diff --git a/llvm/include/llvm/MC/MCPseudoProbe.h b/llvm/include/llvm/MC/MCPseudoProbe.h
--- a/llvm/include/llvm/MC/MCPseudoProbe.h
+++ b/llvm/include/llvm/MC/MCPseudoProbe.h
@@ -36,7 +36,6 @@
 //        A list of NUM_INLINED_FUNCTIONS entries describing each of the inlined
 //        callees.  Each record contains:
 //          INLINE SITE
-//            GUID of the inlinee (uint64)
 //            ID of the callsite probe (ULEB128)
 //          FUNCTION BODY
 //            A FUNCTION BODY entry describing the inlined function.
diff --git a/llvm/test/Transforms/SampleProfile/pseudo-probe-slotindex.s b/llvm/test/Transforms/SampleProfile/pseudo-probe-slotindex.s
new file mode 100644
--- /dev/null
+++ b/llvm/test/Transforms/SampleProfile/pseudo-probe-slotindex.s
@@ -0,0 +1,67 @@
+--- |
+  ; ModuleID = '/home/hoy/src/llvm-project/llvm/test/Transforms/SampleProfile/pseudo-probe-slotindex.ll'
+  source_filename = "/home/hoy/src/llvm-project/llvm/test/Transforms/SampleProfile/pseudo-probe-slotindex.ll"
+  target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
+  target triple = "x86_64-unknown-unknown"
+  
+  define void @foo(i32* %p) {
+    call void @llvm.pseudoprobe(i64 5116412291814990879, i64 1, i32 0, i64 -1)
+    store i32 0, i32* %p, align 4
+    store i32 0, i32* %p, align 4
+    ret void
+  }
+  
+  ; Function Attrs: inaccessiblememonly nounwind willreturn
+  declare void @llvm.pseudoprobe(i64, i64, i32, i64) #0
+  
+  attributes #0 = { inaccessiblememonly nounwind willreturn }
+
+...
+---
+name:            foo
+alignment:       16
+exposesReturnsTwice: false
+legalized:       false
+regBankSelected: false
+selected:        false
+failedISel:      false
+tracksRegLiveness: true
+hasWinCFI:       false
+registers:       []
+liveins:
+  - { reg: '$rdi', virtual-reg: '' }
+frameInfo:
+  isFrameAddressTaken: false
+  isReturnAddressTaken: false
+  hasStackMap:     false
+  hasPatchPoint:   false
+  stackSize:       0
+  offsetAdjustment: 0
+  maxAlignment:    1
+  adjustsStack:    false
+  hasCalls:        false
+  stackProtector:  ''
+  maxCallFrameSize: 0
+  cvBytesOfCalleeSavedRegisters: 0
+  hasOpaqueSPAdjustment: false
+  hasVAStart:      false
+  hasMustTailInVarArgFunc: false
+  hasTailCall:     false
+  localFrameSize:  0
+  savePoint:       ''
+  restorePoint:    ''
+fixedStack:      []
+stack:           []
+callSites:       []
+debugValueSubstitutions: []
+constants:       []
+machineFunctionInfo: {}
+body:             |
+  bb.0 (%ir-block.0):
+    liveins: $rdi
+  
+    PSEUDO_PROBE 5116412291814990879, 1, 0, 0
+    MOV32mi killed renamable $rdi, 1, $noreg, 0, $noreg, 0 :: (store 4 into %ir.p)
+    RETQ
+
+...
diff --git a/llvm/tools/llvm-profgen/PseudoProbe.h b/llvm/tools/llvm-profgen/PseudoProbe.h
--- a/llvm/tools/llvm-profgen/PseudoProbe.h
+++ b/llvm/tools/llvm-profgen/PseudoProbe.h
@@ -71,6 +71,7 @@
     return Ret.first->second.get();
   }
 
+  auto &getChildren() { return Children; }
   void addProbes(PseudoProbe *Probe) { ProbeVector.push_back(Probe); }
   // Return false if it's a dummy inline site
   bool hasInlineSite() const { return std::get<0>(ISite) != 0; }
@@ -91,7 +92,7 @@
 // GUID to PseudoProbeFuncDesc map
 using GUIDProbeFunctionMap = std::unordered_map<uint64_t, PseudoProbeFuncDesc>;
 // Address to pseudo probes map.
-using AddressProbesMap = std::unordered_map<uint64_t, std::vector<PseudoProbe>>;
+using AddressProbesMap = std::unordered_map<uint64_t, std::list<PseudoProbe>>;
 
 /*
 A pseudo probe has the format like below:
diff --git a/llvm/tools/llvm-profgen/PseudoProbe.cpp b/llvm/tools/llvm-profgen/PseudoProbe.cpp
--- a/llvm/tools/llvm-profgen/PseudoProbe.cpp
+++ b/llvm/tools/llvm-profgen/PseudoProbe.cpp
@@ -198,7 +198,6 @@
   //         A list of NUM_INLINED_FUNCTIONS entries describing each of the
   //         inlined callees.  Each record contains:
   //           INLINE SITE
-  //             GUID of the inlinee (uint64)
   //             Index of the callsite probe (ULEB128)
   //           FUNCTION BODY
   //             A FUNCTION BODY entry describing the inlined function.
@@ -214,8 +213,11 @@
   uint32_t Index = 0;
   // A DFS-based decoding
   while (Data < End) {
-    // Read inline site for inlinees
-    if (Root != Cur) {
+    if (Root == Cur) {
+      // Use a sequential id for outline functions.
+      Index = Root->getChildren().size();
+    } else {
+      // Read inline site for inlinees
       Index = readUnsignedNumber<uint32_t>();
     }
     // Switch/add to a new tree node(inlinee)
@@ -243,10 +245,10 @@
         Addr = readUnencodedNumber<int64_t>();
       }
       // Populate Address2ProbesMap
-      std::vector<PseudoProbe> &ProbeVec = Address2ProbesMap[Addr];
-      ProbeVec.emplace_back(Addr, Cur->GUID, Index, PseudoProbeType(Kind), Attr,
-                            Cur);
-      Cur->addProbes(&ProbeVec.back());
+      auto &Probes = Address2ProbesMap[Addr];
+      Probes.emplace_back(Addr, Cur->GUID, Index, PseudoProbeType(Kind), Attr,
+                          Cur);
+      Cur->addProbes(&Probes.back());
       LastAddr = Addr;
     }
 
@@ -298,7 +300,7 @@
   auto It = Address2ProbesMap.find(Address);
   if (It == Address2ProbesMap.end())
     return nullptr;
-  const std::vector<PseudoProbe> &Probes = It->second;
+  const auto &Probes = It->second;
 
   const PseudoProbe *CallProbe = nullptr;
   for (const auto &Probe : Probes) {