[llvm-profgen] Loading binary functions from .symtab when DWARF info is incomplete

llvm/include/llvm/ProfileData/SampleProf.h

@@ -1214,13 +1214,19 @@ class FunctionSamples {
     // Note the sequence of the suffixes in the knownSuffixes array matters.
     // If suffix "A" is appended after the suffix "B", "A" should be in front
     // of "B" in knownSuffixes.
-    const char *KnownSuffixes[] = {LLVMSuffix, PartSuffix, UniqSuffix};
+    const SmallVector<StringRef> KnownSuffixes{LLVMSuffix, PartSuffix,
+                                               UniqSuffix};
+    return getCanonicalFnName(FnName, KnownSuffixes, Attr);
+  }
+
+  static StringRef getCanonicalFnName(StringRef FnName,
+                                      ArrayRef<StringRef> Suffixes,
+                                      StringRef Attr = "selected") {
     if (Attr == "" || Attr == "all")
       return FnName.split('.').first;
     if (Attr == "selected") {
       StringRef Cand(FnName);
-      for (const auto &Suf : KnownSuffixes) {
-        StringRef Suffix(Suf);
+      for (const auto Suffix : Suffixes) {
         // If the profile contains ".__uniq." suffix, don't strip the
         // suffix for names in the IR.
         if (Suffix == UniqSuffix && FunctionSamples::HasUniqSuffix)

llvm/test/tools/llvm-profgen/missing-dwarf.test

@@ -0,0 +1,39 @@
+; RUN: rm -rf %t
+; RUN: mkdir -p %t
+; RUN: cd %t
+
+; RUN: echo -e "1\n401120-40113b:1\n1\n40112f->401110:1" > %t.prof
+
+; Test --load-function-from-symbol=0
+; RUN: llvm-profgen --format=text --unsymbolized-profile=%t.prof --binary=%S/Inputs/missing-dwarf.exe --output=%t1 --fill-zero-for-all-funcs --show-detailed-warning --use-offset=0 --load-function-from-symbol=0 2>&1 | FileCheck %s --check-prefix=CHECK-NO-LOAD-SYMTAB
+
+; CHECK-NO-LOAD-SYMTAB:      warning: 100.00%(1/1) of function range samples do not belong to any function
+; CHECK-NO-LOAD-SYMTAB-NEXT: warning: 100.00%(1/1) of LBR source samples do not belong to any function
+; CHECK-NO-LOAD-SYMTAB-NEXT: warning: 100.00%(1/1) of LBR target samples do not belong to any function
+
+; Test --load-function-from-symbol=1
+; RUN: llvm-profgen --format=text --unsymbolized-profile=%t.prof --binary=%S/Inputs/missing-dwarf.exe --output=%t2 --fill-zero-for-all-funcs --show-detailed-warning --use-offset=0 --load-function-from-symbol=1
+; RUN: FileCheck %s --input-file %t2 --check-prefix=CHECK-LOAD-SYMTAB
+
+; CHECK-LOAD-SYMTAB:      main:2:1
+; CHECK-LOAD-SYMTAB-NEXT:  1: 1
+; CHECK-LOAD-SYMTAB-NEXT:  2: 1 foo:1
+; CHECK-LOAD-SYMTAB-NEXT:  !CFGChecksum: 281479271677951
+; CHECK-LOAD-SYMTAB-NEXT: foo:0:0
+; CHECK-LOAD-SYMTAB-NEXT:  1: 0
+; CHECK-LOAD-SYMTAB-NEXT:  !CFGChecksum: 4294967295
+
+; Build instructions:
+; missing-dwarf.o:       clang -gsplit-dwarf=split -fdebug-compilation-dir=.  test.c   -fdebug-info-for-profiling  -fpseudo-probe-for-profiling  -O0 -g -o missing-dwarf.o -c
+; missing-dwarf.exe:     clang -fdebug-compilation-dir=.  missing-dwarf.o -o missing-dwarf.exe  -fdebug-info-for-profiling  -fpseudo-probe-for-profiling  -O0 -g
+
+; Source code:
+
+int foo() {
+  return 1;
+}
+
+int main() {
+  foo();
+  return 0;
+}

llvm/tools/llvm-profgen/PerfReader.cpp

@@ -1274,16 +1274,13 @@ void PerfScriptReader::warnInvalidRange() {
 
   const char *EndNotBoundaryMsg = "Range is not on instruction boundary, "
                                   "likely due to profile and binary mismatch.";
-  const char *DanglingRangeMsg = "Range does not belong to any functions, "
-                                 "likely from PLT, .init or .fini section.";
   const char *RangeCrossFuncMsg =
       "Fall through range should not cross function boundaries, likely due to "
       "profile and binary mismatch.";
   const char *BogusRangeMsg = "Range start is after or too far from range end.";
 
   uint64_t TotalRangeNum = 0;
   uint64_t InstNotBoundary = 0;
-  uint64_t UnmatchedRange = 0;
   uint64_t RangeCrossFunc = 0;
   uint64_t BogusRange = 0;
 
@@ -1303,11 +1300,8 @@ void PerfScriptReader::warnInvalidRange() {
     }
 
     auto *FRange = Binary->findFuncRange(StartAddress);
-    if (!FRange) {
-      UnmatchedRange += I.second;
-      WarnInvalidRange(StartAddress, EndAddress, DanglingRangeMsg);
+    if (!FRange)
       continue;
-    }
 
     if (EndAddress >= FRange->EndAddress) {
       RangeCrossFunc += I.second;
@@ -1325,9 +1319,6 @@ void PerfScriptReader::warnInvalidRange() {
   emitWarningSummary(
       InstNotBoundary, TotalRangeNum,
       "of samples are from ranges that are not on instruction boundary.");
-  emitWarningSummary(
-      UnmatchedRange, TotalRangeNum,
-      "of samples are from ranges that do not belong to any functions.");
   emitWarningSummary(
       RangeCrossFunc, TotalRangeNum,
       "of samples are from ranges that do cross function boundaries.");

llvm/tools/llvm-profgen/ProfileGenerator.cpp

@@ -449,29 +449,62 @@ bool ProfileGeneratorBase::collectFunctionsFromRawProfile(
   // Go through all the stacks, ranges and branches in sample counters, use
   // the start of the range to look up the function it belongs and record the
   // function.
+  uint64_t ErrStkAddr = 0, ErrFuncRange = 0, ErrSrc = 0, ErrTgt = 0;
+  uint64_t TotalStkAddr = 0, TotalFuncRange = 0, TotalSrc = 0, TotalTgt = 0;
   for (const auto &CI : *SampleCounters) {
     if (const auto *CtxKey = dyn_cast<AddrBasedCtxKey>(CI.first.getPtr())) {
       for (auto StackAddr : CtxKey->Context) {
+        uint64_t Inc = Binary->addressIsCode(StackAddr) ? 1 : 0;
+        TotalStkAddr += Inc;
         if (FuncRange *FRange = Binary->findFuncRange(StackAddr))
           ProfiledFunctions.insert(FRange->Func);
+        else
+          ErrStkAddr += Inc;
       }
     }
 
     for (auto Item : CI.second.RangeCounter) {
       uint64_t StartAddress = Item.first.first;
+      uint64_t Inc = Binary->addressIsCode(StartAddress) ? Item.second : 0;
+      TotalFuncRange += Inc;
       if (FuncRange *FRange = Binary->findFuncRange(StartAddress))
         ProfiledFunctions.insert(FRange->Func);
+      else
+        ErrFuncRange += Inc;
     }
 
     for (auto Item : CI.second.BranchCounter) {
       uint64_t SourceAddress = Item.first.first;
       uint64_t TargetAddress = Item.first.second;
+      uint64_t SrcInc = Binary->addressIsCode(SourceAddress) ? Item.second : 0;
+      uint64_t TgtInc = Binary->addressIsCode(TargetAddress) ? Item.second : 0;
+      TotalSrc += SrcInc;
       if (FuncRange *FRange = Binary->findFuncRange(SourceAddress))
         ProfiledFunctions.insert(FRange->Func);
+      else
+        ErrSrc += SrcInc;
+      TotalTgt += TgtInc;
       if (FuncRange *FRange = Binary->findFuncRange(TargetAddress))
         ProfiledFunctions.insert(FRange->Func);
+      else
+        ErrTgt += TgtInc;
     }
   }
+
+  if (ErrStkAddr)
+    emitWarningSummary(
+        ErrStkAddr, TotalStkAddr,
+        "of stack address samples do not belong to any function");
+  if (ErrFuncRange)
+    emitWarningSummary(
+        ErrFuncRange, TotalFuncRange,
+        "of function range samples do not belong to any function");
+  if (ErrSrc)
+    emitWarningSummary(ErrSrc, TotalSrc,
+                       "of LBR source samples do not belong to any function");
+  if (ErrTgt)
+    emitWarningSummary(ErrTgt, TotalTgt,
+                       "of LBR target samples do not belong to any function");
   return true;
 }
 

llvm/tools/llvm-profgen/ProfiledBinary.cpp

@@ -65,6 +65,13 @@ static cl::list<std::string> DisassembleFunctions(
              "names only. Only work with show-disassembly-only"),
     cl::cat(ProfGenCategory));
 
+static cl::opt<bool>
+    LoadFunctionFromSymbol("load-function-from-symbol", cl::init(true),
+                           cl::desc("Gather additional binary function info "
+                                    "from symbols (e.g. .symtab) in case "
+                                    "dwarf info is incomplete."),
+                           cl::cat(ProfGenCategory));
+
 static cl::opt<bool>
     KernelBinary("kernel",
                  cl::desc("Generate the profile for Linux kernel binary."),
@@ -257,6 +264,9 @@ void ProfiledBinary::load() {
   if (ShowDisassemblyOnly)
     decodePseudoProbe(Obj);
 
+  if (LoadFunctionFromSymbol && UsePseudoProbes)
+    populateSymbolsFromBinary(Obj);
+
   // Disassemble the text sections.
   disassemble(Obj);
 
@@ -604,13 +614,13 @@ bool ProfiledBinary::dissassembleSymbol(std::size_t SI, ArrayRef<uint8_t> Bytes,
       // Record potential call targets for tail frame inference later-on.
       if (InferMissingFrames && FRange) {
         uint64_t Target = 0;
-        MIA->evaluateBranch(Inst, Address, Size, Target);
+        bool Err = MIA->evaluateBranch(Inst, Address, Size, Target);
         if (MCDesc.isCall()) {
           // Indirect call targets are unknown at this point. Recording the
           // unknown target (zero) for further LBR-based refinement.
           MissingContextInferrer->CallEdges[Address].insert(Target);
         } else if (MCDesc.isUnconditionalBranch()) {
-          assert(Target &&
+          assert(Err &&
                  "target should be known for unconditional direct branch");
           // Any inter-function unconditional jump is considered tail call at
           // this point. This is not 100% accurate and could further be
@@ -820,6 +830,54 @@ void ProfiledBinary::populateSymbolAddressList(const ObjectFile *Obj) {
   }
 }
 
+void ProfiledBinary::populateSymbolsFromBinary(const ObjectFile *Obj) {
+  // Load binary functions from symbol table when Debug info is incomplete
+  const SmallVector<StringRef> Suffixes(
+      {".destroy", ".resume", ".llvm.", ".cold", ".warm"});
+  StringRef FileName = Obj->getFileName();
+  for (const SymbolRef &Symbol : Obj->symbols()) {
+    const SymbolRef::Type Type = unwrapOrError(Symbol.getType(), FileName);
+    const uint64_t Addr = unwrapOrError(Symbol.getAddress(), FileName);
+    const StringRef Name = unwrapOrError(Symbol.getName(), FileName);
+    uint64_t Size = 0;
+    if (isa<ELFObjectFileBase>(Symbol.getObject())) {
+      ELFSymbolRef ElfSymbol(Symbol);
+      Size = ElfSymbol.getSize();
+    }
+
+    if (Size == 0 || Type != SymbolRef::ST_Function)
+      continue;
+
+    const StringRef SymName =
+        FunctionSamples::getCanonicalFnName(Name, Suffixes);
+
+    auto Ret = BinaryFunctions.emplace(SymName, BinaryFunction());
+    auto &Func = Ret.first->second;
+    if (Ret.second) {
+      Func.FuncName = Ret.first->first;
+      HashBinaryFunctions[MD5Hash(StringRef(SymName))] = &Func;
+    }
+
+    if (auto Range = findFuncRange(Addr)) {
+      if (Ret.second && ShowDetailedWarning)
+        WithColor::warning()
+            << "Symbol " << Name << " start address "
+            << format("%8" PRIx64, Addr) << " already exists in DWARF at "
+            << format("%8" PRIx64, Range->StartAddress) << " in function "
+            << Range->getFuncName() << "\n";
+    } else {
+      // Store/Update Function Range from SymTab
+      Func.Ranges.emplace_back(Addr, Addr + Size);
+
+      auto R = StartAddrToFuncRangeMap.emplace(Addr, FuncRange());
+      FuncRange &FRange = R.first->second;
+      FRange.Func = &Func;
+      FRange.StartAddress = Addr;
+      FRange.EndAddress = Addr + Size;
+    }
+  }
+}
+
 void ProfiledBinary::loadSymbolsFromDWARFUnit(DWARFUnit &CompilationUnit) {
   for (const auto &DieInfo : CompilationUnit.dies()) {
     llvm::DWARFDie Die(&CompilationUnit, &DieInfo);

llvm/tools/llvm-profgen/ProfiledBinary.h

@@ -356,6 +356,9 @@ class ProfiledBinary {
   // Create symbol to its start address mapping.
   void populateSymbolAddressList(const object::ObjectFile *O);
 
+  // Load functions from its symbol table (when DWARF info is missing).
+  void populateSymbolsFromBinary(const object::ObjectFile *O);
+
   // A function may be spilt into multiple non-continuous address ranges. We use
   // this to set whether start a function range is the real entry of the
   // function and also set false to the non-function label.