diff --git a/llvm/docs/CommandGuide/llvm-profdata.rst b/llvm/docs/CommandGuide/llvm-profdata.rst
--- a/llvm/docs/CommandGuide/llvm-profdata.rst
+++ b/llvm/docs/CommandGuide/llvm-profdata.rst
@@ -161,6 +161,12 @@
  coverage for the optimized target. This option can only be used with
  sample-based profile in extbinary format.
 
+.. option:: --convert-sample-profile-layout=[nest|flat]
+
+ Convert the merged profile into a profile with a new layout. Supported
+ layout are ``nest``(Nested profile, the input should be CS flat profile) and
+ ``flat``(Profile with nested inlinees flattened out).
+
 .. option:: --supplement-instr-with-sample=<file>
 
  Supplement an instrumentation profile with sample profile. The sample profile
diff --git a/llvm/include/llvm/ProfileData/SampleProf.h b/llvm/include/llvm/ProfileData/SampleProf.h
--- a/llvm/include/llvm/ProfileData/SampleProf.h
+++ b/llvm/include/llvm/ProfileData/SampleProf.h
@@ -96,6 +96,12 @@
   SPF_Binary = 0xff
 };
 
+enum SampleProfileLayout {
+  SPL_None = 0,
+  SPL_Nest = 0x1,
+  SPL_Flat = 0x2,
+};
+
 static inline uint64_t SPMagic(SampleProfileFormat Format = SPF_Binary) {
   return uint64_t('S') << (64 - 8) | uint64_t('P') << (64 - 16) |
          uint64_t('R') << (64 - 24) | uint64_t('O') << (64 - 32) |
@@ -747,6 +753,8 @@
 
   void setTotalSamples(uint64_t Num) { TotalSamples = Num; }
 
+  void setHeadSamples(uint64_t Num) { TotalHeadSamples = Num; }
+
   sampleprof_error addHeadSamples(uint64_t Num, uint64_t Weight = 1) {
     bool Overflowed;
     TotalHeadSamples =
@@ -934,6 +942,8 @@
     return CallsiteSamples;
   }
 
+  CallsiteSampleMap &getCallsiteSamples() { return CallsiteSamples; }
+
   /// Return the maximum of sample counts in a function body. When SkipCallSite
   /// is false, which is the default, the return count includes samples in the
   /// inlined functions. When SkipCallSite is true, the return count only
@@ -1274,12 +1284,16 @@
   SampleProfileMap &ProfileMap;
 };
 
-// CSProfileConverter converts a full context-sensitive flat sample profile into
-// a nested context-sensitive sample profile.
-class CSProfileConverter {
+/// Helper class for profile conversion.
+///
+/// It supports full context-sensitive profile to nested profile conversion,
+/// nested profile to flatten profile conversion, etc.
+class ProfileConverter {
 public:
-  CSProfileConverter(SampleProfileMap &Profiles);
-  void convertProfiles();
+  ProfileConverter(SampleProfileMap &Profiles);
+  // Convert a full context-sensitive flat sample profile into a nested sample
+  // profile.
+  void convertCSProfiles();
   struct FrameNode {
     FrameNode(StringRef FName = StringRef(),
               FunctionSamples *FSamples = nullptr,
@@ -1299,9 +1313,85 @@
                                      StringRef CalleeName);
   };
 
+  static void flattenProfile(SampleProfileMap &ProfileMap,
+                             bool ProfileIsCS = false) {
+    SampleProfileMap TmpProfiles;
+    flattenProfile(ProfileMap, TmpProfiles, ProfileIsCS);
+    ProfileMap = std::move(TmpProfiles);
+  }
+
+  static void flattenProfile(const SampleProfileMap &InputProfiles,
+                             SampleProfileMap &OutputProfiles,
+                             bool ProfileIsCS = false) {
+    if (ProfileIsCS) {
+      for (const auto &I : InputProfiles)
+        OutputProfiles[I.second.getName()].merge(I.second);
+      // Retain the profile name and clear the full context for each function
+      // profile.
+      for (auto &I : OutputProfiles)
+        I.second.setContext(SampleContext(I.first));
+    } else {
+      for (const auto &I : InputProfiles)
+        flattenNestedProfile(OutputProfiles, I.second);
+    }
+  }
+
 private:
+  static void flattenNestedProfile(SampleProfileMap &OutputProfiles,
+                                   const FunctionSamples &FS) {
+    // To retain the context, checksum, attributes of the original profile, make
+    // a copy of it if no profile is found.
+    SampleContext &Context = FS.getContext();
+    auto Ret = OutputProfiles.emplace(Context, FS);
+    FunctionSamples &Profile = Ret.first->second;
+    if (Ret.second) {
+      // When it's the copy of the old profile, just clear all the inlinees'
+      // samples.
+      Profile.getCallsiteSamples().clear();
+      // We recompute TotalSamples later, so here set to zero.
+      Profile.setTotalSamples(0);
+    } else {
+      for (const auto &Line : FS.getBodySamples()) {
+        Profile.addBodySamples(Line.first.LineOffset, Line.first.Discriminator,
+                               Line.second.getSamples());
+      }
+    }
+
+    assert(Profile.getCallsiteSamples().empty() &&
+           "There should be no inlinees' profiles after flattening.");
+
+    // TotalSamples might not be equal to the sum of all samples from
+    // BodySamples and CallsiteSamples. So here we use "TotalSamples =
+    // Original_TotalSamples - All_of_Callsite_TotalSamples +
+    // All_of_Callsite_HeadSamples" to compute the new TotalSamples.
+    uint64_t TotalSamples = FS.getTotalSamples();
+
+    for (const auto &I : FS.getCallsiteSamples()) {
+      for (const auto &Callee : I.second) {
+        const auto &CalleeProfile = Callee.second;
+        // Add body sample.
+        Profile.addBodySamples(I.first.LineOffset, I.first.Discriminator,
+                               CalleeProfile.getHeadSamplesEstimate());
+        // Add callsite sample.
+        Profile.addCalledTargetSamples(
+            I.first.LineOffset, I.first.Discriminator, CalleeProfile.getName(),
+            CalleeProfile.getHeadSamplesEstimate());
+        // Update total samples.
+        TotalSamples = TotalSamples >= CalleeProfile.getTotalSamples()
+                           ? TotalSamples - CalleeProfile.getTotalSamples()
+                           : 0;
+        TotalSamples += CalleeProfile.getHeadSamplesEstimate();
+        // Recursively convert callee profile.
+        flattenNestedProfile(OutputProfiles, CalleeProfile);
+      }
+    }
+    Profile.addTotalSamples(TotalSamples);
+
+    Profile.setHeadSamples(Profile.getHeadSamplesEstimate());
+  }
+
   // Nest all children profiles into the profile of Node.
-  void convertProfiles(FrameNode &Node);
+  void convertCSProfiles(FrameNode &Node);
   FrameNode *getOrCreateContextPath(const SampleContext &Context);
 
   SampleProfileMap &ProfileMap;
diff --git a/llvm/lib/ProfileData/SampleProf.cpp b/llvm/lib/ProfileData/SampleProf.cpp
--- a/llvm/lib/ProfileData/SampleProf.cpp
+++ b/llvm/lib/ProfileData/SampleProf.cpp
@@ -461,9 +461,9 @@
     OS << Sym << "\n";
 }
 
-CSProfileConverter::FrameNode *
-CSProfileConverter::FrameNode::getOrCreateChildFrame(
-    const LineLocation &CallSite, StringRef CalleeName) {
+ProfileConverter::FrameNode *
+ProfileConverter::FrameNode::getOrCreateChildFrame(const LineLocation &CallSite,
+                                                   StringRef CalleeName) {
   uint64_t Hash = FunctionSamples::getCallSiteHash(CalleeName, CallSite);
   auto It = AllChildFrames.find(Hash);
   if (It != AllChildFrames.end()) {
@@ -476,7 +476,7 @@
   return &AllChildFrames[Hash];
 }
 
-CSProfileConverter::CSProfileConverter(SampleProfileMap &Profiles)
+ProfileConverter::ProfileConverter(SampleProfileMap &Profiles)
     : ProfileMap(Profiles) {
   for (auto &FuncSample : Profiles) {
     FunctionSamples *FSamples = &FuncSample.second;
@@ -486,8 +486,8 @@
   }
 }
 
-CSProfileConverter::FrameNode *
-CSProfileConverter::getOrCreateContextPath(const SampleContext &Context) {
+ProfileConverter::FrameNode *
+ProfileConverter::getOrCreateContextPath(const SampleContext &Context) {
   auto Node = &RootFrame;
   LineLocation CallSiteLoc(0, 0);
   for (auto &Callsite : Context.getContextFrames()) {
@@ -497,14 +497,14 @@
   return Node;
 }
 
-void CSProfileConverter::convertProfiles(CSProfileConverter::FrameNode &Node) {
+void ProfileConverter::convertCSProfiles(ProfileConverter::FrameNode &Node) {
   // Process each child profile. Add each child profile to callsite profile map
   // of the current node `Node` if `Node` comes with a profile. Otherwise
   // promote the child profile to a standalone profile.
   auto *NodeProfile = Node.FuncSamples;
   for (auto &It : Node.AllChildFrames) {
     auto &ChildNode = It.second;
-    convertProfiles(ChildNode);
+    convertCSProfiles(ChildNode);
     auto *ChildProfile = ChildNode.FuncSamples;
     if (!ChildProfile)
       continue;
@@ -544,4 +544,4 @@
   }
 }
 
-void CSProfileConverter::convertProfiles() { convertProfiles(RootFrame); }
+void ProfileConverter::convertCSProfiles() { convertCSProfiles(RootFrame); }
diff --git a/llvm/lib/Transforms/IPO/SampleProfile.cpp b/llvm/lib/Transforms/IPO/SampleProfile.cpp
--- a/llvm/lib/Transforms/IPO/SampleProfile.cpp
+++ b/llvm/lib/Transforms/IPO/SampleProfile.cpp
@@ -139,6 +139,11 @@
     cl::desc("Compute stale profile statistical metrics and write it into the "
              "native object file(.llvm_stats section)."));
 
+static cl::opt<bool> FlattenProfileForMatching(
+    "flatten-profile-for-matching", cl::Hidden, cl::init(true),
+    cl::desc(
+        "Use flattened profile for stale profile detection and matching."));
+
 static cl::opt<bool> ProfileSampleAccurate(
     "profile-sample-accurate", cl::Hidden, cl::init(false),
     cl::desc("If the sample profile is accurate, we will mark all un-sampled "
@@ -434,6 +439,7 @@
   Module &M;
   SampleProfileReader &Reader;
   const PseudoProbeManager *ProbeManager;
+  SampleProfileMap FlattenedProfiles;
 
   // Profile mismatching statstics.
   uint64_t TotalProfiledCallsites = 0;
@@ -448,7 +454,21 @@
 public:
   SampleProfileMatcher(Module &M, SampleProfileReader &Reader,
                        const PseudoProbeManager *ProbeManager)
-      : M(M), Reader(Reader), ProbeManager(ProbeManager) {}
+      : M(M), Reader(Reader), ProbeManager(ProbeManager) {
+    if (FlattenProfileForMatching) {
+      ProfileConverter::flattenProfile(Reader.getProfiles(), FlattenedProfiles,
+                                       FunctionSamples::ProfileIsCS);
+    }
+  }
+
+  FunctionSamples *getFlattenedSamplesFor(const Function &F) {
+    StringRef CanonFName = FunctionSamples::getCanonicalFnName(F);
+    auto It = FlattenedProfiles.find(CanonFName);
+    if (It != FlattenedProfiles.end())
+      return &It->second;
+    return nullptr;
+  }
+
   void detectProfileMismatch();
   void detectProfileMismatch(const Function &F, const FunctionSamples &FS);
 };
@@ -2156,7 +2176,11 @@
   for (auto &F : M) {
     if (F.isDeclaration() || !F.hasFnAttribute("use-sample-profile"))
       continue;
-    FunctionSamples *FS = Reader.getSamplesFor(F);
+    FunctionSamples *FS = nullptr;
+    if (FlattenProfileForMatching)
+      FS = getFlattenedSamplesFor(F);
+    else
+      FS = Reader.getSamplesFor(F);
     if (!FS)
       continue;
     detectProfileMismatch(F, *FS);
diff --git a/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch-cs.prof b/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch-cs.prof
new file mode 100644
--- /dev/null
+++ b/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch-cs.prof
@@ -0,0 +1,18 @@
+[main]:30:0
+ 0: 0
+ 1.1: 0
+ 3: 10 matched:10
+ 4: 10
+ 5: 10 bar_mismatch:10
+ 7: 5 foo:5
+ 8: 0
+[main:7 @ foo]:15:5
+ 1: 5
+ 2: 5
+ 3: 5 inlinee_mismatch:5
+[bar]:10:10
+ 1: 10
+[matched]:10:10
+ 1: 10
+[main:7 @ foo:3 @ inlinee_mismatch]:5:5
+ 1: 5
diff --git a/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch.prof b/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch.prof
--- a/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch.prof
+++ b/llvm/test/Transforms/SampleProfile/Inputs/profile-mismatch.prof
@@ -5,9 +5,11 @@
  4: 10
  5: 10 bar_mismatch:10
  8: 0
- 7: foo:10
+ 7: foo:15
   1: 5
   2: 5
+  3: inlinee_mismatch:5
+   1: 5
 bar:10:10
  1: 10
 matched:10:10
diff --git a/llvm/test/Transforms/SampleProfile/csspgo-inline.ll b/llvm/test/Transforms/SampleProfile/csspgo-inline.ll
--- a/llvm/test/Transforms/SampleProfile/csspgo-inline.ll
+++ b/llvm/test/Transforms/SampleProfile/csspgo-inline.ll
@@ -10,7 +10,7 @@
 ; RUN: llvm-profdata merge --sample --extbinary --use-md5 %S/Inputs/profile-context-tracker.prof -o %t.md5
 ; RUN: opt < %s -passes=sample-profile -sample-profile-file=%t.md5 -sample-profile-inline-size -sample-profile-prioritized-inline=0 -profile-sample-accurate -S -pass-remarks=inline -o /dev/null 2>&1 | FileCheck %s --check-prefix=INLINE-BASE
 
-; RUN: llvm-profdata merge --sample --text --gen-cs-nested-profile %S/Inputs/profile-context-tracker.prof -o %t.prof
+; RUN: llvm-profdata merge --sample --text --convert-sample-profile-layout=nest  %S/Inputs/profile-context-tracker.prof -o %t.prof
 ; RUN: opt < %s -passes=sample-profile -sample-profile-file=%t.prof -sample-profile-inline-size -sample-profile-prioritized-inline=0 -profile-sample-accurate -S -pass-remarks=inline -o /dev/null 2>&1 | FileCheck %s --check-prefix=INLINE-BASE
 
 ; With new FDO early inliner, callee entry count is used to drive inlining instead of callee total samples, so we get less inlining for given profile
diff --git a/llvm/test/Transforms/SampleProfile/csspgo-use-preinliner.ll b/llvm/test/Transforms/SampleProfile/csspgo-use-preinliner.ll
--- a/llvm/test/Transforms/SampleProfile/csspgo-use-preinliner.ll
+++ b/llvm/test/Transforms/SampleProfile/csspgo-use-preinliner.ll
@@ -3,7 +3,7 @@
 ; RUN: opt < %s -passes=sample-profile -sample-profile-file=%S/Inputs/csspgo-use-preinliner.prof -pass-remarks=inline -sample-profile-prioritized-inline -profile-sample-accurate -sample-profile-use-preinliner=0 -S 2>&1 | FileCheck %s --check-prefix=DEFAULT
 ; RUN: opt < %s -passes=sample-profile -sample-profile-file=%S/Inputs/csspgo-use-preinliner.prof -pass-remarks=inline -sample-profile-prioritized-inline -profile-sample-accurate -sample-profile-use-preinliner=1 -S 2>&1 | FileCheck %s --check-prefix=PREINLINE
 
-; RUN: llvm-profdata merge --sample --text --gen-cs-nested-profile -generate-merged-base-profiles=0 %S/Inputs/csspgo-use-preinliner.prof -o %t.prof
+; RUN: llvm-profdata merge --sample --text --convert-sample-profile-layout=nest  -generate-merged-base-profiles=0 %S/Inputs/csspgo-use-preinliner.prof -o %t.prof
 ; RUN: opt < %s -passes=sample-profile -sample-profile-file=%t.prof -pass-remarks=inline -sample-profile-prioritized-inline -profile-sample-accurate -sample-profile-use-preinliner=0 -S 2>&1 | FileCheck %s --check-prefix=DEFAULT
 ; RUN: opt < %s -passes=sample-profile -sample-profile-file=%t.prof -pass-remarks=inline -sample-profile-prioritized-inline -profile-sample-accurate -sample-profile-use-preinliner=1 -S 2>&1 | FileCheck %s --check-prefix=PREINLINE
 
diff --git a/llvm/test/Transforms/SampleProfile/profile-mismatch-flattened-profile.ll b/llvm/test/Transforms/SampleProfile/profile-mismatch-flattened-profile.ll
new file mode 100644
--- /dev/null
+++ b/llvm/test/Transforms/SampleProfile/profile-mismatch-flattened-profile.ll
@@ -0,0 +1,13 @@
+; REQUIRES: x86_64-linux
+; RUN: opt < %S/profile-mismatch.ll -passes=sample-profile -sample-profile-file=%S/Inputs/profile-mismatch.prof -report-profile-staleness -persist-profile-staleness -flatten-profile-for-matching=1 -S 2>%t -o %t.ll
+; RUN: FileCheck %s --input-file %t
+; RUN: FileCheck %s --input-file %t.ll -check-prefix=CHECK-MD
+
+; RUN: opt < %S/profile-mismatch.ll -passes=sample-profile -sample-profile-file=%S/Inputs/profile-mismatch-cs.prof -report-profile-staleness -persist-profile-staleness -flatten-profile-for-matching=1 -S 2>%t -o %t.ll
+; RUN: FileCheck %s --input-file %t
+; RUN: FileCheck %s --input-file %t.ll -check-prefix=CHECK-MD
+
+
+; CHECK: (3/4) of callsites' profile are invalid and (20/30) of samples are discarded due to callsite location mismatch.
+
+; CHECK-MD: ![[#]] = !{!"NumMismatchedCallsites", i64 3, !"TotalProfiledCallsites", i64 4, !"MismatchedCallsiteSamples", i64 20, !"TotalCallsiteSamples", i64 30}
diff --git a/llvm/test/Transforms/SampleProfile/profile-mismatch.ll b/llvm/test/Transforms/SampleProfile/profile-mismatch.ll
--- a/llvm/test/Transforms/SampleProfile/profile-mismatch.ll
+++ b/llvm/test/Transforms/SampleProfile/profile-mismatch.ll
@@ -1,5 +1,5 @@
 ; REQUIRES: x86_64-linux
-; RUN: opt < %s -passes=sample-profile -sample-profile-file=%S/Inputs/profile-mismatch.prof -report-profile-staleness -persist-profile-staleness -S 2>%t -o %t.ll
+; RUN: opt < %s -passes=sample-profile -sample-profile-file=%S/Inputs/profile-mismatch.prof -report-profile-staleness -persist-profile-staleness -flatten-profile-for-matching=0 -S 2>%t -o %t.ll
 ; RUN: FileCheck %s --input-file %t
 ; RUN: FileCheck %s --input-file %t.ll -check-prefix=CHECK-MD
 ; RUN: llc < %t.ll -filetype=obj -o %t.obj
diff --git a/llvm/test/tools/llvm-profdata/Inputs/sample-flatten-profile-cs.proftext b/llvm/test/tools/llvm-profdata/Inputs/sample-flatten-profile-cs.proftext
new file mode 100644
--- /dev/null
+++ b/llvm/test/tools/llvm-profdata/Inputs/sample-flatten-profile-cs.proftext
@@ -0,0 +1,20 @@
+[baz]:150:10
+ 1: 10
+ 3: 20
+ 5: 20 foo:20
+[foo]:102:1
+ 1: 1
+ 3: 1
+[main]:91:1
+ 4: 1
+ 4.2: 1
+ 7: 1
+ 9: 3 bar:2 foo:1
+ 10: 3 baz:2 foo:1
+[main:10 @ foo]:2:1
+ 3: 1 bar:1
+ 4: 1
+[bar]:1:1
+ 1: 1
+[main:10 @ foo:3 @ bar]:1:1
+ 1: 1
diff --git a/llvm/test/tools/llvm-profdata/Inputs/sample-flatten-profile.proftext b/llvm/test/tools/llvm-profdata/Inputs/sample-flatten-profile.proftext
new file mode 100644
--- /dev/null
+++ b/llvm/test/tools/llvm-profdata/Inputs/sample-flatten-profile.proftext
@@ -0,0 +1,44 @@
+baz:160:10
+ 1: 10
+ 3: 20
+ 5: foo:30
+  1: 20
+  3: bar:10
+   1: 10
+   !CFGChecksum: 4
+   !Attributes: 4
+  !CFGChecksum: 3
+  !Attributes: 3
+ !CFGChecksum: 1
+ !Attributes: 1
+main:110:1
+ 4: 1
+ 4.2: 1
+ 7: 1
+ 9: 3 bar:2 foo:1
+ 10: foo:2
+  4: 1
+  3: bar:1
+   1: 1
+   !CFGChecksum: 4
+   !Attributes: 4
+  !CFGChecksum: 3
+  !Attributes: 3
+ 10: baz:20
+  10: 1
+  6: bar:3
+   1: 2
+   7: 1
+   !CFGChecksum: 4
+   !Attributes: 4
+ !CFGChecksum: 2
+ !Attributes: 2
+foo:102:1
+ 1: 1
+ 3: 1
+ !CFGChecksum: 3
+ !Attributes: 3
+bar:1:1
+ 1: 1
+ !CFGChecksum: 4
+ !Attributes: 4
diff --git a/llvm/test/tools/llvm-profdata/cs-sample-nested-profile.test b/llvm/test/tools/llvm-profdata/cs-sample-nested-profile.test
--- a/llvm/test/tools/llvm-profdata/cs-sample-nested-profile.test
+++ b/llvm/test/tools/llvm-profdata/cs-sample-nested-profile.test
@@ -1,14 +1,14 @@
-RUN: llvm-profdata merge --sample --text -output=%t.proftext %S/Inputs/cs-sample-preinline.proftext --gen-cs-nested-profile=1 -generate-merged-base-profiles=0
+RUN: llvm-profdata merge --sample --text -output=%t.proftext %S/Inputs/cs-sample-preinline.proftext --convert-sample-profile-layout=nest  -generate-merged-base-profiles=0
 RUN: FileCheck %s < %t.proftext --match-full-lines --strict-whitespace
-RUN: llvm-profdata merge --sample --text -output=%t.probe.proftext %S/Inputs/cs-sample-preinline-probe.proftext --gen-cs-nested-profile=1 -generate-merged-base-profiles=0
+RUN: llvm-profdata merge --sample --text -output=%t.probe.proftext %S/Inputs/cs-sample-preinline-probe.proftext --convert-sample-profile-layout=nest  -generate-merged-base-profiles=0
 RUN: FileCheck %s < %t.probe.proftext --match-full-lines --strict-whitespace -check-prefix=PROBE
-RUN: llvm-profdata merge --sample --extbinary -output=%t.profbin %S/Inputs/cs-sample-preinline.proftext --gen-cs-nested-profile=1 -generate-merged-base-profiles=0
+RUN: llvm-profdata merge --sample --extbinary -output=%t.profbin %S/Inputs/cs-sample-preinline.proftext --convert-sample-profile-layout=nest  -generate-merged-base-profiles=0
 RUN: llvm-profdata merge --sample --text -output=%t2.proftext %t.profbin
 RUN: FileCheck %s < %t2.proftext --match-full-lines --strict-whitespace
 RUN: llvm-profdata show --sample -show-sec-info-only %t.profbin | FileCheck %s -check-prefix=PREINLINE
-RUN: llvm-profdata merge --sample --text -output=%t3.proftext %S/Inputs/cs-sample-preinline.proftext --gen-cs-nested-profile=1 -generate-merged-base-profiles=1
+RUN: llvm-profdata merge --sample --text -output=%t3.proftext %S/Inputs/cs-sample-preinline.proftext --convert-sample-profile-layout=nest  -generate-merged-base-profiles=1
 RUN: FileCheck %s < %t3.proftext --match-full-lines --strict-whitespace -check-prefix=RECOUNT
-RUN: llvm-profdata merge --sample --extbinary -output=%t2.profbin %S/Inputs/cs-sample-preinline.proftext --gen-cs-nested-profile=1 -generate-merged-base-profiles=1
+RUN: llvm-profdata merge --sample --extbinary -output=%t2.profbin %S/Inputs/cs-sample-preinline.proftext --convert-sample-profile-layout=nest  -generate-merged-base-profiles=1
 RUN: llvm-profdata show -sample -detailed-summary %S/Inputs/cs-sample-preinline.proftext | FileCheck %s -check-prefix=SUMMARY
 RUN: llvm-profdata show -sample -detailed-summary %t2.profbin | FileCheck %s -check-prefix=SUMMARY-NEST
 RUN: llvm-profdata show -sample -detailed-summary %t3.proftext | FileCheck %s -check-prefix=SUMMARY-NEST
diff --git a/llvm/test/tools/llvm-profdata/sample-flatten-profile.test b/llvm/test/tools/llvm-profdata/sample-flatten-profile.test
new file mode 100644
--- /dev/null
+++ b/llvm/test/tools/llvm-profdata/sample-flatten-profile.test
@@ -0,0 +1,50 @@
+; RUN: llvm-profdata merge --sample --convert-sample-profile-layout=flat --text %S/Inputs/sample-flatten-profile.proftext -o - | FileCheck %s --match-full-lines --strict-whitespace
+; RUN: llvm-profdata merge --sample --extbinary %S/Inputs/sample-flatten-profile.proftext -o %t2 && llvm-profdata merge --sample --convert-sample-profile-layout=flat --text %t2 -o - | FileCheck %s --match-full-lines --strict-whitespace
+
+; RUN: llvm-profdata merge --sample --convert-sample-profile-layout=flat --text %S/Inputs/sample-flatten-profile-cs.proftext -o - | FileCheck %s --match-full-lines --strict-whitespace --check-prefix=CHECK-CS
+; RUN: llvm-profdata merge --sample --extbinary %S/Inputs/sample-flatten-profile-cs.proftext -o %t2 && llvm-profdata merge --sample --convert-sample-profile-layout=flat --text %t2 -o - | FileCheck %s --match-full-lines --strict-whitespace --check-prefix=CHECK-CS
+
+; CHECK:baz:169:10
+; CHECK-NEXT: 1: 10
+; CHECK-NEXT: 3: 20
+; CHECK-NEXT: 5: 20 foo:20
+; CHECK-NEXT: 6: 2 bar:2
+; CHECK-NEXT: 10: 1
+; CHECK-NEXT: !CFGChecksum: 1
+; CHECK-NEXT: !Attributes: 1
+; CHECK-NEXT:foo:134:21
+; CHECK-NEXT: 1: 21
+; CHECK-NEXT: 3: 12 bar:11
+; CHECK-NEXT: 4: 1
+; CHECK-NEXT: !CFGChecksum: 3
+; CHECK-NEXT: !Attributes: 3
+; CHECK-NEXT:main:91:1
+; CHECK-NEXT: 4: 1
+; CHECK-NEXT: 4.2: 1
+; CHECK-NEXT: 7: 1
+; CHECK-NEXT: 9: 3 bar:2 foo:1
+; CHECK-NEXT: 10: 3 baz:2 foo:1
+; CHECK-NEXT: !CFGChecksum: 2
+; CHECK-NEXT: !Attributes: 2
+; CHECK-NEXT:bar:15:14
+; CHECK-NEXT: 1: 14
+; CHECK-NEXT: 7: 1
+; CHECK-NEXT: !CFGChecksum: 4
+; CHECK-NEXT: !Attributes: 4
+
+; CHECK-CS:baz:150:10
+; CHECK-CS-NEXT: 1: 10
+; CHECK-CS-NEXT: 3: 20
+; CHECK-CS-NEXT: 5: 20 foo:20
+; CHECK-CS-NEXT:foo:104:2
+; CHECK-CS-NEXT: 1: 1
+; CHECK-CS-NEXT: 3: 2 bar:1
+; CHECK-CS-NEXT: 4: 1
+; CHECK-CS-NEXT:main:91:1
+; CHECK-CS-NEXT: 4: 1
+; CHECK-CS-NEXT: 4.2: 1
+; CHECK-CS-NEXT: 7: 1
+; CHECK-CS-NEXT: 9: 3 bar:2 foo:1
+; CHECK-CS-NEXT: 10: 3 baz:2 foo:1
+; CHECK-CS-NEXT:bar:2:2
+; CHECK-CS-NEXT: 1: 2
diff --git a/llvm/tools/llvm-profdata/llvm-profdata.cpp b/llvm/tools/llvm-profdata/llvm-profdata.cpp
--- a/llvm/tools/llvm-profdata/llvm-profdata.cpp
+++ b/llvm/tools/llvm-profdata/llvm-profdata.cpp
@@ -968,7 +968,8 @@
 mergeSampleProfile(const WeightedFileVector &Inputs, SymbolRemapper *Remapper,
                    StringRef OutputFilename, ProfileFormat OutputFormat,
                    StringRef ProfileSymbolListFile, bool CompressAllSections,
-                   bool UseMD5, bool GenPartialProfile, bool GenCSNestedProfile,
+                   bool UseMD5, bool GenPartialProfile,
+                   SampleProfileLayout ProfileLayout,
                    bool SampleMergeColdContext, bool SampleTrimColdContext,
                    bool SampleColdContextFrameDepth, FailureMode FailMode,
                    bool DropProfileSymbolList, size_t OutputSizeLimit) {
@@ -1048,9 +1049,12 @@
             SampleMergeColdContext, SampleColdContextFrameDepth, false);
   }
 
-  if (ProfileIsCS && GenCSNestedProfile) {
-    CSProfileConverter CSConverter(ProfileMap);
-    CSConverter.convertProfiles();
+  if (ProfileLayout == llvm::sampleprof::SPL_Flat) {
+    ProfileConverter::flattenProfile(ProfileMap, FunctionSamples::ProfileIsCS);
+    ProfileIsCS = FunctionSamples::ProfileIsCS = false;
+  } else if (ProfileIsCS && ProfileLayout == llvm::sampleprof::SPL_Nest) {
+    ProfileConverter CSConverter(ProfileMap);
+    CSConverter.convertCSProfiles();
     ProfileIsCS = FunctionSamples::ProfileIsCS = false;
   }
 
@@ -1241,9 +1245,15 @@
       "instr-prof-cold-threshold", cl::init(0), cl::Hidden,
       cl::desc("User specified cold threshold for instr profile which will "
                "override the cold threshold got from profile summary. "));
-  cl::opt<bool> GenCSNestedProfile(
-      "gen-cs-nested-profile", cl::Hidden, cl::init(false),
-      cl::desc("Generate nested function profiles for CSSPGO"));
+  cl::opt<SampleProfileLayout> ProfileLayout(
+      "convert-sample-profile-layout",
+      cl::desc("Convert the generated profile to a profile with a new layout"),
+      cl::init(SPL_None),
+      cl::values(
+          clEnumValN(SPL_Nest, "nest",
+                     "Nested profile, the input should be CS flat profile"),
+          clEnumValN(SPL_Flat, "flat",
+                     "Profile with nested inlinee flatten out")));
   cl::opt<std::string> DebugInfoFilename(
       "debug-info", cl::init(""),
       cl::desc("Use the provided debug info to correlate the raw profile."));
@@ -1298,12 +1308,12 @@
                       OutputFilename, OutputFormat, OutputSparse, NumThreads,
                       FailureMode, ProfiledBinary);
   else
-    mergeSampleProfile(
-        WeightedInputs, Remapper.get(), OutputFilename, OutputFormat,
-        ProfileSymbolListFile, CompressAllSections, UseMD5, GenPartialProfile,
-        GenCSNestedProfile, SampleMergeColdContext, SampleTrimColdContext,
-        SampleColdContextFrameDepth, FailureMode, DropProfileSymbolList,
-        OutputSizeLimit);
+    mergeSampleProfile(WeightedInputs, Remapper.get(), OutputFilename,
+                       OutputFormat, ProfileSymbolListFile, CompressAllSections,
+                       UseMD5, GenPartialProfile, ProfileLayout,
+                       SampleMergeColdContext, SampleTrimColdContext,
+                       SampleColdContextFrameDepth, FailureMode,
+                       DropProfileSymbolList, OutputSizeLimit);
   return 0;
 }
 
diff --git a/llvm/tools/llvm-profgen/ProfileGenerator.cpp b/llvm/tools/llvm-profgen/ProfileGenerator.cpp
--- a/llvm/tools/llvm-profgen/ProfileGenerator.cpp
+++ b/llvm/tools/llvm-profgen/ProfileGenerator.cpp
@@ -1026,8 +1026,8 @@
 
   calculateAndShowDensity(ContextLessProfiles);
   if (GenCSNestedProfile) {
-    CSProfileConverter CSConverter(ProfileMap);
-    CSConverter.convertProfiles();
+    ProfileConverter CSConverter(ProfileMap);
+    CSConverter.convertCSProfiles();
     FunctionSamples::ProfileIsCS = false;
   }
 }