1 files changed, 305 insertions, 35 deletions

diff --git a/gnu/llvm/lib/Fuzzer/FuzzerTracePC.cpp b/gnu/llvm/lib/Fuzzer/FuzzerTracePC.cpp
index 46c43d0c17f..39d6e602621 100644
--- a/gnu/llvm/lib/Fuzzer/FuzzerTracePC.cpp
+++ b/gnu/llvm/lib/Fuzzer/FuzzerTracePC.cpp
@@ -7,65 +7,335 @@
 //
 //===----------------------------------------------------------------------===//
 // Trace PCs.
-// This module implements __sanitizer_cov_trace_pc, a callback required
-// for -fsanitize-coverage=trace-pc instrumentation.
+// This module implements __sanitizer_cov_trace_pc_guard[_init],
+// the callback required for -fsanitize-coverage=trace-pc-guard instrumentation.
 //
 //===----------------------------------------------------------------------===//
 
-#include "FuzzerInternal.h"
+#include "FuzzerCorpus.h"
+#include "FuzzerDefs.h"
+#include "FuzzerDictionary.h"
+#include "FuzzerExtFunctions.h"
+#include "FuzzerIO.h"
+#include "FuzzerTracePC.h"
+#include "FuzzerValueBitMap.h"
+#include <map>
+#include <sanitizer/coverage_interface.h>
+#include <set>
+#include <sstream>
 
 namespace fuzzer {
 
-void PcCoverageMap::Reset() { memset(Map, 0, sizeof(Map)); }
+TracePC TPC;
 
-void PcCoverageMap::Update(uintptr_t Addr) {
-  uintptr_t Idx = Addr % kMapSizeInBits;
-  uintptr_t WordIdx = Idx / kBitsInWord;
-  uintptr_t BitIdx = Idx % kBitsInWord;
-  Map[WordIdx] |= 1UL << BitIdx;
+void TracePC::HandleTrace(uint32_t *Guard, uintptr_t PC) {
+  uint32_t Idx = *Guard;
+  if (!Idx) return;
+  PCs[Idx % kNumPCs] = PC;
+  Counters[Idx % kNumCounters]++;
 }
 
-size_t PcCoverageMap::MergeFrom(const PcCoverageMap &Other) {
-  uintptr_t Res = 0;
-  for (size_t i = 0; i < kMapSizeInWords; i++)
-    Res += __builtin_popcountl(Map[i] |= Other.Map[i]);
+size_t TracePC::GetTotalPCCoverage() {
+  size_t Res = 0;
+  for (size_t i = 1; i < GetNumPCs(); i++)
+    if (PCs[i])
+      Res++;
   return Res;
 }
 
-static PcCoverageMap CurrentMap;
-static thread_local uintptr_t Prev;
+void TracePC::HandleInit(uint32_t *Start, uint32_t *Stop) {
+  if (Start == Stop || *Start) return;
+  assert(NumModules < sizeof(Modules) / sizeof(Modules[0]));
+  for (uint32_t *P = Start; P < Stop; P++)
+    *P = ++NumGuards;
+  Modules[NumModules].Start = Start;
+  Modules[NumModules].Stop = Stop;
+  NumModules++;
+}
+
+void TracePC::PrintModuleInfo() {
+  Printf("INFO: Loaded %zd modules (%zd guards): ", NumModules, NumGuards);
+  for (size_t i = 0; i < NumModules; i++)
+    Printf("[%p, %p), ", Modules[i].Start, Modules[i].Stop);
+  Printf("\n");
+}
+
+void TracePC::HandleCallerCallee(uintptr_t Caller, uintptr_t Callee) {
+  const uintptr_t kBits = 12;
+  const uintptr_t kMask = (1 << kBits) - 1;
+  uintptr_t Idx = (Caller & kMask) | ((Callee & kMask) << kBits);
+  HandleValueProfile(Idx);
+}
 
-void PcMapResetCurrent() {
-  if (Prev) {
-    Prev = 0;
-    CurrentMap.Reset();
+static bool IsInterestingCoverageFile(std::string &File) {
+  if (File.find("compiler-rt/lib/") != std::string::npos)
+    return false; // sanitizer internal.
+  if (File.find("/usr/lib/") != std::string::npos)
+    return false;
+  if (File.find("/usr/include/") != std::string::npos)
+    return false;
+  if (File == "<null>")
+    return false;
+  return true;
+}
+
+void TracePC::InitializePrintNewPCs() {
+  if (!DoPrintNewPCs) return;
+  assert(!PrintedPCs);
+  PrintedPCs = new std::set<uintptr_t>;
+  for (size_t i = 1; i < GetNumPCs(); i++)
+    if (PCs[i])
+      PrintedPCs->insert(PCs[i]);
+}
+
+void TracePC::PrintNewPCs() {
+  if (!DoPrintNewPCs) return;
+  assert(PrintedPCs);
+  for (size_t i = 1; i < GetNumPCs(); i++)
+    if (PCs[i] && PrintedPCs->insert(PCs[i]).second)
+      PrintPC("\tNEW_PC: %p %F %L\n", "\tNEW_PC: %p\n", PCs[i]);
+}
+
+void TracePC::PrintCoverage() {
+  if (!EF->__sanitizer_symbolize_pc ||
+      !EF->__sanitizer_get_module_and_offset_for_pc) {
+    Printf("INFO: __sanitizer_symbolize_pc or "
+           "__sanitizer_get_module_and_offset_for_pc is not available,"
+           " not printing coverage\n");
+    return;
+  }
+  std::map<std::string, std::vector<uintptr_t>> CoveredPCsPerModule;
+  std::map<std::string, uintptr_t> ModuleOffsets;
+  std::set<std::string> CoveredDirs, CoveredFiles, CoveredFunctions,
+      CoveredLines;
+  Printf("COVERAGE:\n");
+  for (size_t i = 1; i < GetNumPCs(); i++) {
+    if (!PCs[i]) continue;
+    std::string FileStr = DescribePC("%s", PCs[i]);
+    if (!IsInterestingCoverageFile(FileStr)) continue;
+    std::string FixedPCStr = DescribePC("%p", PCs[i]);
+    std::string FunctionStr = DescribePC("%F", PCs[i]);
+    std::string LineStr = DescribePC("%l", PCs[i]);
+    char ModulePathRaw[4096] = "";  // What's PATH_MAX in portable C++?
+    void *OffsetRaw = nullptr;
+    if (!EF->__sanitizer_get_module_and_offset_for_pc(
+            reinterpret_cast<void *>(PCs[i]), ModulePathRaw,
+            sizeof(ModulePathRaw), &OffsetRaw))
+      continue;
+    std::string Module = ModulePathRaw;
+    uintptr_t FixedPC = std::stol(FixedPCStr, 0, 16);
+    uintptr_t PcOffset = reinterpret_cast<uintptr_t>(OffsetRaw);
+    ModuleOffsets[Module] = FixedPC - PcOffset;
+    CoveredPCsPerModule[Module].push_back(PcOffset);
+    CoveredFunctions.insert(FunctionStr);
+    CoveredFiles.insert(FileStr);
+    CoveredDirs.insert(DirName(FileStr));
+    if (!CoveredLines.insert(FileStr + ":" + LineStr).second)
+      continue;
+    Printf("COVERED: %s %s:%s\n", FunctionStr.c_str(),
+           FileStr.c_str(), LineStr.c_str());
+  }
+
+  std::string CoveredDirsStr;
+  for (auto &Dir : CoveredDirs) {
+    if (!CoveredDirsStr.empty())
+      CoveredDirsStr += ",";
+    CoveredDirsStr += Dir;
+  }
+  Printf("COVERED_DIRS: %s\n", CoveredDirsStr.c_str());
+
+  for (auto &M : CoveredPCsPerModule) {
+    std::set<std::string> UncoveredFiles, UncoveredFunctions;
+    std::map<std::string, std::set<int> > UncoveredLines;  // Func+File => lines
+    auto &ModuleName = M.first;
+    auto &CoveredOffsets = M.second;
+    uintptr_t ModuleOffset = ModuleOffsets[ModuleName];
+    std::sort(CoveredOffsets.begin(), CoveredOffsets.end());
+    Printf("MODULE_WITH_COVERAGE: %s\n", ModuleName.c_str());
+    // sancov does not yet fully support DSOs.
+    // std::string Cmd = "sancov -print-coverage-pcs " + ModuleName;
+    std::string Cmd = "objdump -d " + ModuleName +
+        " | grep 'call.*__sanitizer_cov_trace_pc_guard' | awk -F: '{print $1}'";
+    std::string SanCovOutput;
+    if (!ExecuteCommandAndReadOutput(Cmd, &SanCovOutput)) {
+      Printf("INFO: Command failed: %s\n", Cmd.c_str());
+      continue;
+    }
+    std::istringstream ISS(SanCovOutput);
+    std::string S;
+    while (std::getline(ISS, S, '\n')) {
+      uintptr_t PcOffset = std::stol(S, 0, 16);
+      if (!std::binary_search(CoveredOffsets.begin(), CoveredOffsets.end(),
+                              PcOffset)) {
+        uintptr_t PC = ModuleOffset + PcOffset;
+        auto FileStr = DescribePC("%s", PC);
+        if (!IsInterestingCoverageFile(FileStr)) continue;
+        if (CoveredFiles.count(FileStr) == 0) {
+          UncoveredFiles.insert(FileStr);
+          continue;
+        }
+        auto FunctionStr = DescribePC("%F", PC);
+        if (CoveredFunctions.count(FunctionStr) == 0) {
+          UncoveredFunctions.insert(FunctionStr);
+          continue;
+        }
+        std::string LineStr = DescribePC("%l", PC);
+        uintptr_t Line = std::stoi(LineStr);
+        std::string FileLineStr = FileStr + ":" + LineStr;
+        if (CoveredLines.count(FileLineStr) == 0)
+          UncoveredLines[FunctionStr + " " + FileStr].insert(Line);
+      }
+    }
+    for (auto &FileLine: UncoveredLines)
+      for (int Line : FileLine.second)
+        Printf("UNCOVERED_LINE: %s:%d\n", FileLine.first.c_str(), Line);
+    for (auto &Func : UncoveredFunctions)
+      Printf("UNCOVERED_FUNC: %s\n", Func.c_str());
+    for (auto &File : UncoveredFiles)
+      Printf("UNCOVERED_FILE: %s\n", File.c_str());
   }
 }
 
-size_t PcMapMergeInto(PcCoverageMap *Map) {
-  if (!Prev)
-    return 0;
-  return Map->MergeFrom(CurrentMap);
+void TracePC::DumpCoverage() {
+  __sanitizer_dump_coverage(PCs, GetNumPCs());
 }
 
-static void HandlePC(uint32_t PC) {
-  // We take 12 bits of PC and mix it with the previous PCs.
-  uintptr_t Next = (Prev << 5) ^ (PC & 4095);
-  CurrentMap.Update(Next);
-  Prev = Next;
+// Value profile.
+// We keep track of various values that affect control flow.
+// These values are inserted into a bit-set-based hash map.
+// Every new bit in the map is treated as a new coverage.
+//
+// For memcmp/strcmp/etc the interesting value is the length of the common
+// prefix of the parameters.
+// For cmp instructions the interesting value is a XOR of the parameters.
+// The interesting value is mixed up with the PC and is then added to the map.
+
+ATTRIBUTE_NO_SANITIZE_MEMORY
+void TracePC::AddValueForMemcmp(void *caller_pc, const void *s1, const void *s2,
+                              size_t n) {
+  if (!n) return;
+  size_t Len = std::min(n, (size_t)32);
+  const uint8_t *A1 = reinterpret_cast<const uint8_t *>(s1);
+  const uint8_t *A2 = reinterpret_cast<const uint8_t *>(s2);
+  size_t I = 0;
+  for (; I < Len; I++)
+    if (A1[I] != A2[I])
+      break;
+  size_t PC = reinterpret_cast<size_t>(caller_pc);
+  size_t Idx = I;
+  // if (I < Len)
+  //  Idx += __builtin_popcountl((A1[I] ^ A2[I])) - 1;
+  TPC.HandleValueProfile((PC & 4095) | (Idx << 12));
+}
+
+ATTRIBUTE_NO_SANITIZE_MEMORY
+void TracePC::AddValueForStrcmp(void *caller_pc, const char *s1, const char *s2,
+                              size_t n) {
+  if (!n) return;
+  size_t Len = std::min(n, (size_t)32);
+  const uint8_t *A1 = reinterpret_cast<const uint8_t *>(s1);
+  const uint8_t *A2 = reinterpret_cast<const uint8_t *>(s2);
+  size_t I = 0;
+  for (; I < Len; I++)
+    if (A1[I] != A2[I] || A1[I] == 0)
+      break;
+  size_t PC = reinterpret_cast<size_t>(caller_pc);
+  size_t Idx = I;
+  // if (I < Len && A1[I])
+  //  Idx += __builtin_popcountl((A1[I] ^ A2[I])) - 1;
+  TPC.HandleValueProfile((PC & 4095) | (Idx << 12));
+}
+
+template <class T>
+ATTRIBUTE_TARGET_POPCNT
+#ifdef __clang__  // g++ can't handle this __attribute__ here :(
+__attribute__((always_inline))
+#endif  // __clang__
+void TracePC::HandleCmp(void *PC, T Arg1, T Arg2) {
+  uintptr_t PCuint = reinterpret_cast<uintptr_t>(PC);
+  uint64_t ArgXor = Arg1 ^ Arg2;
+  uint64_t ArgDistance = __builtin_popcountl(ArgXor) + 1; // [1,65]
+  uintptr_t Idx = ((PCuint & 4095) + 1) * ArgDistance;
+  if (sizeof(T) == 4)
+      TORC4.Insert(ArgXor, Arg1, Arg2);
+  else if (sizeof(T) == 8)
+      TORC8.Insert(ArgXor, Arg1, Arg2);
+  HandleValueProfile(Idx);
 }
 
 } // namespace fuzzer
 
 extern "C" {
-void __sanitizer_cov_trace_pc() {
-  fuzzer::HandlePC(static_cast<uint32_t>(
-      reinterpret_cast<uintptr_t>(__builtin_return_address(0))));
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_pc_guard(uint32_t *Guard) {
+  uintptr_t PC = (uintptr_t)__builtin_return_address(0);
+  fuzzer::TPC.HandleTrace(Guard, PC);
+}
+
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_pc_guard_init(uint32_t *Start, uint32_t *Stop) {
+  fuzzer::TPC.HandleInit(Start, Stop);
+}
+
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_pc_indir(uintptr_t Callee) {
+  uintptr_t PC = (uintptr_t)__builtin_return_address(0);
+  fuzzer::TPC.HandleCallerCallee(PC, Callee);
+}
+
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_cmp8(uint64_t Arg1, uint64_t Arg2) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Arg1, Arg2);
 }
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_cmp4(uint32_t Arg1, uint32_t Arg2) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Arg1, Arg2);
+}
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_cmp2(uint16_t Arg1, uint16_t Arg2) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Arg1, Arg2);
+}
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_cmp1(uint8_t Arg1, uint8_t Arg2) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Arg1, Arg2);
+}
+
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_switch(uint64_t Val, uint64_t *Cases) {
+  uint64_t N = Cases[0];
+  uint64_t ValSizeInBits = Cases[1];
+  uint64_t *Vals = Cases + 2;
+  // Skip the most common and the most boring case.
+  if (Vals[N - 1]  < 256 && Val < 256)
+    return;
+  char *PC = (char*)__builtin_return_address(0);
+  size_t i;
+  uint64_t Token = 0;
+  for (i = 0; i < N; i++) {
+    Token = Val ^ Vals[i];
+    if (Val < Vals[i])
+      break;
+  }
 
-void __sanitizer_cov_trace_pc_indir(int *) {
-  // Stub to allow linking with code built with
-  // -fsanitize=indirect-calls,trace-pc.
-  // This isn't used currently.
+  if (ValSizeInBits == 16)
+    fuzzer::TPC.HandleCmp(PC + i, static_cast<uint16_t>(Token), (uint16_t)(0));
+  else if (ValSizeInBits == 32)
+    fuzzer::TPC.HandleCmp(PC + i, static_cast<uint32_t>(Token), (uint32_t)(0));
+  else
+    fuzzer::TPC.HandleCmp(PC + i, Token, (uint64_t)(0));
 }
+
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_div4(uint32_t Val) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Val, (uint32_t)0);
+}
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_div8(uint64_t Val) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Val, (uint64_t)0);
 }
+__attribute__((visibility("default")))
+void __sanitizer_cov_trace_gep(uintptr_t Idx) {
+  fuzzer::TPC.HandleCmp(__builtin_return_address(0), Idx, (uintptr_t)0);
+}
+
+}  // extern "C"