaryInfo.h
+++ include/llvm/Analysis/ProfileSummaryInfo.h
@@ -74,6 +74,12 @@
            Summary->getKind() == ProfileSummary::PSK_Instr;
   }
 
+  /// Returns true if module \c M has context sensitive instrumentation profile.
+  bool hasCSInstrumentationProfile() {
+    return hasProfileSummary() &&
+           Summary->getKind() == ProfileSummary::PSK_CSInstr;
+  }
+
   /// Handle the invalidation of this information.
   ///
   /// When used as a result of \c ProfileSummaryAnalysis this method will be
Index: include/llvm/IR/Module.h
===================================================================
--- include/llvm/IR/Module.h
+++ include/llvm/IR/Module.h
@@ -29,6 +29,7 @@
 #include "llvm/IR/GlobalIFunc.h"
 #include "llvm/IR/GlobalVariable.h"
 #include "llvm/IR/Metadata.h"
+#include "llvm/IR/ProfileSummary.h"
 #include "llvm/IR/SymbolTableListTraits.h"
 #include "llvm/Support/CBindingWrapping.h"
 #include "llvm/Support/CodeGen.h"
@@ -856,10 +857,10 @@
   /// @{
 
   /// Attach profile summary metadata to this module.
-  void setProfileSummary(Metadata *M);
+  void setProfileSummary(Metadata *M, ProfileSummary::Kind Kind);
 
   /// Returns profile summary metadata
-  Metadata *getProfileSummary();
+  Metadata *getProfileSummary(ProfileSummary::Kind Kind);
   /// @}
 
   /// Returns true if PLT should be avoided for RTLib calls.
Index: include/llvm/IR/ProfileSummary.h
===================================================================
--- include/llvm/IR/ProfileSummary.h
+++ include/llvm/IR/ProfileSummary.h
@@ -43,11 +43,10 @@
 
 class ProfileSummary {
 public:
-  enum Kind { PSK_Instr, PSK_Sample };
+  enum Kind { PSK_Instr, PSK_CSInstr, PSK_Sample };
 
 private:
   const Kind PSK;
-  static const char *KindStr[2];
   SummaryEntryVector DetailedSummary;
   uint64_t TotalCount, MaxCount, MaxInternalCount, MaxFunctionCount;
   uint32_t NumCounts, NumFunctions;
Index: include/llvm/LTO/Config.h
===================================================================
--- include/llvm/LTO/Config.h
+++ include/llvm/LTO/Config.h
@@ -56,6 +56,9 @@
   /// Disable entirely the optimizer, including importing for ThinLTO
   bool CodeGenOnly = false;
 
+  /// Run PGO context sensitive IR instrumentation.
+  bool RunCSIRInstr = false;
+
   /// If this field is set, the set of passes run in the middle-end optimizer
   /// will be the one specified by the string. Only works with the new pass
   /// manager as the old one doesn't have this ability.
@@ -74,6 +77,9 @@
   /// with this triple.
   std::string DefaultTriple;
 
+  /// Context Sensitive PGO profile path.
+  std::string CSIRProfile;
+
   /// Sample PGO profile path.
   std::string SampleProfile;
 
Index: include/llvm/Passes/PassBuilder.h
===================================================================
--- include/llvm/Passes/PassBuilder.h
+++ include/llvm/Passes/PassBuilder.h
@@ -32,25 +32,24 @@
 
 /// A struct capturing PGO tunables.
 struct PGOOptions {
+  enum PGOAction { NoAction, IRInstr, CSIRInstr, CSIRUse };
   PGOOptions(std::string ProfileGenFile = "", std::string ProfileUseFile = "",
              std::string SampleProfileFile = "",
-             std::string ProfileRemappingFile = "",
-             bool RunProfileGen = false, bool SamplePGOSupport = false)
+             std::string ProfileRemappingFile = "", PGOAction Action = NoAction,
+             bool SamplePGOSupport = false)
       : ProfileGenFile(ProfileGenFile), ProfileUseFile(ProfileUseFile),
         SampleProfileFile(SampleProfileFile),
-        ProfileRemappingFile(ProfileRemappingFile),
-        RunProfileGen(RunProfileGen),
+        ProfileRemappingFile(ProfileRemappingFile), Action(Action),
         SamplePGOSupport(SamplePGOSupport || !SampleProfileFile.empty()) {
-    assert((RunProfileGen ||
-            !SampleProfileFile.empty() ||
-            !ProfileUseFile.empty() ||
-            SamplePGOSupport) && "Illegal PGOOptions.");
+    assert((Action != NoAction || !SampleProfileFile.empty() ||
+            !ProfileUseFile.empty() || SamplePGOSupport) &&
+           "Illegal PGOOptions.");
   }
   std::string ProfileGenFile;
   std::string ProfileUseFile;
   std::string SampleProfileFile;
   std::string ProfileRemappingFile;
-  bool RunProfileGen;
+  PGOAction Action;
   bool SamplePGOSupport;
 };
 
@@ -606,9 +605,8 @@
                                 bool VerifyEachPass, bool DebugLogging);
 
   void addPGOInstrPasses(ModulePassManager &MPM, bool DebugLogging,
-                         OptimizationLevel Level, bool RunProfileGen,
-                         std::string ProfileGenFile,
-                         std::string ProfileUseFile,
+                         OptimizationLevel Level, bool RunProfileGen, bool IsCS,
+                         std::string ProfileGenFile, std::string ProfileUseFile,
                          std::string ProfileRemappingFile);
 
   void invokePeepholeEPCallbacks(FunctionPassManager &, OptimizationLevel);
Index: include/llvm/ProfileData/InstrProf.h
===================================================================
--- include/llvm/ProfileData/InstrProf.h
+++ include/llvm/ProfileData/InstrProf.h
@@ -767,11 +767,19 @@
 struct NamedInstrProfRecord : InstrProfRecord {
   StringRef Name;
   uint64_t Hash;
+  static const int CS_FLAG_IN_FUNC_HASH = 60;
 
   NamedInstrProfRecord() = default;
   NamedInstrProfRecord(StringRef Name, uint64_t Hash,
                        std::vector<uint64_t> Counts)
       : InstrProfRecord(std::move(Counts)), Name(Name), Hash(Hash) {}
+
+  static bool hasCSFlagInHash(uint64_t FuncHash) {
+    return ((FuncHash >> CS_FLAG_IN_FUNC_HASH) & 1);
+  }
+  static void setCSFlagInHash(uint64_t &FuncHash) {
+    FuncHash |= ((uint64_t)1 << CS_FLAG_IN_FUNC_HASH);
+  }
 };
 
 uint32_t InstrProfRecord::getNumValueKinds() const {
@@ -1005,6 +1013,7 @@
 // from control data struct is changed from raw pointer to Name's MD5 value.
 // Version 4: ValueDataBegin and ValueDataSizes fields are removed from the
 // raw header.
+// Version 5: Bit 60-63 of FuncHash is reserved for storing other information.
 const uint64_t Version = INSTR_PROF_RAW_VERSION;
 
 template <class IntPtrT> inline uint64_t getMagic();
Index: include/llvm/ProfileData/InstrProfData.inc
===================================================================
--- include/llvm/ProfileData/InstrProfData.inc
+++ include/llvm/ProfileData/InstrProfData.inc
@@ -636,10 +636,12 @@
  * version for other variants of profile. We set the lowest bit of the upper 8
  * bits (i.e. bit 56) to 1 to indicate if this is an IR-level instrumentaiton
  * generated profile, and 0 if this is a Clang FE generated profile.
+ * 1 in bit 57 indicates there are context-sesitive records in the profile.
  */
 #define VARIANT_MASKS_ALL 0xff00000000000000ULL
 #define GET_VERSION(V) ((V) & ~VARIANT_MASKS_ALL)
 #define VARIANT_MASK_IR_PROF (0x1ULL << 56)
+#define VARIANT_MASK_CSIR_PROF (0x1ULL << 57)
 #define INSTR_PROF_RAW_VERSION_VAR __llvm_profile_raw_version
 #define INSTR_PROF_PROFILE_RUNTIME_VAR __llvm_profile_runtime
 
Index: include/llvm/ProfileData/InstrProfReader.h
===================================================================
--- include/llvm/ProfileData/InstrProfReader.h
+++ include/llvm/ProfileData/InstrProfReader.h
@@ -78,6 +78,8 @@
 
   virtual bool isIRLevelProfile() const = 0;
 
+  virtual bool hasCSIRLevelProfile() const = 0;
+
   /// Return the PGO symtab. There are three different readers:
   /// Raw, Text, and Indexed profile readers. The first two types
   /// of readers are used only by llvm-profdata tool, while the indexed
@@ -143,6 +145,7 @@
   /// Iterator over the profile data.
   line_iterator Line;
   bool IsIRLevelProfile = false;
+  bool HasCSIRLevelProfile = false;
 
   Error readValueProfileData(InstrProfRecord &Record);
 
@@ -157,6 +160,8 @@
 
   bool isIRLevelProfile() const override { return IsIRLevelProfile; }
 
+  bool hasCSIRLevelProfile() const override { return HasCSIRLevelProfile; }
+
   /// Read the header.
   Error readHeader() override;
 
@@ -213,6 +218,10 @@
     return (Version & VARIANT_MASK_IR_PROF) != 0;
   }
 
+  bool hasCSIRLevelProfile() const override {
+    return (Version & VARIANT_MASK_CSIR_PROF) != 0;
+  }
+
   InstrProfSymtab &getSymtab() override {
     assert(Symtab.get());
     return *Symtab.get();
@@ -342,6 +351,7 @@
   virtual void setValueProfDataEndianness(support::endianness Endianness) = 0;
   virtual uint64_t getVersion() const = 0;
   virtual bool isIRLevelProfile() const = 0;
+  virtual bool hasCSIRLevelProfile() const = 0;
   virtual Error populateSymtab(InstrProfSymtab &) = 0;
 };
 
@@ -386,6 +396,10 @@
     return (FormatVersion & VARIANT_MASK_IR_PROF) != 0;
   }
 
+  bool hasCSIRLevelProfile() const override {
+    return (FormatVersion & VARIANT_MASK_CSIR_PROF) != 0;
+  }
+
   Error populateSymtab(InstrProfSymtab &Symtab) override {
     return Symtab.create(HashTable->keys());
   }
@@ -413,13 +427,15 @@
   std::unique_ptr<InstrProfReaderRemapper> Remapper;
   /// Profile summary data.
   std::unique_ptr<ProfileSummary> Summary;
+  /// Context sensitive profile summary data.
+  std::unique_ptr<ProfileSummary> CS_Summary;
   // Index to the current record in the record array.
   unsigned RecordIndex;
 
   // Read the profile summary. Return a pointer pointing to one byte past the
   // end of the summary data if it exists or the input \c Cur.
   const unsigned char *readSummary(IndexedInstrProf::ProfVersion Version,
-                                   const unsigned char *Cur);
+                                   const unsigned char *Cur, bool IsCS);
 
 public:
   IndexedInstrProfReader(
@@ -433,6 +449,9 @@
   /// Return the profile version.
   uint64_t getVersion() const { return Index->getVersion(); }
   bool isIRLevelProfile() const override { return Index->isIRLevelProfile(); }
+  bool hasCSIRLevelProfile() const override {
+    return Index->hasCSIRLevelProfile();
+  }
 
   /// Return true if the given buffer is in an indexed instrprof format.
   static bool hasFormat(const MemoryBuffer &DataBuffer);
@@ -451,7 +470,15 @@
                           std::vector<uint64_t> &Counts);
 
   /// Return the maximum of all known function counts.
-  uint64_t getMaximumFunctionCount() { return Summary->getMaxFunctionCount(); }
+  uint64_t getMaximumFunctionCount(bool IsCS) {
+    if (IsCS) {
+      assert(CS_Summary && "No context sensitive profile summary");
+      return CS_Summary->getMaxFunctionCount();
+    } else {
+      assert(Summary && "No profile summary");
+      return Summary->getMaxFunctionCount();
+    }
+  }
 
   /// Factory method to create an indexed reader.
   static Expected<std::unique_ptr<IndexedInstrProfReader>>
@@ -470,7 +497,15 @@
   // to be used by llvm-profdata (for dumping). Avoid using this when
   // the client is the compiler.
   InstrProfSymtab &getSymtab() override;
-  ProfileSummary &getSummary() { return *(Summary.get()); }
+  ProfileSummary &getSummary(bool IsCS) {
+    if (IsCS) {
+      assert(CS_Summary && "No context sensitive summary");
+      return *(CS_Summary.get());
+    } else {
+      assert(Summary && "No profile summary");
+      return *(Summary.get());
+    }
+  }
 };
 
 } // end namespace llvm
Index: include/llvm/ProfileData/InstrProfWriter.h
===================================================================
--- include/llvm/ProfileData/InstrProfWriter.h
+++ include/llvm/ProfileData/InstrProfWriter.h
@@ -34,7 +34,7 @@
 class InstrProfWriter {
 public:
   using ProfilingData = SmallDenseMap<uint64_t, InstrProfRecord>;
-  enum ProfKind { PF_Unknown = 0, PF_FE, PF_IRLevel };
+  enum ProfKind { PF_Unknown = 0, PF_FE, PF_IRLevel, PF_IRLevelWithCS };
 
 private:
   bool Sparse;
@@ -75,15 +75,24 @@
   std::unique_ptr<MemoryBuffer> writeBuffer();
 
   /// Set the ProfileKind. Report error if mixing FE and IR level profiles.
-  Error setIsIRLevelProfile(bool IsIRLevel) {
+  Error setIsIRLevelProfile(bool IsIRLevel, bool WithCS) {
     if (ProfileKind == PF_Unknown) {
-      ProfileKind = IsIRLevel ? PF_IRLevel: PF_FE;
+      if (IsIRLevel)
+        ProfileKind = WithCS ? PF_IRLevelWithCS : PF_IRLevel;
+      else
+        ProfileKind = PF_FE;
       return Error::success();
     }
-    return (IsIRLevel == (ProfileKind == PF_IRLevel))
-               ? Error::success()
-               : make_error<InstrProfError>(
-                     instrprof_error::unsupported_version);
+
+    if (((ProfileKind != PF_FE) && !IsIRLevel) ||
+        ((ProfileKind == PF_FE) && IsIRLevel))
+      return make_error<InstrProfError>(instrprof_error::unsupported_version);
+
+    // Promote to PF_IRLevelWithCS if WithCS is true;
+    if (ProfileKind == PF_IRLevel && WithCS)
+      ProfileKind = PF_IRLevelWithCS;
+
+    return Error::success();
   }
 
   // Internal interface for testing purpose only.
Index: include/llvm/Transforms/IPO/PassManagerBuilder.h
===================================================================
--- include/llvm/Transforms/IPO/PassManagerBuilder.h
+++ include/llvm/Transforms/IPO/PassManagerBuilder.h
@@ -160,6 +160,10 @@
 
   /// Enable profile instrumentation pass.
   bool EnablePGOInstrGen;
+  /// Enable profile context sensitive instrumentation pass.
+  bool EnablePGOCSInstrGen;
+  /// Enable profile context sensitive profile use pass.
+  bool EnablePGOCSInstrUse;
   /// Profile data file name that the instrumentation will be written to.
   std::string PGOInstrGen;
   /// Path of the profile data file.
@@ -186,7 +190,7 @@
   void addInitialAliasAnalysisPasses(legacy::PassManagerBase &PM) const;
   void addLTOOptimizationPasses(legacy::PassManagerBase &PM);
   void addLateLTOOptimizationPasses(legacy::PassManagerBase &PM);
-  void addPGOInstrPasses(legacy::PassManagerBase &MPM);
+  void addPGOInstrPasses(legacy::PassManagerBase &MPM, bool IsCS);
   void addFunctionSimplificationPasses(legacy::PassManagerBase &MPM);
   void addInstructionCombiningPass(legacy::PassManagerBase &MPM) const;
 
Index: include/llvm/Transforms/Instrumentation.h
===================================================================
--- include/llvm/Transforms/Instrumentation.h
+++ include/llvm/Transforms/Instrumentation.h
@@ -89,9 +89,10 @@
                                    GCOVOptions::getDefault());
 
 // PGO Instrumention
-ModulePass *createPGOInstrumentationGenLegacyPass();
+ModulePass *createPGOInstrumentationGenLegacyPass(bool IsCS = false);
 ModulePass *
-createPGOInstrumentationUseLegacyPass(StringRef Filename = StringRef(""));
+createPGOInstrumentationUseLegacyPass(StringRef Filename = StringRef(""),
+                                      bool IsCS = false);
 ModulePass *createPGOIndirectCallPromotionLegacyPass(bool InLTO = false,
                                                      bool SamplePGO = false);
 FunctionPass *createPGOMemOPSizeOptLegacyPass();
@@ -133,6 +134,9 @@
   // Use atomic profile counter increments.
   bool Atomic = false;
 
+  // Use BFI to guide register promotion
+  bool UseBFIInPromotion = false;
+
   // Name of the profile file to use as output
   std::string InstrProfileOutput;
 
Index: include/llvm/Transforms/Instrumentation/PGOInstrumentation.h
===================================================================
--- include/llvm/Transforms/Instrumentation/PGOInstrumentation.h
+++ include/llvm/Transforms/Instrumentation/PGOInstrumentation.h
@@ -30,20 +30,27 @@
 /// The instrumentation (profile-instr-gen) pass for IR based PGO.
 class PGOInstrumentationGen : public PassInfoMixin<PGOInstrumentationGen> {
 public:
+  PGOInstrumentationGen(bool IsCS = false) : IsCS(IsCS) {}
   PreservedAnalyses run(Module &M, ModuleAnalysisManager &AM);
+
+private:
+  // If this is a context sensitive instrumentation.
+  bool IsCS;
 };
 
 /// The profile annotation (profile-instr-use) pass for IR based PGO.
 class PGOInstrumentationUse : public PassInfoMixin<PGOInstrumentationUse> {
 public:
   PGOInstrumentationUse(std::string Filename = "",
-                        std::string RemappingFilename = "");
+                        std::string RemappingFilename = "", bool IsCS = false);
 
   PreservedAnalyses run(Module &M, ModuleAnalysisManager &AM);
 
 private:
   std::string ProfileFileName;
   std::string ProfileRemappingFileName;
+  // If this is a context sensitive instrumentation.
+  bool IsCS;
 };
 
 /// The indirect function call promotion pass.
Index: lib/Analysis/ProfileSummaryInfo.cpp
===================================================================
--- lib/Analysis/ProfileSummaryInfo.cpp
+++ lib/Analysis/ProfileSummaryInfo.cpp
@@ -85,7 +85,14 @@
 bool ProfileSummaryInfo::computeSummary() {
   if (Summary)
     return true;
-  auto *SummaryMD = M.getProfileSummary();
+  // First try to get context sensitive ProfileSummary.
+  auto *SummaryMD = M.getProfileSummary(ProfileSummary::PSK_CSInstr);
+  if (SummaryMD) {
+    Summary.reset(ProfileSummary::getFromMD(SummaryMD));
+    return true;
+  }
+  // This will actually return PSK_Instr or PSK_Sample summary.
+  SummaryMD = M.getProfileSummary(ProfileSummary::PSK_Instr);
   if (!SummaryMD)
     return false;
   Summary.reset(ProfileSummary::getFromMD(SummaryMD));
Index: lib/IR/Module.cpp
===================================================================
--- lib/IR/Module.cpp
+++ lib/IR/Module.cpp
@@ -526,12 +526,18 @@
   addModuleFlag(ModFlagBehavior::Error, "Code Model", CL);
 }
 
-void Module::setProfileSummary(Metadata *M) {
-  addModuleFlag(ModFlagBehavior::Error, "ProfileSummary", M);
+void Module::setProfileSummary(Metadata *M, ProfileSummary::Kind Kind) {
+  if (Kind == ProfileSummary::PSK_CSInstr)
+    addModuleFlag(ModFlagBehavior::Error, "CSProfileSummary", M);
+  else
+    addModuleFlag(ModFlagBehavior::Error, "ProfileSummary", M);
 }
 
-Metadata *Module::getProfileSummary() {
-  return getModuleFlag("ProfileSummary");
+Metadata *Module::getProfileSummary(ProfileSummary::Kind Kind) {
+  if (Kind == ProfileSummary::PSK_CSInstr)
+    return getModuleFlag("CSProfileSummary");
+  else
+    return getModuleFlag("ProfileSummary");
 }
 
 void Module::setOwnedMemoryBuffer(std::unique_ptr<MemoryBuffer> MB) {
Index: lib/IR/ProfileSummary.cpp
===================================================================
--- lib/IR/ProfileSummary.cpp
+++ lib/IR/ProfileSummary.cpp
@@ -22,8 +22,6 @@
 
 using namespace llvm;
 
-const char *ProfileSummary::KindStr[2] = {"InstrProf", "SampleProfile"};
-
 // Return an MDTuple with two elements. The first element is a string Key and
 // the second is a uint64_t Value.
 static Metadata *getKeyValMD(LLVMContext &Context, const char *Key,
@@ -69,6 +67,7 @@
 // "SampleProfile"). The rest of the elements of the outer MDTuple are specific
 // to the kind of profile summary as returned by getFormatSpecificMD.
 Metadata *ProfileSummary::getMD(LLVMContext &Context) {
+  const char *KindStr[3] = {"InstrProf", "CSInstrProf", "SampleProfile"};
   Metadata *Components[] = {
     getKeyValMD(Context, "ProfileFormat", KindStr[PSK]),
     getKeyValMD(Context, "TotalCount", getTotalCount()),
@@ -154,6 +153,9 @@
   else if (isKeyValuePair(dyn_cast_or_null<MDTuple>(FormatMD), "ProfileFormat",
                           "InstrProf"))
     SummaryKind = PSK_Instr;
+  else if (isKeyValuePair(dyn_cast_or_null<MDTuple>(FormatMD), "ProfileFormat",
+                          "CSInstrProf"))
+    SummaryKind = PSK_CSInstr;
   else
     return nullptr;
 
Index: lib/LTO/LTOBackend.cpp
===================================================================
--- lib/LTO/LTOBackend.cpp
+++ lib/LTO/LTOBackend.cpp
@@ -156,7 +156,14 @@
   Optional<PGOOptions> PGOOpt;
   if (!Conf.SampleProfile.empty())
     PGOOpt = PGOOptions("", "", Conf.SampleProfile, Conf.ProfileRemapping,
-                        false, true);
+                        PGOOptions::NoAction, true);
+  else if (Conf.RunCSIRInstr) {
+    PGOOpt = PGOOptions(Conf.CSIRProfile, "", "", Conf.ProfileRemapping,
+                        PGOOptions::CSIRInstr);
+  } else if (!Conf.CSIRProfile.empty()) {
+    PGOOpt = PGOOptions("", Conf.CSIRProfile, "", Conf.ProfileRemapping,
+                        PGOOptions::CSIRUse);
+  }
 
   PassBuilder PB(TM, PGOOpt);
   AAManager AA;
Index: lib/Passes/PassBuilder.cpp
===================================================================
--- lib/Passes/PassBuilder.cpp
+++ lib/Passes/PassBuilder.cpp
@@ -506,7 +506,7 @@
 
 void PassBuilder::addPGOInstrPasses(ModulePassManager &MPM, bool DebugLogging,
                                     PassBuilder::OptimizationLevel Level,
-                                    bool RunProfileGen,
+                                    bool RunProfileGen, bool IsCS,
                                     std::string ProfileGenFile,
                                     std::string ProfileUseFile,
                                     std::string ProfileRemappingFile) {
@@ -514,7 +514,7 @@
   // threshold results in smaller executables, but there may be cases where
   // the size grows, so let's be conservative here and skip this simplification
   // at -Os/Oz.
-  if (!isOptimizingForSize(Level)) {
+  if (!isOptimizingForSize(Level) && !IsCS) {
     InlineParams IP;
 
     // In the old pass manager, this is a cl::opt. Should still this be one?
@@ -547,7 +547,7 @@
   MPM.addPass(GlobalDCEPass());
 
   if (RunProfileGen) {
-    MPM.addPass(PGOInstrumentationGen());
+    MPM.addPass(PGOInstrumentationGen(IsCS));
 
     FunctionPassManager FPM;
     FPM.addPass(
@@ -559,11 +559,11 @@
     if (!ProfileGenFile.empty())
       Options.InstrProfileOutput = ProfileGenFile;
     Options.DoCounterPromotion = true;
+    Options.UseBFIInPromotion = IsCS;
     MPM.addPass(InstrProfiling(Options));
-  }
-
-  if (!ProfileUseFile.empty())
-    MPM.addPass(PGOInstrumentationUse(ProfileUseFile, ProfileRemappingFile));
+  } else if (!ProfileUseFile.empty())
+    MPM.addPass(
+        PGOInstrumentationUse(ProfileUseFile, ProfileRemappingFile, IsCS));
 }
 
 static InlineParams
@@ -658,7 +658,8 @@
   // Add all the requested passes for instrumentation PGO, if requested.
   if (PGOOpt && Phase != ThinLTOPhase::PostLink &&
       (!PGOOpt->ProfileGenFile.empty() || !PGOOpt->ProfileUseFile.empty())) {
-    addPGOInstrPasses(MPM, DebugLogging, Level, PGOOpt->RunProfileGen,
+    addPGOInstrPasses(MPM, DebugLogging, Level,
+                      PGOOpt->Action == PGOOptions::IRInstr, false,
                       PGOOpt->ProfileGenFile, PGOOpt->ProfileUseFile,
                       PGOOpt->ProfileRemappingFile);
     MPM.addPass(PGOIndirectCallPromotion(false, false));
@@ -760,6 +761,15 @@
   // FIXME: Is this really an optimization rather than a canonicalization?
   MPM.addPass(ReversePostOrderFunctionAttrsPass());
 
+  // Do a post inline PGO instrumention and use pass. This is a context
+  // sensitive PGO pass.
+  if (PGOOpt && (PGOOpt->Action == PGOOptions::CSIRInstr ||
+                 PGOOpt->Action == PGOOptions::CSIRUse))
+    addPGOInstrPasses(MPM, DebugLogging, Level,
+                      PGOOpt->Action == PGOOptions::CSIRInstr, true,
+                      PGOOpt->ProfileGenFile, PGOOpt->ProfileUseFile,
+                      PGOOpt->ProfileRemappingFile);
+
   // Re-require GloblasAA here prior to function passes. This is particularly
   // useful as the above will have inlined, DCE'ed, and function-attr
   // propagated everything. We should at this point have a reasonably minimal
Index: lib/ProfileData/InstrProfReader.cpp
===================================================================
--- lib/ProfileData/InstrProfReader.cpp
+++ lib/ProfileData/InstrProfReader.cpp
@@ -163,7 +163,10 @@
     IsIRInstr = true;
   else if (Str.equals_lower("fe"))
     IsIRInstr = false;
-  else
+  else if (Str.equals_lower("csir")) {
+    IsIRInstr = true;
+    HasCSIRLevelProfile = true;
+  } else
     return error(instrprof_error::bad_header);
 
   ++Line;
@@ -734,7 +737,7 @@
 
 const unsigned char *
 IndexedInstrProfReader::readSummary(IndexedInstrProf::ProfVersion Version,
-                                    const unsigned char *Cur) {
+                                    const unsigned char *Cur, bool IsCS) {
   using namespace IndexedInstrProf;
   using namespace support;
 
@@ -761,10 +764,13 @@
       DetailedSummary.emplace_back((uint32_t)Ent.Cutoff, Ent.MinBlockCount,
                                    Ent.NumBlocks);
     }
+    std::unique_ptr<llvm::ProfileSummary> &Summary =
+        IsCS ? this->CS_Summary : this->Summary;
+
     // initialize InstrProfSummary using the SummaryData from disk.
-    this->Summary = llvm::make_unique<ProfileSummary>(
-        ProfileSummary::PSK_Instr, DetailedSummary,
-        SummaryData->get(Summary::TotalBlockCount),
+    Summary = llvm::make_unique<ProfileSummary>(
+        IsCS ? ProfileSummary::PSK_CSInstr : ProfileSummary::PSK_Instr,
+        DetailedSummary, SummaryData->get(Summary::TotalBlockCount),
         SummaryData->get(Summary::MaxBlockCount),
         SummaryData->get(Summary::MaxInternalBlockCount),
         SummaryData->get(Summary::MaxFunctionCount),
@@ -806,7 +812,9 @@
       IndexedInstrProf::ProfVersion::CurrentVersion)
     return error(instrprof_error::unsupported_version);
 
-  Cur = readSummary((IndexedInstrProf::ProfVersion)FormatVersion, Cur);
+  Cur = readSummary((IndexedInstrProf::ProfVersion)FormatVersion, Cur, false);
+  if (Header->Version & VARIANT_MASK_CSIR_PROF)
+    Cur = readSummary((IndexedInstrProf::ProfVersion)FormatVersion, Cur, true);
 
   // Read the hash type and start offset.
   IndexedInstrProf::HashT HashType = static_cast<IndexedInstrProf::HashT>(
Index: lib/ProfileData/InstrProfWriter.cpp
===================================================================
--- lib/ProfileData/InstrProfWriter.cpp
+++ lib/ProfileData/InstrProfWriter.cpp
@@ -102,6 +102,7 @@
 
   support::endianness ValueProfDataEndianness = support::little;
   InstrProfSummaryBuilder *SummaryBuilder;
+  InstrProfSummaryBuilder *CSSummaryBuilder;
 
   InstrProfRecordWriterTrait() = default;
 
@@ -143,7 +144,10 @@
     endian::Writer LE(Out, little);
     for (const auto &ProfileData : *V) {
       const InstrProfRecord &ProfRecord = ProfileData.second;
-      SummaryBuilder->addRecord(ProfRecord);
+      if (NamedInstrProfRecord::hasCSFlagInHash(ProfileData.first))
+        CSSummaryBuilder->addRecord(ProfRecord);
+      else
+        SummaryBuilder->addRecord(ProfRecord);
 
       LE.write<uint64_t>(ProfileData.first); // Function hash
       LE.write<uint64_t>(ProfRecord.Counts.size());
@@ -254,6 +258,8 @@
 
   InstrProfSummaryBuilder ISB(ProfileSummaryBuilder::DefaultCutoffs);
   InfoObj->SummaryBuilder = &ISB;
+  InstrProfSummaryBuilder CSISB(ProfileSummaryBuilder::DefaultCutoffs);
+  InfoObj->CSSummaryBuilder = &CSISB;
 
   // Populate the hash table generator.
   for (const auto &I : FunctionData)
@@ -265,6 +271,10 @@
   Header.Version = IndexedInstrProf::ProfVersion::CurrentVersion;
   if (ProfileKind == PF_IRLevel)
     Header.Version |= VARIANT_MASK_IR_PROF;
+  if (ProfileKind == PF_IRLevelWithCS) {
+    Header.Version |= VARIANT_MASK_IR_PROF;
+    Header.Version |= VARIANT_MASK_CSIR_PROF;
+  }
   Header.Unused = 0;
   Header.HashType = static_cast<uint64_t>(IndexedInstrProf::HashType);
   Header.HashOffset = 0;
@@ -288,6 +298,14 @@
   uint64_t SummaryOffset = OS.tell();
   for (unsigned I = 0; I < SummarySize / sizeof(uint64_t); I++)
     OS.write(0);
+  uint64_t CSSummaryOffset = 0;
+  uint64_t CSSummarySize = 0;
+  if (ProfileKind == PF_IRLevelWithCS) {
+    CSSummaryOffset = OS.tell();
+    CSSummarySize = SummarySize / sizeof(uint64_t);
+    for (unsigned I = 0; I < CSSummarySize; I++)
+      OS.write(0);
+  }
 
   // Write the hash table.
   uint64_t HashTableStart = Generator.Emit(OS.OS, *InfoObj);
@@ -301,13 +319,25 @@
   setSummary(TheSummary.get(), *PS);
   InfoObj->SummaryBuilder = nullptr;
 
+  // For Context Sensitive summary.
+  std::unique_ptr<IndexedInstrProf::Summary> TheCSSummary = nullptr;
+  if (ProfileKind == PF_IRLevelWithCS) {
+    TheCSSummary = IndexedInstrProf::allocSummary(SummarySize);
+    std::unique_ptr<ProfileSummary> CSPS = CSISB.getSummary();
+    setSummary(TheCSSummary.get(), *CSPS);
+  }
+  InfoObj->CSSummaryBuilder = nullptr;
+
   // Now do the final patch:
   PatchItem PatchItems[] = {
       // Patch the Header.HashOffset field.
       {HashTableStartFieldOffset, &HashTableStart, 1},
       // Patch the summary data.
       {SummaryOffset, reinterpret_cast<uint64_t *>(TheSummary.get()),
-       (int)(SummarySize / sizeof(uint64_t))}};
+       (int)(SummarySize / sizeof(uint64_t))},
+      {CSSummaryOffset, reinterpret_cast<uint64_t *>(TheCSSummary.get()),
+       (int)CSSummarySize}};
+
   OS.patch(PatchItems, sizeof(PatchItems) / sizeof(*PatchItems));
 }
 
@@ -376,6 +406,8 @@
 Error InstrProfWriter::writeText(raw_fd_ostream &OS) {
   if (ProfileKind == PF_IRLevel)
     OS << "# IR level Instrumentation Flag\n:ir\n";
+  else if (ProfileKind == PF_IRLevelWithCS)
+    OS << "# CSIR level Instrumentation Flag\n:csir\n";
   InstrProfSymtab Symtab;
   for (const auto &I : FunctionData)
     if (shouldEncodeData(I.getValue()))
Index: lib/Transforms/IPO/PassManagerBuilder.cpp
===================================================================
--- lib/Transforms/IPO/PassManagerBuilder.cpp
+++ lib/Transforms/IPO/PassManagerBuilder.cpp
@@ -176,6 +176,8 @@
     MergeFunctions = false;
     PrepareForLTO = false;
     EnablePGOInstrGen = RunPGOInstrGen;
+    EnablePGOCSInstrGen = false;
+    EnablePGOCSInstrUse = false;
     PGOInstrGen = PGOOutputFile;
     PGOInstrUse = RunPGOInstrUse;
     PrepareForThinLTO = EnablePrepareForThinLTO;
@@ -272,13 +274,18 @@
 }
 
 // Do PGO instrumentation generation or use pass as the option specified.
-void PassManagerBuilder::addPGOInstrPasses(legacy::PassManagerBase &MPM) {
-  if (!EnablePGOInstrGen && PGOInstrUse.empty() && PGOSampleUse.empty())
+void PassManagerBuilder::addPGOInstrPasses(legacy::PassManagerBase &MPM,
+                                           bool IsCS = false) {
+  if (IsCS) {
+    if (!EnablePGOCSInstrGen && !EnablePGOCSInstrUse)
+      return;
+  } else if (!EnablePGOInstrGen && PGOInstrUse.empty() && PGOSampleUse.empty())
     return;
+
   // Perform the preinline and cleanup passes for O1 and above.
   // And avoid doing them if optimizing for size.
   if (OptLevel > 0 && SizeLevel == 0 && !DisablePreInliner &&
-      PGOSampleUse.empty()) {
+      PGOSampleUse.empty() && !IsCS) {
     // Create preinline pass. We construct an InlineParams object and specify
     // the threshold here to avoid the command line options of the regular
     // inliner to influence pre-inlining. The only fields of InlineParams we
@@ -296,22 +303,23 @@
     MPM.add(createInstructionCombiningPass()); // Combine silly seq's
     addExtensionsToPM(EP_Peephole, MPM);
   }
-  if (EnablePGOInstrGen) {
-    MPM.add(createPGOInstrumentationGenLegacyPass());
+  if ((EnablePGOInstrGen && !IsCS) || (EnablePGOCSInstrGen && IsCS)) {
+    MPM.add(createPGOInstrumentationGenLegacyPass(IsCS));
     // Add the profile lowering pass.
     InstrProfOptions Options;
     if (!PGOInstrGen.empty())
       Options.InstrProfileOutput = PGOInstrGen;
     Options.DoCounterPromotion = true;
+    Options.UseBFIInPromotion = IsCS;
     MPM.add(createLoopRotatePass());
     MPM.add(createInstrProfilingLegacyPass(Options));
   }
   if (!PGOInstrUse.empty())
-    MPM.add(createPGOInstrumentationUseLegacyPass(PGOInstrUse));
+    MPM.add(createPGOInstrumentationUseLegacyPass(PGOInstrUse, IsCS));
   // Indirect call promotion that promotes intra-module targets only.
   // For ThinLTO this is done earlier due to interactions with globalopt
   // for imported functions. We don't run this at -O0.
-  if (OptLevel > 0)
+  if (OptLevel > 0 && !IsCS)
     MPM.add(
         createPGOIndirectCallPromotionLegacyPass(false, !PGOSampleUse.empty()));
 }
@@ -553,6 +561,9 @@
   if (RunPartialInlining)
     MPM.add(createPartialInliningPass());
 
+  // CSFDO instrumentation and use pass.
+  addPGOInstrPasses(MPM, true);
+
   if (OptLevel > 1 && !PrepareForLTO && !PrepareForThinLTO)
     // Remove avail extern fns and globals definitions if we aren't
     // compiling an object file for later LTO. For LTO we want to preserve
@@ -834,6 +845,9 @@
 
   PM.add(createPruneEHPass());   // Remove dead EH info.
 
+  // CSFDO instrumentation and use pass.
+  addPGOInstrPasses(PM, true);
+
   // Optimize globals again if we ran the inliner.
   if (RunInliner)
     PM.add(createGlobalOptimizerPass());
Index: lib/Transforms/IPO/SampleProfile.cpp
===================================================================
--- lib/Transforms/IPO/SampleProfile.cpp
+++ lib/Transforms/IPO/SampleProfile.cpp
@@ -1559,8 +1559,9 @@
     return false;
 
   PSI = _PSI;
-  if (M.getProfileSummary() == nullptr)
-    M.setProfileSummary(Reader->getSummary().getMD(M.getContext()));
+  if (M.getProfileSummary(ProfileSummary::PSK_Sample) == nullptr)
+    M.setProfileSummary(Reader->getSummary().getMD(M.getContext()),
+                        ProfileSummary::PSK_Sample);
 
   // Compute the total number of samples collected in this profile.
   for (const auto &I : Reader->getProfiles())
Index: lib/Transforms/Instrumentation/InstrProfiling.cpp
===================================================================
--- lib/Transforms/Instrumentation/InstrProfiling.cpp
+++ lib/Transforms/Instrumentation/InstrProfiling.cpp
@@ -19,6 +19,8 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/Triple.h"
 #include "llvm/ADT/Twine.h"
+#include "llvm/Analysis/BlockFrequencyInfo.h"
+#include "llvm/Analysis/BranchProbabilityInfo.h"
 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
 #include "llvm/IR/Attributes.h"
@@ -232,9 +234,9 @@
 public:
   PGOCounterPromoter(
       DenseMap<Loop *, SmallVector<LoadStorePair, 8>> &LoopToCands,
-      Loop &CurLoop, LoopInfo &LI)
+      Loop &CurLoop, LoopInfo &LI, BlockFrequencyInfo *BFI)
       : LoopToCandidates(LoopToCands), ExitBlocks(), InsertPts(), L(CurLoop),
-        LI(LI) {
+        LI(LI), BFI(BFI) {
 
     SmallVector<BasicBlock *, 8> LoopExitBlocks;
     SmallPtrSet<BasicBlock *, 8> BlockSet;
@@ -263,6 +265,18 @@
       SSAUpdater SSA(&NewPHIs);
       Value *InitVal = ConstantInt::get(Cand.first->getType(), 0);
 
+      // If BFI is set, we will use it to guide the promotions.
+      if (BFI) {
+        auto *BB = Cand.first->getParent();
+        auto InstrCount = BFI->getBlockProfileCount(BB);
+        if (!InstrCount)
+          continue;
+        auto PreheaderCount = BFI->getBlockProfileCount(L.getLoopPreheader());
+        if (PreheaderCount &&
+            (PreheaderCount.getValue() * 3) >= (InstrCount.getValue() * 2))
+          continue;
+      }
+
       PGOCounterPromoterHelper Promoter(Cand.first, Cand.second, SSA, InitVal,
                                         L.getLoopPreheader(), ExitBlocks,
                                         InsertPts, LoopToCandidates, LI);
@@ -312,6 +326,11 @@
 
     SmallVector<BasicBlock *, 8> ExitingBlocks;
     LP->getExitingBlocks(ExitingBlocks);
+
+    // If BFI is set, we do more aggressive promotions based on BFI.
+    if (BFI)
+      return (unsigned)-1;
+
     // Not considierered speculative.
     if (ExitingBlocks.size() == 1)
       return MaxNumOfPromotionsPerLoop;
@@ -343,6 +362,7 @@
   SmallVector<Instruction *, 8> InsertPts;
   Loop &L;
   LoopInfo &LI;
+  BlockFrequencyInfo *BFI;
 };
 
 } // end anonymous namespace
@@ -415,6 +435,13 @@
   LoopInfo LI(DT);
   DenseMap<Loop *, SmallVector<LoadStorePair, 8>> LoopPromotionCandidates;
 
+  std::unique_ptr<BlockFrequencyInfo> BFI;
+  if (Options.UseBFIInPromotion) {
+    std::unique_ptr<BranchProbabilityInfo> BPI;
+    BPI.reset(new BranchProbabilityInfo(*F, LI, TLI));
+    BFI.reset(new BlockFrequencyInfo(*F, *BPI, LI));
+  }
+
   for (const auto &LoadStore : PromotionCandidates) {
     auto *CounterLoad = LoadStore.first;
     auto *CounterStore = LoadStore.second;
@@ -430,7 +457,7 @@
   // Do a post-order traversal of the loops so that counter updates can be
   // iteratively hoisted outside the loop nest.
   for (auto *Loop : llvm::reverse(Loops)) {
-    PGOCounterPromoter Promoter(LoopPromotionCandidates, *Loop, LI);
+    PGOCounterPromoter Promoter(LoopPromotionCandidates, *Loop, LI, BFI.get());
     Promoter.run(&TotalCountersPromoted);
   }
 }
Index: lib/Transforms/Instrumentation/PGOInstrumentation.cpp
===================================================================
--- lib/Transforms/Instrumentation/PGOInstrumentation.cpp
+++ lib/Transforms/Instrumentation/PGOInstrumentation.cpp
@@ -66,6 +66,7 @@
 #include "llvm/Analysis/IndirectCallSiteVisitor.h"
 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/Analysis/OptimizationRemarkEmitter.h"
+#include "llvm/Analysis/ProfileSummaryInfo.h"
 #include "llvm/IR/Attributes.h"
 #include "llvm/IR/BasicBlock.h"
 #include "llvm/IR/CFG.h"
@@ -384,7 +385,8 @@
 public:
   static char ID;
 
-  PGOInstrumentationGenLegacyPass() : ModulePass(ID) {
+  PGOInstrumentationGenLegacyPass(bool IsCS = false)
+      : ModulePass(ID), IsCS(IsCS) {
     initializePGOInstrumentationGenLegacyPassPass(
         *PassRegistry::getPassRegistry());
   }
@@ -392,6 +394,7 @@
   StringRef getPassName() const override { return "PGOInstrumentationGenPass"; }
 
 private:
+  bool IsCS;
   bool runOnModule(Module &M) override;
 
   void getAnalysisUsage(AnalysisUsage &AU) const override {
@@ -404,8 +407,8 @@
   static char ID;
 
   // Provide the profile filename as the parameter.
-  PGOInstrumentationUseLegacyPass(std::string Filename = "")
-      : ModulePass(ID), ProfileFileName(std::move(Filename)) {
+  PGOInstrumentationUseLegacyPass(std::string Filename = "", bool IsCS = false)
+      : ModulePass(ID), ProfileFileName(std::move(Filename)), IsCS(IsCS) {
     if (!PGOTestProfileFile.empty())
       ProfileFileName = PGOTestProfileFile;
     initializePGOInstrumentationUseLegacyPassPass(
@@ -416,10 +419,12 @@
 
 private:
   std::string ProfileFileName;
+  bool IsCS;
 
   bool runOnModule(Module &M) override;
 
   void getAnalysisUsage(AnalysisUsage &AU) const override {
+    AU.addRequired<ProfileSummaryInfoWrapperPass>();
     AU.addRequired<BlockFrequencyInfoWrapperPass>();
   }
 };
@@ -435,8 +440,8 @@
 INITIALIZE_PASS_END(PGOInstrumentationGenLegacyPass, "pgo-instr-gen",
                     "PGO instrumentation.", false, false)
 
-ModulePass *llvm::createPGOInstrumentationGenLegacyPass() {
-  return new PGOInstrumentationGenLegacyPass();
+ModulePass *llvm::createPGOInstrumentationGenLegacyPass(bool IsCS) {
+  return new PGOInstrumentationGenLegacyPass(IsCS);
 }
 
 char PGOInstrumentationUseLegacyPass::ID = 0;
@@ -445,11 +450,13 @@
                       "Read PGO instrumentation profile.", false, false)
 INITIALIZE_PASS_DEPENDENCY(BlockFrequencyInfoWrapperPass)
 INITIALIZE_PASS_DEPENDENCY(BranchProbabilityInfoWrapperPass)
+INITIALIZE_PASS_DEPENDENCY(ProfileSummaryInfoWrapperPass)
 INITIALIZE_PASS_END(PGOInstrumentationUseLegacyPass, "pgo-instr-use",
                     "Read PGO instrumentation profile.", false, false)
 
-ModulePass *llvm::createPGOInstrumentationUseLegacyPass(StringRef Filename) {
-  return new PGOInstrumentationUseLegacyPass(Filename.str());
+ModulePass *llvm::createPGOInstrumentationUseLegacyPass(StringRef Filename,
+                                                        bool IsCS) {
+  return new PGOInstrumentationUseLegacyPass(Filename.str(), IsCS);
 }
 
 namespace {
@@ -496,6 +503,7 @@
 template <class Edge, class BBInfo> class FuncPGOInstrumentation {
 private:
   Function &F;
+  bool IsCS;
 
   // A map that stores the Comdat group in function F.
   std::unordered_multimap<Comdat *, GlobalValue *> &ComdatMembers;
@@ -536,15 +544,17 @@
       Function &Func,
       std::unordered_multimap<Comdat *, GlobalValue *> &ComdatMembers,
       bool CreateGlobalVar = false, BranchProbabilityInfo *BPI = nullptr,
-      BlockFrequencyInfo *BFI = nullptr)
-      : F(Func), ComdatMembers(ComdatMembers), ValueSites(IPVK_Last + 1),
-        SIVisitor(Func), MIVisitor(Func), MST(F, BPI, BFI) {
+      BlockFrequencyInfo *BFI = nullptr, bool IsCS = false)
+      : F(Func), IsCS(IsCS), ComdatMembers(ComdatMembers),
+        ValueSites(IPVK_Last + 1), SIVisitor(Func), MIVisitor(Func),
+        MST(F, BPI, BFI) {
     // This should be done before CFG hash computation.
     SIVisitor.countSelects(Func);
     MIVisitor.countMemIntrinsics(Func);
     NumOfPGOSelectInsts += SIVisitor.getNumOfSelectInsts();
     NumOfPGOMemIntrinsics += MIVisitor.getNumOfMemIntrinsics();
-    ValueSites[IPVK_IndirectCallTarget] = findIndirectCallSites(Func);
+    if (!IsCS)
+      ValueSites[IPVK_IndirectCallTarget] = findIndirectCallSites(Func);
     ValueSites[IPVK_MemOPSize] = MIVisitor.findMemIntrinsics(Func);
 
     FuncName = getPGOFuncName(F);
@@ -598,14 +608,21 @@
     }
   }
   JC.update(Indexes);
+
+  // Hash format for context senstive profile. Reserve 4 bits for other
+  // information.
   FunctionHash = (uint64_t)SIVisitor.getNumOfSelectInsts() << 56 |
                  (uint64_t)ValueSites[IPVK_IndirectCallTarget].size() << 48 |
                  (uint64_t)MST.AllEdges.size() << 32 | JC.getCRC();
+  // Reserve bit 60-63 for other information purpose.
+  FunctionHash &= 0x0FFFFFFFFFFFFFFF;
+  if (IsCS)
+    NamedInstrProfRecord::setCSFlagInHash(FunctionHash);
   LLVM_DEBUG(dbgs() << "Function Hash Computation for " << F.getName() << ":\n"
                     << " CRC = " << JC.getCRC()
-                    << ", Selects = " << SIVisitor.getNumOfSelectInsts()
                     << ", Edges = " << MST.AllEdges.size() << ", ICSites = "
                     << ValueSites[IPVK_IndirectCallTarget].size()
+                    << ", Selects = " << SIVisitor.getNumOfSelectInsts()
                     << ", Hash = " << FunctionHash << "\n";);
 }
 
@@ -721,12 +738,14 @@
 // Critical edges will be split.
 static void instrumentOneFunc(
     Function &F, Module *M, BranchProbabilityInfo *BPI, BlockFrequencyInfo *BFI,
-    std::unordered_multimap<Comdat *, GlobalValue *> &ComdatMembers) {
+    std::unordered_multimap<Comdat *, GlobalValue *> &ComdatMembers,
+    bool IsCS) {
   // Split indirectbr critical edges here before computing the MST rather than
   // later in getInstrBB() to avoid invalidating it.
   SplitIndirectBrCriticalEdges(F, BPI, BFI);
+
   FuncPGOInstrumentation<PGOEdge, BBInfo> FuncInfo(F, ComdatMembers, true, BPI,
-                                                   BFI);
+                                                   BFI, IsCS);
   unsigned NumCounters = FuncInfo.getNumCounters();
 
   uint32_t I = 0;
@@ -853,10 +872,10 @@
   PGOUseFunc(Function &Func, Module *Modu,
              std::unordered_multimap<Comdat *, GlobalValue *> &ComdatMembers,
              BranchProbabilityInfo *BPI = nullptr,
-             BlockFrequencyInfo *BFIin = nullptr)
+             BlockFrequencyInfo *BFIin = nullptr, bool IsCS = false)
       : F(Func), M(Modu), BFI(BFIin),
-        FuncInfo(Func, ComdatMembers, false, BPI, BFIin),
-        FreqAttr(FFA_Normal) {}
+        FuncInfo(Func, ComdatMembers, false, BPI, BFIin, IsCS),
+        FreqAttr(FFA_Normal), IsCS(IsCS) {}
 
   // Read counts for the instrumented BB from profile.
   bool readCounters(IndexedInstrProfReader *PGOReader, bool &AllZeros);
@@ -929,6 +948,9 @@
   // Function hotness info derived from profile.
   FuncFreqAttr FreqAttr;
 
+  // Is to use the context sensitive profile.
+  bool IsCS;
+
   // Find the Instrumented BB and set the value.
   void setInstrumentedCounts(const std::vector<uint64_t> &CountFromProfile);
 
@@ -1062,7 +1084,7 @@
   getBBInfo(nullptr).UnknownCountInEdge = 2;
 
   setInstrumentedCounts(CountFromProfile);
-  ProgramMaxCount = PGOReader->getMaximumFunctionCount();
+  ProgramMaxCount = PGOReader->getMaximumFunctionCount(IsCS);
   return true;
 }
 
@@ -1175,6 +1197,7 @@
     if (!(isa<BranchInst>(TI) || isa<SwitchInst>(TI) ||
           isa<IndirectBrInst>(TI)))
       continue;
+
     if (getBBInfo(&BB).CountValue == 0)
       continue;
 
@@ -1354,9 +1377,11 @@
 
 // Create a COMDAT variable INSTR_PROF_RAW_VERSION_VAR to make the runtime
 // aware this is an ir_level profile so it can set the version flag.
-static void createIRLevelProfileFlagVariable(Module &M) {
+static void createIRLevelProfileFlagVariable(Module &M, bool IsCS) {
   Type *IntTy64 = Type::getInt64Ty(M.getContext());
   uint64_t ProfileVersion = (INSTR_PROF_RAW_VERSION | VARIANT_MASK_IR_PROF);
+  if (IsCS)
+    ProfileVersion |= VARIANT_MASK_CSIR_PROF;
   auto IRLevelVersionVariable = new GlobalVariable(
       M, IntTy64, true, GlobalVariable::ExternalLinkage,
       Constant::getIntegerValue(IntTy64, APInt(64, ProfileVersion)),
@@ -1390,8 +1415,8 @@
 
 static bool InstrumentAllFunctions(
     Module &M, function_ref<BranchProbabilityInfo *(Function &)> LookupBPI,
-    function_ref<BlockFrequencyInfo *(Function &)> LookupBFI) {
-  createIRLevelProfileFlagVariable(M);
+    function_ref<BlockFrequencyInfo *(Function &)> LookupBFI, bool IsCS) {
+  createIRLevelProfileFlagVariable(M, IsCS);
   std::unordered_multimap<Comdat *, GlobalValue *> ComdatMembers;
   collectComdatMembers(M, ComdatMembers);
 
@@ -1400,7 +1425,7 @@
       continue;
     auto *BPI = LookupBPI(F);
     auto *BFI = LookupBFI(F);
-    instrumentOneFunc(F, &M, BPI, BFI, ComdatMembers);
+    instrumentOneFunc(F, &M, BPI, BFI, ComdatMembers, IsCS);
   }
   return true;
 }
@@ -1415,7 +1440,7 @@
   auto LookupBFI = [this](Function &F) {
     return &this->getAnalysis<BlockFrequencyInfoWrapperPass>(F).getBFI();
   };
-  return InstrumentAllFunctions(M, LookupBPI, LookupBFI);
+  return InstrumentAllFunctions(M, LookupBPI, LookupBFI, IsCS);
 }
 
 PreservedAnalyses PGOInstrumentationGen::run(Module &M,
@@ -1429,7 +1454,7 @@
     return &FAM.getResult<BlockFrequencyAnalysis>(F);
   };
 
-  if (!InstrumentAllFunctions(M, LookupBPI, LookupBFI))
+  if (!InstrumentAllFunctions(M, LookupBPI, LookupBFI, IsCS))
     return PreservedAnalyses::all();
 
   return PreservedAnalyses::none();
@@ -1438,7 +1463,8 @@
 static bool annotateAllFunctions(
     Module &M, StringRef ProfileFileName, StringRef ProfileRemappingFileName,
     function_ref<BranchProbabilityInfo *(Function &)> LookupBPI,
-    function_ref<BlockFrequencyInfo *(Function &)> LookupBFI) {
+    function_ref<BlockFrequencyInfo *(Function &)> LookupBFI,
+    ProfileSummaryInfo *PSI, bool IsCS) {
   LLVM_DEBUG(dbgs() << "Read in profile counters: ");
   auto &Ctx = M.getContext();
   // Read the counter array from file.
@@ -1459,6 +1485,9 @@
                                           StringRef("Cannot get PGOReader")));
     return false;
   }
+  if (!PGOReader->hasCSIRLevelProfile() && IsCS)
+    return false;
+
   // TODO: might need to change the warning once the clang option is finalized.
   if (!PGOReader->isIRLevelProfile()) {
     Ctx.diagnose(DiagnosticInfoPGOProfile(
@@ -1478,7 +1507,7 @@
     // Split indirectbr critical edges here before computing the MST rather than
     // later in getInstrBB() to avoid invalidating it.
     SplitIndirectBrCriticalEdges(F, BPI, BFI);
-    PGOUseFunc Func(F, &M, ComdatMembers, BPI, BFI);
+    PGOUseFunc Func(F, &M, ComdatMembers, BPI, BFI, IsCS);
     bool AllZeros = false;
     if (!Func.readCounters(PGOReader.get(), AllZeros))
       continue;
@@ -1526,7 +1555,10 @@
       }
     }
   }
-  M.setProfileSummary(PGOReader->getSummary().getMD(M.getContext()));
+  M.setProfileSummary(PGOReader->getSummary(IsCS).getMD(M.getContext()),
+                      IsCS ? ProfileSummary::PSK_CSInstr
+                           : ProfileSummary::PSK_Instr);
+
   // Set function hotness attribute from the profile.
   // We have to apply these attributes at the end because their presence
   // can affect the BranchProbabilityInfo of any callers, resulting in an
@@ -1545,9 +1577,10 @@
 }
 
 PGOInstrumentationUse::PGOInstrumentationUse(std::string Filename,
-                                             std::string RemappingFilename)
+                                             std::string RemappingFilename,
+                                             bool IsCS)
     : ProfileFileName(std::move(Filename)),
-      ProfileRemappingFileName(std::move(RemappingFilename)) {
+      ProfileRemappingFileName(std::move(RemappingFilename)), IsCS(IsCS) {
   if (!PGOTestProfileFile.empty())
     ProfileFileName = PGOTestProfileFile;
   if (!PGOTestProfileRemappingFile.empty())
@@ -1566,8 +1599,12 @@
     return &FAM.getResult<BlockFrequencyAnalysis>(F);
   };
 
+  ProfileSummaryInfo *PSI = nullptr;
+  if (IsCS)
+    PSI = &AM.getResult<ProfileSummaryAnalysis>(M);
+
   if (!annotateAllFunctions(M, ProfileFileName, ProfileRemappingFileName,
-                            LookupBPI, LookupBFI))
+                            LookupBPI, LookupBFI, PSI, IsCS))
     return PreservedAnalyses::all();
 
   return PreservedAnalyses::none();
@@ -1584,7 +1621,12 @@
     return &this->getAnalysis<BlockFrequencyInfoWrapperPass>(F).getBFI();
   };
 
-  return annotateAllFunctions(M, ProfileFileName, "", LookupBPI, LookupBFI);
+  ProfileSummaryInfo *PSI = nullptr;
+  if (IsCS)
+    PSI = &this->getAnalysis<ProfileSummaryInfoWrapperPass>().getPSI();
+
+  return annotateAllFunctions(M, ProfileFileName, "", LookupBPI, LookupBFI, PSI,
+                              IsCS);
 }
 
 static std::string getSimpleNodeName(const BasicBlock *Node) {
Index: tools/gold/gold-plugin.cpp
===================================================================
--- tools/gold/gold-plugin.cpp
+++ tools/gold/gold-plugin.cpp
@@ -209,6 +209,10 @@
   static std::string OptRemarksFilename;
   static bool OptRemarksWithHotness = false;
 
+  // Context sensitive PGO options.
+  static std::string cs_profile_path;
+  static bool cs_pgo_gen = false;
+
   static void process_plugin_option(const char *opt_)
   {
     if (opt_ == nullptr)
@@ -266,7 +270,11 @@
     } else if (opt == "disable-verify") {
       DisableVerify = true;
     } else if (opt.startswith("sample-profile=")) {
-      sample_profile= opt.substr(strlen("sample-profile="));
+      sample_profile = opt.substr(strlen("sample-profile="));
+    } else if (opt == "cs-profile-generate") {
+      cs_pgo_gen = true;
+    } else if (opt.startswith("cs-profile-path=")) {
+      cs_profile_path = opt.substr(strlen("cs-profile-path="));
     } else if (opt == "new-pass-manager") {
       new_pass_manager = true;
     } else if (opt == "debug-pass-manager") {
@@ -885,6 +893,10 @@
   if (!options::sample_profile.empty())
     Conf.SampleProfile = options::sample_profile;
 
+  if (!options::cs_profile_path.empty())
+    Conf.CSIRProfile = options::cs_profile_path;
+  Conf.RunCSIRInstr = options::cs_pgo_gen;
+
   Conf.DwoDir = options::dwo_dir;
 
   // Set up optimization remarks handling.
Index: tools/llvm-profdata/llvm-profdata.cpp
===================================================================
--- tools/llvm-profdata/llvm-profdata.cpp
+++ tools/llvm-profdata/llvm-profdata.cpp
@@ -226,7 +226,10 @@
 
   auto Reader = std::move(ReaderOrErr.get());
   bool IsIRProfile = Reader->isIRLevelProfile();
-  if (WC->Writer.setIsIRLevelProfile(IsIRProfile)) {
+  bool HasCSIRProfile = Reader->hasCSIRLevelProfile();
+  // outs() << "Loadinput for " << Input.Filename << " CS=" << HasCSIRProfile <<
+  // " IR=" << IsIRProfile << "\n";
+  if (WC->Writer.setIsIRLevelProfile(IsIRProfile, HasCSIRProfile)) {
     WC->Err = make_error<StringError>(
         "Merge IR generated profile with Clang generated profile.",
         std::error_code());
@@ -252,6 +255,7 @@
                              FuncName, firstTime);
     });
   }
+
   if (Reader->hasError()) {
     if (Error E = Reader->getError()) {
       instrprof_error IPE = InstrProfError::take(std::move(E));
@@ -662,7 +666,7 @@
                             uint32_t TopN, bool ShowIndirectCallTargets,
                             bool ShowMemOPSizes, bool ShowDetailedSummary,
                             std::vector<uint32_t> DetailedSummaryCutoffs,
-                            bool ShowAllFunctions,
+                            bool ShowAllFunctions, bool ShowCS,
                             const std::string &ShowFunction, bool TextFormat,
                             raw_fd_ostream &OS) {
   auto ReaderOrErr = InstrProfReader::create(Filename);
@@ -695,6 +699,13 @@
     OS << ":ir\n";
 
   for (const auto &Func : *Reader) {
+    if (Reader->isIRLevelProfile() && Reader->hasCSIRLevelProfile()) {
+      bool FuncIsCS = NamedInstrProfRecord::hasCSFlagInHash(Func.Hash);
+      if (FuncIsCS && !ShowCS)
+        continue;
+      if (!FuncIsCS && ShowCS)
+        continue;
+    }
     bool Show =
         ShowAllFunctions || (!ShowFunction.empty() &&
                              Func.Name.find(ShowFunction) != Func.Name.npos);
@@ -868,6 +879,8 @@
       cl::value_desc("800000,901000,999999"));
   cl::opt<bool> ShowAllFunctions("all-functions", cl::init(false),
                                  cl::desc("Details for every function"));
+  cl::opt<bool> ShowCS("showcs", cl::init(false),
+                       cl::desc("Show context sensitive counts"));
   cl::opt<std::string> ShowFunction("function",
                                     cl::desc("Details for matching functions"));
 
@@ -899,10 +912,10 @@
   std::vector<uint32_t> Cutoffs(DetailedSummaryCutoffs.begin(),
                                 DetailedSummaryCutoffs.end());
   if (ProfileKind == instr)
-    return showInstrProfile(Filename, ShowCounts, TopNFunctions,
-                            ShowIndirectCallTargets, ShowMemOPSizes,
-                            ShowDetailedSummary, DetailedSummaryCutoffs,
-                            ShowAllFunctions, ShowFunction, TextFormat, OS);
+    return showInstrProfile(
+        Filename, ShowCounts, TopNFunctions, ShowIndirectCallTargets,
+        ShowMemOPSizes, ShowDetailedSummary, DetailedSummaryCutoffs,
+        ShowAllFunctions, ShowCS, ShowFunction, TextFormat, OS);
   else
     return showSampleProfile(Filename, ShowCounts, ShowAllFunctions,
                              ShowFunction, OS);
Index: tools/opt/NewPMDriver.cpp
===================================================================
--- tools/opt/NewPMDriver.cpp
+++ tools/opt/NewPMDriver.cpp
@@ -228,20 +228,21 @@
                            bool EnableDebugify) {
   bool VerifyEachPass = VK == VK_VerifyEachPass;
 
+  // TODO: handle CSFDO options.
   Optional<PGOOptions> P;
   switch (PGOKindFlag) {
     case InstrGen:
-      P = PGOOptions(ProfileFile, "", "", "", true);
+      P = PGOOptions(ProfileFile, "", "", "", PGOOptions::IRInstr);
       break;
     case InstrUse:
-      P = PGOOptions("", ProfileFile, "", ProfileRemappingFile, false);
+      P = PGOOptions("", ProfileFile, "", ProfileRemappingFile);
       break;
     case SampleUse:
-      P = PGOOptions("", "", ProfileFile, ProfileRemappingFile, false);
+      P = PGOOptions("", "", ProfileFile, ProfileRemappingFile);
       break;
     case NoPGO:
       if (DebugInfoForProfiling)
-        P = PGOOptions("", "", "", "", false, true);
+        P = PGOOptions("", "", "", "", PGOOptions::NoAction, true);
       else
         P = None;
   }
Index: unittests/ProfileData/InstrProfTest.cpp
===================================================================
--- unittests/ProfileData/InstrProfTest.cpp
+++ unittests/ProfileData/InstrProfTest.cpp
@@ -176,7 +176,7 @@
     ASSERT_EQ(288230376151711744U, NinetyFivePerc->MinCount);
     ASSERT_EQ(72057594037927936U, NinetyNinePerc->MinCount);
   };
-  ProfileSummary &PS = Reader->getSummary();
+  ProfileSummary &PS = Reader->getSummary(false);
   VerifySummary(PS);
 
   // Test that conversion of summary to and from Metadata works.
@@ -190,8 +190,8 @@
 
   // Test that summary can be attached to and read back from module.
   Module M("my_module", Context);
-  M.setProfileSummary(MD);
-  MD = M.getProfileSummary();
+  M.setProfileSummary(MD, ProfileSummary::PSK_Instr);
+  MD = M.getProfileSummary(ProfileSummary::PSK_Instr);
   ASSERT_TRUE(MD);
   PSFromMD = ProfileSummary::getFromMD(MD);
   ASSERT_TRUE(PSFromMD);
@@ -802,7 +802,7 @@
   auto Profile = Writer.writeBuffer();
   readProfile(std::move(Profile));
 
-  ASSERT_EQ(1ULL << 63, Reader->getMaximumFunctionCount());
+  ASSERT_EQ(1ULL << 63, Reader->getMaximumFunctionCount(false));
 }
 
 TEST_P(MaybeSparseInstrProfTest, get_weighted_function_counts) {
Index: unittests/ProfileData/SampleProfTest.cpp
===================================================================
--- unittests/ProfileData/SampleProfTest.cpp
+++ unittests/ProfileData/SampleProfTest.cpp
@@ -186,8 +186,8 @@
     delete PS;
 
     // Test that summary can be attached to and read back from module.
-    M.setProfileSummary(MD);
-    MD = M.getProfileSummary();
+    M.setProfileSummary(MD, ProfileSummary::PSK_Sample);
+    MD = M.getProfileSummary(ProfileSummary::PSK_Sample);
     ASSERT_TRUE(MD);
     PS = ProfileSummary::getFromMD(MD);
     ASSERT_TRUE(PS);