add libfuzzer custom mutator, minor enhancements and fixes

GNUmakefile

@@ -465,9 +465,9 @@ code-format:
 	./.custom-format.py -i instrumentation/*.h
 	./.custom-format.py -i instrumentation/*.cc
 	./.custom-format.py -i instrumentation/*.c
-	./.custom-format.py -i custom_mutators/*/*.c
+	./.custom-format.py -i custom_mutators/*/*.c*
 	@#./.custom-format.py -i custom_mutators/*/*.h # destroys input.h :-(
-	./.custom-format.py -i examples/*/*.c
+	./.custom-format.py -i examples/*/*.c*
 	./.custom-format.py -i examples/*/*.h
 	./.custom-format.py -i test/*.c
 	./.custom-format.py -i qemu_mode/libcompcov/*.c

README.md

@@ -379,6 +379,11 @@ How to do this is described below.
 
 Then build the target. (Usually with `make`)
 
+**NOTE**: sometimes configure and build systems are fickle and do not like
+stderr output (and think this means a test failure) - which is something
+afl++ like to do to show statistics. It is recommended to disable them via
+`export AFL_QUIET=1`.
+
 ##### configure
 
 For `configure` build systems this is usually done by:

custom_mutators/README.md

@@ -12,9 +12,7 @@ git submodule init
 git submodule update
 ```
 
-otherwise just checkout the repository here with either
-`git clone https://github.com/AFLplusplus/Grammar-Mutator` or
-`svn co https://github.com/AFLplusplus/Grammar-Mutator`.
+otherwise just use the script: `grammar_mutator/build_grammar_mutator.sh`
 
 Read the [Grammar-Mutator/README.md](Grammar-Mutator/README.md) on how to use
 it.

custom_mutators/grammar_mutator/README.md

@@ -0,0 +1,6 @@
+# Grammar-Mutator
+
+This is just a stub directory that will clone the real grammar mutator
+directory.
+
+Execute `./build_grammar_mutator.sh` to set everything up.

custom_mutators/grammar_mutator/build_grammar_mutator.sh

@@ -0,0 +1,17 @@
+#!/bin/sh
+
+test -d Grammar-Mutator || git clone --depth=1 https://github.com/AFLplusplus/Grammar-Mutator
+
+cd Grammar-Mutator || exit 1
+git stash ; git pull
+
+wget -c https://www.antlr.org/download/antlr-4.8-complete.jar
+
+echo
+echo
+echo "All successfully prepared!"
+echo "To build for your grammar just do:"
+echo "  cd Grammar_Mutator"
+echo "  make GRAMMAR_FILE=/path/to/your/grammar"
+echo "You will find a JSON and RUBY grammar in Grammar_Mutator/grammars to play with."
+echo

custom_mutators/honggfuzz/Makefile

@@ -1,10 +1,10 @@
 
 CFLAGS = -O3 -funroll-loops -fPIC -Wl,-Bsymbolic
 
-all: honggfuzz.so
+all: honggfuzz-mutator.so
 
-honggfuzz.so:	honggfuzz.c input.h mangle.c ../../src/afl-performance.c
-	$(CC) $(CFLAGS) -I../../include -I. -shared -o honggfuzz.so honggfuzz.c mangle.c ../../src/afl-performance.c
+honggfuzz-mutator.so:	honggfuzz.c input.h mangle.c ../../src/afl-performance.c
+	$(CC) $(CFLAGS) -I../../include -I. -shared -o honggfuzz-mutator.so honggfuzz.c mangle.c ../../src/afl-performance.c
 
 update:
 	@# seriously? --unlink is a dud option? sigh ...

custom_mutators/honggfuzz/README.md

@@ -1,12 +1,12 @@
 # custum mutator: honggfuzz mangle
 
-this is the very good honggfuzz mutator in mangle.c as a custom mutator
+this is the honggfuzz mutator in mangle.c as a custom mutator
 module for afl++. It is the original mangle.c, mangle.h and honggfuzz.h
 with a lot of mocking around it :-)
 
 just type `make` to build
 
-```AFL_CUSTOM_MUTATOR_LIBRARY=custom_mutators/honggfuzz/honggfuzz.so afl-fuzz ...```
+```AFL_CUSTOM_MUTATOR_LIBRARY=custom_mutators/honggfuzz/honggfuzz-mutator.so afl-fuzz ...```
 
 > Original repository: https://github.com/google/honggfuzz
 > Source commit: d0fbcb0373c32436b8fb922e6937da93b17291f5

custom_mutators/libfuzzer/FuzzerBuiltins.h

@@ -0,0 +1,35 @@
+//===- FuzzerBuiltins.h - Internal header for builtins ----------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Wrapper functions and marcos around builtin functions.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_BUILTINS_H
+#define LLVM_FUZZER_BUILTINS_H
+
+#include "FuzzerPlatform.h"
+
+#if !LIBFUZZER_MSVC
+#include <cstdint>
+
+#define GET_CALLER_PC() __builtin_return_address(0)
+
+namespace fuzzer {
+
+inline uint8_t  Bswap(uint8_t x)  { return x; }
+inline uint16_t Bswap(uint16_t x) { return __builtin_bswap16(x); }
+inline uint32_t Bswap(uint32_t x) { return __builtin_bswap32(x); }
+inline uint64_t Bswap(uint64_t x) { return __builtin_bswap64(x); }
+
+inline uint32_t Clzll(unsigned long long X) { return __builtin_clzll(X); }
+inline uint32_t Clz(unsigned long long X) { return __builtin_clz(X); }
+inline int Popcountll(unsigned long long X) { return __builtin_popcountll(X); }
+
+}  // namespace fuzzer
+
+#endif  // !LIBFUZZER_MSVC
+#endif  // LLVM_FUZZER_BUILTINS_H

custom_mutators/libfuzzer/FuzzerBuiltinsMsvc.h

@@ -0,0 +1,72 @@
+//===- FuzzerBuiltinsMSVC.h - Internal header for builtins ------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Wrapper functions and marcos that use intrinsics instead of builtin functions
+// which cannot be compiled by MSVC.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_BUILTINS_MSVC_H
+#define LLVM_FUZZER_BUILTINS_MSVC_H
+
+#include "FuzzerPlatform.h"
+
+#if LIBFUZZER_MSVC
+#include <intrin.h>
+#include <cstdint>
+#include <cstdlib>
+
+// __builtin_return_address() cannot be compiled with MSVC. Use the equivalent
+// from <intrin.h>
+#define GET_CALLER_PC() _ReturnAddress()
+
+namespace fuzzer {
+
+inline uint8_t  Bswap(uint8_t x)  { return x; }
+// Use alternatives to __builtin functions from <stdlib.h> and <intrin.h> on
+// Windows since the builtins are not supported by MSVC.
+inline uint16_t Bswap(uint16_t x) { return _byteswap_ushort(x); }
+inline uint32_t Bswap(uint32_t x) { return _byteswap_ulong(x); }
+inline uint64_t Bswap(uint64_t x) { return _byteswap_uint64(x); }
+
+// The functions below were mostly copied from
+// compiler-rt/lib/builtins/int_lib.h which defines the __builtin functions used
+// outside of Windows.
+inline uint32_t Clzll(uint64_t X) {
+  unsigned long LeadZeroIdx = 0;
+
+#if !defined(_M_ARM) && !defined(_M_X64)
+  // Scan the high 32 bits.
+  if (_BitScanReverse(&LeadZeroIdx, static_cast<unsigned long>(X >> 32)))
+    return static_cast<int>(63 - (LeadZeroIdx + 32)); // Create a bit offset from the MSB.
+  // Scan the low 32 bits.
+  if (_BitScanReverse(&LeadZeroIdx, static_cast<unsigned long>(X)))
+    return static_cast<int>(63 - LeadZeroIdx);
+
+#else
+  if (_BitScanReverse64(&LeadZeroIdx, X)) return 63 - LeadZeroIdx;
+#endif
+  return 64;
+}
+
+inline uint32_t Clz(uint32_t X) {
+  unsigned long LeadZeroIdx = 0;
+  if (_BitScanReverse(&LeadZeroIdx, X)) return 31 - LeadZeroIdx;
+  return 32;
+}
+
+inline int Popcountll(unsigned long long X) {
+#if !defined(_M_ARM) && !defined(_M_X64)
+  return __popcnt(X) + __popcnt(X >> 32);
+#else
+  return __popcnt64(X);
+#endif
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZER_MSVC
+#endif  // LLVM_FUZZER_BUILTINS_MSVC_H

custom_mutators/libfuzzer/FuzzerCommand.h

@@ -0,0 +1,178 @@
+//===- FuzzerCommand.h - Interface representing a process -------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// FuzzerCommand represents a command to run in a subprocess.  It allows callers
+// to manage command line arguments and output and error streams.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_COMMAND_H
+#define LLVM_FUZZER_COMMAND_H
+
+#include "FuzzerDefs.h"
+#include "FuzzerIO.h"
+
+#include <algorithm>
+#include <sstream>
+#include <string>
+#include <vector>
+
+namespace fuzzer {
+
+class Command final {
+public:
+  // This command line flag is used to indicate that the remaining command line
+  // is immutable, meaning this flag effectively marks the end of the mutable
+  // argument list.
+  static inline const char *ignoreRemainingArgs() {
+    return "-ignore_remaining_args=1";
+  }
+
+  Command() : CombinedOutAndErr(false) {}
+
+  explicit Command(const Vector<std::string> &ArgsToAdd)
+      : Args(ArgsToAdd), CombinedOutAndErr(false) {}
+
+  explicit Command(const Command &Other)
+      : Args(Other.Args), CombinedOutAndErr(Other.CombinedOutAndErr),
+        OutputFile(Other.OutputFile) {}
+
+  Command &operator=(const Command &Other) {
+    Args = Other.Args;
+    CombinedOutAndErr = Other.CombinedOutAndErr;
+    OutputFile = Other.OutputFile;
+    return *this;
+  }
+
+  ~Command() {}
+
+  // Returns true if the given Arg is present in Args.  Only checks up to
+  // "-ignore_remaining_args=1".
+  bool hasArgument(const std::string &Arg) const {
+    auto i = endMutableArgs();
+    return std::find(Args.begin(), i, Arg) != i;
+  }
+
+  // Gets all of the current command line arguments, **including** those after
+  // "-ignore-remaining-args=1".
+  const Vector<std::string> &getArguments() const { return Args; }
+
+  // Adds the given argument before "-ignore_remaining_args=1", or at the end
+  // if that flag isn't present.
+  void addArgument(const std::string &Arg) {
+    Args.insert(endMutableArgs(), Arg);
+  }
+
+  // Adds all given arguments before "-ignore_remaining_args=1", or at the end
+  // if that flag isn't present.
+  void addArguments(const Vector<std::string> &ArgsToAdd) {
+    Args.insert(endMutableArgs(), ArgsToAdd.begin(), ArgsToAdd.end());
+  }
+
+  // Removes the given argument from the command argument list.  Ignores any
+  // occurrences after "-ignore_remaining_args=1", if present.
+  void removeArgument(const std::string &Arg) {
+    auto i = endMutableArgs();
+    Args.erase(std::remove(Args.begin(), i, Arg), i);
+  }
+
+  // Like hasArgument, but checks for "-[Flag]=...".
+  bool hasFlag(const std::string &Flag) const {
+    std::string Arg("-" + Flag + "=");
+    auto IsMatch = [&](const std::string &Other) {
+      return Arg.compare(0, std::string::npos, Other, 0, Arg.length()) == 0;
+    };
+    return std::any_of(Args.begin(), endMutableArgs(), IsMatch);
+  }
+
+  // Returns the value of the first instance of a given flag, or an empty string
+  // if the flag isn't present.  Ignores any occurrences after
+  // "-ignore_remaining_args=1", if present.
+  std::string getFlagValue(const std::string &Flag) const {
+    std::string Arg("-" + Flag + "=");
+    auto IsMatch = [&](const std::string &Other) {
+      return Arg.compare(0, std::string::npos, Other, 0, Arg.length()) == 0;
+    };
+    auto i = endMutableArgs();
+    auto j = std::find_if(Args.begin(), i, IsMatch);
+    std::string result;
+    if (j != i) {
+      result = j->substr(Arg.length());
+    }
+    return result;
+  }
+
+  // Like AddArgument, but adds "-[Flag]=[Value]".
+  void addFlag(const std::string &Flag, const std::string &Value) {
+    addArgument("-" + Flag + "=" + Value);
+  }
+
+  // Like RemoveArgument, but removes "-[Flag]=...".
+  void removeFlag(const std::string &Flag) {
+    std::string Arg("-" + Flag + "=");
+    auto IsMatch = [&](const std::string &Other) {
+      return Arg.compare(0, std::string::npos, Other, 0, Arg.length()) == 0;
+    };
+    auto i = endMutableArgs();
+    Args.erase(std::remove_if(Args.begin(), i, IsMatch), i);
+  }
+
+  // Returns whether the command's stdout is being written to an output file.
+  bool hasOutputFile() const { return !OutputFile.empty(); }
+
+  // Returns the currently set output file.
+  const std::string &getOutputFile() const { return OutputFile; }
+
+  // Configures the command to redirect its output to the name file.
+  void setOutputFile(const std::string &FileName) { OutputFile = FileName; }
+
+  // Returns whether the command's stderr is redirected to stdout.
+  bool isOutAndErrCombined() const { return CombinedOutAndErr; }
+
+  // Sets whether to redirect the command's stderr to its stdout.
+  void combineOutAndErr(bool combine = true) { CombinedOutAndErr = combine; }
+
+  // Returns a string representation of the command.  On many systems this will
+  // be the equivalent command line.
+  std::string toString() const {
+    std::stringstream SS;
+    for (auto arg : getArguments())
+      SS << arg << " ";
+    if (hasOutputFile())
+      SS << ">" << getOutputFile() << " ";
+    if (isOutAndErrCombined())
+      SS << "2>&1 ";
+    std::string result = SS.str();
+    if (!result.empty())
+      result = result.substr(0, result.length() - 1);
+    return result;
+  }
+
+private:
+  Command(Command &&Other) = delete;
+  Command &operator=(Command &&Other) = delete;
+
+  Vector<std::string>::iterator endMutableArgs() {
+    return std::find(Args.begin(), Args.end(), ignoreRemainingArgs());
+  }
+
+  Vector<std::string>::const_iterator endMutableArgs() const {
+    return std::find(Args.begin(), Args.end(), ignoreRemainingArgs());
+  }
+
+  // The command arguments.  Args[0] is the command name.
+  Vector<std::string> Args;
+
+  // True indicates stderr is redirected to stdout.
+  bool CombinedOutAndErr;
+
+  // If not empty, stdout is redirected to the named file.
+  std::string OutputFile;
+};
+
+} // namespace fuzzer
+
+#endif // LLVM_FUZZER_COMMAND_H

custom_mutators/libfuzzer/FuzzerCorpus.h

@@ -0,0 +1,581 @@
+//===- FuzzerCorpus.h - Internal header for the Fuzzer ----------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::InputCorpus
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_CORPUS
+#define LLVM_FUZZER_CORPUS
+
+#include "FuzzerDataFlowTrace.h"
+#include "FuzzerDefs.h"
+#include "FuzzerIO.h"
+#include "FuzzerRandom.h"
+#include "FuzzerSHA1.h"
+#include "FuzzerTracePC.h"
+#include <algorithm>
+#include <chrono>
+#include <numeric>
+#include <random>
+#include <unordered_set>
+
+namespace fuzzer {
+
+struct InputInfo {
+  Unit U;  // The actual input data.
+  std::chrono::microseconds TimeOfUnit;
+  uint8_t Sha1[kSHA1NumBytes];  // Checksum.
+  // Number of features that this input has and no smaller input has.
+  size_t NumFeatures = 0;
+  size_t Tmp = 0; // Used by ValidateFeatureSet.
+  // Stats.
+  size_t NumExecutedMutations = 0;
+  size_t NumSuccessfullMutations = 0;
+  bool NeverReduce = false;
+  bool MayDeleteFile = false;
+  bool Reduced = false;
+  bool HasFocusFunction = false;
+  Vector<uint32_t> UniqFeatureSet;
+  Vector<uint8_t> DataFlowTraceForFocusFunction;
+  // Power schedule.
+  bool NeedsEnergyUpdate = false;
+  double Energy = 0.0;
+  size_t SumIncidence = 0;
+  Vector<std::pair<uint32_t, uint16_t>> FeatureFreqs;
+
+  // Delete feature Idx and its frequency from FeatureFreqs.
+  bool DeleteFeatureFreq(uint32_t Idx) {
+    if (FeatureFreqs.empty())
+      return false;
+
+    // Binary search over local feature frequencies sorted by index.
+    auto Lower = std::lower_bound(FeatureFreqs.begin(), FeatureFreqs.end(),
+                                  std::pair<uint32_t, uint16_t>(Idx, 0));
+
+    if (Lower != FeatureFreqs.end() && Lower->first == Idx) {
+      FeatureFreqs.erase(Lower);
+      return true;
+    }
+    return false;
+  }
+
+  // Assign more energy to a high-entropy seed, i.e., that reveals more
+  // information about the globally rare features in the neighborhood of the
+  // seed. Since we do not know the entropy of a seed that has never been
+  // executed we assign fresh seeds maximum entropy and let II->Energy approach
+  // the true entropy from above. If ScalePerExecTime is true, the computed
+  // entropy is scaled based on how fast this input executes compared to the
+  // average execution time of inputs. The faster an input executes, the more
+  // energy gets assigned to the input.
+  void UpdateEnergy(size_t GlobalNumberOfFeatures, bool ScalePerExecTime,
+                    std::chrono::microseconds AverageUnitExecutionTime) {
+    Energy = 0.0;
+    SumIncidence = 0;
+
+    // Apply add-one smoothing to locally discovered features.
+    for (auto F : FeatureFreqs) {
+      size_t LocalIncidence = F.second + 1;
+      Energy -= LocalIncidence * logl(LocalIncidence);
+      SumIncidence += LocalIncidence;
+    }
+
+    // Apply add-one smoothing to locally undiscovered features.
+    //   PreciseEnergy -= 0; // since logl(1.0) == 0)
+    SumIncidence += (GlobalNumberOfFeatures - FeatureFreqs.size());
+
+    // Add a single locally abundant feature apply add-one smoothing.
+    size_t AbdIncidence = NumExecutedMutations + 1;
+    Energy -= AbdIncidence * logl(AbdIncidence);
+    SumIncidence += AbdIncidence;
+
+    // Normalize.
+    if (SumIncidence != 0)
+      Energy = (Energy / SumIncidence) + logl(SumIncidence);
+
+    if (ScalePerExecTime) {
+      // Scaling to favor inputs with lower execution time.
+      uint32_t PerfScore = 100;
+      if (TimeOfUnit.count() > AverageUnitExecutionTime.count() * 10)
+        PerfScore = 10;
+      else if (TimeOfUnit.count() > AverageUnitExecutionTime.count() * 4)
+        PerfScore = 25;
+      else if (TimeOfUnit.count() > AverageUnitExecutionTime.count() * 2)
+        PerfScore = 50;
+      else if (TimeOfUnit.count() * 3 > AverageUnitExecutionTime.count() * 4)
+        PerfScore = 75;
+      else if (TimeOfUnit.count() * 4 < AverageUnitExecutionTime.count())
+        PerfScore = 300;
+      else if (TimeOfUnit.count() * 3 < AverageUnitExecutionTime.count())
+        PerfScore = 200;
+      else if (TimeOfUnit.count() * 2 < AverageUnitExecutionTime.count())
+        PerfScore = 150;
+
+      Energy *= PerfScore;
+    }
+  }
+
+  // Increment the frequency of the feature Idx.
+  void UpdateFeatureFrequency(uint32_t Idx) {
+    NeedsEnergyUpdate = true;
+
+    // The local feature frequencies is an ordered vector of pairs.
+    // If there are no local feature frequencies, push_back preserves order.
+    // Set the feature frequency for feature Idx32 to 1.
+    if (FeatureFreqs.empty()) {
+      FeatureFreqs.push_back(std::pair<uint32_t, uint16_t>(Idx, 1));
+      return;
+    }
+
+    // Binary search over local feature frequencies sorted by index.
+    auto Lower = std::lower_bound(FeatureFreqs.begin(), FeatureFreqs.end(),
+                                  std::pair<uint32_t, uint16_t>(Idx, 0));
+
+    // If feature Idx32 already exists, increment its frequency.
+    // Otherwise, insert a new pair right after the next lower index.
+    if (Lower != FeatureFreqs.end() && Lower->first == Idx) {
+      Lower->second++;
+    } else {
+      FeatureFreqs.insert(Lower, std::pair<uint32_t, uint16_t>(Idx, 1));
+    }
+  }
+};
+
+struct EntropicOptions {
+  bool Enabled;
+  size_t NumberOfRarestFeatures;
+  size_t FeatureFrequencyThreshold;
+  bool ScalePerExecTime;
+};
+
+class InputCorpus {
+  static const uint32_t kFeatureSetSize = 1 << 21;
+  static const uint8_t kMaxMutationFactor = 20;
+  static const size_t kSparseEnergyUpdates = 100;
+
+  size_t NumExecutedMutations = 0;
+
+  EntropicOptions Entropic;
+
+public:
+  InputCorpus(const std::string &OutputCorpus, EntropicOptions Entropic)
+      : Entropic(Entropic), OutputCorpus(OutputCorpus) {
+    memset(InputSizesPerFeature, 0, sizeof(InputSizesPerFeature));
+    memset(SmallestElementPerFeature, 0, sizeof(SmallestElementPerFeature));
+  }
+  ~InputCorpus() {
+    for (auto II : Inputs)
+      delete II;
+  }
+  size_t size() const { return Inputs.size(); }
+  size_t SizeInBytes() const {
+    size_t Res = 0;
+    for (auto II : Inputs)
+      Res += II->U.size();
+    return Res;
+  }
+  size_t NumActiveUnits() const {
+    size_t Res = 0;
+    for (auto II : Inputs)
+      Res += !II->U.empty();
+    return Res;
+  }
+  size_t MaxInputSize() const {
+    size_t Res = 0;
+    for (auto II : Inputs)
+        Res = std::max(Res, II->U.size());
+    return Res;
+  }
+  void IncrementNumExecutedMutations() { NumExecutedMutations++; }
+
+  size_t NumInputsThatTouchFocusFunction() {
+    return std::count_if(Inputs.begin(), Inputs.end(), [](const InputInfo *II) {
+      return II->HasFocusFunction;
+    });
+  }
+
+  size_t NumInputsWithDataFlowTrace() {
+    return std::count_if(Inputs.begin(), Inputs.end(), [](const InputInfo *II) {
+      return !II->DataFlowTraceForFocusFunction.empty();
+    });
+  }
+
+  bool empty() const { return Inputs.empty(); }
+  const Unit &operator[] (size_t Idx) const { return Inputs[Idx]->U; }
+  InputInfo *AddToCorpus(const Unit &U, size_t NumFeatures, bool MayDeleteFile,
+                         bool HasFocusFunction, bool NeverReduce,
+                         std::chrono::microseconds TimeOfUnit,
+                         const Vector<uint32_t> &FeatureSet,
+                         const DataFlowTrace &DFT, const InputInfo *BaseII) {
+    assert(!U.empty());
+    if (FeatureDebug)
+      Printf("ADD_TO_CORPUS %zd NF %zd\n", Inputs.size(), NumFeatures);
+    Inputs.push_back(new InputInfo());
+    InputInfo &II = *Inputs.back();
+    II.U = U;
+    II.NumFeatures = NumFeatures;
+    II.NeverReduce = NeverReduce;
+    II.TimeOfUnit = TimeOfUnit;
+    II.MayDeleteFile = MayDeleteFile;
+    II.UniqFeatureSet = FeatureSet;
+    II.HasFocusFunction = HasFocusFunction;
+    // Assign maximal energy to the new seed.
+    II.Energy = RareFeatures.empty() ? 1.0 : log(RareFeatures.size());
+    II.SumIncidence = RareFeatures.size();
+    II.NeedsEnergyUpdate = false;
+    std::sort(II.UniqFeatureSet.begin(), II.UniqFeatureSet.end());
+    ComputeSHA1(U.data(), U.size(), II.Sha1);
+    auto Sha1Str = Sha1ToString(II.Sha1);
+    Hashes.insert(Sha1Str);
+    if (HasFocusFunction)
+      if (auto V = DFT.Get(Sha1Str))
+        II.DataFlowTraceForFocusFunction = *V;
+    // This is a gross heuristic.
+    // Ideally, when we add an element to a corpus we need to know its DFT.
+    // But if we don't, we'll use the DFT of its base input.
+    if (II.DataFlowTraceForFocusFunction.empty() && BaseII)
+      II.DataFlowTraceForFocusFunction = BaseII->DataFlowTraceForFocusFunction;
+    DistributionNeedsUpdate = true;
+    PrintCorpus();
+    // ValidateFeatureSet();
+    return &II;
+  }
+
+  // Debug-only
+  void PrintUnit(const Unit &U) {
+    if (!FeatureDebug) return;
+    for (uint8_t C : U) {
+      if (C != 'F' && C != 'U' && C != 'Z')
+        C = '.';
+      Printf("%c", C);
+    }
+  }
+
+  // Debug-only
+  void PrintFeatureSet(const Vector<uint32_t> &FeatureSet) {
+    if (!FeatureDebug) return;
+    Printf("{");
+    for (uint32_t Feature: FeatureSet)
+      Printf("%u,", Feature);
+    Printf("}");
+  }
+
+  // Debug-only
+  void PrintCorpus() {
+    if (!FeatureDebug) return;
+    Printf("======= CORPUS:\n");
+    int i = 0;
+    for (auto II : Inputs) {
+      if (std::find(II->U.begin(), II->U.end(), 'F') != II->U.end()) {
+        Printf("[%2d] ", i);
+        Printf("%s sz=%zd ", Sha1ToString(II->Sha1).c_str(), II->U.size());
+        PrintUnit(II->U);
+        Printf(" ");
+        PrintFeatureSet(II->UniqFeatureSet);
+        Printf("\n");
+      }
+      i++;
+    }
+  }
+
+  void Replace(InputInfo *II, const Unit &U) {
+    assert(II->U.size() > U.size());
+    Hashes.erase(Sha1ToString(II->Sha1));
+    DeleteFile(*II);
+    ComputeSHA1(U.data(), U.size(), II->Sha1);
+    Hashes.insert(Sha1ToString(II->Sha1));
+    II->U = U;
+    II->Reduced = true;
+    DistributionNeedsUpdate = true;
+  }
+
+  bool HasUnit(const Unit &U) { return Hashes.count(Hash(U)); }
+  bool HasUnit(const std::string &H) { return Hashes.count(H); }
+  InputInfo &ChooseUnitToMutate(Random &Rand) {
+    InputInfo &II = *Inputs[ChooseUnitIdxToMutate(Rand)];
+    assert(!II.U.empty());
+    return II;
+  }
+
+  InputInfo &ChooseUnitToCrossOverWith(Random &Rand, bool UniformDist) {
+    if (!UniformDist) {
+      return ChooseUnitToMutate(Rand);
+    }
+    InputInfo &II = *Inputs[Rand(Inputs.size())];
+    assert(!II.U.empty());
+    return II;
+  }
+
+  // Returns an index of random unit from the corpus to mutate.
+  size_t ChooseUnitIdxToMutate(Random &Rand) {
+    UpdateCorpusDistribution(Rand);
+    size_t Idx = static_cast<size_t>(CorpusDistribution(Rand));
+    assert(Idx < Inputs.size());
+    return Idx;
+  }
+
+  void PrintStats() {
+    for (size_t i = 0; i < Inputs.size(); i++) {
+      const auto &II = *Inputs[i];
+      Printf("  [% 3zd %s] sz: % 5zd runs: % 5zd succ: % 5zd focus: %d\n", i,
+             Sha1ToString(II.Sha1).c_str(), II.U.size(),
+             II.NumExecutedMutations, II.NumSuccessfullMutations, II.HasFocusFunction);
+    }
+  }
+
+  void PrintFeatureSet() {
+    for (size_t i = 0; i < kFeatureSetSize; i++) {
+      if(size_t Sz = GetFeature(i))
+        Printf("[%zd: id %zd sz%zd] ", i, SmallestElementPerFeature[i], Sz);
+    }
+    Printf("\n\t");
+    for (size_t i = 0; i < Inputs.size(); i++)
+      if (size_t N = Inputs[i]->NumFeatures)
+        Printf(" %zd=>%zd ", i, N);
+    Printf("\n");
+  }
+
+  void DeleteFile(const InputInfo &II) {
+    if (!OutputCorpus.empty() && II.MayDeleteFile)
+      RemoveFile(DirPlusFile(OutputCorpus, Sha1ToString(II.Sha1)));
+  }
+
+  void DeleteInput(size_t Idx) {
+    InputInfo &II = *Inputs[Idx];
+    DeleteFile(II);
+    Unit().swap(II.U);
+    II.Energy = 0.0;
+    II.NeedsEnergyUpdate = false;
+    DistributionNeedsUpdate = true;
+    if (FeatureDebug)
+      Printf("EVICTED %zd\n", Idx);
+  }
+
+  void AddRareFeature(uint32_t Idx) {
+    // Maintain *at least* TopXRarestFeatures many rare features
+    // and all features with a frequency below ConsideredRare.
+    // Remove all other features.
+    while (RareFeatures.size() > Entropic.NumberOfRarestFeatures &&
+           FreqOfMostAbundantRareFeature > Entropic.FeatureFrequencyThreshold) {
+
+      // Find most and second most abbundant feature.
+      uint32_t MostAbundantRareFeatureIndices[2] = {RareFeatures[0],
+                                                    RareFeatures[0]};
+      size_t Delete = 0;
+      for (size_t i = 0; i < RareFeatures.size(); i++) {
+        uint32_t Idx2 = RareFeatures[i];
+        if (GlobalFeatureFreqs[Idx2] >=
+            GlobalFeatureFreqs[MostAbundantRareFeatureIndices[0]]) {
+          MostAbundantRareFeatureIndices[1] = MostAbundantRareFeatureIndices[0];
+          MostAbundantRareFeatureIndices[0] = Idx2;
+          Delete = i;
+        }
+      }
+
+      // Remove most abundant rare feature.
+      RareFeatures[Delete] = RareFeatures.back();
+      RareFeatures.pop_back();
+
+      for (auto II : Inputs) {
+        if (II->DeleteFeatureFreq(MostAbundantRareFeatureIndices[0]))
+          II->NeedsEnergyUpdate = true;
+      }
+
+      // Set 2nd most abundant as the new most abundant feature count.
+      FreqOfMostAbundantRareFeature =
+          GlobalFeatureFreqs[MostAbundantRareFeatureIndices[1]];
+    }
+
+    // Add rare feature, handle collisions, and update energy.
+    RareFeatures.push_back(Idx);
+    GlobalFeatureFreqs[Idx] = 0;
+    for (auto II : Inputs) {
+      II->DeleteFeatureFreq(Idx);
+
+      // Apply add-one smoothing to this locally undiscovered feature.
+      // Zero energy seeds will never be fuzzed and remain zero energy.
+      if (II->Energy > 0.0) {
+        II->SumIncidence += 1;
+        II->Energy += logl(II->SumIncidence) / II->SumIncidence;
+      }
+    }
+
+    DistributionNeedsUpdate = true;
+  }
+
+  bool AddFeature(size_t Idx, uint32_t NewSize, bool Shrink) {
+    assert(NewSize);
+    Idx = Idx % kFeatureSetSize;
+    uint32_t OldSize = GetFeature(Idx);
+    if (OldSize == 0 || (Shrink && OldSize > NewSize)) {
+      if (OldSize > 0) {
+        size_t OldIdx = SmallestElementPerFeature[Idx];
+        InputInfo &II = *Inputs[OldIdx];
+        assert(II.NumFeatures > 0);
+        II.NumFeatures--;
+        if (II.NumFeatures == 0)
+          DeleteInput(OldIdx);
+      } else {
+        NumAddedFeatures++;
+        if (Entropic.Enabled)
+          AddRareFeature((uint32_t)Idx);
+      }
+      NumUpdatedFeatures++;
+      if (FeatureDebug)
+        Printf("ADD FEATURE %zd sz %d\n", Idx, NewSize);
+      SmallestElementPerFeature[Idx] = Inputs.size();
+      InputSizesPerFeature[Idx] = NewSize;
+      return true;
+    }
+    return false;
+  }
+
+  // Increment frequency of feature Idx globally and locally.
+  void UpdateFeatureFrequency(InputInfo *II, size_t Idx) {
+    uint32_t Idx32 = Idx % kFeatureSetSize;
+
+    // Saturated increment.
+    if (GlobalFeatureFreqs[Idx32] == 0xFFFF)
+      return;
+    uint16_t Freq = GlobalFeatureFreqs[Idx32]++;
+
+    // Skip if abundant.
+    if (Freq > FreqOfMostAbundantRareFeature ||
+        std::find(RareFeatures.begin(), RareFeatures.end(), Idx32) ==
+            RareFeatures.end())
+      return;
+
+    // Update global frequencies.
+    if (Freq == FreqOfMostAbundantRareFeature)
+      FreqOfMostAbundantRareFeature++;
+
+    // Update local frequencies.
+    if (II)
+      II->UpdateFeatureFrequency(Idx32);
+  }
+
+  size_t NumFeatures() const { return NumAddedFeatures; }
+  size_t NumFeatureUpdates() const { return NumUpdatedFeatures; }
+
+private:
+
+  static const bool FeatureDebug = false;
+
+  size_t GetFeature(size_t Idx) const { return InputSizesPerFeature[Idx]; }
+
+  void ValidateFeatureSet() {
+    if (FeatureDebug)
+      PrintFeatureSet();
+    for (size_t Idx = 0; Idx < kFeatureSetSize; Idx++)
+      if (GetFeature(Idx))
+        Inputs[SmallestElementPerFeature[Idx]]->Tmp++;
+    for (auto II: Inputs) {
+      if (II->Tmp != II->NumFeatures)
+        Printf("ZZZ %zd %zd\n", II->Tmp, II->NumFeatures);
+      assert(II->Tmp == II->NumFeatures);
+      II->Tmp = 0;
+    }
+  }
+
+  // Updates the probability distribution for the units in the corpus.
+  // Must be called whenever the corpus or unit weights are changed.
+  //
+  // Hypothesis: inputs that maximize information about globally rare features
+  // are interesting.
+  void UpdateCorpusDistribution(Random &Rand) {
+    // Skip update if no seeds or rare features were added/deleted.
+    // Sparse updates for local change of feature frequencies,
+    // i.e., randomly do not skip.
+    if (!DistributionNeedsUpdate &&
+        (!Entropic.Enabled || Rand(kSparseEnergyUpdates)))
+      return;
+
+    DistributionNeedsUpdate = false;
+
+    size_t N = Inputs.size();
+    assert(N);
+    Intervals.resize(N + 1);
+    Weights.resize(N);
+    std::iota(Intervals.begin(), Intervals.end(), 0);
+
+    std::chrono::microseconds AverageUnitExecutionTime(0);
+    for (auto II : Inputs) {
+      AverageUnitExecutionTime += II->TimeOfUnit;
+    }
+    AverageUnitExecutionTime /= N;
+
+    bool VanillaSchedule = true;
+    if (Entropic.Enabled) {
+      for (auto II : Inputs) {
+        if (II->NeedsEnergyUpdate && II->Energy != 0.0) {
+          II->NeedsEnergyUpdate = false;
+          II->UpdateEnergy(RareFeatures.size(), Entropic.ScalePerExecTime,
+                           AverageUnitExecutionTime);
+        }
+      }
+
+      for (size_t i = 0; i < N; i++) {
+
+        if (Inputs[i]->NumFeatures == 0) {
+          // If the seed doesn't represent any features, assign zero energy.
+          Weights[i] = 0.;
+        } else if (Inputs[i]->NumExecutedMutations / kMaxMutationFactor >
+                   NumExecutedMutations / Inputs.size()) {
+          // If the seed was fuzzed a lot more than average, assign zero energy.
+          Weights[i] = 0.;
+        } else {
+          // Otherwise, simply assign the computed energy.
+          Weights[i] = Inputs[i]->Energy;
+        }
+
+        // If energy for all seeds is zero, fall back to vanilla schedule.
+        if (Weights[i] > 0.0)
+          VanillaSchedule = false;
+      }
+    }
+
+    if (VanillaSchedule) {
+      for (size_t i = 0; i < N; i++)
+        Weights[i] = Inputs[i]->NumFeatures
+                         ? (i + 1) * (Inputs[i]->HasFocusFunction ? 1000 : 1)
+                         : 0.;
+    }
+
+    if (FeatureDebug) {
+      for (size_t i = 0; i < N; i++)
+        Printf("%zd ", Inputs[i]->NumFeatures);
+      Printf("SCORE\n");
+      for (size_t i = 0; i < N; i++)
+        Printf("%f ", Weights[i]);
+      Printf("Weights\n");
+    }
+    CorpusDistribution = std::piecewise_constant_distribution<double>(
+        Intervals.begin(), Intervals.end(), Weights.begin());
+  }
+  std::piecewise_constant_distribution<double> CorpusDistribution;
+
+  Vector<double> Intervals;
+  Vector<double> Weights;
+
+  std::unordered_set<std::string> Hashes;
+  Vector<InputInfo*> Inputs;
+
+  size_t NumAddedFeatures = 0;
+  size_t NumUpdatedFeatures = 0;
+  uint32_t InputSizesPerFeature[kFeatureSetSize];
+  uint32_t SmallestElementPerFeature[kFeatureSetSize];
+
+  bool DistributionNeedsUpdate = true;
+  uint16_t FreqOfMostAbundantRareFeature = 0;
+  uint16_t GlobalFeatureFreqs[kFeatureSetSize] = {};
+  Vector<uint32_t> RareFeatures;
+
+  std::string OutputCorpus;
+};
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_CORPUS

custom_mutators/libfuzzer/FuzzerCrossOver.cpp

@@ -0,0 +1,60 @@
+//===- FuzzerCrossOver.cpp - Cross over two test inputs -------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Cross over test inputs.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerDefs.h"
+#include "FuzzerMutate.h"
+#include "FuzzerRandom.h"
+#include <cstring>
+
+namespace fuzzer {
+
+// Cross Data1 and Data2, store the result (up to MaxOutSize bytes) in Out.
+size_t MutationDispatcher::CrossOver(const uint8_t *Data1, size_t Size1,
+                                     const uint8_t *Data2, size_t Size2,
+                                     uint8_t *Out, size_t MaxOutSize) {
+
+  assert(Size1 || Size2);
+  MaxOutSize = Rand(MaxOutSize) + 1;
+  size_t         OutPos = 0;
+  size_t         Pos1 = 0;
+  size_t         Pos2 = 0;
+  size_t *       InPos = &Pos1;
+  size_t         InSize = Size1;
+  const uint8_t *Data = Data1;
+  bool           CurrentlyUsingFirstData = true;
+  while (OutPos < MaxOutSize && (Pos1 < Size1 || Pos2 < Size2)) {
+
+    // Merge a part of Data into Out.
+    size_t OutSizeLeft = MaxOutSize - OutPos;
+    if (*InPos < InSize) {
+
+      size_t InSizeLeft = InSize - *InPos;
+      size_t MaxExtraSize = std::min(OutSizeLeft, InSizeLeft);
+      size_t ExtraSize = Rand(MaxExtraSize) + 1;
+      memcpy(Out + OutPos, Data + *InPos, ExtraSize);
+      OutPos += ExtraSize;
+      (*InPos) += ExtraSize;
+
+    }
+
+    // Use the other input data on the next iteration.
+    InPos = CurrentlyUsingFirstData ? &Pos2 : &Pos1;
+    InSize = CurrentlyUsingFirstData ? Size2 : Size1;
+    Data = CurrentlyUsingFirstData ? Data2 : Data1;
+    CurrentlyUsingFirstData = !CurrentlyUsingFirstData;
+
+  }
+
+  return OutPos;
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerDataFlowTrace.cpp

@@ -0,0 +1,344 @@
+//===- FuzzerDataFlowTrace.cpp - DataFlowTrace                ---*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::DataFlowTrace
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerDataFlowTrace.h"
+
+#include "FuzzerCommand.h"
+#include "FuzzerIO.h"
+#include "FuzzerRandom.h"
+#include "FuzzerSHA1.h"
+#include "FuzzerUtil.h"
+
+#include <cstdlib>
+#include <fstream>
+#include <numeric>
+#include <queue>
+#include <sstream>
+#include <string>
+#include <unordered_map>
+#include <unordered_set>
+#include <vector>
+
+namespace fuzzer {
+
+static const char *kFunctionsTxt = "functions.txt";
+
+bool BlockCoverage::AppendCoverage(const std::string &S) {
+
+  std::stringstream SS(S);
+  return AppendCoverage(SS);
+
+}
+
+// Coverage lines have this form:
+// CN X Y Z T
+// where N is the number of the function, T is the total number of instrumented
+// BBs, and X,Y,Z, if present, are the indecies of covered BB.
+// BB #0, which is the entry block, is not explicitly listed.
+bool BlockCoverage::AppendCoverage(std::istream &IN) {
+
+  std::string L;
+  while (std::getline(IN, L, '\n')) {
+
+    if (L.empty()) continue;
+    std::stringstream SS(L.c_str() + 1);
+    size_t            FunctionId = 0;
+    SS >> FunctionId;
+    if (L[0] == 'F') {
+
+      FunctionsWithDFT.insert(FunctionId);
+      continue;
+
+    }
+
+    if (L[0] != 'C') continue;
+    Vector<uint32_t> CoveredBlocks;
+    while (true) {
+
+      uint32_t BB = 0;
+      SS >> BB;
+      if (!SS) break;
+      CoveredBlocks.push_back(BB);
+
+    }
+
+    if (CoveredBlocks.empty()) return false;
+    uint32_t NumBlocks = CoveredBlocks.back();
+    CoveredBlocks.pop_back();
+    for (auto BB : CoveredBlocks)
+      if (BB >= NumBlocks) return false;
+    auto  It = Functions.find(FunctionId);
+    auto &Counters =
+        It == Functions.end()
+            ? Functions.insert({FunctionId, Vector<uint32_t>(NumBlocks)})
+                  .first->second
+            : It->second;
+
+    if (Counters.size() != NumBlocks) return false;  // wrong number of blocks.
+
+    Counters[0]++;
+    for (auto BB : CoveredBlocks)
+      Counters[BB]++;
+
+  }
+
+  return true;
+
+}
+
+// Assign weights to each function.
+// General principles:
+//   * any uncovered function gets weight 0.
+//   * a function with lots of uncovered blocks gets bigger weight.
+//   * a function with a less frequently executed code gets bigger weight.
+Vector<double> BlockCoverage::FunctionWeights(size_t NumFunctions) const {
+
+  Vector<double> Res(NumFunctions);
+  for (auto It : Functions) {
+
+    auto FunctionID = It.first;
+    auto Counters = It.second;
+    assert(FunctionID < NumFunctions);
+    auto &Weight = Res[FunctionID];
+    // Give higher weight if the function has a DFT.
+    Weight = FunctionsWithDFT.count(FunctionID) ? 1000. : 1;
+    // Give higher weight to functions with less frequently seen basic blocks.
+    Weight /= SmallestNonZeroCounter(Counters);
+    // Give higher weight to functions with the most uncovered basic blocks.
+    Weight *= NumberOfUncoveredBlocks(Counters) + 1;
+
+  }
+
+  return Res;
+
+}
+
+void DataFlowTrace::ReadCoverage(const std::string &DirPath) {
+
+  Vector<SizedFile> Files;
+  GetSizedFilesFromDir(DirPath, &Files);
+  for (auto &SF : Files) {
+
+    auto Name = Basename(SF.File);
+    if (Name == kFunctionsTxt) continue;
+    if (!CorporaHashes.count(Name)) continue;
+    std::ifstream IF(SF.File);
+    Coverage.AppendCoverage(IF);
+
+  }
+
+}
+
+static void DFTStringAppendToVector(Vector<uint8_t> *  DFT,
+                                    const std::string &DFTString) {
+
+  assert(DFT->size() == DFTString.size());
+  for (size_t I = 0, Len = DFT->size(); I < Len; I++)
+    (*DFT)[I] = DFTString[I] == '1';
+
+}
+
+// converts a string of '0' and '1' into a Vector<uint8_t>
+static Vector<uint8_t> DFTStringToVector(const std::string &DFTString) {
+
+  Vector<uint8_t> DFT(DFTString.size());
+  DFTStringAppendToVector(&DFT, DFTString);
+  return DFT;
+
+}
+
+static bool ParseError(const char *Err, const std::string &Line) {
+
+  Printf("DataFlowTrace: parse error: %s: Line: %s\n", Err, Line.c_str());
+  return false;
+
+}
+
+// TODO(metzman): replace std::string with std::string_view for
+// better performance. Need to figure our how to use string_view on Windows.
+static bool ParseDFTLine(const std::string &Line, size_t *FunctionNum,
+                         std::string *DFTString) {
+
+  if (!Line.empty() && Line[0] != 'F') return false;  // Ignore coverage.
+  size_t SpacePos = Line.find(' ');
+  if (SpacePos == std::string::npos)
+    return ParseError("no space in the trace line", Line);
+  if (Line.empty() || Line[0] != 'F')
+    return ParseError("the trace line doesn't start with 'F'", Line);
+  *FunctionNum = std::atol(Line.c_str() + 1);
+  const char *Beg = Line.c_str() + SpacePos + 1;
+  const char *End = Line.c_str() + Line.size();
+  assert(Beg < End);
+  size_t Len = End - Beg;
+  for (size_t I = 0; I < Len; I++) {
+
+    if (Beg[I] != '0' && Beg[I] != '1')
+      return ParseError("the trace should contain only 0 or 1", Line);
+
+  }
+
+  *DFTString = Beg;
+  return true;
+
+}
+
+bool DataFlowTrace::Init(const std::string &DirPath, std::string *FocusFunction,
+                         Vector<SizedFile> &CorporaFiles, Random &Rand) {
+
+  if (DirPath.empty()) return false;
+  Printf("INFO: DataFlowTrace: reading from '%s'\n", DirPath.c_str());
+  Vector<SizedFile> Files;
+  GetSizedFilesFromDir(DirPath, &Files);
+  std::string         L;
+  size_t              FocusFuncIdx = SIZE_MAX;
+  Vector<std::string> FunctionNames;
+
+  // Collect the hashes of the corpus files.
+  for (auto &SF : CorporaFiles)
+    CorporaHashes.insert(Hash(FileToVector(SF.File)));
+
+  // Read functions.txt
+  std::ifstream IF(DirPlusFile(DirPath, kFunctionsTxt));
+  size_t        NumFunctions = 0;
+  while (std::getline(IF, L, '\n')) {
+
+    FunctionNames.push_back(L);
+    NumFunctions++;
+    if (*FocusFunction == L) FocusFuncIdx = NumFunctions - 1;
+
+  }
+
+  if (!NumFunctions) return false;
+
+  if (*FocusFunction == "auto") {
+
+    // AUTOFOCUS works like this:
+    // * reads the coverage data from the DFT files.
+    // * assigns weights to functions based on coverage.
+    // * chooses a random function according to the weights.
+    ReadCoverage(DirPath);
+    auto           Weights = Coverage.FunctionWeights(NumFunctions);
+    Vector<double> Intervals(NumFunctions + 1);
+    std::iota(Intervals.begin(), Intervals.end(), 0);
+    auto Distribution = std::piecewise_constant_distribution<double>(
+        Intervals.begin(), Intervals.end(), Weights.begin());
+    FocusFuncIdx = static_cast<size_t>(Distribution(Rand));
+    *FocusFunction = FunctionNames[FocusFuncIdx];
+    assert(FocusFuncIdx < NumFunctions);
+    Printf("INFO: AUTOFOCUS: %zd %s\n", FocusFuncIdx,
+           FunctionNames[FocusFuncIdx].c_str());
+    for (size_t i = 0; i < NumFunctions; i++) {
+
+      if (!Weights[i]) continue;
+      Printf("  [%zd] W %g\tBB-tot %u\tBB-cov %u\tEntryFreq %u:\t%s\n", i,
+             Weights[i], Coverage.GetNumberOfBlocks(i),
+             Coverage.GetNumberOfCoveredBlocks(i), Coverage.GetCounter(i, 0),
+             FunctionNames[i].c_str());
+
+    }
+
+  }
+
+  if (!NumFunctions || FocusFuncIdx == SIZE_MAX || Files.size() <= 1)
+    return false;
+
+  // Read traces.
+  size_t NumTraceFiles = 0;
+  size_t NumTracesWithFocusFunction = 0;
+  for (auto &SF : Files) {
+
+    auto Name = Basename(SF.File);
+    if (Name == kFunctionsTxt) continue;
+    if (!CorporaHashes.count(Name)) continue;  // not in the corpus.
+    NumTraceFiles++;
+    // Printf("=== %s\n", Name.c_str());
+    std::ifstream IF(SF.File);
+    while (std::getline(IF, L, '\n')) {
+
+      size_t      FunctionNum = 0;
+      std::string DFTString;
+      if (ParseDFTLine(L, &FunctionNum, &DFTString) &&
+          FunctionNum == FocusFuncIdx) {
+
+        NumTracesWithFocusFunction++;
+
+        if (FunctionNum >= NumFunctions)
+          return ParseError("N is greater than the number of functions", L);
+        Traces[Name] = DFTStringToVector(DFTString);
+        // Print just a few small traces.
+        if (NumTracesWithFocusFunction <= 3 && DFTString.size() <= 16)
+          Printf("%s => |%s|\n", Name.c_str(), std::string(DFTString).c_str());
+        break;  // No need to parse the following lines.
+
+      }
+
+    }
+
+  }
+
+  Printf(
+      "INFO: DataFlowTrace: %zd trace files, %zd functions, "
+      "%zd traces with focus function\n",
+      NumTraceFiles, NumFunctions, NumTracesWithFocusFunction);
+  return NumTraceFiles > 0;
+
+}
+
+int CollectDataFlow(const std::string &DFTBinary, const std::string &DirPath,
+                    const Vector<SizedFile> &CorporaFiles) {
+
+  Printf("INFO: collecting data flow: bin: %s dir: %s files: %zd\n",
+         DFTBinary.c_str(), DirPath.c_str(), CorporaFiles.size());
+  if (CorporaFiles.empty()) {
+
+    Printf("ERROR: can't collect data flow without corpus provided.");
+    return 1;
+
+  }
+
+  static char DFSanEnv[] = "DFSAN_OPTIONS=warn_unimplemented=0";
+  putenv(DFSanEnv);
+  MkDir(DirPath);
+  for (auto &F : CorporaFiles) {
+
+    // For every input F we need to collect the data flow and the coverage.
+    // Data flow collection may fail if we request too many DFSan tags at once.
+    // So, we start from requesting all tags in range [0,Size) and if that fails
+    // we then request tags in [0,Size/2) and [Size/2, Size), and so on.
+    // Function number => DFT.
+    auto OutPath = DirPlusFile(DirPath, Hash(FileToVector(F.File)));
+    std::unordered_map<size_t, Vector<uint8_t>> DFTMap;
+    std::unordered_set<std::string>             Cov;
+    Command                                     Cmd;
+    Cmd.addArgument(DFTBinary);
+    Cmd.addArgument(F.File);
+    Cmd.addArgument(OutPath);
+    Printf("CMD: %s\n", Cmd.toString().c_str());
+    ExecuteCommand(Cmd);
+
+  }
+
+  // Write functions.txt if it's currently empty or doesn't exist.
+  auto FunctionsTxtPath = DirPlusFile(DirPath, kFunctionsTxt);
+  if (FileToString(FunctionsTxtPath).empty()) {
+
+    Command Cmd;
+    Cmd.addArgument(DFTBinary);
+    Cmd.setOutputFile(FunctionsTxtPath);
+    ExecuteCommand(Cmd);
+
+  }
+
+  return 0;
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerDataFlowTrace.h

@@ -0,0 +1,135 @@
+//===- FuzzerDataFlowTrace.h - Internal header for the Fuzzer ---*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::DataFlowTrace; reads and handles a data-flow trace.
+//
+// A data flow trace is generated by e.g. dataflow/DataFlow.cpp
+// and is stored on disk in a separate directory.
+//
+// The trace dir contains a file 'functions.txt' which lists function names,
+// oner per line, e.g.
+// ==> functions.txt <==
+// Func2
+// LLVMFuzzerTestOneInput
+// Func1
+//
+// All other files in the dir are the traces, see dataflow/DataFlow.cpp.
+// The name of the file is sha1 of the input used to generate the trace.
+//
+// Current status:
+//   the data is parsed and the summary is printed, but the data is not yet
+//   used in any other way.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_DATA_FLOW_TRACE
+#define LLVM_FUZZER_DATA_FLOW_TRACE
+
+#include "FuzzerDefs.h"
+#include "FuzzerIO.h"
+
+#include <unordered_map>
+#include <unordered_set>
+#include <vector>
+#include <string>
+
+namespace fuzzer {
+
+int CollectDataFlow(const std::string &DFTBinary, const std::string &DirPath,
+                    const Vector<SizedFile> &CorporaFiles);
+
+class BlockCoverage {
+ public:
+  bool AppendCoverage(std::istream &IN);
+  bool AppendCoverage(const std::string &S);
+
+  size_t NumCoveredFunctions() const { return Functions.size(); }
+
+  uint32_t GetCounter(size_t FunctionId, size_t BasicBlockId) {
+    auto It = Functions.find(FunctionId);
+    if (It == Functions.end()) return 0;
+    const auto &Counters = It->second;
+    if (BasicBlockId < Counters.size())
+      return Counters[BasicBlockId];
+    return 0;
+  }
+
+  uint32_t GetNumberOfBlocks(size_t FunctionId) {
+    auto It = Functions.find(FunctionId);
+    if (It == Functions.end()) return 0;
+    const auto &Counters = It->second;
+    return Counters.size();
+  }
+
+  uint32_t GetNumberOfCoveredBlocks(size_t FunctionId) {
+    auto It = Functions.find(FunctionId);
+    if (It == Functions.end()) return 0;
+    const auto &Counters = It->second;
+    uint32_t Result = 0;
+    for (auto Cnt: Counters)
+      if (Cnt)
+        Result++;
+    return Result;
+  }
+
+  Vector<double> FunctionWeights(size_t NumFunctions) const;
+  void clear() { Functions.clear(); }
+
+ private:
+
+  typedef Vector<uint32_t> CoverageVector;
+
+  uint32_t NumberOfCoveredBlocks(const CoverageVector &Counters) const {
+    uint32_t Res = 0;
+    for (auto Cnt : Counters)
+      if (Cnt)
+        Res++;
+    return Res;
+  }
+
+  uint32_t NumberOfUncoveredBlocks(const CoverageVector &Counters) const {
+    return Counters.size() - NumberOfCoveredBlocks(Counters);
+  }
+
+  uint32_t SmallestNonZeroCounter(const CoverageVector &Counters) const {
+    assert(!Counters.empty());
+    uint32_t Res = Counters[0];
+    for (auto Cnt : Counters)
+      if (Cnt)
+        Res = Min(Res, Cnt);
+    assert(Res);
+    return Res;
+  }
+
+  // Function ID => vector of counters.
+  // Each counter represents how many input files trigger the given basic block.
+  std::unordered_map<size_t, CoverageVector> Functions;
+  // Functions that have DFT entry.
+  std::unordered_set<size_t> FunctionsWithDFT;
+};
+
+class DataFlowTrace {
+ public:
+  void ReadCoverage(const std::string &DirPath);
+  bool Init(const std::string &DirPath, std::string *FocusFunction,
+            Vector<SizedFile> &CorporaFiles, Random &Rand);
+  void Clear() { Traces.clear(); }
+  const Vector<uint8_t> *Get(const std::string &InputSha1) const {
+    auto It = Traces.find(InputSha1);
+    if (It != Traces.end())
+      return &It->second;
+    return nullptr;
+  }
+
+ private:
+  // Input's sha1 => DFT for the FocusFunction.
+  std::unordered_map<std::string, Vector<uint8_t> > Traces;
+  BlockCoverage Coverage;
+  std::unordered_set<std::string> CorporaHashes;
+};
+}  // namespace fuzzer
+
+#endif // LLVM_FUZZER_DATA_FLOW_TRACE

custom_mutators/libfuzzer/FuzzerDefs.h

@@ -0,0 +1,75 @@
+//===- FuzzerDefs.h - Internal header for the Fuzzer ------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Basic definitions.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_DEFS_H
+#define LLVM_FUZZER_DEFS_H
+
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <cstring>
+#include <memory>
+#include <set>
+#include <string>
+#include <vector>
+
+
+namespace fuzzer {
+
+template <class T> T Min(T a, T b) { return a < b ? a : b; }
+template <class T> T Max(T a, T b) { return a > b ? a : b; }
+
+class Random;
+class Dictionary;
+class DictionaryEntry;
+class MutationDispatcher;
+struct FuzzingOptions;
+class InputCorpus;
+struct InputInfo;
+struct ExternalFunctions;
+
+// Global interface to functions that may or may not be available.
+extern ExternalFunctions *EF;
+
+// We are using a custom allocator to give a different symbol name to STL
+// containers in order to avoid ODR violations.
+template<typename T>
+  class fuzzer_allocator: public std::allocator<T> {
+    public:
+      fuzzer_allocator() = default;
+
+      template<class U>
+      fuzzer_allocator(const fuzzer_allocator<U>&) {}
+
+      template<class Other>
+      struct rebind { typedef fuzzer_allocator<Other> other;  };
+  };
+
+template<typename T>
+using Vector = std::vector<T, fuzzer_allocator<T>>;
+
+template<typename T>
+using Set = std::set<T, std::less<T>, fuzzer_allocator<T>>;
+
+typedef Vector<uint8_t> Unit;
+typedef Vector<Unit> UnitVector;
+typedef int (*UserCallback)(const uint8_t *Data, size_t Size);
+
+int FuzzerDriver(int *argc, char ***argv, UserCallback Callback);
+
+uint8_t *ExtraCountersBegin();
+uint8_t *ExtraCountersEnd();
+void ClearExtraCounters();
+
+extern bool RunningUserCallback;
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_DEFS_H

custom_mutators/libfuzzer/FuzzerDictionary.h

@@ -0,0 +1,118 @@
+//===- FuzzerDictionary.h - Internal header for the Fuzzer ------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::Dictionary
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_DICTIONARY_H
+#define LLVM_FUZZER_DICTIONARY_H
+
+#include "FuzzerDefs.h"
+#include "FuzzerIO.h"
+#include "FuzzerUtil.h"
+#include <algorithm>
+#include <limits>
+
+namespace fuzzer {
+// A simple POD sized array of bytes.
+template <size_t kMaxSizeT> class FixedWord {
+public:
+  static const size_t kMaxSize = kMaxSizeT;
+  FixedWord() {}
+  FixedWord(const uint8_t *B, uint8_t S) { Set(B, S); }
+
+  void Set(const uint8_t *B, uint8_t S) {
+    assert(S <= kMaxSize);
+    memcpy(Data, B, S);
+    Size = S;
+  }
+
+  bool operator==(const FixedWord<kMaxSize> &w) const {
+    return Size == w.Size && 0 == memcmp(Data, w.Data, Size);
+  }
+
+  static size_t GetMaxSize() { return kMaxSize; }
+  const uint8_t *data() const { return Data; }
+  uint8_t size() const { return Size; }
+
+private:
+  uint8_t Size = 0;
+  uint8_t Data[kMaxSize];
+};
+
+typedef FixedWord<64> Word;
+
+class DictionaryEntry {
+ public:
+  DictionaryEntry() {}
+  DictionaryEntry(Word W) : W(W) {}
+  DictionaryEntry(Word W, size_t PositionHint) : W(W), PositionHint(PositionHint) {}
+  const Word &GetW() const { return W; }
+
+  bool HasPositionHint() const { return PositionHint != std::numeric_limits<size_t>::max(); }
+  size_t GetPositionHint() const {
+    assert(HasPositionHint());
+    return PositionHint;
+  }
+  void IncUseCount() { UseCount++; }
+  void IncSuccessCount() { SuccessCount++; }
+  size_t GetUseCount() const { return UseCount; }
+  size_t GetSuccessCount() const {return SuccessCount; }
+
+  void Print(const char *PrintAfter = "\n") {
+    PrintASCII(W.data(), W.size());
+    if (HasPositionHint())
+      Printf("@%zd", GetPositionHint());
+    Printf("%s", PrintAfter);
+  }
+
+private:
+  Word W;
+  size_t PositionHint = std::numeric_limits<size_t>::max();
+  size_t UseCount = 0;
+  size_t SuccessCount = 0;
+};
+
+class Dictionary {
+ public:
+  static const size_t kMaxDictSize = 1 << 14;
+
+  bool ContainsWord(const Word &W) const {
+    return std::any_of(begin(), end(), [&](const DictionaryEntry &DE) {
+      return DE.GetW() == W;
+    });
+  }
+  const DictionaryEntry *begin() const { return &DE[0]; }
+  const DictionaryEntry *end() const { return begin() + Size; }
+  DictionaryEntry & operator[] (size_t Idx) {
+    assert(Idx < Size);
+    return DE[Idx];
+  }
+  void push_back(DictionaryEntry DE) {
+    if (Size < kMaxDictSize)
+      this->DE[Size++] = DE;
+  }
+  void clear() { Size = 0; }
+  bool empty() const { return Size == 0; }
+  size_t size() const { return Size; }
+
+private:
+  DictionaryEntry DE[kMaxDictSize];
+  size_t Size = 0;
+};
+
+// Parses one dictionary entry.
+// If successful, write the enty to Unit and returns true,
+// otherwise returns false.
+bool ParseOneDictionaryEntry(const std::string &Str, Unit *U);
+// Parses the dictionary file, fills Units, returns true iff all lines
+// were parsed successfully.
+bool ParseDictionaryFile(const std::string &Text, Vector<Unit> *Units);
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_DICTIONARY_H

custom_mutators/libfuzzer/FuzzerExtFunctions.def

@@ -0,0 +1,50 @@
+//===- FuzzerExtFunctions.def - External functions --------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// This defines the external function pointers that
+// ``fuzzer::ExternalFunctions`` should contain and try to initialize.  The
+// EXT_FUNC macro must be defined at the point of inclusion. The signature of
+// the macro is:
+//
+// EXT_FUNC(<name>, <return_type>, <function_signature>, <warn_if_missing>)
+//===----------------------------------------------------------------------===//
+
+// Optional user functions
+EXT_FUNC(LLVMFuzzerInitialize, int, (int *argc, char ***argv), false);
+EXT_FUNC(LLVMFuzzerCustomMutator, size_t,
+         (uint8_t *Data, size_t Size, size_t MaxSize, unsigned int Seed),
+         false);
+EXT_FUNC(LLVMFuzzerCustomCrossOver, size_t,
+         (const uint8_t *Data1, size_t Size1,
+          const uint8_t *Data2, size_t Size2,
+          uint8_t *Out, size_t MaxOutSize, unsigned int Seed),
+         false);
+
+// Sanitizer functions
+EXT_FUNC(__lsan_enable, void, (), false);
+EXT_FUNC(__lsan_disable, void, (), false);
+EXT_FUNC(__lsan_do_recoverable_leak_check, int, (), false);
+EXT_FUNC(__sanitizer_acquire_crash_state, int, (), true);
+EXT_FUNC(__sanitizer_install_malloc_and_free_hooks, int,
+         (void (*malloc_hook)(const volatile void *, size_t),
+          void (*free_hook)(const volatile void *)),
+         false);
+EXT_FUNC(__sanitizer_log_write, void, (const char *buf, size_t len), false);
+EXT_FUNC(__sanitizer_purge_allocator, void, (), false);
+EXT_FUNC(__sanitizer_print_memory_profile, void, (size_t, size_t), false);
+EXT_FUNC(__sanitizer_print_stack_trace, void, (), true);
+EXT_FUNC(__sanitizer_symbolize_pc, void,
+         (void *, const char *fmt, char *out_buf, size_t out_buf_size), false);
+EXT_FUNC(__sanitizer_get_module_and_offset_for_pc, int,
+         (void *pc, char *module_path,
+         size_t module_path_len,void **pc_offset), false);
+EXT_FUNC(__sanitizer_set_death_callback, void, (void (*)(void)), true);
+EXT_FUNC(__sanitizer_set_report_fd, void, (void*), false);
+EXT_FUNC(__msan_scoped_disable_interceptor_checks, void, (), false);
+EXT_FUNC(__msan_scoped_enable_interceptor_checks, void, (), false);
+EXT_FUNC(__msan_unpoison, void, (const volatile void *, size_t size), false);
+EXT_FUNC(__msan_unpoison_param, void, (size_t n), false);

custom_mutators/libfuzzer/FuzzerExtFunctions.h

@@ -0,0 +1,34 @@
+//===- FuzzerExtFunctions.h - Interface to external functions ---*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Defines an interface to (possibly optional) functions.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_EXT_FUNCTIONS_H
+#define LLVM_FUZZER_EXT_FUNCTIONS_H
+
+#include <stddef.h>
+#include <stdint.h>
+
+namespace fuzzer {
+
+struct ExternalFunctions {
+  // Initialize function pointers. Functions that are not available will be set
+  // to nullptr.  Do not call this constructor  before ``main()`` has been
+  // entered.
+  ExternalFunctions();
+
+#define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN)                            \
+  RETURN_TYPE(*NAME) FUNC_SIG = nullptr
+
+#include "FuzzerExtFunctions.def"
+
+#undef EXT_FUNC
+};
+} // namespace fuzzer
+
+#endif

custom_mutators/libfuzzer/FuzzerExtFunctionsDlsym.cpp

@@ -0,0 +1,60 @@
+//===- FuzzerExtFunctionsDlsym.cpp - Interface to external functions ------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Implementation for operating systems that support dlsym(). We only use it on
+// Apple platforms for now. We don't use this approach on Linux because it
+// requires that clients of LibFuzzer pass ``--export-dynamic`` to the linker.
+// That is a complication we don't wish to expose to clients right now.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_APPLE
+
+  #include "FuzzerExtFunctions.h"
+  #include "FuzzerIO.h"
+  #include <dlfcn.h>
+
+using namespace fuzzer;
+
+template <typename T>
+static T GetFnPtr(const char *FnName, bool WarnIfMissing) {
+
+  dlerror();  // Clear any previous errors.
+  void *Fn = dlsym(RTLD_DEFAULT, FnName);
+  if (Fn == nullptr) {
+
+    if (WarnIfMissing) {
+
+      const char *ErrorMsg = dlerror();
+      Printf("WARNING: Failed to find function \"%s\".", FnName);
+      if (ErrorMsg) Printf(" Reason %s.", ErrorMsg);
+      Printf("\n");
+
+    }
+
+  }
+
+  return reinterpret_cast<T>(Fn);
+
+}
+
+namespace fuzzer {
+
+ExternalFunctions::ExternalFunctions() {
+\
+  #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \
+    this->NAME = GetFnPtr<decltype(ExternalFunctions::NAME)>(#NAME, WARN)
+
+  #include "FuzzerExtFunctions.def"
+
+  #undef EXT_FUNC
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_APPLE
+

custom_mutators/libfuzzer/FuzzerExtFunctionsWeak.cpp

@@ -0,0 +1,63 @@
+//===- FuzzerExtFunctionsWeak.cpp - Interface to external functions -------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Implementation for Linux. This relies on the linker's support for weak
+// symbols. We don't use this approach on Apple platforms because it requires
+// clients of LibFuzzer to pass ``-U _<symbol_name>`` to the linker to allow
+// weak symbols to be undefined. That is a complication we don't want to expose
+// to clients right now.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_LINUX || LIBFUZZER_NETBSD || LIBFUZZER_FUCHSIA || \
+    LIBFUZZER_FREEBSD || LIBFUZZER_OPENBSD || LIBFUZZER_EMSCRIPTEN
+
+  #include "FuzzerExtFunctions.h"
+  #include "FuzzerIO.h"
+
+extern "C" {
+
+  // Declare these symbols as weak to allow them to be optionally defined.
+  #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \
+    __attribute__((weak, visibility("default"))) RETURN_TYPE NAME FUNC_SIG
+
+  #include "FuzzerExtFunctions.def"
+
+  #undef EXT_FUNC
+
+}
+
+using namespace fuzzer;
+
+static void CheckFnPtr(void *FnPtr, const char *FnName, bool WarnIfMissing) {
+
+  if (FnPtr == nullptr && WarnIfMissing) {
+
+    Printf("WARNING: Failed to find function \"%s\".\n", FnName);
+
+  }
+
+}
+
+namespace fuzzer {
+
+ExternalFunctions::ExternalFunctions() {
+\
+  #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN)                         \
+    this->NAME = ::NAME;                                                      \
+    CheckFnPtr(reinterpret_cast<void *>(reinterpret_cast<uintptr_t>(::NAME)), \
+               #NAME, WARN);
+
+  #include "FuzzerExtFunctions.def"
+
+  #undef EXT_FUNC
+
+}
+
+}  // namespace fuzzer
+
+#endif
+

custom_mutators/libfuzzer/FuzzerExtFunctionsWindows.cpp

@@ -0,0 +1,95 @@
+//=== FuzzerExtWindows.cpp - Interface to external functions --------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Implementation of FuzzerExtFunctions for Windows. Uses alternatename when
+// compiled with MSVC. Uses weak aliases when compiled with clang. Unfortunately
+// the method each compiler supports is not supported by the other.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_WINDOWS
+
+  #include "FuzzerExtFunctions.h"
+  #include "FuzzerIO.h"
+
+using namespace fuzzer;
+
+  // Intermediate macro to ensure the parameter is expanded before stringified.
+  #define STRINGIFY_(A) #A
+  #define STRINGIFY(A) STRINGIFY_(A)
+
+  #if LIBFUZZER_MSVC
+    // Copied from compiler-rt/lib/sanitizer_common/sanitizer_win_defs.h
+    #if defined(_M_IX86) || defined(__i386__)
+      #define WIN_SYM_PREFIX "_"
+    #else
+      #define WIN_SYM_PREFIX
+    #endif
+
+    // Declare external functions as having alternativenames, so that we can
+    // determine if they are not defined.
+    #define EXTERNAL_FUNC(Name, Default)                              \
+      __pragma(                                                       \
+          comment(linker, "/alternatename:" WIN_SYM_PREFIX STRINGIFY( \
+                              Name) "=" WIN_SYM_PREFIX STRINGIFY(Default)))
+  #else
+    // Declare external functions as weak to allow them to default to a
+    // specified function if not defined explicitly. We must use weak symbols
+    // because clang's support for alternatename is not 100%, see
+    // https://bugs.llvm.org/show_bug.cgi?id=40218 for more details.
+    #define EXTERNAL_FUNC(Name, Default) \
+      __attribute__((weak, alias(STRINGIFY(Default))))
+  #endif  // LIBFUZZER_MSVC
+
+extern "C" {
+\
+  #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN)         \
+    RETURN_TYPE NAME##Def FUNC_SIG {                          \
+                                                              \
+      Printf("ERROR: Function \"%s\" not defined.\n", #NAME); \
+      exit(1);                                                \
+                                                              \
+    }                                                         \
+    EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG
+
+  #include "FuzzerExtFunctions.def"
+
+  #undef EXT_FUNC
+
+}
+
+template <typename T>
+static T *GetFnPtr(T *Fun, T *FunDef, const char *FnName, bool WarnIfMissing) {
+
+  if (Fun == FunDef) {
+
+    if (WarnIfMissing)
+      Printf("WARNING: Failed to find function \"%s\".\n", FnName);
+    return nullptr;
+
+  }
+
+  return Fun;
+
+}
+
+namespace fuzzer {
+
+ExternalFunctions::ExternalFunctions() {
+\
+  #define EXT_FUNC(NAME, RETURN_TYPE, FUNC_SIG, WARN) \
+    this->NAME = GetFnPtr<decltype(::NAME)>(::NAME, ::NAME##Def, #NAME, WARN);
+
+  #include "FuzzerExtFunctions.def"
+
+  #undef EXT_FUNC
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_WINDOWS
+

custom_mutators/libfuzzer/FuzzerExtraCounters.cpp

@@ -0,0 +1,71 @@
+//===- FuzzerExtraCounters.cpp - Extra coverage counters ------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Extra coverage counters defined by user code.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerPlatform.h"
+#include <cstdint>
+
+#if LIBFUZZER_LINUX || LIBFUZZER_NETBSD || LIBFUZZER_FREEBSD || \
+    LIBFUZZER_OPENBSD || LIBFUZZER_FUCHSIA || LIBFUZZER_EMSCRIPTEN
+__attribute__((weak)) extern uint8_t __start___libfuzzer_extra_counters;
+__attribute__((weak)) extern uint8_t __stop___libfuzzer_extra_counters;
+
+namespace fuzzer {
+
+uint8_t *ExtraCountersBegin() {
+
+  return &__start___libfuzzer_extra_counters;
+
+}
+
+uint8_t *ExtraCountersEnd() {
+
+  return &__stop___libfuzzer_extra_counters;
+
+}
+
+ATTRIBUTE_NO_SANITIZE_ALL
+void ClearExtraCounters() {  // hand-written memset, don't asan-ify.
+  uintptr_t *Beg = reinterpret_cast<uintptr_t *>(ExtraCountersBegin());
+  uintptr_t *End = reinterpret_cast<uintptr_t *>(ExtraCountersEnd());
+  for (; Beg < End; Beg++) {
+
+    *Beg = 0;
+    __asm__ __volatile__("" : : : "memory");
+
+  }
+
+}
+
+}  // namespace fuzzer
+
+#else
+// TODO: implement for other platforms.
+namespace fuzzer {
+
+uint8_t *ExtraCountersBegin() {
+
+  return nullptr;
+
+}
+
+uint8_t *ExtraCountersEnd() {
+
+  return nullptr;
+
+}
+
+void ClearExtraCounters() {
+
+}
+
+}  // namespace fuzzer
+
+#endif
+

custom_mutators/libfuzzer/FuzzerFlags.def

@@ -0,0 +1,197 @@
+//===- FuzzerFlags.def - Run-time flags -------------------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Flags. FUZZER_FLAG_INT/FUZZER_FLAG_STRING macros should be defined at the
+// point of inclusion. We are not using any flag parsing library for better
+// portability and independence.
+//===----------------------------------------------------------------------===//
+FUZZER_FLAG_INT(verbosity, 1, "Verbosity level.")
+FUZZER_FLAG_UNSIGNED(seed, 0, "Random seed. If 0, seed is generated.")
+FUZZER_FLAG_INT(runs, -1,
+            "Number of individual test runs (-1 for infinite runs).")
+FUZZER_FLAG_INT(max_len, 0, "Maximum length of the test input. "
+    "If 0, libFuzzer tries to guess a good value based on the corpus "
+    "and reports it. ")
+FUZZER_FLAG_INT(len_control, 100, "Try generating small inputs first, "
+  "then try larger inputs over time.  Specifies the rate at which the length "
+  "limit is increased (smaller == faster).  If 0, immediately try inputs with "
+  "size up to max_len. Default value is 0, if LLVMFuzzerCustomMutator is used.")
+FUZZER_FLAG_STRING(seed_inputs, "A comma-separated list of input files "
+  "to use as an additional seed corpus. Alternatively, an \"@\" followed by "
+  "the name of a file containing the comma-separated list.")
+FUZZER_FLAG_INT(keep_seed, 0, "If 1, keep seed inputs in the corpus even if "
+  "they do not produce new coverage. When used with |reduce_inputs==1|, the "
+  "seed inputs will never be reduced. This option can be useful when seeds are"
+  "not properly formed for the fuzz target but still have useful snippets.")
+FUZZER_FLAG_INT(cross_over, 1, "If 1, cross over inputs.")
+FUZZER_FLAG_INT(cross_over_uniform_dist, 0, "Experimental. If 1, use a "
+  "uniform probability distribution when choosing inputs to cross over with. "
+  "Some of the inputs in the corpus may never get chosen for mutation "
+  "depending on the input mutation scheduling policy. With this flag, all "
+  "inputs, regardless of the input mutation scheduling policy, can be chosen "
+  "as an input to cross over with. This can be particularly useful with "
+  "|keep_seed==1|; all the initial seed inputs, even though they do not "
+  "increase coverage because they are not properly formed, will still be "
+  "chosen as an input to cross over with.")
+
+FUZZER_FLAG_INT(mutate_depth, 5,
+            "Apply this number of consecutive mutations to each input.")
+FUZZER_FLAG_INT(reduce_depth, 0, "Experimental/internal. "
+                "Reduce depth if mutations lose unique features")
+FUZZER_FLAG_INT(shuffle, 1, "Shuffle inputs at startup")
+FUZZER_FLAG_INT(prefer_small, 1,
+    "If 1, always prefer smaller inputs during the corpus shuffle.")
+FUZZER_FLAG_INT(
+    timeout, 1200,
+    "Timeout in seconds (if positive). "
+    "If one unit runs more than this number of seconds the process will abort.")
+FUZZER_FLAG_INT(error_exitcode, 77, "When libFuzzer itself reports a bug "
+  "this exit code will be used.")
+FUZZER_FLAG_INT(timeout_exitcode, 70, "When libFuzzer reports a timeout "
+  "this exit code will be used.")
+FUZZER_FLAG_INT(max_total_time, 0, "If positive, indicates the maximal total "
+                                   "time in seconds to run the fuzzer.")
+FUZZER_FLAG_INT(help, 0, "Print help.")
+FUZZER_FLAG_INT(fork, 0, "Experimental mode where fuzzing happens "
+                "in a subprocess")
+FUZZER_FLAG_INT(ignore_timeouts, 1, "Ignore timeouts in fork mode")
+FUZZER_FLAG_INT(ignore_ooms, 1, "Ignore OOMs in fork mode")
+FUZZER_FLAG_INT(ignore_crashes, 0, "Ignore crashes in fork mode")
+FUZZER_FLAG_INT(merge, 0, "If 1, the 2-nd, 3-rd, etc corpora will be "
+  "merged into the 1-st corpus. Only interesting units will be taken. "
+  "This flag can be used to minimize a corpus.")
+FUZZER_FLAG_STRING(stop_file, "Stop fuzzing ASAP if this file exists")
+FUZZER_FLAG_STRING(merge_inner, "internal flag")
+FUZZER_FLAG_STRING(merge_control_file,
+                   "Specify a control file used for the merge process. "
+                   "If a merge process gets killed it tries to leave this file "
+                   "in a state suitable for resuming the merge. "
+                   "By default a temporary file will be used."
+                   "The same file can be used for multistep merge process.")
+FUZZER_FLAG_INT(minimize_crash, 0, "If 1, minimizes the provided"
+  " crash input. Use with -runs=N or -max_total_time=N to limit "
+  "the number attempts."
+  " Use with -exact_artifact_path to specify the output."
+  " Combine with ASAN_OPTIONS=dedup_token_length=3 (or similar) to ensure that"
+  " the minimized input triggers the same crash."
+  )
+FUZZER_FLAG_INT(cleanse_crash, 0, "If 1, tries to cleanse the provided"
+  " crash input to make it contain fewer original bytes."
+  " Use with -exact_artifact_path to specify the output."
+  )
+FUZZER_FLAG_INT(minimize_crash_internal_step, 0, "internal flag")
+FUZZER_FLAG_STRING(features_dir, "internal flag. Used to dump feature sets on disk."
+  "Every time a new input is added to the corpus, a corresponding file in the features_dir"
+  " is created containing the unique features of that input."
+  " Features are stored in binary format.")
+FUZZER_FLAG_STRING(mutation_graph_file, "Saves a graph (in DOT format) to"
+  " mutation_graph_file. The graph contains a vertex for each input that has"
+  " unique coverage; directed edges are provided between parents and children"
+  " where the child has unique coverage, and are recorded with the type of"
+  " mutation that caused the child.")
+FUZZER_FLAG_INT(use_counters, 1, "Use coverage counters")
+FUZZER_FLAG_INT(use_memmem, 1,
+                "Use hints from intercepting memmem, strstr, etc")
+FUZZER_FLAG_INT(use_value_profile, 0,
+                "Experimental. Use value profile to guide fuzzing.")
+FUZZER_FLAG_INT(use_cmp, 1, "Use CMP traces to guide mutations")
+FUZZER_FLAG_INT(shrink, 0, "Experimental. Try to shrink corpus inputs.")
+FUZZER_FLAG_INT(reduce_inputs, 1,
+  "Try to reduce the size of inputs while preserving their full feature sets")
+FUZZER_FLAG_UNSIGNED(jobs, 0, "Number of jobs to run. If jobs >= 1 we spawn"
+                          " this number of jobs in separate worker processes"
+                          " with stdout/stderr redirected to fuzz-JOB.log.")
+FUZZER_FLAG_UNSIGNED(workers, 0,
+            "Number of simultaneous worker processes to run the jobs."
+            " If zero, \"min(jobs,NumberOfCpuCores()/2)\" is used.")
+FUZZER_FLAG_INT(reload, 1,
+                "Reload the main corpus every <N> seconds to get new units"
+                " discovered by other processes. If 0, disabled")
+FUZZER_FLAG_INT(report_slow_units, 10,
+    "Report slowest units if they run for more than this number of seconds.")
+FUZZER_FLAG_INT(only_ascii, 0,
+                "If 1, generate only ASCII (isprint+isspace) inputs.")
+FUZZER_FLAG_STRING(dict, "Experimental. Use the dictionary file.")
+FUZZER_FLAG_STRING(artifact_prefix, "Write fuzzing artifacts (crash, "
+                                    "timeout, or slow inputs) as "
+                                    "$(artifact_prefix)file")
+FUZZER_FLAG_STRING(exact_artifact_path,
+                   "Write the single artifact on failure (crash, timeout) "
+                   "as $(exact_artifact_path). This overrides -artifact_prefix "
+                   "and will not use checksum in the file name. Do not "
+                   "use the same path for several parallel processes.")
+FUZZER_FLAG_INT(print_pcs, 0, "If 1, print out newly covered PCs.")
+FUZZER_FLAG_INT(print_funcs, 2, "If >=1, print out at most this number of "
+                                "newly covered functions.")
+FUZZER_FLAG_INT(print_final_stats, 0, "If 1, print statistics at exit.")
+FUZZER_FLAG_INT(print_corpus_stats, 0,
+  "If 1, print statistics on corpus elements at exit.")
+FUZZER_FLAG_INT(print_coverage, 0, "If 1, print coverage information as text"
+                                   " at exit.")
+FUZZER_FLAG_INT(dump_coverage, 0, "Deprecated.")
+FUZZER_FLAG_INT(handle_segv, 1, "If 1, try to intercept SIGSEGV.")
+FUZZER_FLAG_INT(handle_bus, 1, "If 1, try to intercept SIGBUS.")
+FUZZER_FLAG_INT(handle_abrt, 1, "If 1, try to intercept SIGABRT.")
+FUZZER_FLAG_INT(handle_ill, 1, "If 1, try to intercept SIGILL.")
+FUZZER_FLAG_INT(handle_fpe, 1, "If 1, try to intercept SIGFPE.")
+FUZZER_FLAG_INT(handle_int, 1, "If 1, try to intercept SIGINT.")
+FUZZER_FLAG_INT(handle_term, 1, "If 1, try to intercept SIGTERM.")
+FUZZER_FLAG_INT(handle_xfsz, 1, "If 1, try to intercept SIGXFSZ.")
+FUZZER_FLAG_INT(handle_usr1, 1, "If 1, try to intercept SIGUSR1.")
+FUZZER_FLAG_INT(handle_usr2, 1, "If 1, try to intercept SIGUSR2.")
+FUZZER_FLAG_INT(close_fd_mask, 0, "If 1, close stdout at startup; "
+    "if 2, close stderr; if 3, close both. "
+    "Be careful, this will also close e.g. stderr of asan.")
+FUZZER_FLAG_INT(detect_leaks, 1, "If 1, and if LeakSanitizer is enabled "
+    "try to detect memory leaks during fuzzing (i.e. not only at shut down).")
+FUZZER_FLAG_INT(purge_allocator_interval, 1, "Purge allocator caches and "
+    "quarantines every <N> seconds. When rss_limit_mb is specified (>0), "
+    "purging starts when RSS exceeds 50% of rss_limit_mb. Pass "
+    "purge_allocator_interval=-1 to disable this functionality.")
+FUZZER_FLAG_INT(trace_malloc, 0, "If >= 1 will print all mallocs/frees. "
+    "If >= 2 will also print stack traces.")
+FUZZER_FLAG_INT(rss_limit_mb, 2048, "If non-zero, the fuzzer will exit upon"
+    "reaching this limit of RSS memory usage.")
+FUZZER_FLAG_INT(malloc_limit_mb, 0, "If non-zero, the fuzzer will exit "
+    "if the target tries to allocate this number of Mb with one malloc call. "
+    "If zero (default) same limit as rss_limit_mb is applied.")
+FUZZER_FLAG_STRING(exit_on_src_pos, "Exit if a newly found PC originates"
+    " from the given source location. Example: -exit_on_src_pos=foo.cc:123. "
+    "Used primarily for testing libFuzzer itself.")
+FUZZER_FLAG_STRING(exit_on_item, "Exit if an item with a given sha1 sum"
+    " was added to the corpus. "
+    "Used primarily for testing libFuzzer itself.")
+FUZZER_FLAG_INT(ignore_remaining_args, 0, "If 1, ignore all arguments passed "
+                "after this one. Useful for fuzzers that need to do their own "
+                "argument parsing.")
+FUZZER_FLAG_STRING(focus_function, "Experimental. "
+     "Fuzzing will focus on inputs that trigger calls to this function. "
+     "If -focus_function=auto and -data_flow_trace is used, libFuzzer "
+     "will choose the focus functions automatically.")
+FUZZER_FLAG_INT(entropic, 0, "Experimental. Enables entropic power schedule.")
+FUZZER_FLAG_INT(entropic_feature_frequency_threshold, 0xFF, "Experimental. If "
+     "entropic is enabled, all features which are observed less often than "
+     "the specified value are considered as rare.")
+FUZZER_FLAG_INT(entropic_number_of_rarest_features, 100, "Experimental. If "
+     "entropic is enabled, we keep track of the frequencies only for the "
+     "Top-X least abundant features (union features that are considered as "
+     "rare).")
+FUZZER_FLAG_INT(entropic_scale_per_exec_time, 0, "Experimental. If 1, "
+     "the Entropic power schedule gets scaled based on the input execution "
+     "time. Inputs with lower execution time get scheduled more (up to 30x). "
+     "Note that, if 1, fuzzer stops from being deterministic even if a "
+     "non-zero random seed is given.")
+
+FUZZER_FLAG_INT(analyze_dict, 0, "Experimental")
+FUZZER_DEPRECATED_FLAG(use_clang_coverage)
+FUZZER_FLAG_STRING(data_flow_trace, "Experimental: use the data flow trace")
+FUZZER_FLAG_STRING(collect_data_flow,
+                   "Experimental: collect the data flow trace")
+
+FUZZER_FLAG_INT(create_missing_dirs, 0, "Automatically attempt to create "
+     "directories for arguments that would normally expect them to already "
+     "exist (i.e. artifact_prefix, exact_artifact_path, features_dir, corpus)")

custom_mutators/libfuzzer/FuzzerFork.cpp

@@ -0,0 +1,501 @@
+//===- FuzzerFork.cpp - run fuzzing in separate subprocesses --------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Spawn and orchestrate separate fuzzing processes.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerCommand.h"
+#include "FuzzerFork.h"
+#include "FuzzerIO.h"
+#include "FuzzerInternal.h"
+#include "FuzzerMerge.h"
+#include "FuzzerSHA1.h"
+#include "FuzzerTracePC.h"
+#include "FuzzerUtil.h"
+
+#include <atomic>
+#include <chrono>
+#include <condition_variable>
+#include <fstream>
+#include <memory>
+#include <mutex>
+#include <queue>
+#include <sstream>
+#include <thread>
+
+namespace fuzzer {
+
+struct Stats {
+
+  size_t number_of_executed_units = 0;
+  size_t peak_rss_mb = 0;
+  size_t average_exec_per_sec = 0;
+
+};
+
+static Stats ParseFinalStatsFromLog(const std::string &LogPath) {
+
+  std::ifstream In(LogPath);
+  std::string   Line;
+  Stats         Res;
+  struct {
+
+    const char *Name;
+    size_t *    Var;
+
+  } NameVarPairs[] = {
+
+      {"stat::number_of_executed_units:", &Res.number_of_executed_units},
+      {"stat::peak_rss_mb:", &Res.peak_rss_mb},
+      {"stat::average_exec_per_sec:", &Res.average_exec_per_sec},
+      {nullptr, nullptr},
+
+  };
+
+  while (std::getline(In, Line, '\n')) {
+
+    if (Line.find("stat::") != 0) continue;
+    std::istringstream ISS(Line);
+    std::string        Name;
+    size_t             Val;
+    ISS >> Name >> Val;
+    for (size_t i = 0; NameVarPairs[i].Name; i++)
+      if (Name == NameVarPairs[i].Name) *NameVarPairs[i].Var = Val;
+
+  }
+
+  return Res;
+
+}
+
+struct FuzzJob {
+
+  // Inputs.
+  Command     Cmd;
+  std::string CorpusDir;
+  std::string FeaturesDir;
+  std::string LogPath;
+  std::string SeedListPath;
+  std::string CFPath;
+  size_t      JobId;
+
+  int DftTimeInSeconds = 0;
+
+  // Fuzzing Outputs.
+  int ExitCode;
+
+  ~FuzzJob() {
+
+    RemoveFile(CFPath);
+    RemoveFile(LogPath);
+    RemoveFile(SeedListPath);
+    RmDirRecursive(CorpusDir);
+    RmDirRecursive(FeaturesDir);
+
+  }
+
+};
+
+struct GlobalEnv {
+
+  Vector<std::string>                   Args;
+  Vector<std::string>                   CorpusDirs;
+  std::string                           MainCorpusDir;
+  std::string                           TempDir;
+  std::string                           DFTDir;
+  std::string                           DataFlowBinary;
+  Set<uint32_t>                         Features, Cov;
+  Set<std::string>                      FilesWithDFT;
+  Vector<std::string>                   Files;
+  Random *                              Rand;
+  std::chrono::system_clock::time_point ProcessStartTime;
+  int                                   Verbosity = 0;
+
+  size_t NumTimeouts = 0;
+  size_t NumOOMs = 0;
+  size_t NumCrashes = 0;
+
+  size_t NumRuns = 0;
+
+  std::string StopFile() {
+
+    return DirPlusFile(TempDir, "STOP");
+
+  }
+
+  size_t secondsSinceProcessStartUp() const {
+
+    return std::chrono::duration_cast<std::chrono::seconds>(
+               std::chrono::system_clock::now() - ProcessStartTime)
+        .count();
+
+  }
+
+  FuzzJob *CreateNewJob(size_t JobId) {
+
+    Command Cmd(Args);
+    Cmd.removeFlag("fork");
+    Cmd.removeFlag("runs");
+    Cmd.removeFlag("collect_data_flow");
+    for (auto &C : CorpusDirs)  // Remove all corpora from the args.
+      Cmd.removeArgument(C);
+    Cmd.addFlag("reload", "0");  // working in an isolated dir, no reload.
+    Cmd.addFlag("print_final_stats", "1");
+    Cmd.addFlag("print_funcs", "0");  // no need to spend time symbolizing.
+    Cmd.addFlag("max_total_time", std::to_string(std::min((size_t)300, JobId)));
+    Cmd.addFlag("stop_file", StopFile());
+    if (!DataFlowBinary.empty()) {
+
+      Cmd.addFlag("data_flow_trace", DFTDir);
+      if (!Cmd.hasFlag("focus_function")) Cmd.addFlag("focus_function", "auto");
+
+    }
+
+    auto        Job = new FuzzJob;
+    std::string Seeds;
+    if (size_t CorpusSubsetSize =
+            std::min(Files.size(), (size_t)sqrt(Files.size() + 2))) {
+
+      auto Time1 = std::chrono::system_clock::now();
+      for (size_t i = 0; i < CorpusSubsetSize; i++) {
+
+        auto &SF = Files[Rand->SkewTowardsLast(Files.size())];
+        Seeds += (Seeds.empty() ? "" : ",") + SF;
+        CollectDFT(SF);
+
+      }
+
+      auto Time2 = std::chrono::system_clock::now();
+      Job->DftTimeInSeconds = duration_cast<seconds>(Time2 - Time1).count();
+
+    }
+
+    if (!Seeds.empty()) {
+
+      Job->SeedListPath =
+          DirPlusFile(TempDir, std::to_string(JobId) + ".seeds");
+      WriteToFile(Seeds, Job->SeedListPath);
+      Cmd.addFlag("seed_inputs", "@" + Job->SeedListPath);
+
+    }
+
+    Job->LogPath = DirPlusFile(TempDir, std::to_string(JobId) + ".log");
+    Job->CorpusDir = DirPlusFile(TempDir, "C" + std::to_string(JobId));
+    Job->FeaturesDir = DirPlusFile(TempDir, "F" + std::to_string(JobId));
+    Job->CFPath = DirPlusFile(TempDir, std::to_string(JobId) + ".merge");
+    Job->JobId = JobId;
+
+    Cmd.addArgument(Job->CorpusDir);
+    Cmd.addFlag("features_dir", Job->FeaturesDir);
+
+    for (auto &D : {Job->CorpusDir, Job->FeaturesDir}) {
+
+      RmDirRecursive(D);
+      MkDir(D);
+
+    }
+
+    Cmd.setOutputFile(Job->LogPath);
+    Cmd.combineOutAndErr();
+
+    Job->Cmd = Cmd;
+
+    if (Verbosity >= 2)
+      Printf("Job %zd/%p Created: %s\n", JobId, Job,
+             Job->Cmd.toString().c_str());
+    // Start from very short runs and gradually increase them.
+    return Job;
+
+  }
+
+  void RunOneMergeJob(FuzzJob *Job) {
+
+    auto Stats = ParseFinalStatsFromLog(Job->LogPath);
+    NumRuns += Stats.number_of_executed_units;
+
+    Vector<SizedFile> TempFiles, MergeCandidates;
+    // Read all newly created inputs and their feature sets.
+    // Choose only those inputs that have new features.
+    GetSizedFilesFromDir(Job->CorpusDir, &TempFiles);
+    std::sort(TempFiles.begin(), TempFiles.end());
+    for (auto &F : TempFiles) {
+
+      auto FeatureFile = F.File;
+      FeatureFile.replace(0, Job->CorpusDir.size(), Job->FeaturesDir);
+      auto FeatureBytes = FileToVector(FeatureFile, 0, false);
+      assert((FeatureBytes.size() % sizeof(uint32_t)) == 0);
+      Vector<uint32_t> NewFeatures(FeatureBytes.size() / sizeof(uint32_t));
+      memcpy(NewFeatures.data(), FeatureBytes.data(), FeatureBytes.size());
+      for (auto Ft : NewFeatures) {
+
+        if (!Features.count(Ft)) {
+
+          MergeCandidates.push_back(F);
+          break;
+
+        }
+
+      }
+
+    }
+
+    // if (!FilesToAdd.empty() || Job->ExitCode != 0)
+    Printf(
+        "#%zd: cov: %zd ft: %zd corp: %zd exec/s %zd "
+        "oom/timeout/crash: %zd/%zd/%zd time: %zds job: %zd dft_time: %d\n",
+        NumRuns, Cov.size(), Features.size(), Files.size(),
+        Stats.average_exec_per_sec, NumOOMs, NumTimeouts, NumCrashes,
+        secondsSinceProcessStartUp(), Job->JobId, Job->DftTimeInSeconds);
+
+    if (MergeCandidates.empty()) return;
+
+    Vector<std::string> FilesToAdd;
+    Set<uint32_t>       NewFeatures, NewCov;
+    CrashResistantMerge(Args, {}, MergeCandidates, &FilesToAdd, Features,
+                        &NewFeatures, Cov, &NewCov, Job->CFPath, false);
+    for (auto &Path : FilesToAdd) {
+
+      auto U = FileToVector(Path);
+      auto NewPath = DirPlusFile(MainCorpusDir, Hash(U));
+      WriteToFile(U, NewPath);
+      Files.push_back(NewPath);
+
+    }
+
+    Features.insert(NewFeatures.begin(), NewFeatures.end());
+    Cov.insert(NewCov.begin(), NewCov.end());
+    for (auto Idx : NewCov)
+      if (auto *TE = TPC.PCTableEntryByIdx(Idx))
+        if (TPC.PcIsFuncEntry(TE))
+          PrintPC("  NEW_FUNC: %p %F %L\n", "",
+                  TPC.GetNextInstructionPc(TE->PC));
+
+  }
+
+  void CollectDFT(const std::string &InputPath) {
+
+    if (DataFlowBinary.empty()) return;
+    if (!FilesWithDFT.insert(InputPath).second) return;
+    Command Cmd(Args);
+    Cmd.removeFlag("fork");
+    Cmd.removeFlag("runs");
+    Cmd.addFlag("data_flow_trace", DFTDir);
+    Cmd.addArgument(InputPath);
+    for (auto &C : CorpusDirs)  // Remove all corpora from the args.
+      Cmd.removeArgument(C);
+    Cmd.setOutputFile(DirPlusFile(TempDir, "dft.log"));
+    Cmd.combineOutAndErr();
+    // Printf("CollectDFT: %s\n", Cmd.toString().c_str());
+    ExecuteCommand(Cmd);
+
+  }
+
+};
+
+struct JobQueue {
+
+  std::queue<FuzzJob *>   Qu;
+  std::mutex              Mu;
+  std::condition_variable Cv;
+
+  void Push(FuzzJob *Job) {
+
+    {
+
+      std::lock_guard<std::mutex> Lock(Mu);
+      Qu.push(Job);
+
+    }
+
+    Cv.notify_one();
+
+  }
+
+  FuzzJob *Pop() {
+
+    std::unique_lock<std::mutex> Lk(Mu);
+    // std::lock_guard<std::mutex> Lock(Mu);
+    Cv.wait(Lk, [&] { return !Qu.empty(); });
+    assert(!Qu.empty());
+    auto Job = Qu.front();
+    Qu.pop();
+    return Job;
+
+  }
+
+};
+
+void WorkerThread(JobQueue *FuzzQ, JobQueue *MergeQ) {
+
+  while (auto Job = FuzzQ->Pop()) {
+
+    // Printf("WorkerThread: job %p\n", Job);
+    Job->ExitCode = ExecuteCommand(Job->Cmd);
+    MergeQ->Push(Job);
+
+  }
+
+}
+
+// This is just a skeleton of an experimental -fork=1 feature.
+void FuzzWithFork(Random &Rand, const FuzzingOptions &Options,
+                  const Vector<std::string> &Args,
+                  const Vector<std::string> &CorpusDirs, int NumJobs) {
+
+  Printf("INFO: -fork=%d: fuzzing in separate process(s)\n", NumJobs);
+
+  GlobalEnv Env;
+  Env.Args = Args;
+  Env.CorpusDirs = CorpusDirs;
+  Env.Rand = &Rand;
+  Env.Verbosity = Options.Verbosity;
+  Env.ProcessStartTime = std::chrono::system_clock::now();
+  Env.DataFlowBinary = Options.CollectDataFlow;
+
+  Vector<SizedFile> SeedFiles;
+  for (auto &Dir : CorpusDirs)
+    GetSizedFilesFromDir(Dir, &SeedFiles);
+  std::sort(SeedFiles.begin(), SeedFiles.end());
+  Env.TempDir = TempPath("FuzzWithFork", ".dir");
+  Env.DFTDir = DirPlusFile(Env.TempDir, "DFT");
+  RmDirRecursive(Env.TempDir);  // in case there is a leftover from old runs.
+  MkDir(Env.TempDir);
+  MkDir(Env.DFTDir);
+
+  if (CorpusDirs.empty())
+    MkDir(Env.MainCorpusDir = DirPlusFile(Env.TempDir, "C"));
+  else
+    Env.MainCorpusDir = CorpusDirs[0];
+
+  if (Options.KeepSeed) {
+
+    for (auto &File : SeedFiles)
+      Env.Files.push_back(File.File);
+
+  } else {
+
+    auto CFPath = DirPlusFile(Env.TempDir, "merge.txt");
+    CrashResistantMerge(Env.Args, {}, SeedFiles, &Env.Files, {}, &Env.Features,
+                        {}, &Env.Cov, CFPath, false);
+    RemoveFile(CFPath);
+
+  }
+
+  Printf("INFO: -fork=%d: %zd seed inputs, starting to fuzz in %s\n", NumJobs,
+         Env.Files.size(), Env.TempDir.c_str());
+
+  int ExitCode = 0;
+
+  JobQueue FuzzQ, MergeQ;
+
+  auto StopJobs = [&]() {
+
+    for (int i = 0; i < NumJobs; i++)
+      FuzzQ.Push(nullptr);
+    MergeQ.Push(nullptr);
+    WriteToFile(Unit({1}), Env.StopFile());
+
+  };
+
+  size_t              JobId = 1;
+  Vector<std::thread> Threads;
+  for (int t = 0; t < NumJobs; t++) {
+
+    Threads.push_back(std::thread(WorkerThread, &FuzzQ, &MergeQ));
+    FuzzQ.Push(Env.CreateNewJob(JobId++));
+
+  }
+
+  while (true) {
+
+    std::unique_ptr<FuzzJob> Job(MergeQ.Pop());
+    if (!Job) break;
+    ExitCode = Job->ExitCode;
+    if (ExitCode == Options.InterruptExitCode) {
+
+      Printf("==%lu== libFuzzer: a child was interrupted; exiting\n", GetPid());
+      StopJobs();
+      break;
+
+    }
+
+    Fuzzer::MaybeExitGracefully();
+
+    Env.RunOneMergeJob(Job.get());
+
+    // Continue if our crash is one of the ignorred ones.
+    if (Options.IgnoreTimeouts && ExitCode == Options.TimeoutExitCode)
+      Env.NumTimeouts++;
+    else if (Options.IgnoreOOMs && ExitCode == Options.OOMExitCode)
+      Env.NumOOMs++;
+    else if (ExitCode != 0) {
+
+      Env.NumCrashes++;
+      if (Options.IgnoreCrashes) {
+
+        std::ifstream In(Job->LogPath);
+        std::string   Line;
+        while (std::getline(In, Line, '\n'))
+          if (Line.find("ERROR:") != Line.npos ||
+              Line.find("runtime error:") != Line.npos)
+            Printf("%s\n", Line.c_str());
+
+      } else {
+
+        // And exit if we don't ignore this crash.
+        Printf("INFO: log from the inner process:\n%s",
+               FileToString(Job->LogPath).c_str());
+        StopJobs();
+        break;
+
+      }
+
+    }
+
+    // Stop if we are over the time budget.
+    // This is not precise, since other threads are still running
+    // and we will wait while joining them.
+    // We also don't stop instantly: other jobs need to finish.
+    if (Options.MaxTotalTimeSec > 0 &&
+        Env.secondsSinceProcessStartUp() >= (size_t)Options.MaxTotalTimeSec) {
+
+      Printf("INFO: fuzzed for %zd seconds, wrapping up soon\n",
+             Env.secondsSinceProcessStartUp());
+      StopJobs();
+      break;
+
+    }
+
+    if (Env.NumRuns >= Options.MaxNumberOfRuns) {
+
+      Printf("INFO: fuzzed for %zd iterations, wrapping up soon\n",
+             Env.NumRuns);
+      StopJobs();
+      break;
+
+    }
+
+    FuzzQ.Push(Env.CreateNewJob(JobId++));
+
+  }
+
+  for (auto &T : Threads)
+    T.join();
+
+  // The workers have terminated. Don't try to remove the directory before they
+  // terminate to avoid a race condition preventing cleanup on Windows.
+  RmDirRecursive(Env.TempDir);
+
+  // Use the exit code from the last child process.
+  Printf("INFO: exiting: %d time: %zds\n", ExitCode,
+         Env.secondsSinceProcessStartUp());
+  exit(ExitCode);
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerFork.h

@@ -0,0 +1,24 @@
+//===- FuzzerFork.h - run fuzzing in sub-processes --------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_FORK_H
+#define LLVM_FUZZER_FORK_H
+
+#include "FuzzerDefs.h"
+#include "FuzzerOptions.h"
+#include "FuzzerRandom.h"
+
+#include <string>
+
+namespace fuzzer {
+void FuzzWithFork(Random &Rand, const FuzzingOptions &Options,
+                  const Vector<std::string> &Args,
+                  const Vector<std::string> &CorpusDirs, int NumJobs);
+} // namespace fuzzer
+
+#endif // LLVM_FUZZER_FORK_H

custom_mutators/libfuzzer/FuzzerIO.cpp

@@ -0,0 +1,248 @@
+//===- FuzzerIO.cpp - IO utils. -------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// IO functions.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerDefs.h"
+#include "FuzzerExtFunctions.h"
+#include "FuzzerIO.h"
+#include "FuzzerUtil.h"
+#include <algorithm>
+#include <cstdarg>
+#include <fstream>
+#include <iterator>
+#include <sys/stat.h>
+#include <sys/types.h>
+
+namespace fuzzer {
+
+static FILE *OutputFile = stderr;
+
+long GetEpoch(const std::string &Path) {
+
+  struct stat St;
+  if (stat(Path.c_str(), &St)) return 0;  // Can't stat, be conservative.
+  return St.st_mtime;
+
+}
+
+Unit FileToVector(const std::string &Path, size_t MaxSize, bool ExitOnError) {
+
+  std::ifstream T(Path, std::ios::binary);
+  if (ExitOnError && !T) {
+
+    Printf("No such directory: %s; exiting\n", Path.c_str());
+    exit(1);
+
+  }
+
+  T.seekg(0, T.end);
+  auto EndPos = T.tellg();
+  if (EndPos < 0) return {};
+  size_t FileLen = EndPos;
+  if (MaxSize) FileLen = std::min(FileLen, MaxSize);
+
+  T.seekg(0, T.beg);
+  Unit Res(FileLen);
+  T.read(reinterpret_cast<char *>(Res.data()), FileLen);
+  return Res;
+
+}
+
+std::string FileToString(const std::string &Path) {
+
+  std::ifstream T(Path, std::ios::binary);
+  return std::string((std::istreambuf_iterator<char>(T)),
+                     std::istreambuf_iterator<char>());
+
+}
+
+void CopyFileToErr(const std::string &Path) {
+
+  Printf("%s", FileToString(Path).c_str());
+
+}
+
+void WriteToFile(const Unit &U, const std::string &Path) {
+
+  WriteToFile(U.data(), U.size(), Path);
+
+}
+
+void WriteToFile(const std::string &Data, const std::string &Path) {
+
+  WriteToFile(reinterpret_cast<const uint8_t *>(Data.c_str()), Data.size(),
+              Path);
+
+}
+
+void WriteToFile(const uint8_t *Data, size_t Size, const std::string &Path) {
+
+  // Use raw C interface because this function may be called from a sig handler.
+  FILE *Out = fopen(Path.c_str(), "wb");
+  if (!Out) return;
+  fwrite(Data, sizeof(Data[0]), Size, Out);
+  fclose(Out);
+
+}
+
+void AppendToFile(const std::string &Data, const std::string &Path) {
+
+  AppendToFile(reinterpret_cast<const uint8_t *>(Data.data()), Data.size(),
+               Path);
+
+}
+
+void AppendToFile(const uint8_t *Data, size_t Size, const std::string &Path) {
+
+  FILE *Out = fopen(Path.c_str(), "a");
+  if (!Out) return;
+  fwrite(Data, sizeof(Data[0]), Size, Out);
+  fclose(Out);
+
+}
+
+void ReadDirToVectorOfUnits(const char *Path, Vector<Unit> *V, long *Epoch,
+                            size_t MaxSize, bool ExitOnError) {
+
+  long                E = Epoch ? *Epoch : 0;
+  Vector<std::string> Files;
+  ListFilesInDirRecursive(Path, Epoch, &Files, /*TopDir*/ true);
+  size_t NumLoaded = 0;
+  for (size_t i = 0; i < Files.size(); i++) {
+
+    auto &X = Files[i];
+    if (Epoch && GetEpoch(X) < E) continue;
+    NumLoaded++;
+    if ((NumLoaded & (NumLoaded - 1)) == 0 && NumLoaded >= 1024)
+      Printf("Loaded %zd/%zd files from %s\n", NumLoaded, Files.size(), Path);
+    auto S = FileToVector(X, MaxSize, ExitOnError);
+    if (!S.empty()) V->push_back(S);
+
+  }
+
+}
+
+void GetSizedFilesFromDir(const std::string &Dir, Vector<SizedFile> *V) {
+
+  Vector<std::string> Files;
+  ListFilesInDirRecursive(Dir, 0, &Files, /*TopDir*/ true);
+  for (auto &File : Files)
+    if (size_t Size = FileSize(File)) V->push_back({File, Size});
+
+}
+
+std::string DirPlusFile(const std::string &DirPath,
+                        const std::string &FileName) {
+
+  return DirPath + GetSeparator() + FileName;
+
+}
+
+void DupAndCloseStderr() {
+
+  int OutputFd = DuplicateFile(2);
+  if (OutputFd >= 0) {
+
+    FILE *NewOutputFile = OpenFile(OutputFd, "w");
+    if (NewOutputFile) {
+
+      OutputFile = NewOutputFile;
+      if (EF->__sanitizer_set_report_fd)
+        EF->__sanitizer_set_report_fd(
+            reinterpret_cast<void *>(GetHandleFromFd(OutputFd)));
+      DiscardOutput(2);
+
+    }
+
+  }
+
+}
+
+void CloseStdout() {
+
+  DiscardOutput(1);
+
+}
+
+void Printf(const char *Fmt, ...) {
+
+  va_list ap;
+  va_start(ap, Fmt);
+  vfprintf(OutputFile, Fmt, ap);
+  va_end(ap);
+  fflush(OutputFile);
+
+}
+
+void VPrintf(bool Verbose, const char *Fmt, ...) {
+
+  if (!Verbose) return;
+  va_list ap;
+  va_start(ap, Fmt);
+  vfprintf(OutputFile, Fmt, ap);
+  va_end(ap);
+  fflush(OutputFile);
+
+}
+
+static bool MkDirRecursiveInner(const std::string &Leaf) {
+
+  // Prevent chance of potential infinite recursion
+  if (Leaf == ".") return true;
+
+  const std::string &Dir = DirName(Leaf);
+
+  if (IsDirectory(Dir)) {
+
+    MkDir(Leaf);
+    return IsDirectory(Leaf);
+
+  }
+
+  bool ret = MkDirRecursiveInner(Dir);
+  if (!ret) {
+
+    // Give up early if a previous MkDir failed
+    return ret;
+
+  }
+
+  MkDir(Leaf);
+  return IsDirectory(Leaf);
+
+}
+
+bool MkDirRecursive(const std::string &Dir) {
+
+  if (Dir.empty()) return false;
+
+  if (IsDirectory(Dir)) return true;
+
+  return MkDirRecursiveInner(Dir);
+
+}
+
+void RmDirRecursive(const std::string &Dir) {
+
+  IterateDirRecursive(
+      Dir, [](const std::string &Path) {},
+      [](const std::string &Path) { RmDir(Path); },
+      [](const std::string &Path) { RemoveFile(Path); });
+
+}
+
+std::string TempPath(const char *Prefix, const char *Extension) {
+
+  return DirPlusFile(TmpDir(), std::string("libFuzzerTemp.") + Prefix +
+                                   std::to_string(GetPid()) + Extension);
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerIO.h

@@ -0,0 +1,112 @@
+//===- FuzzerIO.h - Internal header for IO utils ----------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// IO interface.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_IO_H
+#define LLVM_FUZZER_IO_H
+
+#include "FuzzerDefs.h"
+
+namespace fuzzer {
+
+long GetEpoch(const std::string &Path);
+
+Unit FileToVector(const std::string &Path, size_t MaxSize = 0,
+                  bool ExitOnError = true);
+
+std::string FileToString(const std::string &Path);
+
+void CopyFileToErr(const std::string &Path);
+
+void WriteToFile(const uint8_t *Data, size_t Size, const std::string &Path);
+// Write Data.c_str() to the file without terminating null character.
+void WriteToFile(const std::string &Data, const std::string &Path);
+void WriteToFile(const Unit &U, const std::string &Path);
+
+void AppendToFile(const uint8_t *Data, size_t Size, const std::string &Path);
+void AppendToFile(const std::string &Data, const std::string &Path);
+
+void ReadDirToVectorOfUnits(const char *Path, Vector<Unit> *V,
+                            long *Epoch, size_t MaxSize, bool ExitOnError);
+
+// Returns "Dir/FileName" or equivalent for the current OS.
+std::string DirPlusFile(const std::string &DirPath,
+                        const std::string &FileName);
+
+// Returns the name of the dir, similar to the 'dirname' utility.
+std::string DirName(const std::string &FileName);
+
+// Returns path to a TmpDir.
+std::string TmpDir();
+
+std::string TempPath(const char *Prefix, const char *Extension);
+
+bool IsInterestingCoverageFile(const std::string &FileName);
+
+void DupAndCloseStderr();
+
+void CloseStdout();
+
+void Printf(const char *Fmt, ...);
+void VPrintf(bool Verbose, const char *Fmt, ...);
+
+// Print using raw syscalls, useful when printing at early init stages.
+void RawPrint(const char *Str);
+
+// Platform specific functions:
+bool IsFile(const std::string &Path);
+bool IsDirectory(const std::string &Path);
+size_t FileSize(const std::string &Path);
+
+void ListFilesInDirRecursive(const std::string &Dir, long *Epoch,
+                             Vector<std::string> *V, bool TopDir);
+
+bool MkDirRecursive(const std::string &Dir);
+void RmDirRecursive(const std::string &Dir);
+
+// Iterate files and dirs inside Dir, recursively.
+// Call DirPreCallback/DirPostCallback on dirs before/after
+// calling FileCallback on files.
+void IterateDirRecursive(const std::string &Dir,
+                         void (*DirPreCallback)(const std::string &Dir),
+                         void (*DirPostCallback)(const std::string &Dir),
+                         void (*FileCallback)(const std::string &Dir));
+
+struct SizedFile {
+  std::string File;
+  size_t Size;
+  bool operator<(const SizedFile &B) const { return Size < B.Size; }
+};
+
+void GetSizedFilesFromDir(const std::string &Dir, Vector<SizedFile> *V);
+
+char GetSeparator();
+bool IsSeparator(char C);
+// Similar to the basename utility: returns the file name w/o the dir prefix.
+std::string Basename(const std::string &Path);
+
+FILE* OpenFile(int Fd, const char *Mode);
+
+int CloseFile(int Fd);
+
+int DuplicateFile(int Fd);
+
+void RemoveFile(const std::string &Path);
+void RenameFile(const std::string &OldPath, const std::string &NewPath);
+
+intptr_t GetHandleFromFd(int fd);
+
+void MkDir(const std::string &Path);
+void RmDir(const std::string &Path);
+
+const std::string &getDevNull();
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_IO_H

custom_mutators/libfuzzer/FuzzerIOPosix.cpp

@@ -0,0 +1,223 @@
+//===- FuzzerIOPosix.cpp - IO utils for Posix. ----------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// IO functions implementation using Posix API.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_POSIX || LIBFUZZER_FUCHSIA
+
+  #include "FuzzerExtFunctions.h"
+  #include "FuzzerIO.h"
+  #include <cstdarg>
+  #include <cstdio>
+  #include <dirent.h>
+  #include <fstream>
+  #include <iterator>
+  #include <libgen.h>
+  #include <sys/stat.h>
+  #include <sys/types.h>
+  #include <unistd.h>
+
+namespace fuzzer {
+
+bool IsFile(const std::string &Path) {
+
+  struct stat St;
+  if (stat(Path.c_str(), &St)) return false;
+  return S_ISREG(St.st_mode);
+
+}
+
+bool IsDirectory(const std::string &Path) {
+
+  struct stat St;
+  if (stat(Path.c_str(), &St)) return false;
+  return S_ISDIR(St.st_mode);
+
+}
+
+size_t FileSize(const std::string &Path) {
+
+  struct stat St;
+  if (stat(Path.c_str(), &St)) return 0;
+  return St.st_size;
+
+}
+
+std::string Basename(const std::string &Path) {
+
+  size_t Pos = Path.rfind(GetSeparator());
+  if (Pos == std::string::npos) return Path;
+  assert(Pos < Path.size());
+  return Path.substr(Pos + 1);
+
+}
+
+void ListFilesInDirRecursive(const std::string &Dir, long *Epoch,
+                             Vector<std::string> *V, bool TopDir) {
+
+  auto E = GetEpoch(Dir);
+  if (Epoch)
+    if (E && *Epoch >= E) return;
+
+  DIR *D = opendir(Dir.c_str());
+  if (!D) {
+
+    Printf("%s: %s; exiting\n", strerror(errno), Dir.c_str());
+    exit(1);
+
+  }
+
+  while (auto E = readdir(D)) {
+
+    std::string Path = DirPlusFile(Dir, E->d_name);
+    if (E->d_type == DT_REG || E->d_type == DT_LNK ||
+        (E->d_type == DT_UNKNOWN && IsFile(Path)))
+      V->push_back(Path);
+    else if ((E->d_type == DT_DIR ||
+              (E->d_type == DT_UNKNOWN && IsDirectory(Path))) &&
+             *E->d_name != '.')
+      ListFilesInDirRecursive(Path, Epoch, V, false);
+
+  }
+
+  closedir(D);
+  if (Epoch && TopDir) *Epoch = E;
+
+}
+
+void IterateDirRecursive(const std::string &Dir,
+                         void (*DirPreCallback)(const std::string &Dir),
+                         void (*DirPostCallback)(const std::string &Dir),
+                         void (*FileCallback)(const std::string &Dir)) {
+
+  DirPreCallback(Dir);
+  DIR *D = opendir(Dir.c_str());
+  if (!D) return;
+  while (auto E = readdir(D)) {
+
+    std::string Path = DirPlusFile(Dir, E->d_name);
+    if (E->d_type == DT_REG || E->d_type == DT_LNK ||
+        (E->d_type == DT_UNKNOWN && IsFile(Path)))
+      FileCallback(Path);
+    else if ((E->d_type == DT_DIR ||
+              (E->d_type == DT_UNKNOWN && IsDirectory(Path))) &&
+             *E->d_name != '.')
+      IterateDirRecursive(Path, DirPreCallback, DirPostCallback, FileCallback);
+
+  }
+
+  closedir(D);
+  DirPostCallback(Dir);
+
+}
+
+char GetSeparator() {
+
+  return '/';
+
+}
+
+bool IsSeparator(char C) {
+
+  return C == '/';
+
+}
+
+FILE *OpenFile(int Fd, const char *Mode) {
+
+  return fdopen(Fd, Mode);
+
+}
+
+int CloseFile(int fd) {
+
+  return close(fd);
+
+}
+
+int DuplicateFile(int Fd) {
+
+  return dup(Fd);
+
+}
+
+void RemoveFile(const std::string &Path) {
+
+  unlink(Path.c_str());
+
+}
+
+void RenameFile(const std::string &OldPath, const std::string &NewPath) {
+
+  rename(OldPath.c_str(), NewPath.c_str());
+
+}
+
+intptr_t GetHandleFromFd(int fd) {
+
+  return static_cast<intptr_t>(fd);
+
+}
+
+std::string DirName(const std::string &FileName) {
+
+  char *Tmp = new char[FileName.size() + 1];
+  memcpy(Tmp, FileName.c_str(), FileName.size() + 1);
+  std::string Res = dirname(Tmp);
+  delete[] Tmp;
+  return Res;
+
+}
+
+std::string TmpDir() {
+
+  if (auto Env = getenv("TMPDIR")) return Env;
+  return "/tmp";
+
+}
+
+bool IsInterestingCoverageFile(const std::string &FileName) {
+
+  if (FileName.find("compiler-rt/lib/") != std::string::npos)
+    return false;  // sanitizer internal.
+  if (FileName.find("/usr/lib/") != std::string::npos) return false;
+  if (FileName.find("/usr/include/") != std::string::npos) return false;
+  if (FileName == "<null>") return false;
+  return true;
+
+}
+
+void RawPrint(const char *Str) {
+
+  write(2, Str, strlen(Str));
+
+}
+
+void MkDir(const std::string &Path) {
+
+  mkdir(Path.c_str(), 0700);
+
+}
+
+void RmDir(const std::string &Path) {
+
+  rmdir(Path.c_str());
+
+}
+
+const std::string &getDevNull() {
+
+  static const std::string devNull = "/dev/null";
+  return devNull;
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_POSIX
+

custom_mutators/libfuzzer/FuzzerIOWindows.cpp

@@ -0,0 +1,513 @@
+//===- FuzzerIOWindows.cpp - IO utils for Windows. ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// IO functions implementation for Windows.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_WINDOWS
+
+  #include "FuzzerExtFunctions.h"
+  #include "FuzzerIO.h"
+  #include <cstdarg>
+  #include <cstdio>
+  #include <fstream>
+  #include <io.h>
+  #include <iterator>
+  #include <sys/stat.h>
+  #include <sys/types.h>
+  #include <windows.h>
+
+namespace fuzzer {
+
+static bool IsFile(const std::string &Path, const DWORD &FileAttributes) {
+
+  if (FileAttributes & FILE_ATTRIBUTE_NORMAL) return true;
+
+  if (FileAttributes & FILE_ATTRIBUTE_DIRECTORY) return false;
+
+  HANDLE FileHandle(CreateFileA(Path.c_str(), 0, FILE_SHARE_READ, NULL,
+                                OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, 0));
+
+  if (FileHandle == INVALID_HANDLE_VALUE) {
+
+    Printf("CreateFileA() failed for \"%s\" (Error code: %lu).\n", Path.c_str(),
+           GetLastError());
+    return false;
+
+  }
+
+  DWORD FileType = GetFileType(FileHandle);
+
+  if (FileType == FILE_TYPE_UNKNOWN) {
+
+    Printf("GetFileType() failed for \"%s\" (Error code: %lu).\n", Path.c_str(),
+           GetLastError());
+    CloseHandle(FileHandle);
+    return false;
+
+  }
+
+  if (FileType != FILE_TYPE_DISK) {
+
+    CloseHandle(FileHandle);
+    return false;
+
+  }
+
+  CloseHandle(FileHandle);
+  return true;
+
+}
+
+bool IsFile(const std::string &Path) {
+
+  DWORD Att = GetFileAttributesA(Path.c_str());
+
+  if (Att == INVALID_FILE_ATTRIBUTES) {
+
+    Printf("GetFileAttributesA() failed for \"%s\" (Error code: %lu).\n",
+           Path.c_str(), GetLastError());
+    return false;
+
+  }
+
+  return IsFile(Path, Att);
+
+}
+
+static bool IsDir(DWORD FileAttrs) {
+
+  if (FileAttrs == INVALID_FILE_ATTRIBUTES) return false;
+  return FileAttrs & FILE_ATTRIBUTE_DIRECTORY;
+
+}
+
+bool IsDirectory(const std::string &Path) {
+
+  DWORD Att = GetFileAttributesA(Path.c_str());
+
+  if (Att == INVALID_FILE_ATTRIBUTES) {
+
+    Printf("GetFileAttributesA() failed for \"%s\" (Error code: %lu).\n",
+           Path.c_str(), GetLastError());
+    return false;
+
+  }
+
+  return IsDir(Att);
+
+}
+
+std::string Basename(const std::string &Path) {
+
+  size_t Pos = Path.find_last_of("/\\");
+  if (Pos == std::string::npos) return Path;
+  assert(Pos < Path.size());
+  return Path.substr(Pos + 1);
+
+}
+
+size_t FileSize(const std::string &Path) {
+
+  WIN32_FILE_ATTRIBUTE_DATA attr;
+  if (!GetFileAttributesExA(Path.c_str(), GetFileExInfoStandard, &attr)) {
+
+    DWORD LastError = GetLastError();
+    if (LastError != ERROR_FILE_NOT_FOUND)
+      Printf("GetFileAttributesExA() failed for \"%s\" (Error code: %lu).\n",
+             Path.c_str(), LastError);
+    return 0;
+
+  }
+
+  ULARGE_INTEGER size;
+  size.HighPart = attr.nFileSizeHigh;
+  size.LowPart = attr.nFileSizeLow;
+  return size.QuadPart;
+
+}
+
+void ListFilesInDirRecursive(const std::string &Dir, long *Epoch,
+                             Vector<std::string> *V, bool TopDir) {
+
+  auto E = GetEpoch(Dir);
+  if (Epoch)
+    if (E && *Epoch >= E) return;
+
+  std::string Path(Dir);
+  assert(!Path.empty());
+  if (Path.back() != '\\') Path.push_back('\\');
+  Path.push_back('*');
+
+  // Get the first directory entry.
+  WIN32_FIND_DATAA FindInfo;
+  HANDLE           FindHandle(FindFirstFileA(Path.c_str(), &FindInfo));
+  if (FindHandle == INVALID_HANDLE_VALUE) {
+
+    if (GetLastError() == ERROR_FILE_NOT_FOUND) return;
+    Printf("No such file or directory: %s; exiting\n", Dir.c_str());
+    exit(1);
+
+  }
+
+  do {
+
+    std::string FileName = DirPlusFile(Dir, FindInfo.cFileName);
+
+    if (FindInfo.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) {
+
+      size_t FilenameLen = strlen(FindInfo.cFileName);
+      if ((FilenameLen == 1 && FindInfo.cFileName[0] == '.') ||
+          (FilenameLen == 2 && FindInfo.cFileName[0] == '.' &&
+           FindInfo.cFileName[1] == '.'))
+        continue;
+
+      ListFilesInDirRecursive(FileName, Epoch, V, false);
+
+    } else if (IsFile(FileName, FindInfo.dwFileAttributes))
+
+      V->push_back(FileName);
+
+  } while (FindNextFileA(FindHandle, &FindInfo));
+
+  DWORD LastError = GetLastError();
+  if (LastError != ERROR_NO_MORE_FILES)
+    Printf("FindNextFileA failed (Error code: %lu).\n", LastError);
+
+  FindClose(FindHandle);
+
+  if (Epoch && TopDir) *Epoch = E;
+
+}
+
+void IterateDirRecursive(const std::string &Dir,
+                         void (*DirPreCallback)(const std::string &Dir),
+                         void (*DirPostCallback)(const std::string &Dir),
+                         void (*FileCallback)(const std::string &Dir)) {
+
+  // TODO(metzman): Implement ListFilesInDirRecursive via this function.
+  DirPreCallback(Dir);
+
+  DWORD DirAttrs = GetFileAttributesA(Dir.c_str());
+  if (!IsDir(DirAttrs)) return;
+
+  std::string TargetDir(Dir);
+  assert(!TargetDir.empty());
+  if (TargetDir.back() != '\\') TargetDir.push_back('\\');
+  TargetDir.push_back('*');
+
+  WIN32_FIND_DATAA FindInfo;
+  // Find the directory's first file.
+  HANDLE FindHandle = FindFirstFileA(TargetDir.c_str(), &FindInfo);
+  if (FindHandle == INVALID_HANDLE_VALUE) {
+
+    DWORD LastError = GetLastError();
+    if (LastError != ERROR_FILE_NOT_FOUND) {
+
+      // If the directory isn't empty, then something abnormal is going on.
+      Printf("FindFirstFileA failed for %s (Error code: %lu).\n", Dir.c_str(),
+             LastError);
+
+    }
+
+    return;
+
+  }
+
+  do {
+
+    std::string Path = DirPlusFile(Dir, FindInfo.cFileName);
+    DWORD       PathAttrs = FindInfo.dwFileAttributes;
+    if (IsDir(PathAttrs)) {
+
+      // Is Path the current directory (".") or the parent ("..")?
+      if (strcmp(FindInfo.cFileName, ".") == 0 ||
+          strcmp(FindInfo.cFileName, "..") == 0)
+        continue;
+      IterateDirRecursive(Path, DirPreCallback, DirPostCallback, FileCallback);
+
+    } else if (PathAttrs != INVALID_FILE_ATTRIBUTES) {
+
+      FileCallback(Path);
+
+    }
+
+  } while (FindNextFileA(FindHandle, &FindInfo));
+
+  DWORD LastError = GetLastError();
+  if (LastError != ERROR_NO_MORE_FILES)
+    Printf("FindNextFileA failed for %s (Error code: %lu).\n", Dir.c_str(),
+           LastError);
+
+  FindClose(FindHandle);
+  DirPostCallback(Dir);
+
+}
+
+char GetSeparator() {
+
+  return '\\';
+
+}
+
+FILE *OpenFile(int Fd, const char *Mode) {
+
+  return _fdopen(Fd, Mode);
+
+}
+
+int CloseFile(int Fd) {
+
+  return _close(Fd);
+
+}
+
+int DuplicateFile(int Fd) {
+
+  return _dup(Fd);
+
+}
+
+void RemoveFile(const std::string &Path) {
+
+  _unlink(Path.c_str());
+
+}
+
+void RenameFile(const std::string &OldPath, const std::string &NewPath) {
+
+  rename(OldPath.c_str(), NewPath.c_str());
+
+}
+
+intptr_t GetHandleFromFd(int fd) {
+
+  return _get_osfhandle(fd);
+
+}
+
+bool IsSeparator(char C) {
+
+  return C == '\\' || C == '/';
+
+}
+
+// Parse disk designators, like "C:\". If Relative == true, also accepts: "C:".
+// Returns number of characters considered if successful.
+static size_t ParseDrive(const std::string &FileName, const size_t Offset,
+                         bool Relative = true) {
+
+  if (Offset + 1 >= FileName.size() || FileName[Offset + 1] != ':') return 0;
+  if (Offset + 2 >= FileName.size() || !IsSeparator(FileName[Offset + 2])) {
+
+    if (!Relative)  // Accept relative path?
+      return 0;
+    else
+      return 2;
+
+  }
+
+  return 3;
+
+}
+
+// Parse a file name, like: SomeFile.txt
+// Returns number of characters considered if successful.
+static size_t ParseFileName(const std::string &FileName, const size_t Offset) {
+
+  size_t       Pos = Offset;
+  const size_t End = FileName.size();
+  for (; Pos < End && !IsSeparator(FileName[Pos]); ++Pos)
+    ;
+  return Pos - Offset;
+
+}
+
+// Parse a directory ending in separator, like: `SomeDir\`
+// Returns number of characters considered if successful.
+static size_t ParseDir(const std::string &FileName, const size_t Offset) {
+
+  size_t       Pos = Offset;
+  const size_t End = FileName.size();
+  if (Pos >= End || IsSeparator(FileName[Pos])) return 0;
+  for (; Pos < End && !IsSeparator(FileName[Pos]); ++Pos)
+    ;
+  if (Pos >= End) return 0;
+  ++Pos;  // Include separator.
+  return Pos - Offset;
+
+}
+
+// Parse a servername and share, like: `SomeServer\SomeShare\`
+// Returns number of characters considered if successful.
+static size_t ParseServerAndShare(const std::string &FileName,
+                                  const size_t       Offset) {
+
+  size_t Pos = Offset, Res;
+  if (!(Res = ParseDir(FileName, Pos))) return 0;
+  Pos += Res;
+  if (!(Res = ParseDir(FileName, Pos))) return 0;
+  Pos += Res;
+  return Pos - Offset;
+
+}
+
+// Parse the given Ref string from the position Offset, to exactly match the
+// given string Patt. Returns number of characters considered if successful.
+static size_t ParseCustomString(const std::string &Ref, size_t Offset,
+                                const char *Patt) {
+
+  size_t Len = strlen(Patt);
+  if (Offset + Len > Ref.size()) return 0;
+  return Ref.compare(Offset, Len, Patt) == 0 ? Len : 0;
+
+}
+
+// Parse a location, like:
+// \\?\UNC\Server\Share\  \\?\C:\  \\Server\Share\  \  C:\  C:
+// Returns number of characters considered if successful.
+static size_t ParseLocation(const std::string &FileName) {
+
+  size_t Pos = 0, Res;
+
+  if ((Res = ParseCustomString(FileName, Pos, R"(\\?\)"))) {
+
+    Pos += Res;
+    if ((Res = ParseCustomString(FileName, Pos, R"(UNC\)"))) {
+
+      Pos += Res;
+      if ((Res = ParseServerAndShare(FileName, Pos))) return Pos + Res;
+      return 0;
+
+    }
+
+    if ((Res = ParseDrive(FileName, Pos, false))) return Pos + Res;
+    return 0;
+
+  }
+
+  if (Pos < FileName.size() && IsSeparator(FileName[Pos])) {
+
+    ++Pos;
+    if (Pos < FileName.size() && IsSeparator(FileName[Pos])) {
+
+      ++Pos;
+      if ((Res = ParseServerAndShare(FileName, Pos))) return Pos + Res;
+      return 0;
+
+    }
+
+    return Pos;
+
+  }
+
+  if ((Res = ParseDrive(FileName, Pos))) return Pos + Res;
+
+  return Pos;
+
+}
+
+std::string DirName(const std::string &FileName) {
+
+  size_t LocationLen = ParseLocation(FileName);
+  size_t DirLen = 0, Res;
+  while ((Res = ParseDir(FileName, LocationLen + DirLen)))
+    DirLen += Res;
+  size_t FileLen = ParseFileName(FileName, LocationLen + DirLen);
+
+  if (LocationLen + DirLen + FileLen != FileName.size()) {
+
+    Printf("DirName() failed for \"%s\", invalid path.\n", FileName.c_str());
+    exit(1);
+
+  }
+
+  if (DirLen) {
+
+    --DirLen;        // Remove trailing separator.
+    if (!FileLen) {  // Path ended in separator.
+      assert(DirLen);
+      // Remove file name from Dir.
+      while (DirLen && !IsSeparator(FileName[LocationLen + DirLen - 1]))
+        --DirLen;
+      if (DirLen)  // Remove trailing separator.
+        --DirLen;
+
+    }
+
+  }
+
+  if (!LocationLen) {  // Relative path.
+    if (!DirLen) return ".";
+    return std::string(".\\").append(FileName, 0, DirLen);
+
+  }
+
+  return FileName.substr(0, LocationLen + DirLen);
+
+}
+
+std::string TmpDir() {
+
+  std::string Tmp;
+  Tmp.resize(MAX_PATH + 1);
+  DWORD Size = GetTempPathA(Tmp.size(), &Tmp[0]);
+  if (Size == 0) {
+
+    Printf("Couldn't get Tmp path.\n");
+    exit(1);
+
+  }
+
+  Tmp.resize(Size);
+  return Tmp;
+
+}
+
+bool IsInterestingCoverageFile(const std::string &FileName) {
+
+  if (FileName.find("Program Files") != std::string::npos) return false;
+  if (FileName.find("compiler-rt\\lib\\") != std::string::npos)
+    return false;  // sanitizer internal.
+  if (FileName == "<null>") return false;
+  return true;
+
+}
+
+void RawPrint(const char *Str) {
+
+  _write(2, Str, strlen(Str));
+
+}
+
+void MkDir(const std::string &Path) {
+
+  if (CreateDirectoryA(Path.c_str(), nullptr)) return;
+  Printf("CreateDirectoryA failed for %s (Error code: %lu).\n", Path.c_str(),
+         GetLastError());
+
+}
+
+void RmDir(const std::string &Path) {
+
+  if (RemoveDirectoryA(Path.c_str())) return;
+  Printf("RemoveDirectoryA failed for %s (Error code: %lu).\n", Path.c_str(),
+         GetLastError());
+
+}
+
+const std::string &getDevNull() {
+
+  static const std::string devNull = "NUL";
+  return devNull;
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_WINDOWS
+

custom_mutators/libfuzzer/FuzzerInterface.h

@@ -0,0 +1,79 @@
+//===- FuzzerInterface.h - Interface header for the Fuzzer ------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Define the interface between libFuzzer and the library being tested.
+//===----------------------------------------------------------------------===//
+
+// NOTE: the libFuzzer interface is thin and in the majority of cases
+// you should not include this file into your target. In 95% of cases
+// all you need is to define the following function in your file:
+// extern "C" int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size);
+
+// WARNING: keep the interface in C.
+
+#ifndef LLVM_FUZZER_INTERFACE_H
+#define LLVM_FUZZER_INTERFACE_H
+
+#include <stddef.h>
+#include <stdint.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif  // __cplusplus
+
+// Define FUZZER_INTERFACE_VISIBILITY to set default visibility in a way that
+// doesn't break MSVC.
+#if defined(_WIN32)
+#define FUZZER_INTERFACE_VISIBILITY __declspec(dllexport)
+#else
+#define FUZZER_INTERFACE_VISIBILITY __attribute__((visibility("default")))
+#endif
+
+// Mandatory user-provided target function.
+// Executes the code under test with [Data, Data+Size) as the input.
+// libFuzzer will invoke this function *many* times with different inputs.
+// Must return 0.
+FUZZER_INTERFACE_VISIBILITY int
+LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size);
+
+// Optional user-provided initialization function.
+// If provided, this function will be called by libFuzzer once at startup.
+// It may read and modify argc/argv.
+// Must return 0.
+FUZZER_INTERFACE_VISIBILITY int LLVMFuzzerInitialize(int *argc, char ***argv);
+
+// Optional user-provided custom mutator.
+// Mutates raw data in [Data, Data+Size) inplace.
+// Returns the new size, which is not greater than MaxSize.
+// Given the same Seed produces the same mutation.
+FUZZER_INTERFACE_VISIBILITY size_t
+LLVMFuzzerCustomMutator(uint8_t *Data, size_t Size, size_t MaxSize,
+                        unsigned int Seed);
+
+// Optional user-provided custom cross-over function.
+// Combines pieces of Data1 & Data2 together into Out.
+// Returns the new size, which is not greater than MaxOutSize.
+// Should produce the same mutation given the same Seed.
+FUZZER_INTERFACE_VISIBILITY size_t
+LLVMFuzzerCustomCrossOver(const uint8_t *Data1, size_t Size1,
+                          const uint8_t *Data2, size_t Size2, uint8_t *Out,
+                          size_t MaxOutSize, unsigned int Seed);
+
+// Experimental, may go away in future.
+// libFuzzer-provided function to be used inside LLVMFuzzerCustomMutator.
+// Mutates raw data in [Data, Data+Size) inplace.
+// Returns the new size, which is not greater than MaxSize.
+FUZZER_INTERFACE_VISIBILITY size_t
+LLVMFuzzerMutate(uint8_t *Data, size_t Size, size_t MaxSize);
+
+#undef FUZZER_INTERFACE_VISIBILITY
+
+#ifdef __cplusplus
+}  // extern "C"
+#endif  // __cplusplus
+
+#endif  // LLVM_FUZZER_INTERFACE_H

custom_mutators/libfuzzer/FuzzerInternal.h

@@ -0,0 +1,173 @@
+//===- FuzzerInternal.h - Internal header for the Fuzzer --------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Define the main class fuzzer::Fuzzer and most functions.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_INTERNAL_H
+#define LLVM_FUZZER_INTERNAL_H
+
+#include "FuzzerDataFlowTrace.h"
+#include "FuzzerDefs.h"
+#include "FuzzerExtFunctions.h"
+#include "FuzzerInterface.h"
+#include "FuzzerOptions.h"
+#include "FuzzerSHA1.h"
+#include "FuzzerValueBitMap.h"
+#include <algorithm>
+#include <atomic>
+#include <chrono>
+#include <climits>
+#include <cstdlib>
+#include <string.h>
+
+namespace fuzzer {
+
+using namespace std::chrono;
+
+class Fuzzer {
+public:
+
+  Fuzzer(UserCallback CB, InputCorpus &Corpus, MutationDispatcher &MD,
+         FuzzingOptions Options);
+  ~Fuzzer();
+  void Loop(Vector<SizedFile> &CorporaFiles);
+  void ReadAndExecuteSeedCorpora(Vector<SizedFile> &CorporaFiles);
+  void MinimizeCrashLoop(const Unit &U);
+  void RereadOutputCorpus(size_t MaxSize);
+
+  size_t secondsSinceProcessStartUp() {
+    return duration_cast<seconds>(system_clock::now() - ProcessStartTime)
+        .count();
+  }
+
+  bool TimedOut() {
+    return Options.MaxTotalTimeSec > 0 &&
+           secondsSinceProcessStartUp() >
+               static_cast<size_t>(Options.MaxTotalTimeSec);
+  }
+
+  size_t execPerSec() {
+    size_t Seconds = secondsSinceProcessStartUp();
+    return Seconds ? TotalNumberOfRuns / Seconds : 0;
+  }
+
+  size_t getTotalNumberOfRuns() { return TotalNumberOfRuns; }
+
+  static void StaticAlarmCallback();
+  static void StaticCrashSignalCallback();
+  static void StaticExitCallback();
+  static void StaticInterruptCallback();
+  static void StaticFileSizeExceedCallback();
+  static void StaticGracefulExitCallback();
+
+  void ExecuteCallback(const uint8_t *Data, size_t Size);
+  bool RunOne(const uint8_t *Data, size_t Size, bool MayDeleteFile = false,
+              InputInfo *II = nullptr, bool ForceAddToCorpus = false,
+              bool *FoundUniqFeatures = nullptr);
+
+  // Merge Corpora[1:] into Corpora[0].
+  void Merge(const Vector<std::string> &Corpora);
+  void CrashResistantMergeInternalStep(const std::string &ControlFilePath);
+  MutationDispatcher &GetMD() { return MD; }
+  void PrintFinalStats();
+  void SetMaxInputLen(size_t MaxInputLen);
+  void SetMaxMutationLen(size_t MaxMutationLen);
+  void RssLimitCallback();
+
+  bool InFuzzingThread() const { return IsMyThread; }
+  size_t GetCurrentUnitInFuzzingThead(const uint8_t **Data) const;
+  void TryDetectingAMemoryLeak(const uint8_t *Data, size_t Size,
+                               bool DuringInitialCorpusExecution);
+
+  void HandleMalloc(size_t Size);
+  static void MaybeExitGracefully();
+  std::string WriteToOutputCorpus(const Unit &U);
+
+private:
+  void AlarmCallback();
+  void CrashCallback();
+  void ExitCallback();
+  void CrashOnOverwrittenData();
+  void InterruptCallback();
+  void MutateAndTestOne();
+  void PurgeAllocator();
+  void ReportNewCoverage(InputInfo *II, const Unit &U);
+  void PrintPulseAndReportSlowInput(const uint8_t *Data, size_t Size);
+  void WriteUnitToFileWithPrefix(const Unit &U, const char *Prefix);
+  void PrintStats(const char *Where, const char *End = "\n", size_t Units = 0,
+                  size_t Features = 0);
+  void PrintStatusForNewUnit(const Unit &U, const char *Text);
+  void CheckExitOnSrcPosOrItem();
+
+  static void StaticDeathCallback();
+  void DumpCurrentUnit(const char *Prefix);
+  void DeathCallback();
+
+  void AllocateCurrentUnitData();
+  uint8_t *CurrentUnitData = nullptr;
+  std::atomic<size_t> CurrentUnitSize;
+  uint8_t BaseSha1[kSHA1NumBytes];  // Checksum of the base unit.
+
+  bool GracefulExitRequested = false;
+
+  size_t TotalNumberOfRuns = 0;
+  size_t NumberOfNewUnitsAdded = 0;
+
+  size_t LastCorpusUpdateRun = 0;
+
+  bool HasMoreMallocsThanFrees = false;
+  size_t NumberOfLeakDetectionAttempts = 0;
+
+  system_clock::time_point LastAllocatorPurgeAttemptTime = system_clock::now();
+
+  UserCallback CB;
+  InputCorpus &Corpus;
+  MutationDispatcher &MD;
+  FuzzingOptions Options;
+  DataFlowTrace DFT;
+
+  system_clock::time_point ProcessStartTime = system_clock::now();
+  system_clock::time_point UnitStartTime, UnitStopTime;
+  long TimeOfLongestUnitInSeconds = 0;
+  long EpochOfLastReadOfOutputCorpus = 0;
+
+  size_t MaxInputLen = 0;
+  size_t MaxMutationLen = 0;
+  size_t TmpMaxMutationLen = 0;
+
+  Vector<uint32_t> UniqFeatureSetTmp;
+
+  // Need to know our own thread.
+  static thread_local bool IsMyThread;
+};
+
+struct ScopedEnableMsanInterceptorChecks {
+  ScopedEnableMsanInterceptorChecks() {
+    if (EF->__msan_scoped_enable_interceptor_checks)
+      EF->__msan_scoped_enable_interceptor_checks();
+  }
+  ~ScopedEnableMsanInterceptorChecks() {
+    if (EF->__msan_scoped_disable_interceptor_checks)
+      EF->__msan_scoped_disable_interceptor_checks();
+  }
+};
+
+struct ScopedDisableMsanInterceptorChecks {
+  ScopedDisableMsanInterceptorChecks() {
+    if (EF->__msan_scoped_disable_interceptor_checks)
+      EF->__msan_scoped_disable_interceptor_checks();
+  }
+  ~ScopedDisableMsanInterceptorChecks() {
+    if (EF->__msan_scoped_enable_interceptor_checks)
+      EF->__msan_scoped_enable_interceptor_checks();
+  }
+};
+
+} // namespace fuzzer
+
+#endif // LLVM_FUZZER_INTERNAL_H

custom_mutators/libfuzzer/FuzzerMerge.cpp

@@ -0,0 +1,485 @@
+//===- FuzzerMerge.cpp - merging corpora ----------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Merging corpora.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerCommand.h"
+#include "FuzzerMerge.h"
+#include "FuzzerIO.h"
+#include "FuzzerInternal.h"
+#include "FuzzerTracePC.h"
+#include "FuzzerUtil.h"
+
+#include <fstream>
+#include <iterator>
+#include <set>
+#include <sstream>
+#include <unordered_set>
+
+namespace fuzzer {
+
+bool Merger::Parse(const std::string &Str, bool ParseCoverage) {
+
+  std::istringstream SS(Str);
+  return Parse(SS, ParseCoverage);
+
+}
+
+void Merger::ParseOrExit(std::istream &IS, bool ParseCoverage) {
+
+  if (!Parse(IS, ParseCoverage)) {
+
+    Printf("MERGE: failed to parse the control file (unexpected error)\n");
+    exit(1);
+
+  }
+
+}
+
+// The control file example:
+//
+// 3 # The number of inputs
+// 1 # The number of inputs in the first corpus, <= the previous number
+// file0
+// file1
+// file2  # One file name per line.
+// STARTED 0 123  # FileID, file size
+// FT 0 1 4 6 8  # FileID COV1 COV2 ...
+// COV 0 7 8 9 # FileID COV1 COV1
+// STARTED 1 456  # If FT is missing, the input crashed while processing.
+// STARTED 2 567
+// FT 2 8 9
+// COV 2 11 12
+bool Merger::Parse(std::istream &IS, bool ParseCoverage) {
+
+  LastFailure.clear();
+  std::string Line;
+
+  // Parse NumFiles.
+  if (!std::getline(IS, Line, '\n')) return false;
+  std::istringstream L1(Line);
+  size_t             NumFiles = 0;
+  L1 >> NumFiles;
+  if (NumFiles == 0 || NumFiles > 10000000) return false;
+
+  // Parse NumFilesInFirstCorpus.
+  if (!std::getline(IS, Line, '\n')) return false;
+  std::istringstream L2(Line);
+  NumFilesInFirstCorpus = NumFiles + 1;
+  L2 >> NumFilesInFirstCorpus;
+  if (NumFilesInFirstCorpus > NumFiles) return false;
+
+  // Parse file names.
+  Files.resize(NumFiles);
+  for (size_t i = 0; i < NumFiles; i++)
+    if (!std::getline(IS, Files[i].Name, '\n')) return false;
+
+  // Parse STARTED, FT, and COV lines.
+  size_t           ExpectedStartMarker = 0;
+  const size_t     kInvalidStartMarker = -1;
+  size_t           LastSeenStartMarker = kInvalidStartMarker;
+  Vector<uint32_t> TmpFeatures;
+  Set<uint32_t>    PCs;
+  while (std::getline(IS, Line, '\n')) {
+
+    std::istringstream ISS1(Line);
+    std::string        Marker;
+    size_t             N;
+    ISS1 >> Marker;
+    ISS1 >> N;
+    if (Marker == "STARTED") {
+
+      // STARTED FILE_ID FILE_SIZE
+      if (ExpectedStartMarker != N) return false;
+      ISS1 >> Files[ExpectedStartMarker].Size;
+      LastSeenStartMarker = ExpectedStartMarker;
+      assert(ExpectedStartMarker < Files.size());
+      ExpectedStartMarker++;
+
+    } else if (Marker == "FT") {
+
+      // FT FILE_ID COV1 COV2 COV3 ...
+      size_t CurrentFileIdx = N;
+      if (CurrentFileIdx != LastSeenStartMarker) return false;
+      LastSeenStartMarker = kInvalidStartMarker;
+      if (ParseCoverage) {
+
+        TmpFeatures.clear();  // use a vector from outer scope to avoid resizes.
+        while (ISS1 >> N)
+          TmpFeatures.push_back(N);
+        std::sort(TmpFeatures.begin(), TmpFeatures.end());
+        Files[CurrentFileIdx].Features = TmpFeatures;
+
+      }
+
+    } else if (Marker == "COV") {
+
+      size_t CurrentFileIdx = N;
+      if (ParseCoverage)
+        while (ISS1 >> N)
+          if (PCs.insert(N).second) Files[CurrentFileIdx].Cov.push_back(N);
+
+    } else {
+
+      return false;
+
+    }
+
+  }
+
+  if (LastSeenStartMarker != kInvalidStartMarker)
+    LastFailure = Files[LastSeenStartMarker].Name;
+
+  FirstNotProcessedFile = ExpectedStartMarker;
+  return true;
+
+}
+
+size_t Merger::ApproximateMemoryConsumption() const {
+
+  size_t Res = 0;
+  for (const auto &F : Files)
+    Res += sizeof(F) + F.Features.size() * sizeof(F.Features[0]);
+  return Res;
+
+}
+
+// Decides which files need to be merged (add those to NewFiles).
+// Returns the number of new features added.
+size_t Merger::Merge(const Set<uint32_t> &InitialFeatures,
+                     Set<uint32_t> *      NewFeatures,
+                     const Set<uint32_t> &InitialCov, Set<uint32_t> *NewCov,
+                     Vector<std::string> *NewFiles) {
+
+  NewFiles->clear();
+  assert(NumFilesInFirstCorpus <= Files.size());
+  Set<uint32_t> AllFeatures = InitialFeatures;
+
+  // What features are in the initial corpus?
+  for (size_t i = 0; i < NumFilesInFirstCorpus; i++) {
+
+    auto &Cur = Files[i].Features;
+    AllFeatures.insert(Cur.begin(), Cur.end());
+
+  }
+
+  // Remove all features that we already know from all other inputs.
+  for (size_t i = NumFilesInFirstCorpus; i < Files.size(); i++) {
+
+    auto &           Cur = Files[i].Features;
+    Vector<uint32_t> Tmp;
+    std::set_difference(Cur.begin(), Cur.end(), AllFeatures.begin(),
+                        AllFeatures.end(), std::inserter(Tmp, Tmp.begin()));
+    Cur.swap(Tmp);
+
+  }
+
+  // Sort. Give preference to
+  //   * smaller files
+  //   * files with more features.
+  std::sort(Files.begin() + NumFilesInFirstCorpus, Files.end(),
+            [&](const MergeFileInfo &a, const MergeFileInfo &b) -> bool {
+
+              if (a.Size != b.Size) return a.Size < b.Size;
+              return a.Features.size() > b.Features.size();
+
+            });
+
+  // One greedy pass: add the file's features to AllFeatures.
+  // If new features were added, add this file to NewFiles.
+  for (size_t i = NumFilesInFirstCorpus; i < Files.size(); i++) {
+
+    auto &Cur = Files[i].Features;
+    // Printf("%s -> sz %zd ft %zd\n", Files[i].Name.c_str(),
+    //       Files[i].Size, Cur.size());
+    bool FoundNewFeatures = false;
+    for (auto Fe : Cur) {
+
+      if (AllFeatures.insert(Fe).second) {
+
+        FoundNewFeatures = true;
+        NewFeatures->insert(Fe);
+
+      }
+
+    }
+
+    if (FoundNewFeatures) NewFiles->push_back(Files[i].Name);
+    for (auto Cov : Files[i].Cov)
+      if (InitialCov.find(Cov) == InitialCov.end()) NewCov->insert(Cov);
+
+  }
+
+  return NewFeatures->size();
+
+}
+
+Set<uint32_t> Merger::AllFeatures() const {
+
+  Set<uint32_t> S;
+  for (auto &File : Files)
+    S.insert(File.Features.begin(), File.Features.end());
+  return S;
+
+}
+
+// Inner process. May crash if the target crashes.
+void Fuzzer::CrashResistantMergeInternalStep(const std::string &CFPath) {
+
+  Printf("MERGE-INNER: using the control file '%s'\n", CFPath.c_str());
+  Merger        M;
+  std::ifstream IF(CFPath);
+  M.ParseOrExit(IF, false);
+  IF.close();
+  if (!M.LastFailure.empty())
+    Printf("MERGE-INNER: '%s' caused a failure at the previous merge step\n",
+           M.LastFailure.c_str());
+
+  Printf(
+      "MERGE-INNER: %zd total files;"
+      " %zd processed earlier; will process %zd files now\n",
+      M.Files.size(), M.FirstNotProcessedFile,
+      M.Files.size() - M.FirstNotProcessedFile);
+
+  std::ofstream OF(CFPath, std::ofstream::out | std::ofstream::app);
+  Set<size_t>   AllFeatures;
+  auto          PrintStatsWrapper = [this, &AllFeatures](const char *Where) {
+
+    this->PrintStats(Where, "\n", 0, AllFeatures.size());
+
+  };
+
+  Set<const TracePC::PCTableEntry *> AllPCs;
+  for (size_t i = M.FirstNotProcessedFile; i < M.Files.size(); i++) {
+
+    Fuzzer::MaybeExitGracefully();
+    auto U = FileToVector(M.Files[i].Name);
+    if (U.size() > MaxInputLen) {
+
+      U.resize(MaxInputLen);
+      U.shrink_to_fit();
+
+    }
+
+    // Write the pre-run marker.
+    OF << "STARTED " << i << " " << U.size() << "\n";
+    OF.flush();  // Flush is important since Command::Execute may crash.
+    // Run.
+    TPC.ResetMaps();
+    ExecuteCallback(U.data(), U.size());
+    // Collect coverage. We are iterating over the files in this order:
+    // * First, files in the initial corpus ordered by size, smallest first.
+    // * Then, all other files, smallest first.
+    // So it makes no sense to record all features for all files, instead we
+    // only record features that were not seen before.
+    Set<size_t> UniqFeatures;
+    TPC.CollectFeatures([&](size_t Feature) {
+
+      if (AllFeatures.insert(Feature).second) UniqFeatures.insert(Feature);
+
+    });
+
+    TPC.UpdateObservedPCs();
+    // Show stats.
+    if (!(TotalNumberOfRuns & (TotalNumberOfRuns - 1)))
+      PrintStatsWrapper("pulse ");
+    if (TotalNumberOfRuns == M.NumFilesInFirstCorpus)
+      PrintStatsWrapper("LOADED");
+    // Write the post-run marker and the coverage.
+    OF << "FT " << i;
+    for (size_t F : UniqFeatures)
+      OF << " " << F;
+    OF << "\n";
+    OF << "COV " << i;
+    TPC.ForEachObservedPC([&](const TracePC::PCTableEntry *TE) {
+
+      if (AllPCs.insert(TE).second) OF << " " << TPC.PCTableEntryIdx(TE);
+
+    });
+
+    OF << "\n";
+    OF.flush();
+
+  }
+
+  PrintStatsWrapper("DONE  ");
+
+}
+
+static size_t WriteNewControlFile(const std::string &          CFPath,
+                                  const Vector<SizedFile> &    OldCorpus,
+                                  const Vector<SizedFile> &    NewCorpus,
+                                  const Vector<MergeFileInfo> &KnownFiles) {
+
+  std::unordered_set<std::string> FilesToSkip;
+  for (auto &SF : KnownFiles)
+    FilesToSkip.insert(SF.Name);
+
+  Vector<std::string> FilesToUse;
+  auto                MaybeUseFile = [=, &FilesToUse](std::string Name) {
+
+    if (FilesToSkip.find(Name) == FilesToSkip.end()) FilesToUse.push_back(Name);
+
+  };
+
+  for (auto &SF : OldCorpus)
+    MaybeUseFile(SF.File);
+  auto FilesToUseFromOldCorpus = FilesToUse.size();
+  for (auto &SF : NewCorpus)
+    MaybeUseFile(SF.File);
+
+  RemoveFile(CFPath);
+  std::ofstream ControlFile(CFPath);
+  ControlFile << FilesToUse.size() << "\n";
+  ControlFile << FilesToUseFromOldCorpus << "\n";
+  for (auto &FN : FilesToUse)
+    ControlFile << FN << "\n";
+
+  if (!ControlFile) {
+
+    Printf("MERGE-OUTER: failed to write to the control file: %s\n",
+           CFPath.c_str());
+    exit(1);
+
+  }
+
+  return FilesToUse.size();
+
+}
+
+// Outer process. Does not call the target code and thus should not fail.
+void CrashResistantMerge(const Vector<std::string> &Args,
+                         const Vector<SizedFile> &  OldCorpus,
+                         const Vector<SizedFile> &  NewCorpus,
+                         Vector<std::string> *      NewFiles,
+                         const Set<uint32_t> &      InitialFeatures,
+                         Set<uint32_t> *            NewFeatures,
+                         const Set<uint32_t> &InitialCov, Set<uint32_t> *NewCov,
+                         const std::string &CFPath, bool V /*Verbose*/) {
+
+  if (NewCorpus.empty() && OldCorpus.empty()) return;  // Nothing to merge.
+  size_t                NumAttempts = 0;
+  Vector<MergeFileInfo> KnownFiles;
+  if (FileSize(CFPath)) {
+
+    VPrintf(V, "MERGE-OUTER: non-empty control file provided: '%s'\n",
+            CFPath.c_str());
+    Merger        M;
+    std::ifstream IF(CFPath);
+    if (M.Parse(IF, /*ParseCoverage=*/true)) {
+
+      VPrintf(V,
+              "MERGE-OUTER: control file ok, %zd files total,"
+              " first not processed file %zd\n",
+              M.Files.size(), M.FirstNotProcessedFile);
+      if (!M.LastFailure.empty())
+        VPrintf(V,
+                "MERGE-OUTER: '%s' will be skipped as unlucky "
+                "(merge has stumbled on it the last time)\n",
+                M.LastFailure.c_str());
+      if (M.FirstNotProcessedFile >= M.Files.size()) {
+
+        // Merge has already been completed with the given merge control file.
+        if (M.Files.size() == OldCorpus.size() + NewCorpus.size()) {
+
+          VPrintf(
+              V,
+              "MERGE-OUTER: nothing to do, merge has been completed before\n");
+          exit(0);
+
+        }
+
+        // Number of input files likely changed, start merge from scratch, but
+        // reuse coverage information from the given merge control file.
+        VPrintf(
+            V,
+            "MERGE-OUTER: starting merge from scratch, but reusing coverage "
+            "information from the given control file\n");
+        KnownFiles = M.Files;
+
+      } else {
+
+        // There is a merge in progress, continue.
+        NumAttempts = M.Files.size() - M.FirstNotProcessedFile;
+
+      }
+
+    } else {
+
+      VPrintf(V, "MERGE-OUTER: bad control file, will overwrite it\n");
+
+    }
+
+  }
+
+  if (!NumAttempts) {
+
+    // The supplied control file is empty or bad, create a fresh one.
+    VPrintf(V,
+            "MERGE-OUTER: "
+            "%zd files, %zd in the initial corpus, %zd processed earlier\n",
+            OldCorpus.size() + NewCorpus.size(), OldCorpus.size(),
+            KnownFiles.size());
+    NumAttempts = WriteNewControlFile(CFPath, OldCorpus, NewCorpus, KnownFiles);
+
+  }
+
+  // Execute the inner process until it passes.
+  // Every inner process should execute at least one input.
+  Command BaseCmd(Args);
+  BaseCmd.removeFlag("merge");
+  BaseCmd.removeFlag("fork");
+  BaseCmd.removeFlag("collect_data_flow");
+  for (size_t Attempt = 1; Attempt <= NumAttempts; Attempt++) {
+
+    Fuzzer::MaybeExitGracefully();
+    VPrintf(V, "MERGE-OUTER: attempt %zd\n", Attempt);
+    Command Cmd(BaseCmd);
+    Cmd.addFlag("merge_control_file", CFPath);
+    Cmd.addFlag("merge_inner", "1");
+    if (!V) {
+
+      Cmd.setOutputFile(getDevNull());
+      Cmd.combineOutAndErr();
+
+    }
+
+    auto ExitCode = ExecuteCommand(Cmd);
+    if (!ExitCode) {
+
+      VPrintf(V, "MERGE-OUTER: succesfull in %zd attempt(s)\n", Attempt);
+      break;
+
+    }
+
+  }
+
+  // Read the control file and do the merge.
+  Merger        M;
+  std::ifstream IF(CFPath);
+  IF.seekg(0, IF.end);
+  VPrintf(V, "MERGE-OUTER: the control file has %zd bytes\n",
+          (size_t)IF.tellg());
+  IF.seekg(0, IF.beg);
+  M.ParseOrExit(IF, true);
+  IF.close();
+  VPrintf(V,
+          "MERGE-OUTER: consumed %zdMb (%zdMb rss) to parse the control file\n",
+          M.ApproximateMemoryConsumption() >> 20, GetPeakRSSMb());
+
+  M.Files.insert(M.Files.end(), KnownFiles.begin(), KnownFiles.end());
+  M.Merge(InitialFeatures, NewFeatures, InitialCov, NewCov, NewFiles);
+  VPrintf(V,
+          "MERGE-OUTER: %zd new files with %zd new features added; "
+          "%zd new coverage edges\n",
+          NewFiles->size(), NewFeatures->size(), NewCov->size());
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerMerge.h

@@ -0,0 +1,87 @@
+//===- FuzzerMerge.h - merging corpa ----------------------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Merging Corpora.
+//
+// The task:
+//   Take the existing corpus (possibly empty) and merge new inputs into
+//   it so that only inputs with new coverage ('features') are added.
+//   The process should tolerate the crashes, OOMs, leaks, etc.
+//
+// Algorithm:
+//   The outer process collects the set of files and writes their names
+//   into a temporary "control" file, then repeatedly launches the inner
+//   process until all inputs are processed.
+//   The outer process does not actually execute the target code.
+//
+//   The inner process reads the control file and sees a) list of all the inputs
+//   and b) the last processed input. Then it starts processing the inputs one
+//   by one. Before processing every input it writes one line to control file:
+//   STARTED INPUT_ID INPUT_SIZE
+//   After processing an input it writes the following lines:
+//   FT INPUT_ID Feature1 Feature2 Feature3 ...
+//   COV INPUT_ID Coverage1 Coverage2 Coverage3 ...
+//   If a crash happens while processing an input the last line in the control
+//   file will be "STARTED INPUT_ID" and so the next process will know
+//   where to resume.
+//
+//   Once all inputs are processed by the inner process(es) the outer process
+//   reads the control files and does the merge based entirely on the contents
+//   of control file.
+//   It uses a single pass greedy algorithm choosing first the smallest inputs
+//   within the same size the inputs that have more new features.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_MERGE_H
+#define LLVM_FUZZER_MERGE_H
+
+#include "FuzzerDefs.h"
+
+#include <istream>
+#include <ostream>
+#include <set>
+#include <vector>
+
+namespace fuzzer {
+
+struct MergeFileInfo {
+  std::string Name;
+  size_t Size = 0;
+  Vector<uint32_t> Features, Cov;
+};
+
+struct Merger {
+  Vector<MergeFileInfo> Files;
+  size_t NumFilesInFirstCorpus = 0;
+  size_t FirstNotProcessedFile = 0;
+  std::string LastFailure;
+
+  bool Parse(std::istream &IS, bool ParseCoverage);
+  bool Parse(const std::string &Str, bool ParseCoverage);
+  void ParseOrExit(std::istream &IS, bool ParseCoverage);
+  size_t Merge(const Set<uint32_t> &InitialFeatures, Set<uint32_t> *NewFeatures,
+               const Set<uint32_t> &InitialCov, Set<uint32_t> *NewCov,
+               Vector<std::string> *NewFiles);
+  size_t ApproximateMemoryConsumption() const;
+  Set<uint32_t> AllFeatures() const;
+};
+
+void CrashResistantMerge(const Vector<std::string> &Args,
+                         const Vector<SizedFile> &OldCorpus,
+                         const Vector<SizedFile> &NewCorpus,
+                         Vector<std::string> *NewFiles,
+                         const Set<uint32_t> &InitialFeatures,
+                         Set<uint32_t> *NewFeatures,
+                         const Set<uint32_t> &InitialCov,
+                         Set<uint32_t> *NewCov,
+                         const std::string &CFPath,
+                         bool Verbose);
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_MERGE_H

custom_mutators/libfuzzer/FuzzerMutate.cpp

@@ -0,0 +1,720 @@
+//===- FuzzerMutate.cpp - Mutate a test input -----------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Mutate a test input.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerDefs.h"
+#include "FuzzerExtFunctions.h"
+#include "FuzzerIO.h"
+#include "FuzzerMutate.h"
+#include "FuzzerOptions.h"
+#include "FuzzerTracePC.h"
+
+namespace fuzzer {
+
+const size_t Dictionary::kMaxDictSize;
+
+static void PrintASCII(const Word &W, const char *PrintAfter) {
+
+  PrintASCII(W.data(), W.size(), PrintAfter);
+
+}
+
+MutationDispatcher::MutationDispatcher(Random &              Rand,
+                                       const FuzzingOptions &Options)
+    : Rand(Rand), Options(Options) {
+
+  DefaultMutators.insert(
+      DefaultMutators.begin(),
+      {
+
+          {&MutationDispatcher::Mutate_EraseBytes, "EraseBytes"},
+          {&MutationDispatcher::Mutate_InsertByte, "InsertByte"},
+          {&MutationDispatcher::Mutate_InsertRepeatedBytes,
+           "InsertRepeatedBytes"},
+          {&MutationDispatcher::Mutate_ChangeByte, "ChangeByte"},
+          {&MutationDispatcher::Mutate_ChangeBit, "ChangeBit"},
+          {&MutationDispatcher::Mutate_ShuffleBytes, "ShuffleBytes"},
+          {&MutationDispatcher::Mutate_ChangeASCIIInteger, "ChangeASCIIInt"},
+          {&MutationDispatcher::Mutate_ChangeBinaryInteger, "ChangeBinInt"},
+          {&MutationDispatcher::Mutate_CopyPart, "CopyPart"},
+          {&MutationDispatcher::Mutate_CrossOver, "CrossOver"},
+          {&MutationDispatcher::Mutate_AddWordFromManualDictionary,
+           "ManualDict"},
+          {&MutationDispatcher::Mutate_AddWordFromPersistentAutoDictionary,
+           "PersAutoDict"},
+
+      });
+
+  if (Options.UseCmp)
+    DefaultMutators.push_back(
+        {&MutationDispatcher::Mutate_AddWordFromTORC, "CMP"});
+
+  if (EF->LLVMFuzzerCustomMutator)
+    Mutators.push_back({&MutationDispatcher::Mutate_Custom, "Custom"});
+  else
+    Mutators = DefaultMutators;
+
+  if (EF->LLVMFuzzerCustomCrossOver)
+    Mutators.push_back(
+        {&MutationDispatcher::Mutate_CustomCrossOver, "CustomCrossOver"});
+
+}
+
+static char RandCh(Random &Rand) {
+
+  if (Rand.RandBool()) return Rand(256);
+  const char Special[] = "!*'();:@&=+$,/?%#[]012Az-`~.\xff\x00";
+  return Special[Rand(sizeof(Special) - 1)];
+
+}
+
+size_t MutationDispatcher::Mutate_Custom(uint8_t *Data, size_t Size,
+                                         size_t MaxSize) {
+
+  return EF->LLVMFuzzerCustomMutator(Data, Size, MaxSize, Rand.Rand());
+
+}
+
+size_t MutationDispatcher::Mutate_CustomCrossOver(uint8_t *Data, size_t Size,
+                                                  size_t MaxSize) {
+
+  if (Size == 0) return 0;
+  if (!CrossOverWith) return 0;
+  const Unit &Other = *CrossOverWith;
+  if (Other.empty()) return 0;
+  CustomCrossOverInPlaceHere.resize(MaxSize);
+  auto & U = CustomCrossOverInPlaceHere;
+  size_t NewSize = EF->LLVMFuzzerCustomCrossOver(
+      Data, Size, Other.data(), Other.size(), U.data(), U.size(), Rand.Rand());
+  if (!NewSize) return 0;
+  assert(NewSize <= MaxSize && "CustomCrossOver returned overisized unit");
+  memcpy(Data, U.data(), NewSize);
+  return NewSize;
+
+}
+
+size_t MutationDispatcher::Mutate_ShuffleBytes(uint8_t *Data, size_t Size,
+                                               size_t MaxSize) {
+
+  if (Size > MaxSize || Size == 0) return 0;
+  size_t ShuffleAmount =
+      Rand(std::min(Size, (size_t)8)) + 1;  // [1,8] and <= Size.
+  size_t ShuffleStart = Rand(Size - ShuffleAmount);
+  assert(ShuffleStart + ShuffleAmount <= Size);
+  std::shuffle(Data + ShuffleStart, Data + ShuffleStart + ShuffleAmount, Rand);
+  return Size;
+
+}
+
+size_t MutationDispatcher::Mutate_EraseBytes(uint8_t *Data, size_t Size,
+                                             size_t MaxSize) {
+
+  if (Size <= 1) return 0;
+  size_t N = Rand(Size / 2) + 1;
+  assert(N < Size);
+  size_t Idx = Rand(Size - N + 1);
+  // Erase Data[Idx:Idx+N].
+  memmove(Data + Idx, Data + Idx + N, Size - Idx - N);
+  // Printf("Erase: %zd %zd => %zd; Idx %zd\n", N, Size, Size - N, Idx);
+  return Size - N;
+
+}
+
+size_t MutationDispatcher::Mutate_InsertByte(uint8_t *Data, size_t Size,
+                                             size_t MaxSize) {
+
+  if (Size >= MaxSize) return 0;
+  size_t Idx = Rand(Size + 1);
+  // Insert new value at Data[Idx].
+  memmove(Data + Idx + 1, Data + Idx, Size - Idx);
+  Data[Idx] = RandCh(Rand);
+  return Size + 1;
+
+}
+
+size_t MutationDispatcher::Mutate_InsertRepeatedBytes(uint8_t *Data,
+                                                      size_t   Size,
+                                                      size_t   MaxSize) {
+
+  const size_t kMinBytesToInsert = 3;
+  if (Size + kMinBytesToInsert >= MaxSize) return 0;
+  size_t MaxBytesToInsert = std::min(MaxSize - Size, (size_t)128);
+  size_t N = Rand(MaxBytesToInsert - kMinBytesToInsert + 1) + kMinBytesToInsert;
+  assert(Size + N <= MaxSize && N);
+  size_t Idx = Rand(Size + 1);
+  // Insert new values at Data[Idx].
+  memmove(Data + Idx + N, Data + Idx, Size - Idx);
+  // Give preference to 0x00 and 0xff.
+  uint8_t Byte = Rand.RandBool() ? Rand(256) : (Rand.RandBool() ? 0 : 255);
+  for (size_t i = 0; i < N; i++)
+    Data[Idx + i] = Byte;
+  return Size + N;
+
+}
+
+size_t MutationDispatcher::Mutate_ChangeByte(uint8_t *Data, size_t Size,
+                                             size_t MaxSize) {
+
+  if (Size > MaxSize) return 0;
+  size_t Idx = Rand(Size);
+  Data[Idx] = RandCh(Rand);
+  return Size;
+
+}
+
+size_t MutationDispatcher::Mutate_ChangeBit(uint8_t *Data, size_t Size,
+                                            size_t MaxSize) {
+
+  if (Size > MaxSize) return 0;
+  size_t Idx = Rand(Size);
+  Data[Idx] ^= 1 << Rand(8);
+  return Size;
+
+}
+
+size_t MutationDispatcher::Mutate_AddWordFromManualDictionary(uint8_t *Data,
+                                                              size_t   Size,
+                                                              size_t MaxSize) {
+
+  return AddWordFromDictionary(ManualDictionary, Data, Size, MaxSize);
+
+}
+
+size_t MutationDispatcher::ApplyDictionaryEntry(uint8_t *Data, size_t Size,
+                                                size_t           MaxSize,
+                                                DictionaryEntry &DE) {
+
+  const Word &W = DE.GetW();
+  bool        UsePositionHint = DE.HasPositionHint() &&
+                         DE.GetPositionHint() + W.size() < Size &&
+                         Rand.RandBool();
+  if (Rand.RandBool()) {  // Insert W.
+    if (Size + W.size() > MaxSize) return 0;
+    size_t Idx = UsePositionHint ? DE.GetPositionHint() : Rand(Size + 1);
+    memmove(Data + Idx + W.size(), Data + Idx, Size - Idx);
+    memcpy(Data + Idx, W.data(), W.size());
+    Size += W.size();
+
+  } else {  // Overwrite some bytes with W.
+
+    if (W.size() > Size) return 0;
+    size_t Idx = UsePositionHint ? DE.GetPositionHint() : Rand(Size - W.size());
+    memcpy(Data + Idx, W.data(), W.size());
+
+  }
+
+  return Size;
+
+}
+
+// Somewhere in the past we have observed a comparison instructions
+// with arguments Arg1 Arg2. This function tries to guess a dictionary
+// entry that will satisfy that comparison.
+// It first tries to find one of the arguments (possibly swapped) in the
+// input and if it succeeds it creates a DE with a position hint.
+// Otherwise it creates a DE with one of the arguments w/o a position hint.
+DictionaryEntry MutationDispatcher::MakeDictionaryEntryFromCMP(
+    const void *Arg1, const void *Arg2, const void *Arg1Mutation,
+    const void *Arg2Mutation, size_t ArgSize, const uint8_t *Data,
+    size_t Size) {
+
+  bool           HandleFirst = Rand.RandBool();
+  const void *   ExistingBytes, *DesiredBytes;
+  Word           W;
+  const uint8_t *End = Data + Size;
+  for (int Arg = 0; Arg < 2; Arg++) {
+
+    ExistingBytes = HandleFirst ? Arg1 : Arg2;
+    DesiredBytes = HandleFirst ? Arg2Mutation : Arg1Mutation;
+    HandleFirst = !HandleFirst;
+    W.Set(reinterpret_cast<const uint8_t *>(DesiredBytes), ArgSize);
+    const size_t kMaxNumPositions = 8;
+    size_t       Positions[kMaxNumPositions];
+    size_t       NumPositions = 0;
+    for (const uint8_t *Cur = Data;
+         Cur < End && NumPositions < kMaxNumPositions; Cur++) {
+
+      Cur =
+          (const uint8_t *)SearchMemory(Cur, End - Cur, ExistingBytes, ArgSize);
+      if (!Cur) break;
+      Positions[NumPositions++] = Cur - Data;
+
+    }
+
+    if (!NumPositions) continue;
+    return DictionaryEntry(W, Positions[Rand(NumPositions)]);
+
+  }
+
+  DictionaryEntry DE(W);
+  return DE;
+
+}
+
+template <class T>
+DictionaryEntry MutationDispatcher::MakeDictionaryEntryFromCMP(
+    T Arg1, T Arg2, const uint8_t *Data, size_t Size) {
+
+  if (Rand.RandBool()) Arg1 = Bswap(Arg1);
+  if (Rand.RandBool()) Arg2 = Bswap(Arg2);
+  T Arg1Mutation = Arg1 + Rand(-1, 1);
+  T Arg2Mutation = Arg2 + Rand(-1, 1);
+  return MakeDictionaryEntryFromCMP(&Arg1, &Arg2, &Arg1Mutation, &Arg2Mutation,
+                                    sizeof(Arg1), Data, Size);
+
+}
+
+DictionaryEntry MutationDispatcher::MakeDictionaryEntryFromCMP(
+    const Word &Arg1, const Word &Arg2, const uint8_t *Data, size_t Size) {
+
+  return MakeDictionaryEntryFromCMP(Arg1.data(), Arg2.data(), Arg1.data(),
+                                    Arg2.data(), Arg1.size(), Data, Size);
+
+}
+
+size_t MutationDispatcher::Mutate_AddWordFromTORC(uint8_t *Data, size_t Size,
+                                                  size_t MaxSize) {
+
+  Word            W;
+  DictionaryEntry DE;
+  switch (Rand(4)) {
+
+    case 0: {
+
+      auto X = TPC.TORC8.Get(Rand.Rand());
+      DE = MakeDictionaryEntryFromCMP(X.A, X.B, Data, Size);
+
+    } break;
+
+    case 1: {
+
+      auto X = TPC.TORC4.Get(Rand.Rand());
+      if ((X.A >> 16) == 0 && (X.B >> 16) == 0 && Rand.RandBool())
+        DE = MakeDictionaryEntryFromCMP((uint16_t)X.A, (uint16_t)X.B, Data,
+                                        Size);
+      else
+        DE = MakeDictionaryEntryFromCMP(X.A, X.B, Data, Size);
+
+    } break;
+
+    case 2: {
+
+      auto X = TPC.TORCW.Get(Rand.Rand());
+      DE = MakeDictionaryEntryFromCMP(X.A, X.B, Data, Size);
+
+    } break;
+
+    case 3:
+      if (Options.UseMemmem) {
+
+        auto X = TPC.MMT.Get(Rand.Rand());
+        DE = DictionaryEntry(X);
+
+      }
+
+      break;
+    default:
+      assert(0);
+
+  }
+
+  if (!DE.GetW().size()) return 0;
+  Size = ApplyDictionaryEntry(Data, Size, MaxSize, DE);
+  if (!Size) return 0;
+  DictionaryEntry &DERef =
+      CmpDictionaryEntriesDeque[CmpDictionaryEntriesDequeIdx++ %
+                                kCmpDictionaryEntriesDequeSize];
+  DERef = DE;
+  CurrentDictionaryEntrySequence.push_back(&DERef);
+  return Size;
+
+}
+
+size_t MutationDispatcher::Mutate_AddWordFromPersistentAutoDictionary(
+    uint8_t *Data, size_t Size, size_t MaxSize) {
+
+  return AddWordFromDictionary(PersistentAutoDictionary, Data, Size, MaxSize);
+
+}
+
+size_t MutationDispatcher::AddWordFromDictionary(Dictionary &D, uint8_t *Data,
+                                                 size_t Size, size_t MaxSize) {
+
+  if (Size > MaxSize) return 0;
+  if (D.empty()) return 0;
+  DictionaryEntry &DE = D[Rand(D.size())];
+  Size = ApplyDictionaryEntry(Data, Size, MaxSize, DE);
+  if (!Size) return 0;
+  DE.IncUseCount();
+  CurrentDictionaryEntrySequence.push_back(&DE);
+  return Size;
+
+}
+
+// Overwrites part of To[0,ToSize) with a part of From[0,FromSize).
+// Returns ToSize.
+size_t MutationDispatcher::CopyPartOf(const uint8_t *From, size_t FromSize,
+                                      uint8_t *To, size_t ToSize) {
+
+  // Copy From[FromBeg, FromBeg + CopySize) into To[ToBeg, ToBeg + CopySize).
+  size_t ToBeg = Rand(ToSize);
+  size_t CopySize = Rand(ToSize - ToBeg) + 1;
+  assert(ToBeg + CopySize <= ToSize);
+  CopySize = std::min(CopySize, FromSize);
+  size_t FromBeg = Rand(FromSize - CopySize + 1);
+  assert(FromBeg + CopySize <= FromSize);
+  memmove(To + ToBeg, From + FromBeg, CopySize);
+  return ToSize;
+
+}
+
+// Inserts part of From[0,ToSize) into To.
+// Returns new size of To on success or 0 on failure.
+size_t MutationDispatcher::InsertPartOf(const uint8_t *From, size_t FromSize,
+                                        uint8_t *To, size_t ToSize,
+                                        size_t MaxToSize) {
+
+  if (ToSize >= MaxToSize) return 0;
+  size_t AvailableSpace = MaxToSize - ToSize;
+  size_t MaxCopySize = std::min(AvailableSpace, FromSize);
+  size_t CopySize = Rand(MaxCopySize) + 1;
+  size_t FromBeg = Rand(FromSize - CopySize + 1);
+  assert(FromBeg + CopySize <= FromSize);
+  size_t ToInsertPos = Rand(ToSize + 1);
+  assert(ToInsertPos + CopySize <= MaxToSize);
+  size_t TailSize = ToSize - ToInsertPos;
+  if (To == From) {
+
+    MutateInPlaceHere.resize(MaxToSize);
+    memcpy(MutateInPlaceHere.data(), From + FromBeg, CopySize);
+    memmove(To + ToInsertPos + CopySize, To + ToInsertPos, TailSize);
+    memmove(To + ToInsertPos, MutateInPlaceHere.data(), CopySize);
+
+  } else {
+
+    memmove(To + ToInsertPos + CopySize, To + ToInsertPos, TailSize);
+    memmove(To + ToInsertPos, From + FromBeg, CopySize);
+
+  }
+
+  return ToSize + CopySize;
+
+}
+
+size_t MutationDispatcher::Mutate_CopyPart(uint8_t *Data, size_t Size,
+                                           size_t MaxSize) {
+
+  if (Size > MaxSize || Size == 0) return 0;
+  // If Size == MaxSize, `InsertPartOf(...)` will
+  // fail so there's no point using it in this case.
+  if (Size == MaxSize || Rand.RandBool())
+    return CopyPartOf(Data, Size, Data, Size);
+  else
+    return InsertPartOf(Data, Size, Data, Size, MaxSize);
+
+}
+
+size_t MutationDispatcher::Mutate_ChangeASCIIInteger(uint8_t *Data, size_t Size,
+                                                     size_t MaxSize) {
+
+  if (Size > MaxSize) return 0;
+  size_t B = Rand(Size);
+  while (B < Size && !isdigit(Data[B]))
+    B++;
+  if (B == Size) return 0;
+  size_t E = B;
+  while (E < Size && isdigit(Data[E]))
+    E++;
+  assert(B < E);
+  // now we have digits in [B, E).
+  // strtol and friends don't accept non-zero-teminated data, parse it manually.
+  uint64_t Val = Data[B] - '0';
+  for (size_t i = B + 1; i < E; i++)
+    Val = Val * 10 + Data[i] - '0';
+
+  // Mutate the integer value.
+  switch (Rand(5)) {
+
+    case 0:
+      Val++;
+      break;
+    case 1:
+      Val--;
+      break;
+    case 2:
+      Val /= 2;
+      break;
+    case 3:
+      Val *= 2;
+      break;
+    case 4:
+      Val = Rand(Val * Val);
+      break;
+    default:
+      assert(0);
+
+  }
+
+  // Just replace the bytes with the new ones, don't bother moving bytes.
+  for (size_t i = B; i < E; i++) {
+
+    size_t Idx = E + B - i - 1;
+    assert(Idx >= B && Idx < E);
+    Data[Idx] = (Val % 10) + '0';
+    Val /= 10;
+
+  }
+
+  return Size;
+
+}
+
+template <class T>
+size_t ChangeBinaryInteger(uint8_t *Data, size_t Size, Random &Rand) {
+
+  if (Size < sizeof(T)) return 0;
+  size_t Off = Rand(Size - sizeof(T) + 1);
+  assert(Off + sizeof(T) <= Size);
+  T Val;
+  if (Off < 64 && !Rand(4)) {
+
+    Val = Size;
+    if (Rand.RandBool()) Val = Bswap(Val);
+
+  } else {
+
+    memcpy(&Val, Data + Off, sizeof(Val));
+    T Add = Rand(21);
+    Add -= 10;
+    if (Rand.RandBool())
+      Val = Bswap(T(Bswap(Val) + Add));  // Add assuming different endiannes.
+    else
+      Val = Val + Add;                // Add assuming current endiannes.
+    if (Add == 0 || Rand.RandBool())  // Maybe negate.
+      Val = -Val;
+
+  }
+
+  memcpy(Data + Off, &Val, sizeof(Val));
+  return Size;
+
+}
+
+size_t MutationDispatcher::Mutate_ChangeBinaryInteger(uint8_t *Data,
+                                                      size_t   Size,
+                                                      size_t   MaxSize) {
+
+  if (Size > MaxSize) return 0;
+  switch (Rand(4)) {
+
+    case 3:
+      return ChangeBinaryInteger<uint64_t>(Data, Size, Rand);
+    case 2:
+      return ChangeBinaryInteger<uint32_t>(Data, Size, Rand);
+    case 1:
+      return ChangeBinaryInteger<uint16_t>(Data, Size, Rand);
+    case 0:
+      return ChangeBinaryInteger<uint8_t>(Data, Size, Rand);
+    default:
+      assert(0);
+
+  }
+
+  return 0;
+
+}
+
+size_t MutationDispatcher::Mutate_CrossOver(uint8_t *Data, size_t Size,
+                                            size_t MaxSize) {
+
+  if (Size > MaxSize) return 0;
+  if (Size == 0) return 0;
+  if (!CrossOverWith) return 0;
+  const Unit &O = *CrossOverWith;
+  if (O.empty()) return 0;
+  size_t NewSize = 0;
+  switch (Rand(3)) {
+
+    case 0:
+      MutateInPlaceHere.resize(MaxSize);
+      NewSize = CrossOver(Data, Size, O.data(), O.size(),
+                          MutateInPlaceHere.data(), MaxSize);
+      memcpy(Data, MutateInPlaceHere.data(), NewSize);
+      break;
+    case 1:
+      NewSize = InsertPartOf(O.data(), O.size(), Data, Size, MaxSize);
+      if (!NewSize) NewSize = CopyPartOf(O.data(), O.size(), Data, Size);
+      break;
+    case 2:
+      NewSize = CopyPartOf(O.data(), O.size(), Data, Size);
+      break;
+    default:
+      assert(0);
+
+  }
+
+  assert(NewSize > 0 && "CrossOver returned empty unit");
+  assert(NewSize <= MaxSize && "CrossOver returned overisized unit");
+  return NewSize;
+
+}
+
+void MutationDispatcher::StartMutationSequence() {
+
+  CurrentMutatorSequence.clear();
+  CurrentDictionaryEntrySequence.clear();
+
+}
+
+// Copy successful dictionary entries to PersistentAutoDictionary.
+void MutationDispatcher::RecordSuccessfulMutationSequence() {
+
+  for (auto DE : CurrentDictionaryEntrySequence) {
+
+    // PersistentAutoDictionary.AddWithSuccessCountOne(DE);
+    DE->IncSuccessCount();
+    assert(DE->GetW().size());
+    // Linear search is fine here as this happens seldom.
+    if (!PersistentAutoDictionary.ContainsWord(DE->GetW()))
+      PersistentAutoDictionary.push_back({DE->GetW(), 1});
+
+  }
+
+}
+
+void MutationDispatcher::PrintRecommendedDictionary() {
+
+  Vector<DictionaryEntry> V;
+  for (auto &DE : PersistentAutoDictionary)
+    if (!ManualDictionary.ContainsWord(DE.GetW())) V.push_back(DE);
+  if (V.empty()) return;
+  Printf("###### Recommended dictionary. ######\n");
+  for (auto &DE : V) {
+
+    assert(DE.GetW().size());
+    Printf("\"");
+    PrintASCII(DE.GetW(), "\"");
+    Printf(" # Uses: %zd\n", DE.GetUseCount());
+
+  }
+
+  Printf("###### End of recommended dictionary. ######\n");
+
+}
+
+void MutationDispatcher::PrintMutationSequence() {
+
+  Printf("MS: %zd ", CurrentMutatorSequence.size());
+  for (auto M : CurrentMutatorSequence)
+    Printf("%s-", M.Name);
+  if (!CurrentDictionaryEntrySequence.empty()) {
+
+    Printf(" DE: ");
+    for (auto DE : CurrentDictionaryEntrySequence) {
+
+      Printf("\"");
+      PrintASCII(DE->GetW(), "\"-");
+
+    }
+
+  }
+
+}
+
+std::string MutationDispatcher::MutationSequence() {
+
+  std::string MS;
+  for (auto M : CurrentMutatorSequence) {
+
+    MS += M.Name;
+    MS += "-";
+
+  }
+
+  return MS;
+
+}
+
+size_t MutationDispatcher::Mutate(uint8_t *Data, size_t Size, size_t MaxSize) {
+
+  return MutateImpl(Data, Size, MaxSize, Mutators);
+
+}
+
+size_t MutationDispatcher::DefaultMutate(uint8_t *Data, size_t Size,
+                                         size_t MaxSize) {
+
+  return MutateImpl(Data, Size, MaxSize, DefaultMutators);
+
+}
+
+// Mutates Data in place, returns new size.
+size_t MutationDispatcher::MutateImpl(uint8_t *Data, size_t Size,
+                                      size_t           MaxSize,
+                                      Vector<Mutator> &Mutators) {
+
+  assert(MaxSize > 0);
+  // Some mutations may fail (e.g. can't insert more bytes if Size == MaxSize),
+  // in which case they will return 0.
+  // Try several times before returning un-mutated data.
+  for (int Iter = 0; Iter < 100; Iter++) {
+
+    auto   M = Mutators[Rand(Mutators.size())];
+    size_t NewSize = (this->*(M.Fn))(Data, Size, MaxSize);
+    if (NewSize && NewSize <= MaxSize) {
+
+      if (Options.OnlyASCII) ToASCII(Data, NewSize);
+      CurrentMutatorSequence.push_back(M);
+      return NewSize;
+
+    }
+
+  }
+
+  *Data = ' ';
+  return 1;  // Fallback, should not happen frequently.
+
+}
+
+// Mask represents the set of Data bytes that are worth mutating.
+size_t MutationDispatcher::MutateWithMask(uint8_t *Data, size_t Size,
+                                          size_t                 MaxSize,
+                                          const Vector<uint8_t> &Mask) {
+
+  size_t MaskedSize = std::min(Size, Mask.size());
+  // * Copy the worthy bytes into a temporary array T
+  // * Mutate T
+  // * Copy T back.
+  // This is totally unoptimized.
+  auto &T = MutateWithMaskTemp;
+  if (T.size() < Size) T.resize(Size);
+  size_t OneBits = 0;
+  for (size_t I = 0; I < MaskedSize; I++)
+    if (Mask[I]) T[OneBits++] = Data[I];
+
+  if (!OneBits) return 0;
+  assert(!T.empty());
+  size_t NewSize = Mutate(T.data(), OneBits, OneBits);
+  assert(NewSize <= OneBits);
+  (void)NewSize;
+  // Even if NewSize < OneBits we still use all OneBits bytes.
+  for (size_t I = 0, J = 0; I < MaskedSize; I++)
+    if (Mask[I]) Data[I] = T[J++];
+  return Size;
+
+}
+
+void MutationDispatcher::AddWordToManualDictionary(const Word &W) {
+
+  ManualDictionary.push_back({W, std::numeric_limits<size_t>::max()});
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerMutate.h

@@ -0,0 +1,158 @@
+//===- FuzzerMutate.h - Internal header for the Fuzzer ----------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::MutationDispatcher
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_MUTATE_H
+#define LLVM_FUZZER_MUTATE_H
+
+#include "FuzzerDefs.h"
+#include "FuzzerDictionary.h"
+#include "FuzzerOptions.h"
+#include "FuzzerRandom.h"
+
+namespace fuzzer {
+
+class MutationDispatcher {
+public:
+  MutationDispatcher(Random &Rand, const FuzzingOptions &Options);
+  ~MutationDispatcher() {}
+  /// Indicate that we are about to start a new sequence of mutations.
+  void StartMutationSequence();
+  /// Print the current sequence of mutations.
+  void PrintMutationSequence();
+  /// Return the current sequence of mutations.
+  std::string MutationSequence();
+  /// Indicate that the current sequence of mutations was successful.
+  void RecordSuccessfulMutationSequence();
+  /// Mutates data by invoking user-provided mutator.
+  size_t Mutate_Custom(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by invoking user-provided crossover.
+  size_t Mutate_CustomCrossOver(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by shuffling bytes.
+  size_t Mutate_ShuffleBytes(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by erasing bytes.
+  size_t Mutate_EraseBytes(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by inserting a byte.
+  size_t Mutate_InsertByte(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by inserting several repeated bytes.
+  size_t Mutate_InsertRepeatedBytes(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by chanding one byte.
+  size_t Mutate_ChangeByte(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by chanding one bit.
+  size_t Mutate_ChangeBit(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Mutates data by copying/inserting a part of data into a different place.
+  size_t Mutate_CopyPart(uint8_t *Data, size_t Size, size_t MaxSize);
+
+  /// Mutates data by adding a word from the manual dictionary.
+  size_t Mutate_AddWordFromManualDictionary(uint8_t *Data, size_t Size,
+                                            size_t MaxSize);
+
+  /// Mutates data by adding a word from the TORC.
+  size_t Mutate_AddWordFromTORC(uint8_t *Data, size_t Size, size_t MaxSize);
+
+  /// Mutates data by adding a word from the persistent automatic dictionary.
+  size_t Mutate_AddWordFromPersistentAutoDictionary(uint8_t *Data, size_t Size,
+                                                    size_t MaxSize);
+
+  /// Tries to find an ASCII integer in Data, changes it to another ASCII int.
+  size_t Mutate_ChangeASCIIInteger(uint8_t *Data, size_t Size, size_t MaxSize);
+  /// Change a 1-, 2-, 4-, or 8-byte integer in interesting ways.
+  size_t Mutate_ChangeBinaryInteger(uint8_t *Data, size_t Size, size_t MaxSize);
+
+  /// CrossOver Data with CrossOverWith.
+  size_t Mutate_CrossOver(uint8_t *Data, size_t Size, size_t MaxSize);
+
+  /// Applies one of the configured mutations.
+  /// Returns the new size of data which could be up to MaxSize.
+  size_t Mutate(uint8_t *Data, size_t Size, size_t MaxSize);
+
+  /// Applies one of the configured mutations to the bytes of Data
+  /// that have '1' in Mask.
+  /// Mask.size() should be >= Size.
+  size_t MutateWithMask(uint8_t *Data, size_t Size, size_t MaxSize,
+                        const Vector<uint8_t> &Mask);
+
+  /// Applies one of the default mutations. Provided as a service
+  /// to mutation authors.
+  size_t DefaultMutate(uint8_t *Data, size_t Size, size_t MaxSize);
+
+  /// Creates a cross-over of two pieces of Data, returns its size.
+  size_t CrossOver(const uint8_t *Data1, size_t Size1, const uint8_t *Data2,
+                   size_t Size2, uint8_t *Out, size_t MaxOutSize);
+
+  void AddWordToManualDictionary(const Word &W);
+
+  void PrintRecommendedDictionary();
+
+  void SetCrossOverWith(const Unit *U) { CrossOverWith = U; }
+
+  Random &GetRand() { return Rand; }
+
+ private:
+  struct Mutator {
+    size_t (MutationDispatcher::*Fn)(uint8_t *Data, size_t Size, size_t Max);
+    const char *Name;
+  };
+
+  size_t AddWordFromDictionary(Dictionary &D, uint8_t *Data, size_t Size,
+                               size_t MaxSize);
+  size_t MutateImpl(uint8_t *Data, size_t Size, size_t MaxSize,
+                    Vector<Mutator> &Mutators);
+
+  size_t InsertPartOf(const uint8_t *From, size_t FromSize, uint8_t *To,
+                      size_t ToSize, size_t MaxToSize);
+  size_t CopyPartOf(const uint8_t *From, size_t FromSize, uint8_t *To,
+                    size_t ToSize);
+  size_t ApplyDictionaryEntry(uint8_t *Data, size_t Size, size_t MaxSize,
+                              DictionaryEntry &DE);
+
+  template <class T>
+  DictionaryEntry MakeDictionaryEntryFromCMP(T Arg1, T Arg2,
+                                             const uint8_t *Data, size_t Size);
+  DictionaryEntry MakeDictionaryEntryFromCMP(const Word &Arg1, const Word &Arg2,
+                                             const uint8_t *Data, size_t Size);
+  DictionaryEntry MakeDictionaryEntryFromCMP(const void *Arg1, const void *Arg2,
+                                             const void *Arg1Mutation,
+                                             const void *Arg2Mutation,
+                                             size_t ArgSize,
+                                             const uint8_t *Data, size_t Size);
+
+  Random &Rand;
+  const FuzzingOptions Options;
+
+  // Dictionary provided by the user via -dict=DICT_FILE.
+  Dictionary ManualDictionary;
+  // Temporary dictionary modified by the fuzzer itself,
+  // recreated periodically.
+  Dictionary TempAutoDictionary;
+  // Persistent dictionary modified by the fuzzer, consists of
+  // entries that led to successful discoveries in the past mutations.
+  Dictionary PersistentAutoDictionary;
+
+  Vector<DictionaryEntry *> CurrentDictionaryEntrySequence;
+
+  static const size_t kCmpDictionaryEntriesDequeSize = 16;
+  DictionaryEntry CmpDictionaryEntriesDeque[kCmpDictionaryEntriesDequeSize];
+  size_t CmpDictionaryEntriesDequeIdx = 0;
+
+  const Unit *CrossOverWith = nullptr;
+  Vector<uint8_t> MutateInPlaceHere;
+  Vector<uint8_t> MutateWithMaskTemp;
+  // CustomCrossOver needs its own buffer as a custom implementation may call
+  // LLVMFuzzerMutate, which in turn may resize MutateInPlaceHere.
+  Vector<uint8_t> CustomCrossOverInPlaceHere;
+
+  Vector<Mutator> Mutators;
+  Vector<Mutator> DefaultMutators;
+  Vector<Mutator> CurrentMutatorSequence;
+};
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_MUTATE_H

custom_mutators/libfuzzer/FuzzerOptions.h

@@ -0,0 +1,90 @@
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::FuzzingOptions
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_OPTIONS_H
+#define LLVM_FUZZER_OPTIONS_H
+
+#include "FuzzerDefs.h"
+
+namespace fuzzer {
+
+struct FuzzingOptions {
+  int Verbosity = 1;
+  size_t MaxLen = 0;
+  size_t LenControl = 1000;
+  bool KeepSeed = false;
+  int UnitTimeoutSec = 300;
+  int TimeoutExitCode = 70;
+  int OOMExitCode = 71;
+  int InterruptExitCode = 72;
+  int ErrorExitCode = 77;
+  bool IgnoreTimeouts = true;
+  bool IgnoreOOMs = true;
+  bool IgnoreCrashes = false;
+  int MaxTotalTimeSec = 0;
+  int RssLimitMb = 0;
+  int MallocLimitMb = 0;
+  bool DoCrossOver = true;
+  bool CrossOverUniformDist = false;
+  int MutateDepth = 5;
+  bool ReduceDepth = false;
+  bool UseCounters = false;
+  bool UseMemmem = true;
+  bool UseCmp = false;
+  int UseValueProfile = false;
+  bool Shrink = false;
+  bool ReduceInputs = false;
+  int ReloadIntervalSec = 1;
+  bool ShuffleAtStartUp = true;
+  bool PreferSmall = true;
+  size_t MaxNumberOfRuns = -1L;
+  int ReportSlowUnits = 10;
+  bool OnlyASCII = false;
+  bool Entropic = false;
+  size_t EntropicFeatureFrequencyThreshold = 0xFF;
+  size_t EntropicNumberOfRarestFeatures = 100;
+  bool EntropicScalePerExecTime = false;
+  std::string OutputCorpus;
+  std::string ArtifactPrefix = "./";
+  std::string ExactArtifactPath;
+  std::string ExitOnSrcPos;
+  std::string ExitOnItem;
+  std::string FocusFunction;
+  std::string DataFlowTrace;
+  std::string CollectDataFlow;
+  std::string FeaturesDir;
+  std::string MutationGraphFile;
+  std::string StopFile;
+  bool SaveArtifacts = true;
+  bool PrintNEW = true; // Print a status line when new units are found;
+  bool PrintNewCovPcs = false;
+  int PrintNewCovFuncs = 0;
+  bool PrintFinalStats = false;
+  bool PrintCorpusStats = false;
+  bool PrintCoverage = false;
+  bool DumpCoverage = false;
+  bool DetectLeaks = true;
+  int PurgeAllocatorIntervalSec = 1;
+  int  TraceMalloc = 0;
+  bool HandleAbrt = false;
+  bool HandleAlrm = false;
+  bool HandleBus = false;
+  bool HandleFpe = false;
+  bool HandleIll = false;
+  bool HandleInt = false;
+  bool HandleSegv = false;
+  bool HandleTerm = false;
+  bool HandleXfsz = false;
+  bool HandleUsr1 = false;
+  bool HandleUsr2 = false;
+};
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_OPTIONS_H

custom_mutators/libfuzzer/FuzzerPlatform.h

@@ -0,0 +1,163 @@
+//===-- FuzzerPlatform.h --------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Common platform macros.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_PLATFORM_H
+#define LLVM_FUZZER_PLATFORM_H
+
+// Platform detection.
+#ifdef __linux__
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 1
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif __APPLE__
+#define LIBFUZZER_APPLE 1
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif __NetBSD__
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 1
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif __FreeBSD__
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 1
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif __OpenBSD__
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 1
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif _WIN32
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 1
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif __Fuchsia__
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 1
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 0
+#elif __EMSCRIPTEN__
+#define LIBFUZZER_APPLE 0
+#define LIBFUZZER_FUCHSIA 0
+#define LIBFUZZER_LINUX 0
+#define LIBFUZZER_NETBSD 0
+#define LIBFUZZER_FREEBSD 0
+#define LIBFUZZER_OPENBSD 0
+#define LIBFUZZER_WINDOWS 0
+#define LIBFUZZER_EMSCRIPTEN 1
+#else
+#error "Support for your platform has not been implemented"
+#endif
+
+#if defined(_MSC_VER) && !defined(__clang__)
+// MSVC compiler is being used.
+#define LIBFUZZER_MSVC 1
+#else
+#define LIBFUZZER_MSVC 0
+#endif
+
+#ifndef __has_attribute
+#define __has_attribute(x) 0
+#endif
+
+#define LIBFUZZER_POSIX                                                        \
+  (LIBFUZZER_APPLE || LIBFUZZER_LINUX || LIBFUZZER_NETBSD ||                   \
+   LIBFUZZER_FREEBSD || LIBFUZZER_OPENBSD || LIBFUZZER_EMSCRIPTEN)
+
+#ifdef __x86_64
+#if __has_attribute(target)
+#define ATTRIBUTE_TARGET_POPCNT __attribute__((target("popcnt")))
+#else
+#define ATTRIBUTE_TARGET_POPCNT
+#endif
+#else
+#define ATTRIBUTE_TARGET_POPCNT
+#endif
+
+#ifdef __clang__ // avoid gcc warning.
+#if __has_attribute(no_sanitize)
+#define ATTRIBUTE_NO_SANITIZE_MEMORY __attribute__((no_sanitize("memory")))
+#else
+#define ATTRIBUTE_NO_SANITIZE_MEMORY
+#endif
+#define ALWAYS_INLINE __attribute__((always_inline))
+#else
+#define ATTRIBUTE_NO_SANITIZE_MEMORY
+#define ALWAYS_INLINE
+#endif // __clang__
+
+#if LIBFUZZER_WINDOWS
+#define ATTRIBUTE_NO_SANITIZE_ADDRESS
+#else
+#define ATTRIBUTE_NO_SANITIZE_ADDRESS __attribute__((no_sanitize_address))
+#endif
+
+#if LIBFUZZER_WINDOWS
+#define ATTRIBUTE_ALIGNED(X) __declspec(align(X))
+#define ATTRIBUTE_INTERFACE __declspec(dllexport)
+// This is used for __sancov_lowest_stack which is needed for
+// -fsanitize-coverage=stack-depth. That feature is not yet available on
+// Windows, so make the symbol static to avoid linking errors.
+#define ATTRIBUTES_INTERFACE_TLS_INITIAL_EXEC static
+#define ATTRIBUTE_NOINLINE __declspec(noinline)
+#else
+#define ATTRIBUTE_ALIGNED(X) __attribute__((aligned(X)))
+#define ATTRIBUTE_INTERFACE __attribute__((visibility("default")))
+#define ATTRIBUTES_INTERFACE_TLS_INITIAL_EXEC                                  \
+  ATTRIBUTE_INTERFACE __attribute__((tls_model("initial-exec"))) thread_local
+
+#define ATTRIBUTE_NOINLINE __attribute__((noinline))
+#endif
+
+#if defined(__has_feature)
+#if __has_feature(address_sanitizer)
+#define ATTRIBUTE_NO_SANITIZE_ALL ATTRIBUTE_NO_SANITIZE_ADDRESS
+#elif __has_feature(memory_sanitizer)
+#define ATTRIBUTE_NO_SANITIZE_ALL ATTRIBUTE_NO_SANITIZE_MEMORY
+#else
+#define ATTRIBUTE_NO_SANITIZE_ALL
+#endif
+#else
+#define ATTRIBUTE_NO_SANITIZE_ALL
+#endif
+
+#endif // LLVM_FUZZER_PLATFORM_H

custom_mutators/libfuzzer/FuzzerRandom.h

@@ -0,0 +1,38 @@
+//===- FuzzerRandom.h - Internal header for the Fuzzer ----------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::Random
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_RANDOM_H
+#define LLVM_FUZZER_RANDOM_H
+
+#include <random>
+
+namespace fuzzer {
+class Random : public std::minstd_rand {
+ public:
+  Random(unsigned int seed) : std::minstd_rand(seed) {}
+  result_type operator()() { return this->std::minstd_rand::operator()(); }
+  size_t Rand() { return this->operator()(); }
+  size_t RandBool() { return Rand() % 2; }
+  size_t SkewTowardsLast(size_t n) {
+    size_t T = this->operator()(n * n);
+    size_t Res = sqrt(T);
+    return Res;
+  }
+  size_t operator()(size_t n) { return n ? Rand() % n : 0; }
+  intptr_t operator()(intptr_t From, intptr_t To) {
+    assert(From < To);
+    intptr_t RangeSize = To - From + 1;
+    return operator()(RangeSize) + From;
+  }
+};
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_RANDOM_H

custom_mutators/libfuzzer/FuzzerSHA1.cpp

@@ -0,0 +1,269 @@
+//===- FuzzerSHA1.h - Private copy of the SHA1 implementation ---*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// This code is taken from public domain
+// (http://oauth.googlecode.com/svn/code/c/liboauth/src/sha1.c)
+// and modified by adding anonymous namespace, adding an interface
+// function fuzzer::ComputeSHA1() and removing unnecessary code.
+//
+// lib/Fuzzer can not use SHA1 implementation from openssl because
+// openssl may not be available and because we may be fuzzing openssl itself.
+// For the same reason we do not want to depend on SHA1 from LLVM tree.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerSHA1.h"
+#include "FuzzerDefs.h"
+#include "FuzzerPlatform.h"
+
+/* This code is public-domain - it is based on libcrypt
+ * placed in the public domain by Wei Dai and other contributors.
+ */
+
+#include <iomanip>
+#include <sstream>
+#include <stdint.h>
+#include <string.h>
+
+namespace {  // Added for LibFuzzer
+
+#ifdef __BIG_ENDIAN__
+  #define SHA_BIG_ENDIAN
+// Windows is always little endian and MSVC doesn't have <endian.h>
+#elif defined __LITTLE_ENDIAN__ || LIBFUZZER_WINDOWS
+/* override */
+#elif defined __BYTE_ORDER
+  #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+    #define SHA_BIG_ENDIAN
+  #endif
+#else                  // ! defined __LITTLE_ENDIAN__
+  #include <endian.h>  // machine/endian.h
+  #if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+    #define SHA_BIG_ENDIAN
+  #endif
+#endif
+
+/* header */
+
+#define HASH_LENGTH 20
+#define BLOCK_LENGTH 64
+
+typedef struct sha1nfo {
+
+  uint32_t buffer[BLOCK_LENGTH / 4];
+  uint32_t state[HASH_LENGTH / 4];
+  uint32_t byteCount;
+  uint8_t  bufferOffset;
+  uint8_t  keyBuffer[BLOCK_LENGTH];
+  uint8_t  innerHash[HASH_LENGTH];
+
+} sha1nfo;
+
+/* public API - prototypes - TODO: doxygen*/
+
+/**
+ */
+void sha1_init(sha1nfo *s);
+/**
+ */
+void sha1_writebyte(sha1nfo *s, uint8_t data);
+/**
+ */
+void sha1_write(sha1nfo *s, const char *data, size_t len);
+/**
+ */
+uint8_t *sha1_result(sha1nfo *s);
+
+/* code */
+#define SHA1_K0 0x5a827999
+#define SHA1_K20 0x6ed9eba1
+#define SHA1_K40 0x8f1bbcdc
+#define SHA1_K60 0xca62c1d6
+
+void sha1_init(sha1nfo *s) {
+
+  s->state[0] = 0x67452301;
+  s->state[1] = 0xefcdab89;
+  s->state[2] = 0x98badcfe;
+  s->state[3] = 0x10325476;
+  s->state[4] = 0xc3d2e1f0;
+  s->byteCount = 0;
+  s->bufferOffset = 0;
+
+}
+
+uint32_t sha1_rol32(uint32_t number, uint8_t bits) {
+
+  return ((number << bits) | (number >> (32 - bits)));
+
+}
+
+void sha1_hashBlock(sha1nfo *s) {
+
+  uint8_t  i;
+  uint32_t a, b, c, d, e, t;
+
+  a = s->state[0];
+  b = s->state[1];
+  c = s->state[2];
+  d = s->state[3];
+  e = s->state[4];
+  for (i = 0; i < 80; i++) {
+
+    if (i >= 16) {
+
+      t = s->buffer[(i + 13) & 15] ^ s->buffer[(i + 8) & 15] ^
+          s->buffer[(i + 2) & 15] ^ s->buffer[i & 15];
+      s->buffer[i & 15] = sha1_rol32(t, 1);
+
+    }
+
+    if (i < 20) {
+
+      t = (d ^ (b & (c ^ d))) + SHA1_K0;
+
+    } else if (i < 40) {
+
+      t = (b ^ c ^ d) + SHA1_K20;
+
+    } else if (i < 60) {
+
+      t = ((b & c) | (d & (b | c))) + SHA1_K40;
+
+    } else {
+
+      t = (b ^ c ^ d) + SHA1_K60;
+
+    }
+
+    t += sha1_rol32(a, 5) + e + s->buffer[i & 15];
+    e = d;
+    d = c;
+    c = sha1_rol32(b, 30);
+    b = a;
+    a = t;
+
+  }
+
+  s->state[0] += a;
+  s->state[1] += b;
+  s->state[2] += c;
+  s->state[3] += d;
+  s->state[4] += e;
+
+}
+
+void sha1_addUncounted(sha1nfo *s, uint8_t data) {
+
+  uint8_t *const b = (uint8_t *)s->buffer;
+#ifdef SHA_BIG_ENDIAN
+  b[s->bufferOffset] = data;
+#else
+  b[s->bufferOffset ^ 3] = data;
+#endif
+  s->bufferOffset++;
+  if (s->bufferOffset == BLOCK_LENGTH) {
+
+    sha1_hashBlock(s);
+    s->bufferOffset = 0;
+
+  }
+
+}
+
+void sha1_writebyte(sha1nfo *s, uint8_t data) {
+
+  ++s->byteCount;
+  sha1_addUncounted(s, data);
+
+}
+
+void sha1_write(sha1nfo *s, const char *data, size_t len) {
+
+  for (; len--;)
+    sha1_writebyte(s, (uint8_t)*data++);
+
+}
+
+void sha1_pad(sha1nfo *s) {
+
+  // Implement SHA-1 padding (fips180-2 §5.1.1)
+
+  // Pad with 0x80 followed by 0x00 until the end of the block
+  sha1_addUncounted(s, 0x80);
+  while (s->bufferOffset != 56)
+    sha1_addUncounted(s, 0x00);
+
+  // Append length in the last 8 bytes
+  sha1_addUncounted(s, 0);  // We're only using 32 bit lengths
+  sha1_addUncounted(s, 0);  // But SHA-1 supports 64 bit lengths
+  sha1_addUncounted(s, 0);  // So zero pad the top bits
+  sha1_addUncounted(s, s->byteCount >> 29);  // Shifting to multiply by 8
+  sha1_addUncounted(
+      s, s->byteCount >> 21);  // as SHA-1 supports bitstreams as well as
+  sha1_addUncounted(s, s->byteCount >> 13);  // byte.
+  sha1_addUncounted(s, s->byteCount >> 5);
+  sha1_addUncounted(s, s->byteCount << 3);
+
+}
+
+uint8_t *sha1_result(sha1nfo *s) {
+
+  // Pad to complete the last block
+  sha1_pad(s);
+
+#ifndef SHA_BIG_ENDIAN
+  // Swap byte order back
+  int i;
+  for (i = 0; i < 5; i++) {
+
+    s->state[i] = (((s->state[i]) << 24) & 0xff000000) |
+                  (((s->state[i]) << 8) & 0x00ff0000) |
+                  (((s->state[i]) >> 8) & 0x0000ff00) |
+                  (((s->state[i]) >> 24) & 0x000000ff);
+
+  }
+
+#endif
+
+  // Return pointer to hash (20 characters)
+  return (uint8_t *)s->state;
+
+}
+
+}  // namespace
+
+namespace fuzzer {
+
+// The rest is added for LibFuzzer
+void ComputeSHA1(const uint8_t *Data, size_t Len, uint8_t *Out) {
+
+  sha1nfo s;
+  sha1_init(&s);
+  sha1_write(&s, (const char *)Data, Len);
+  memcpy(Out, sha1_result(&s), HASH_LENGTH);
+
+}
+
+std::string Sha1ToString(const uint8_t Sha1[kSHA1NumBytes]) {
+
+  std::stringstream SS;
+  for (int i = 0; i < kSHA1NumBytes; i++)
+    SS << std::hex << std::setfill('0') << std::setw(2) << (unsigned)Sha1[i];
+  return SS.str();
+
+}
+
+std::string Hash(const Unit &U) {
+
+  uint8_t Hash[kSHA1NumBytes];
+  ComputeSHA1(U.data(), U.size(), Hash);
+  return Sha1ToString(Hash);
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerSHA1.h

@@ -0,0 +1,32 @@
+//===- FuzzerSHA1.h - Internal header for the SHA1 utils --------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// SHA1 utils.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_SHA1_H
+#define LLVM_FUZZER_SHA1_H
+
+#include "FuzzerDefs.h"
+#include <cstddef>
+#include <stdint.h>
+
+namespace fuzzer {
+
+// Private copy of SHA1 implementation.
+static const int kSHA1NumBytes = 20;
+
+// Computes SHA1 hash of 'Len' bytes in 'Data', writes kSHA1NumBytes to 'Out'.
+void ComputeSHA1(const uint8_t *Data, size_t Len, uint8_t *Out);
+
+std::string Sha1ToString(const uint8_t Sha1[kSHA1NumBytes]);
+
+std::string Hash(const Unit &U);
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_SHA1_H

custom_mutators/libfuzzer/FuzzerTracePC.cpp

@@ -0,0 +1,819 @@
+//===- FuzzerTracePC.cpp - PC tracing--------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Trace PCs.
+// This module implements __sanitizer_cov_trace_pc_guard[_init],
+// the callback required for -fsanitize-coverage=trace-pc-guard instrumentation.
+//
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerTracePC.h"
+#include "FuzzerBuiltins.h"
+#include "FuzzerBuiltinsMsvc.h"
+#include "FuzzerCorpus.h"
+#include "FuzzerDefs.h"
+#include "FuzzerDictionary.h"
+#include "FuzzerExtFunctions.h"
+#include "FuzzerIO.h"
+#include "FuzzerPlatform.h"
+#include "FuzzerUtil.h"
+#include "FuzzerValueBitMap.h"
+#include <set>
+
+// Used by -fsanitize-coverage=stack-depth to track stack depth
+ATTRIBUTES_INTERFACE_TLS_INITIAL_EXEC uintptr_t __sancov_lowest_stack;
+
+namespace fuzzer {
+
+TracePC TPC;
+
+size_t TracePC::GetTotalPCCoverage() {
+
+  return ObservedPCs.size();
+
+}
+
+void TracePC::HandleInline8bitCountersInit(uint8_t *Start, uint8_t *Stop) {
+
+  if (Start == Stop) return;
+  if (NumModules && Modules[NumModules - 1].Start() == Start) return;
+  assert(NumModules < sizeof(Modules) / sizeof(Modules[0]));
+  auto &   M = Modules[NumModules++];
+  uint8_t *AlignedStart = RoundUpByPage(Start);
+  uint8_t *AlignedStop = RoundDownByPage(Stop);
+  size_t   NumFullPages = AlignedStop > AlignedStart
+                            ? (AlignedStop - AlignedStart) / PageSize()
+                            : 0;
+  bool NeedFirst = Start < AlignedStart || !NumFullPages;
+  bool NeedLast = Stop > AlignedStop && AlignedStop >= AlignedStart;
+  M.NumRegions = NumFullPages + NeedFirst + NeedLast;
+  ;
+  assert(M.NumRegions > 0);
+  M.Regions = new Module::Region[M.NumRegions];
+  assert(M.Regions);
+  size_t R = 0;
+  if (NeedFirst)
+    M.Regions[R++] = {Start, std::min(Stop, AlignedStart), true, false};
+  for (uint8_t *P = AlignedStart; P < AlignedStop; P += PageSize())
+    M.Regions[R++] = {P, P + PageSize(), true, true};
+  if (NeedLast) M.Regions[R++] = {AlignedStop, Stop, true, false};
+  assert(R == M.NumRegions);
+  assert(M.Size() == (size_t)(Stop - Start));
+  assert(M.Stop() == Stop);
+  assert(M.Start() == Start);
+  NumInline8bitCounters += M.Size();
+
+}
+
+void TracePC::HandlePCsInit(const uintptr_t *Start, const uintptr_t *Stop) {
+
+  const PCTableEntry *B = reinterpret_cast<const PCTableEntry *>(Start);
+  const PCTableEntry *E = reinterpret_cast<const PCTableEntry *>(Stop);
+  if (NumPCTables && ModulePCTable[NumPCTables - 1].Start == B) return;
+  assert(NumPCTables < sizeof(ModulePCTable) / sizeof(ModulePCTable[0]));
+  ModulePCTable[NumPCTables++] = {B, E};
+  NumPCsInPCTables += E - B;
+
+}
+
+void TracePC::PrintModuleInfo() {
+
+  if (NumModules) {
+
+    Printf("INFO: Loaded %zd modules   (%zd inline 8-bit counters): ",
+           NumModules, NumInline8bitCounters);
+    for (size_t i = 0; i < NumModules; i++)
+      Printf("%zd [%p, %p), ", Modules[i].Size(), Modules[i].Start(),
+             Modules[i].Stop());
+    Printf("\n");
+
+  }
+
+  if (NumPCTables) {
+
+    Printf("INFO: Loaded %zd PC tables (%zd PCs): ", NumPCTables,
+           NumPCsInPCTables);
+    for (size_t i = 0; i < NumPCTables; i++) {
+
+      Printf("%zd [%p,%p), ", ModulePCTable[i].Stop - ModulePCTable[i].Start,
+             ModulePCTable[i].Start, ModulePCTable[i].Stop);
+
+    }
+
+    Printf("\n");
+
+    if (NumInline8bitCounters && NumInline8bitCounters != NumPCsInPCTables) {
+
+      Printf(
+          "ERROR: The size of coverage PC tables does not match the\n"
+          "number of instrumented PCs. This might be a compiler bug,\n"
+          "please contact the libFuzzer developers.\n"
+          "Also check https://bugs.llvm.org/show_bug.cgi?id=34636\n"
+          "for possible workarounds (tl;dr: don't use the old GNU ld)\n");
+      _Exit(1);
+
+    }
+
+  }
+
+  if (size_t NumExtraCounters = ExtraCountersEnd() - ExtraCountersBegin())
+    Printf("INFO: %zd Extra Counters\n", NumExtraCounters);
+
+}
+
+ATTRIBUTE_NO_SANITIZE_ALL
+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);
+  ValueProfileMap.AddValueModPrime(Idx);
+
+}
+
+/// \return the address of the previous instruction.
+/// Note: the logic is copied from `sanitizer_common/sanitizer_stacktrace.h`
+inline ALWAYS_INLINE uintptr_t GetPreviousInstructionPc(uintptr_t PC) {
+
+#if defined(__arm__)
+  // T32 (Thumb) branch instructions might be 16 or 32 bit long,
+  // so we return (pc-2) in that case in order to be safe.
+  // For A32 mode we return (pc-4) because all instructions are 32 bit long.
+  return (PC - 3) & (~1);
+#elif defined(__powerpc__) || defined(__powerpc64__) || defined(__aarch64__)
+  // PCs are always 4 byte aligned.
+  return PC - 4;
+#elif defined(__sparc__) || defined(__mips__)
+  return PC - 8;
+#else
+  return PC - 1;
+#endif
+
+}
+
+/// \return the address of the next instruction.
+/// Note: the logic is copied from `sanitizer_common/sanitizer_stacktrace.cpp`
+ALWAYS_INLINE uintptr_t TracePC::GetNextInstructionPc(uintptr_t PC) {
+
+#if defined(__mips__)
+  return PC + 8;
+#elif defined(__powerpc__) || defined(__sparc__) || defined(__arm__) || \
+    defined(__aarch64__)
+  return PC + 4;
+#else
+  return PC + 1;
+#endif
+
+}
+
+void TracePC::UpdateObservedPCs() {
+
+  Vector<uintptr_t> CoveredFuncs;
+  auto              ObservePC = [&](const PCTableEntry *TE) {
+
+    if (ObservedPCs.insert(TE).second && DoPrintNewPCs) {
+
+      PrintPC("\tNEW_PC: %p %F %L", "\tNEW_PC: %p",
+              GetNextInstructionPc(TE->PC));
+      Printf("\n");
+
+    }
+
+  };
+
+  auto Observe = [&](const PCTableEntry *TE) {
+
+    if (PcIsFuncEntry(TE))
+      if (++ObservedFuncs[TE->PC] == 1 && NumPrintNewFuncs)
+        CoveredFuncs.push_back(TE->PC);
+    ObservePC(TE);
+
+  };
+
+  if (NumPCsInPCTables) {
+
+    if (NumInline8bitCounters == NumPCsInPCTables) {
+
+      for (size_t i = 0; i < NumModules; i++) {
+
+        auto &M = Modules[i];
+        assert(M.Size() ==
+               (size_t)(ModulePCTable[i].Stop - ModulePCTable[i].Start));
+        for (size_t r = 0; r < M.NumRegions; r++) {
+
+          auto &R = M.Regions[r];
+          if (!R.Enabled) continue;
+          for (uint8_t *P = R.Start; P < R.Stop; P++)
+            if (*P) Observe(&ModulePCTable[i].Start[M.Idx(P)]);
+
+        }
+
+      }
+
+    }
+
+  }
+
+  for (size_t i = 0, N = Min(CoveredFuncs.size(), NumPrintNewFuncs); i < N;
+       i++) {
+
+    Printf("\tNEW_FUNC[%zd/%zd]: ", i + 1, CoveredFuncs.size());
+    PrintPC("%p %F %L", "%p", GetNextInstructionPc(CoveredFuncs[i]));
+    Printf("\n");
+
+  }
+
+}
+
+uintptr_t TracePC::PCTableEntryIdx(const PCTableEntry *TE) {
+
+  size_t TotalTEs = 0;
+  for (size_t i = 0; i < NumPCTables; i++) {
+
+    auto &M = ModulePCTable[i];
+    if (TE >= M.Start && TE < M.Stop) return TotalTEs + TE - M.Start;
+    TotalTEs += M.Stop - M.Start;
+
+  }
+
+  assert(0);
+  return 0;
+
+}
+
+const TracePC::PCTableEntry *TracePC::PCTableEntryByIdx(uintptr_t Idx) {
+
+  for (size_t i = 0; i < NumPCTables; i++) {
+
+    auto & M = ModulePCTable[i];
+    size_t Size = M.Stop - M.Start;
+    if (Idx < Size) return &M.Start[Idx];
+    Idx -= Size;
+
+  }
+
+  return nullptr;
+
+}
+
+static std::string GetModuleName(uintptr_t PC) {
+
+  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 *>(PC), ModulePathRaw, sizeof(ModulePathRaw),
+          &OffsetRaw))
+    return "";
+  return ModulePathRaw;
+
+}
+
+template <class CallBack>
+void TracePC::IterateCoveredFunctions(CallBack CB) {
+
+  for (size_t i = 0; i < NumPCTables; i++) {
+
+    auto &M = ModulePCTable[i];
+    assert(M.Start < M.Stop);
+    auto ModuleName = GetModuleName(M.Start->PC);
+    for (auto NextFE = M.Start; NextFE < M.Stop;) {
+
+      auto FE = NextFE;
+      assert(PcIsFuncEntry(FE) && "Not a function entry point");
+      do {
+
+        NextFE++;
+
+      } while (NextFE < M.Stop && !(PcIsFuncEntry(NextFE)));
+
+      CB(FE, NextFE, ObservedFuncs[FE->PC]);
+
+    }
+
+  }
+
+}
+
+void TracePC::SetFocusFunction(const std::string &FuncName) {
+
+  // This function should be called once.
+  assert(!FocusFunctionCounterPtr);
+  // "auto" is not a valid function name. If this function is called with "auto"
+  // that means the auto focus functionality failed.
+  if (FuncName.empty() || FuncName == "auto") return;
+  for (size_t M = 0; M < NumModules; M++) {
+
+    auto & PCTE = ModulePCTable[M];
+    size_t N = PCTE.Stop - PCTE.Start;
+    for (size_t I = 0; I < N; I++) {
+
+      if (!(PcIsFuncEntry(&PCTE.Start[I]))) continue;  // not a function entry.
+      auto Name = DescribePC("%F", GetNextInstructionPc(PCTE.Start[I].PC));
+      if (Name[0] == 'i' && Name[1] == 'n' && Name[2] == ' ')
+        Name = Name.substr(3, std::string::npos);
+      if (FuncName != Name) continue;
+      Printf("INFO: Focus function is set to '%s'\n", Name.c_str());
+      FocusFunctionCounterPtr = Modules[M].Start() + I;
+      return;
+
+    }
+
+  }
+
+  Printf(
+      "ERROR: Failed to set focus function. Make sure the function name is "
+      "valid (%s) and symbolization is enabled.\n",
+      FuncName.c_str());
+  exit(1);
+
+}
+
+bool TracePC::ObservedFocusFunction() {
+
+  return FocusFunctionCounterPtr && *FocusFunctionCounterPtr;
+
+}
+
+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;
+
+  }
+
+  Printf("COVERAGE:\n");
+  auto CoveredFunctionCallback = [&](const PCTableEntry *First,
+                                     const PCTableEntry *Last,
+                                     uintptr_t           Counter) {
+
+    assert(First < Last);
+    auto        VisualizePC = GetNextInstructionPc(First->PC);
+    std::string FileStr = DescribePC("%s", VisualizePC);
+    if (!IsInterestingCoverageFile(FileStr)) return;
+    std::string FunctionStr = DescribePC("%F", VisualizePC);
+    if (FunctionStr.find("in ") == 0) FunctionStr = FunctionStr.substr(3);
+    std::string       LineStr = DescribePC("%l", VisualizePC);
+    size_t            NumEdges = Last - First;
+    Vector<uintptr_t> UncoveredPCs;
+    for (auto TE = First; TE < Last; TE++)
+      if (!ObservedPCs.count(TE)) UncoveredPCs.push_back(TE->PC);
+    Printf("%sCOVERED_FUNC: hits: %zd", Counter ? "" : "UN", Counter);
+    Printf(" edges: %zd/%zd", NumEdges - UncoveredPCs.size(), NumEdges);
+    Printf(" %s %s:%s\n", FunctionStr.c_str(), FileStr.c_str(),
+           LineStr.c_str());
+    if (Counter)
+      for (auto PC : UncoveredPCs)
+        Printf("  UNCOVERED_PC: %s\n",
+               DescribePC("%s:%l", GetNextInstructionPc(PC)).c_str());
+
+  };
+
+  IterateCoveredFunctions(CoveredFunctionCallback);
+
+}
+
+// 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_ALL
+void TracePC::AddValueForMemcmp(void *caller_pc, const void *s1, const void *s2,
+                                size_t n, bool StopAtZero) {
+
+  if (!n) return;
+  size_t         Len = std::min(n, Word::GetMaxSize());
+  const uint8_t *A1 = reinterpret_cast<const uint8_t *>(s1);
+  const uint8_t *A2 = reinterpret_cast<const uint8_t *>(s2);
+  uint8_t        B1[Word::kMaxSize];
+  uint8_t        B2[Word::kMaxSize];
+  // Copy the data into locals in this non-msan-instrumented function
+  // to avoid msan complaining further.
+  size_t Hash = 0;  // Compute some simple hash of both strings.
+  for (size_t i = 0; i < Len; i++) {
+
+    B1[i] = A1[i];
+    B2[i] = A2[i];
+    size_t T = B1[i];
+    Hash ^= (T << 8) | B2[i];
+
+  }
+
+  size_t  I = 0;
+  uint8_t HammingDistance = 0;
+  for (; I < Len; I++) {
+
+    if (B1[I] != B2[I] || (StopAtZero && B1[I] == 0)) {
+
+      HammingDistance = Popcountll(B1[I] ^ B2[I]);
+      break;
+
+    }
+
+  }
+
+  size_t PC = reinterpret_cast<size_t>(caller_pc);
+  size_t Idx = (PC & 4095) | (I << 12);
+  Idx += HammingDistance;
+  ValueProfileMap.AddValue(Idx);
+  TORCW.Insert(Idx ^ Hash, Word(B1, Len), Word(B2, Len));
+
+}
+
+template <class T>
+ATTRIBUTE_TARGET_POPCNT ALWAYS_INLINE ATTRIBUTE_NO_SANITIZE_ALL void
+TracePC::HandleCmp(uintptr_t PC, T Arg1, T Arg2) {
+
+  uint64_t ArgXor = Arg1 ^ Arg2;
+  if (sizeof(T) == 4)
+    TORC4.Insert(ArgXor, Arg1, Arg2);
+  else if (sizeof(T) == 8)
+    TORC8.Insert(ArgXor, Arg1, Arg2);
+  uint64_t HammingDistance = Popcountll(ArgXor);  // [0,64]
+  uint64_t AbsoluteDistance = (Arg1 == Arg2 ? 0 : Clzll(Arg1 - Arg2) + 1);
+  ValueProfileMap.AddValue(PC * 128 + HammingDistance);
+  ValueProfileMap.AddValue(PC * 128 + 64 + AbsoluteDistance);
+
+}
+
+static size_t InternalStrnlen(const char *S, size_t MaxLen) {
+
+  size_t Len = 0;
+  for (; Len < MaxLen && S[Len]; Len++) {}
+  return Len;
+
+}
+
+// Finds min of (strlen(S1), strlen(S2)).
+// Needed bacause one of these strings may actually be non-zero terminated.
+static size_t InternalStrnlen2(const char *S1, const char *S2) {
+
+  size_t Len = 0;
+  for (; S1[Len] && S2[Len]; Len++) {}
+  return Len;
+
+}
+
+void TracePC::ClearInlineCounters() {
+
+  IterateCounterRegions([](const Module::Region &R) {
+
+    if (R.Enabled) memset(R.Start, 0, R.Stop - R.Start);
+
+  });
+
+}
+
+ATTRIBUTE_NO_SANITIZE_ALL
+void TracePC::RecordInitialStack() {
+
+  int stack;
+  __sancov_lowest_stack = InitialStack = reinterpret_cast<uintptr_t>(&stack);
+
+}
+
+uintptr_t TracePC::GetMaxStackOffset() const {
+
+  return InitialStack - __sancov_lowest_stack;  // Stack grows down
+
+}
+
+void WarnAboutDeprecatedInstrumentation(const char *flag) {
+
+  // Use RawPrint because Printf cannot be used on Windows before OutputFile is
+  // initialized.
+  RawPrint(flag);
+  RawPrint(
+      " is no longer supported by libFuzzer.\n"
+      "Please either migrate to a compiler that supports -fsanitize=fuzzer\n"
+      "or use an older version of libFuzzer\n");
+  exit(1);
+
+}
+
+}  // namespace fuzzer
+
+extern "C" {
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+void __sanitizer_cov_trace_pc_guard(uint32_t *Guard) {
+
+  fuzzer::WarnAboutDeprecatedInstrumentation(
+      "-fsanitize-coverage=trace-pc-guard");
+
+}
+
+// Best-effort support for -fsanitize-coverage=trace-pc, which is available
+// in both Clang and GCC.
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+void __sanitizer_cov_trace_pc() {
+
+  fuzzer::WarnAboutDeprecatedInstrumentation("-fsanitize-coverage=trace-pc");
+
+}
+
+ATTRIBUTE_INTERFACE
+void __sanitizer_cov_trace_pc_guard_init(uint32_t *Start, uint32_t *Stop) {
+
+  fuzzer::WarnAboutDeprecatedInstrumentation(
+      "-fsanitize-coverage=trace-pc-guard");
+
+}
+
+ATTRIBUTE_INTERFACE
+void __sanitizer_cov_8bit_counters_init(uint8_t *Start, uint8_t *Stop) {
+
+  fuzzer::TPC.HandleInline8bitCountersInit(Start, Stop);
+
+}
+
+ATTRIBUTE_INTERFACE
+void __sanitizer_cov_pcs_init(const uintptr_t *pcs_beg,
+                              const uintptr_t *pcs_end) {
+
+  fuzzer::TPC.HandlePCsInit(pcs_beg, pcs_end);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+void __sanitizer_cov_trace_pc_indir(uintptr_t Callee) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCallerCallee(PC, Callee);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_cmp8(uint64_t Arg1, uint64_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+// Now the __sanitizer_cov_trace_const_cmp[1248] callbacks just mimic
+// the behaviour of __sanitizer_cov_trace_cmp[1248] ones. This, however,
+// should be changed later to make full use of instrumentation.
+void __sanitizer_cov_trace_const_cmp8(uint64_t Arg1, uint64_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_cmp4(uint32_t Arg1, uint32_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_const_cmp4(uint32_t Arg1, uint32_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_cmp2(uint16_t Arg1, uint16_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_const_cmp2(uint16_t Arg1, uint16_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_cmp1(uint8_t Arg1, uint8_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_const_cmp1(uint8_t Arg1, uint8_t Arg2) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Arg1, Arg2);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+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: all switch values are small.
+  // We may want to skip this at compile-time, but it will make the
+  // instrumentation less general.
+  if (Vals[N - 1] < 256) return;
+  // Also skip small inputs values, they won't give good signal.
+  if (Val < 256) return;
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  size_t    i;
+  uint64_t  Smaller = 0;
+  uint64_t  Larger = ~(uint64_t)0;
+  // Find two switch values such that Smaller < Val < Larger.
+  // Use 0 and 0xfff..f as the defaults.
+  for (i = 0; i < N; i++) {
+
+    if (Val < Vals[i]) {
+
+      Larger = Vals[i];
+      break;
+
+    }
+
+    if (Val > Vals[i]) Smaller = Vals[i];
+
+  }
+
+  // Apply HandleCmp to {Val,Smaller} and {Val, Larger},
+  // use i as the PC modifier for HandleCmp.
+  if (ValSizeInBits == 16) {
+
+    fuzzer::TPC.HandleCmp(PC + 2 * i, static_cast<uint16_t>(Val),
+                          (uint16_t)(Smaller));
+    fuzzer::TPC.HandleCmp(PC + 2 * i + 1, static_cast<uint16_t>(Val),
+                          (uint16_t)(Larger));
+
+  } else if (ValSizeInBits == 32) {
+
+    fuzzer::TPC.HandleCmp(PC + 2 * i, static_cast<uint32_t>(Val),
+                          (uint32_t)(Smaller));
+    fuzzer::TPC.HandleCmp(PC + 2 * i + 1, static_cast<uint32_t>(Val),
+                          (uint32_t)(Larger));
+
+  } else {
+
+    fuzzer::TPC.HandleCmp(PC + 2 * i, Val, Smaller);
+    fuzzer::TPC.HandleCmp(PC + 2 * i + 1, Val, Larger);
+
+  }
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_div4(uint32_t Val) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Val, (uint32_t)0);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_div8(uint64_t Val) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Val, (uint64_t)0);
+
+}
+
+ATTRIBUTE_INTERFACE
+ATTRIBUTE_NO_SANITIZE_ALL
+ATTRIBUTE_TARGET_POPCNT
+void __sanitizer_cov_trace_gep(uintptr_t Idx) {
+
+  uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
+  fuzzer::TPC.HandleCmp(PC, Idx, (uintptr_t)0);
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_memcmp(void *caller_pc, const void *s1, const void *s2,
+                             size_t n, int result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  if (result == 0) return;  // No reason to mutate.
+  if (n <= 1) return;       // Not interesting.
+  fuzzer::TPC.AddValueForMemcmp(caller_pc, s1, s2, n, /*StopAtZero*/ false);
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_strncmp(void *caller_pc, const char *s1, const char *s2,
+                              size_t n, int result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  if (result == 0) return;  // No reason to mutate.
+  size_t Len1 = fuzzer::InternalStrnlen(s1, n);
+  size_t Len2 = fuzzer::InternalStrnlen(s2, n);
+  n = std::min(n, Len1);
+  n = std::min(n, Len2);
+  if (n <= 1) return;  // Not interesting.
+  fuzzer::TPC.AddValueForMemcmp(caller_pc, s1, s2, n, /*StopAtZero*/ true);
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_strcmp(void *caller_pc, const char *s1, const char *s2,
+                             int result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  if (result == 0) return;  // No reason to mutate.
+  size_t N = fuzzer::InternalStrnlen2(s1, s2);
+  if (N <= 1) return;  // Not interesting.
+  fuzzer::TPC.AddValueForMemcmp(caller_pc, s1, s2, N, /*StopAtZero*/ true);
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_strncasecmp(void *called_pc, const char *s1,
+                                  const char *s2, size_t n, int result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  return __sanitizer_weak_hook_strncmp(called_pc, s1, s2, n, result);
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_strcasecmp(void *called_pc, const char *s1,
+                                 const char *s2, int result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  return __sanitizer_weak_hook_strcmp(called_pc, s1, s2, result);
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_strstr(void *called_pc, const char *s1, const char *s2,
+                             char *result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  fuzzer::TPC.MMT.Add(reinterpret_cast<const uint8_t *>(s2), strlen(s2));
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_strcasestr(void *called_pc, const char *s1,
+                                 const char *s2, char *result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  fuzzer::TPC.MMT.Add(reinterpret_cast<const uint8_t *>(s2), strlen(s2));
+
+}
+
+ATTRIBUTE_INTERFACE ATTRIBUTE_NO_SANITIZE_MEMORY void
+__sanitizer_weak_hook_memmem(void *called_pc, const void *s1, size_t len1,
+                             const void *s2, size_t len2, void *result) {
+
+  if (!fuzzer::RunningUserCallback) return;
+  fuzzer::TPC.MMT.Add(reinterpret_cast<const uint8_t *>(s2), len2);
+
+}
+
+}  // extern "C"
+

custom_mutators/libfuzzer/FuzzerTracePC.h

@@ -0,0 +1,291 @@
+//===- FuzzerTracePC.h - Internal header for the Fuzzer ---------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// fuzzer::TracePC
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_TRACE_PC
+#define LLVM_FUZZER_TRACE_PC
+
+#include "FuzzerDefs.h"
+#include "FuzzerDictionary.h"
+#include "FuzzerValueBitMap.h"
+
+#include <set>
+#include <unordered_map>
+
+namespace fuzzer {
+
+// TableOfRecentCompares (TORC) remembers the most recently performed
+// comparisons of type T.
+// We record the arguments of CMP instructions in this table unconditionally
+// because it seems cheaper this way than to compute some expensive
+// conditions inside __sanitizer_cov_trace_cmp*.
+// After the unit has been executed we may decide to use the contents of
+// this table to populate a Dictionary.
+template<class T, size_t kSizeT>
+struct TableOfRecentCompares {
+  static const size_t kSize = kSizeT;
+  struct Pair {
+    T A, B;
+  };
+  ATTRIBUTE_NO_SANITIZE_ALL
+  void Insert(size_t Idx, const T &Arg1, const T &Arg2) {
+    Idx = Idx % kSize;
+    Table[Idx].A = Arg1;
+    Table[Idx].B = Arg2;
+  }
+
+  Pair Get(size_t I) { return Table[I % kSize]; }
+
+  Pair Table[kSize];
+};
+
+template <size_t kSizeT>
+struct MemMemTable {
+  static const size_t kSize = kSizeT;
+  Word MemMemWords[kSize];
+  Word EmptyWord;
+
+  void Add(const uint8_t *Data, size_t Size) {
+    if (Size <= 2) return;
+    Size = std::min(Size, Word::GetMaxSize());
+    size_t Idx = SimpleFastHash(Data, Size) % kSize;
+    MemMemWords[Idx].Set(Data, Size);
+  }
+  const Word &Get(size_t Idx) {
+    for (size_t i = 0; i < kSize; i++) {
+      const Word &W = MemMemWords[(Idx + i) % kSize];
+      if (W.size()) return W;
+    }
+    EmptyWord.Set(nullptr, 0);
+    return EmptyWord;
+  }
+};
+
+class TracePC {
+ public:
+  void HandleInline8bitCountersInit(uint8_t *Start, uint8_t *Stop);
+  void HandlePCsInit(const uintptr_t *Start, const uintptr_t *Stop);
+  void HandleCallerCallee(uintptr_t Caller, uintptr_t Callee);
+  template <class T> void HandleCmp(uintptr_t PC, T Arg1, T Arg2);
+  size_t GetTotalPCCoverage();
+  void SetUseCounters(bool UC) { UseCounters = UC; }
+  void SetUseValueProfileMask(uint32_t VPMask) { UseValueProfileMask = VPMask; }
+  void SetPrintNewPCs(bool P) { DoPrintNewPCs = P; }
+  void SetPrintNewFuncs(size_t P) { NumPrintNewFuncs = P; }
+  void UpdateObservedPCs();
+  template <class Callback> void CollectFeatures(Callback CB) const;
+
+  void ResetMaps() {
+    ValueProfileMap.Reset();
+    ClearExtraCounters();
+    ClearInlineCounters();
+  }
+
+  void ClearInlineCounters();
+
+  void UpdateFeatureSet(size_t CurrentElementIdx, size_t CurrentElementSize);
+  void PrintFeatureSet();
+
+  void PrintModuleInfo();
+
+  void PrintCoverage();
+
+  template<class CallBack>
+  void IterateCoveredFunctions(CallBack CB);
+
+  void AddValueForMemcmp(void *caller_pc, const void *s1, const void *s2,
+                         size_t n, bool StopAtZero);
+
+  TableOfRecentCompares<uint32_t, 32> TORC4;
+  TableOfRecentCompares<uint64_t, 32> TORC8;
+  TableOfRecentCompares<Word, 32> TORCW;
+  MemMemTable<1024> MMT;
+
+  void RecordInitialStack();
+  uintptr_t GetMaxStackOffset() const;
+
+  template<class CallBack>
+  void ForEachObservedPC(CallBack CB) {
+    for (auto PC : ObservedPCs)
+      CB(PC);
+  }
+
+  void SetFocusFunction(const std::string &FuncName);
+  bool ObservedFocusFunction();
+
+  struct PCTableEntry {
+    uintptr_t PC, PCFlags;
+  };
+
+  uintptr_t PCTableEntryIdx(const PCTableEntry *TE);
+  const PCTableEntry *PCTableEntryByIdx(uintptr_t Idx);
+  static uintptr_t GetNextInstructionPc(uintptr_t PC);
+  bool PcIsFuncEntry(const PCTableEntry *TE) { return TE->PCFlags & 1; }
+
+private:
+  bool UseCounters = false;
+  uint32_t UseValueProfileMask = false;
+  bool DoPrintNewPCs = false;
+  size_t NumPrintNewFuncs = 0;
+
+  // Module represents the array of 8-bit counters split into regions
+  // such that every region, except maybe the first and the last one, is one
+  // full page.
+  struct Module {
+    struct Region {
+      uint8_t *Start, *Stop;
+      bool Enabled;
+      bool OneFullPage;
+    };
+    Region *Regions;
+    size_t NumRegions;
+    uint8_t *Start() { return Regions[0].Start; }
+    uint8_t *Stop()  { return Regions[NumRegions - 1].Stop; }
+    size_t Size()   { return Stop() - Start(); }
+    size_t  Idx(uint8_t *P) {
+      assert(P >= Start() && P < Stop());
+      return P - Start();
+    }
+  };
+
+  Module Modules[4096];
+  size_t NumModules;  // linker-initialized.
+  size_t NumInline8bitCounters;
+
+  template <class Callback>
+  void IterateCounterRegions(Callback CB) {
+    for (size_t m = 0; m < NumModules; m++)
+      for (size_t r = 0; r < Modules[m].NumRegions; r++)
+        CB(Modules[m].Regions[r]);
+  }
+
+  struct { const PCTableEntry *Start, *Stop; } ModulePCTable[4096];
+  size_t NumPCTables;
+  size_t NumPCsInPCTables;
+
+  Set<const PCTableEntry*> ObservedPCs;
+  std::unordered_map<uintptr_t, uintptr_t> ObservedFuncs;  // PC => Counter.
+
+  uint8_t *FocusFunctionCounterPtr = nullptr;
+
+  ValueBitMap ValueProfileMap;
+  uintptr_t InitialStack;
+};
+
+template <class Callback>
+// void Callback(size_t FirstFeature, size_t Idx, uint8_t Value);
+ATTRIBUTE_NO_SANITIZE_ALL
+size_t ForEachNonZeroByte(const uint8_t *Begin, const uint8_t *End,
+                        size_t FirstFeature, Callback Handle8bitCounter) {
+  typedef uintptr_t LargeType;
+  const size_t Step = sizeof(LargeType) / sizeof(uint8_t);
+  const size_t StepMask = Step - 1;
+  auto P = Begin;
+  // Iterate by 1 byte until either the alignment boundary or the end.
+  for (; reinterpret_cast<uintptr_t>(P) & StepMask && P < End; P++)
+    if (uint8_t V = *P)
+      Handle8bitCounter(FirstFeature, P - Begin, V);
+
+  // Iterate by Step bytes at a time.
+  for (; P < End; P += Step)
+    if (LargeType Bundle = *reinterpret_cast<const LargeType *>(P)) {
+      Bundle = HostToLE(Bundle);
+      for (size_t I = 0; I < Step; I++, Bundle >>= 8)
+        if (uint8_t V = Bundle & 0xff)
+          Handle8bitCounter(FirstFeature, P - Begin + I, V);
+    }
+
+  // Iterate by 1 byte until the end.
+  for (; P < End; P++)
+    if (uint8_t V = *P)
+      Handle8bitCounter(FirstFeature, P - Begin, V);
+  return End - Begin;
+}
+
+// Given a non-zero Counter returns a number in the range [0,7].
+template<class T>
+unsigned CounterToFeature(T Counter) {
+    // Returns a feature number by placing Counters into buckets as illustrated
+    // below.
+    //
+    // Counter bucket: [1] [2] [3] [4-7] [8-15] [16-31] [32-127] [128+]
+    // Feature number:  0   1   2    3     4       5       6       7
+    //
+    // This is a heuristic taken from AFL (see
+    // http://lcamtuf.coredump.cx/afl/technical_details.txt).
+    //
+    // This implementation may change in the future so clients should
+    // not rely on it.
+    assert(Counter);
+    unsigned Bit = 0;
+    /**/ if (Counter >= 128) Bit = 7;
+    else if (Counter >= 32) Bit = 6;
+    else if (Counter >= 16) Bit = 5;
+    else if (Counter >= 8) Bit = 4;
+    else if (Counter >= 4) Bit = 3;
+    else if (Counter >= 3) Bit = 2;
+    else if (Counter >= 2) Bit = 1;
+    return Bit;
+}
+
+template <class Callback>  // void Callback(size_t Feature)
+ATTRIBUTE_NO_SANITIZE_ADDRESS
+ATTRIBUTE_NOINLINE
+void TracePC::CollectFeatures(Callback HandleFeature) const {
+  auto Handle8bitCounter = [&](size_t FirstFeature,
+                               size_t Idx, uint8_t Counter) {
+    if (UseCounters)
+      HandleFeature(FirstFeature + Idx * 8 + CounterToFeature(Counter));
+    else
+      HandleFeature(FirstFeature + Idx);
+  };
+
+  size_t FirstFeature = 0;
+
+  for (size_t i = 0; i < NumModules; i++) {
+    for (size_t r = 0; r < Modules[i].NumRegions; r++) {
+      if (!Modules[i].Regions[r].Enabled) continue;
+      FirstFeature += 8 * ForEachNonZeroByte(Modules[i].Regions[r].Start,
+                                             Modules[i].Regions[r].Stop,
+                                             FirstFeature, Handle8bitCounter);
+    }
+  }
+
+  FirstFeature +=
+      8 * ForEachNonZeroByte(ExtraCountersBegin(), ExtraCountersEnd(),
+                             FirstFeature, Handle8bitCounter);
+
+  if (UseValueProfileMask) {
+    ValueProfileMap.ForEach([&](size_t Idx) {
+      HandleFeature(FirstFeature + Idx);
+    });
+    FirstFeature += ValueProfileMap.SizeInBits();
+  }
+
+  // Step function, grows similar to 8 * Log_2(A).
+  auto StackDepthStepFunction = [](uint32_t A) -> uint32_t {
+    if (!A) return A;
+    uint32_t Log2 = Log(A);
+    if (Log2 < 3) return A;
+    Log2 -= 3;
+    return (Log2 + 1) * 8 + ((A >> Log2) & 7);
+  };
+  assert(StackDepthStepFunction(1024) == 64);
+  assert(StackDepthStepFunction(1024 * 4) == 80);
+  assert(StackDepthStepFunction(1024 * 1024) == 144);
+
+  if (auto MaxStackOffset = GetMaxStackOffset())
+    HandleFeature(FirstFeature + StackDepthStepFunction(MaxStackOffset / 8));
+}
+
+extern TracePC TPC;
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_TRACE_PC

custom_mutators/libfuzzer/FuzzerUtil.cpp

@@ -0,0 +1,314 @@
+//===- FuzzerUtil.cpp - Misc utils ----------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Misc utils.
+//===----------------------------------------------------------------------===//
+
+#include "FuzzerUtil.h"
+#include "FuzzerIO.h"
+#include "FuzzerInternal.h"
+#include <cassert>
+#include <chrono>
+#include <cstring>
+#include <errno.h>
+#include <mutex>
+#include <signal.h>
+#include <sstream>
+#include <stdio.h>
+#include <sys/types.h>
+#include <thread>
+
+namespace fuzzer {
+
+void PrintHexArray(const uint8_t *Data, size_t Size, const char *PrintAfter) {
+
+  for (size_t i = 0; i < Size; i++)
+    Printf("0x%x,", (unsigned)Data[i]);
+  Printf("%s", PrintAfter);
+
+}
+
+void Print(const Unit &v, const char *PrintAfter) {
+
+  PrintHexArray(v.data(), v.size(), PrintAfter);
+
+}
+
+void PrintASCIIByte(uint8_t Byte) {
+
+  if (Byte == '\\')
+    Printf("\\\\");
+  else if (Byte == '"')
+    Printf("\\\"");
+  else if (Byte >= 32 && Byte < 127)
+    Printf("%c", Byte);
+  else
+    Printf("\\x%02x", Byte);
+
+}
+
+void PrintASCII(const uint8_t *Data, size_t Size, const char *PrintAfter) {
+
+  for (size_t i = 0; i < Size; i++)
+    PrintASCIIByte(Data[i]);
+  Printf("%s", PrintAfter);
+
+}
+
+void PrintASCII(const Unit &U, const char *PrintAfter) {
+
+  PrintASCII(U.data(), U.size(), PrintAfter);
+
+}
+
+bool ToASCII(uint8_t *Data, size_t Size) {
+
+  bool Changed = false;
+  for (size_t i = 0; i < Size; i++) {
+
+    uint8_t &X = Data[i];
+    auto     NewX = X;
+    NewX &= 127;
+    if (!isspace(NewX) && !isprint(NewX)) NewX = ' ';
+    Changed |= NewX != X;
+    X = NewX;
+
+  }
+
+  return Changed;
+
+}
+
+bool IsASCII(const Unit &U) {
+
+  return IsASCII(U.data(), U.size());
+
+}
+
+bool IsASCII(const uint8_t *Data, size_t Size) {
+
+  for (size_t i = 0; i < Size; i++)
+    if (!(isprint(Data[i]) || isspace(Data[i]))) return false;
+  return true;
+
+}
+
+bool ParseOneDictionaryEntry(const std::string &Str, Unit *U) {
+
+  U->clear();
+  if (Str.empty()) return false;
+  size_t L = 0, R = Str.size() - 1;  // We are parsing the range [L,R].
+  // Skip spaces from both sides.
+  while (L < R && isspace(Str[L]))
+    L++;
+  while (R > L && isspace(Str[R]))
+    R--;
+  if (R - L < 2) return false;
+  // Check the closing "
+  if (Str[R] != '"') return false;
+  R--;
+  // Find the opening "
+  while (L < R && Str[L] != '"')
+    L++;
+  if (L >= R) return false;
+  assert(Str[L] == '\"');
+  L++;
+  assert(L <= R);
+  for (size_t Pos = L; Pos <= R; Pos++) {
+
+    uint8_t V = (uint8_t)Str[Pos];
+    if (!isprint(V) && !isspace(V)) return false;
+    if (V == '\\') {
+
+      // Handle '\\'
+      if (Pos + 1 <= R && (Str[Pos + 1] == '\\' || Str[Pos + 1] == '"')) {
+
+        U->push_back(Str[Pos + 1]);
+        Pos++;
+        continue;
+
+      }
+
+      // Handle '\xAB'
+      if (Pos + 3 <= R && Str[Pos + 1] == 'x' && isxdigit(Str[Pos + 2]) &&
+          isxdigit(Str[Pos + 3])) {
+
+        char Hex[] = "0xAA";
+        Hex[2] = Str[Pos + 2];
+        Hex[3] = Str[Pos + 3];
+        U->push_back(strtol(Hex, nullptr, 16));
+        Pos += 3;
+        continue;
+
+      }
+
+      return false;  // Invalid escape.
+
+    } else {
+
+      // Any other character.
+      U->push_back(V);
+
+    }
+
+  }
+
+  return true;
+
+}
+
+bool ParseDictionaryFile(const std::string &Text, Vector<Unit> *Units) {
+
+  if (Text.empty()) {
+
+    Printf("ParseDictionaryFile: file does not exist or is empty\n");
+    return false;
+
+  }
+
+  std::istringstream ISS(Text);
+  Units->clear();
+  Unit        U;
+  int         LineNo = 0;
+  std::string S;
+  while (std::getline(ISS, S, '\n')) {
+
+    LineNo++;
+    size_t Pos = 0;
+    while (Pos < S.size() && isspace(S[Pos]))
+      Pos++;                        // Skip spaces.
+    if (Pos == S.size()) continue;  // Empty line.
+    if (S[Pos] == '#') continue;    // Comment line.
+    if (ParseOneDictionaryEntry(S, &U)) {
+
+      Units->push_back(U);
+
+    } else {
+
+      Printf("ParseDictionaryFile: error in line %d\n\t\t%s\n", LineNo,
+             S.c_str());
+      return false;
+
+    }
+
+  }
+
+  return true;
+
+}
+
+// Code duplicated (and tested) in llvm/include/llvm/Support/Base64.h
+std::string Base64(const Unit &U) {
+
+  static const char Table[] =
+      "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
+      "abcdefghijklmnopqrstuvwxyz"
+      "0123456789+/";
+  std::string Buffer;
+  Buffer.resize(((U.size() + 2) / 3) * 4);
+
+  size_t i = 0, j = 0;
+  for (size_t n = U.size() / 3 * 3; i < n; i += 3, j += 4) {
+
+    uint32_t x = ((unsigned char)U[i] << 16) | ((unsigned char)U[i + 1] << 8) |
+                 (unsigned char)U[i + 2];
+    Buffer[j + 0] = Table[(x >> 18) & 63];
+    Buffer[j + 1] = Table[(x >> 12) & 63];
+    Buffer[j + 2] = Table[(x >> 6) & 63];
+    Buffer[j + 3] = Table[x & 63];
+
+  }
+
+  if (i + 1 == U.size()) {
+
+    uint32_t x = ((unsigned char)U[i] << 16);
+    Buffer[j + 0] = Table[(x >> 18) & 63];
+    Buffer[j + 1] = Table[(x >> 12) & 63];
+    Buffer[j + 2] = '=';
+    Buffer[j + 3] = '=';
+
+  } else if (i + 2 == U.size()) {
+
+    uint32_t x = ((unsigned char)U[i] << 16) | ((unsigned char)U[i + 1] << 8);
+    Buffer[j + 0] = Table[(x >> 18) & 63];
+    Buffer[j + 1] = Table[(x >> 12) & 63];
+    Buffer[j + 2] = Table[(x >> 6) & 63];
+    Buffer[j + 3] = '=';
+
+  }
+
+  return Buffer;
+
+}
+
+static std::mutex SymbolizeMutex;
+
+std::string DescribePC(const char *SymbolizedFMT, uintptr_t PC) {
+
+  std::unique_lock<std::mutex> l(SymbolizeMutex, std::try_to_lock);
+  if (!EF->__sanitizer_symbolize_pc || !l.owns_lock())
+    return "<can not symbolize>";
+  char PcDescr[1024] = {};
+  EF->__sanitizer_symbolize_pc(reinterpret_cast<void *>(PC), SymbolizedFMT,
+                               PcDescr, sizeof(PcDescr));
+  PcDescr[sizeof(PcDescr) - 1] = 0;  // Just in case.
+  return PcDescr;
+
+}
+
+void PrintPC(const char *SymbolizedFMT, const char *FallbackFMT, uintptr_t PC) {
+
+  if (EF->__sanitizer_symbolize_pc)
+    Printf("%s", DescribePC(SymbolizedFMT, PC).c_str());
+  else
+    Printf(FallbackFMT, PC);
+
+}
+
+void PrintStackTrace() {
+
+  std::unique_lock<std::mutex> l(SymbolizeMutex, std::try_to_lock);
+  if (EF->__sanitizer_print_stack_trace && l.owns_lock())
+    EF->__sanitizer_print_stack_trace();
+
+}
+
+void PrintMemoryProfile() {
+
+  std::unique_lock<std::mutex> l(SymbolizeMutex, std::try_to_lock);
+  if (EF->__sanitizer_print_memory_profile && l.owns_lock())
+    EF->__sanitizer_print_memory_profile(95, 8);
+
+}
+
+unsigned NumberOfCpuCores() {
+
+  unsigned N = std::thread::hardware_concurrency();
+  if (!N) {
+
+    Printf(
+        "WARNING: std::thread::hardware_concurrency not well defined for "
+        "your platform. Assuming CPU count of 1.\n");
+    N = 1;
+
+  }
+
+  return N;
+
+}
+
+size_t SimpleFastHash(const uint8_t *Data, size_t Size) {
+
+  size_t Res = 0;
+  for (size_t i = 0; i < Size; i++)
+    Res = Res * 11 + Data[i];
+  return Res;
+
+}
+
+}  // namespace fuzzer
+

custom_mutators/libfuzzer/FuzzerUtil.h

@@ -0,0 +1,117 @@
+//===- FuzzerUtil.h - Internal header for the Fuzzer Utils ------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Util functions.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_UTIL_H
+#define LLVM_FUZZER_UTIL_H
+
+#include "FuzzerBuiltins.h"
+#include "FuzzerBuiltinsMsvc.h"
+#include "FuzzerCommand.h"
+#include "FuzzerDefs.h"
+
+namespace fuzzer {
+
+void PrintHexArray(const Unit &U, const char *PrintAfter = "");
+
+void PrintHexArray(const uint8_t *Data, size_t Size,
+                   const char *PrintAfter = "");
+
+void PrintASCII(const uint8_t *Data, size_t Size, const char *PrintAfter = "");
+
+void PrintASCII(const Unit &U, const char *PrintAfter = "");
+
+// Changes U to contain only ASCII (isprint+isspace) characters.
+// Returns true iff U has been changed.
+bool ToASCII(uint8_t *Data, size_t Size);
+
+bool IsASCII(const Unit &U);
+
+bool IsASCII(const uint8_t *Data, size_t Size);
+
+std::string Base64(const Unit &U);
+
+void PrintPC(const char *SymbolizedFMT, const char *FallbackFMT, uintptr_t PC);
+
+std::string DescribePC(const char *SymbolizedFMT, uintptr_t PC);
+
+void PrintStackTrace();
+
+void PrintMemoryProfile();
+
+unsigned NumberOfCpuCores();
+
+// Platform specific functions.
+void SetSignalHandler(const FuzzingOptions& Options);
+
+void SleepSeconds(int Seconds);
+
+unsigned long GetPid();
+
+size_t GetPeakRSSMb();
+
+int ExecuteCommand(const Command &Cmd);
+bool ExecuteCommand(const Command &Cmd, std::string *CmdOutput);
+
+// Fuchsia does not have popen/pclose.
+FILE *OpenProcessPipe(const char *Command, const char *Mode);
+int CloseProcessPipe(FILE *F);
+
+const void *SearchMemory(const void *haystack, size_t haystacklen,
+                         const void *needle, size_t needlelen);
+
+std::string CloneArgsWithoutX(const Vector<std::string> &Args,
+                              const char *X1, const char *X2);
+
+inline std::string CloneArgsWithoutX(const Vector<std::string> &Args,
+                                     const char *X) {
+  return CloneArgsWithoutX(Args, X, X);
+}
+
+inline std::pair<std::string, std::string> SplitBefore(std::string X,
+                                                       std::string S) {
+  auto Pos = S.find(X);
+  if (Pos == std::string::npos)
+    return std::make_pair(S, "");
+  return std::make_pair(S.substr(0, Pos), S.substr(Pos));
+}
+
+void DiscardOutput(int Fd);
+
+std::string DisassembleCmd(const std::string &FileName);
+
+std::string SearchRegexCmd(const std::string &Regex);
+
+size_t SimpleFastHash(const uint8_t *Data, size_t Size);
+
+inline uint32_t Log(uint32_t X) { return 32 - Clz(X) - 1; }
+
+inline size_t PageSize() { return 4096; }
+inline uint8_t *RoundUpByPage(uint8_t *P) {
+  uintptr_t X = reinterpret_cast<uintptr_t>(P);
+  size_t Mask = PageSize() - 1;
+  X = (X + Mask) & ~Mask;
+  return reinterpret_cast<uint8_t *>(X);
+}
+inline uint8_t *RoundDownByPage(uint8_t *P) {
+  uintptr_t X = reinterpret_cast<uintptr_t>(P);
+  size_t Mask = PageSize() - 1;
+  X = X & ~Mask;
+  return reinterpret_cast<uint8_t *>(X);
+}
+
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+template <typename T> T HostToLE(T X) { return X; }
+#else
+template <typename T> T HostToLE(T X) { return Bswap(X); }
+#endif
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_UTIL_H

custom_mutators/libfuzzer/FuzzerUtilDarwin.cpp

@@ -0,0 +1,205 @@
+//===- FuzzerUtilDarwin.cpp - Misc utils ----------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Misc utils for Darwin.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_APPLE
+  #include "FuzzerCommand.h"
+  #include "FuzzerIO.h"
+  #include <mutex>
+  #include <signal.h>
+  #include <spawn.h>
+  #include <stdlib.h>
+  #include <string.h>
+  #include <sys/wait.h>
+  #include <unistd.h>
+
+// There is no header for this on macOS so declare here
+extern "C" char **environ;
+
+namespace fuzzer {
+
+static std::mutex SignalMutex;
+// Global variables used to keep track of how signal handling should be
+// restored. They should **not** be accessed without holding `SignalMutex`.
+static int              ActiveThreadCount = 0;
+static struct sigaction OldSigIntAction;
+static struct sigaction OldSigQuitAction;
+static sigset_t         OldBlockedSignalsSet;
+
+// This is a reimplementation of Libc's `system()`. On Darwin the Libc
+// implementation contains a mutex which prevents it from being used
+// concurrently. This implementation **can** be used concurrently. It sets the
+// signal handlers when the first thread enters and restores them when the last
+// thread finishes execution of the function and ensures this is not racey by
+// using a mutex.
+int ExecuteCommand(const Command &Cmd) {
+
+  std::string       CmdLine = Cmd.toString();
+  posix_spawnattr_t SpawnAttributes;
+  if (posix_spawnattr_init(&SpawnAttributes)) return -1;
+  // Block and ignore signals of the current process when the first thread
+  // enters.
+  {
+
+    std::lock_guard<std::mutex> Lock(SignalMutex);
+    if (ActiveThreadCount == 0) {
+
+      static struct sigaction IgnoreSignalAction;
+      sigset_t                BlockedSignalsSet;
+      memset(&IgnoreSignalAction, 0, sizeof(IgnoreSignalAction));
+      IgnoreSignalAction.sa_handler = SIG_IGN;
+
+      if (sigaction(SIGINT, &IgnoreSignalAction, &OldSigIntAction) == -1) {
+
+        Printf("Failed to ignore SIGINT\n");
+        (void)posix_spawnattr_destroy(&SpawnAttributes);
+        return -1;
+
+      }
+
+      if (sigaction(SIGQUIT, &IgnoreSignalAction, &OldSigQuitAction) == -1) {
+
+        Printf("Failed to ignore SIGQUIT\n");
+        // Try our best to restore the signal handlers.
+        (void)sigaction(SIGINT, &OldSigIntAction, NULL);
+        (void)posix_spawnattr_destroy(&SpawnAttributes);
+        return -1;
+
+      }
+
+      (void)sigemptyset(&BlockedSignalsSet);
+      (void)sigaddset(&BlockedSignalsSet, SIGCHLD);
+      if (sigprocmask(SIG_BLOCK, &BlockedSignalsSet, &OldBlockedSignalsSet) ==
+          -1) {
+
+        Printf("Failed to block SIGCHLD\n");
+        // Try our best to restore the signal handlers.
+        (void)sigaction(SIGQUIT, &OldSigQuitAction, NULL);
+        (void)sigaction(SIGINT, &OldSigIntAction, NULL);
+        (void)posix_spawnattr_destroy(&SpawnAttributes);
+        return -1;
+
+      }
+
+    }
+
+    ++ActiveThreadCount;
+
+  }
+
+  // NOTE: Do not introduce any new `return` statements past this
+  // point. It is important that `ActiveThreadCount` always be decremented
+  // when leaving this function.
+
+  // Make sure the child process uses the default handlers for the
+  // following signals rather than inheriting what the parent has.
+  sigset_t DefaultSigSet;
+  (void)sigemptyset(&DefaultSigSet);
+  (void)sigaddset(&DefaultSigSet, SIGQUIT);
+  (void)sigaddset(&DefaultSigSet, SIGINT);
+  (void)posix_spawnattr_setsigdefault(&SpawnAttributes, &DefaultSigSet);
+  // Make sure the child process doesn't block SIGCHLD
+  (void)posix_spawnattr_setsigmask(&SpawnAttributes, &OldBlockedSignalsSet);
+  short SpawnFlags = POSIX_SPAWN_SETSIGDEF | POSIX_SPAWN_SETSIGMASK;
+  (void)posix_spawnattr_setflags(&SpawnAttributes, SpawnFlags);
+
+  pid_t       Pid;
+  char **     Environ = environ;  // Read from global
+  const char *CommandCStr = CmdLine.c_str();
+  char *const Argv[] = {strdup("sh"), strdup("-c"), strdup(CommandCStr), NULL};
+  int         ErrorCode = 0, ProcessStatus = 0;
+  // FIXME: We probably shouldn't hardcode the shell path.
+  ErrorCode =
+      posix_spawn(&Pid, "/bin/sh", NULL, &SpawnAttributes, Argv, Environ);
+  (void)posix_spawnattr_destroy(&SpawnAttributes);
+  if (!ErrorCode) {
+
+    pid_t SavedPid = Pid;
+    do {
+
+      // Repeat until call completes uninterrupted.
+      Pid = waitpid(SavedPid, &ProcessStatus, /*options=*/0);
+
+    } while (Pid == -1 && errno == EINTR);
+
+    if (Pid == -1) {
+
+      // Fail for some other reason.
+      ProcessStatus = -1;
+
+    }
+
+  } else if (ErrorCode == ENOMEM || ErrorCode == EAGAIN) {
+
+    // Fork failure.
+    ProcessStatus = -1;
+
+  } else {
+
+    // Shell execution failure.
+    ProcessStatus = W_EXITCODE(127, 0);
+
+  }
+
+  for (unsigned i = 0, n = sizeof(Argv) / sizeof(Argv[0]); i < n; ++i)
+    free(Argv[i]);
+
+  // Restore the signal handlers of the current process when the last thread
+  // using this function finishes.
+  {
+
+    std::lock_guard<std::mutex> Lock(SignalMutex);
+    --ActiveThreadCount;
+    if (ActiveThreadCount == 0) {
+
+      bool FailedRestore = false;
+      if (sigaction(SIGINT, &OldSigIntAction, NULL) == -1) {
+
+        Printf("Failed to restore SIGINT handling\n");
+        FailedRestore = true;
+
+      }
+
+      if (sigaction(SIGQUIT, &OldSigQuitAction, NULL) == -1) {
+
+        Printf("Failed to restore SIGQUIT handling\n");
+        FailedRestore = true;
+
+      }
+
+      if (sigprocmask(SIG_BLOCK, &OldBlockedSignalsSet, NULL) == -1) {
+
+        Printf("Failed to unblock SIGCHLD\n");
+        FailedRestore = true;
+
+      }
+
+      if (FailedRestore) ProcessStatus = -1;
+
+    }
+
+  }
+
+  return ProcessStatus;
+
+}
+
+void DiscardOutput(int Fd) {
+
+  FILE *Temp = fopen("/dev/null", "w");
+  if (!Temp) return;
+  dup2(fileno(Temp), Fd);
+  fclose(Temp);
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_APPLE
+

custom_mutators/libfuzzer/FuzzerUtilFuchsia.cpp

@@ -0,0 +1,658 @@
+//===- FuzzerUtilFuchsia.cpp - Misc utils for Fuchsia. --------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Misc utils implementation using Fuchsia/Zircon APIs.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+
+#if LIBFUZZER_FUCHSIA
+
+  #include "FuzzerInternal.h"
+  #include "FuzzerUtil.h"
+  #include <cassert>
+  #include <cerrno>
+  #include <cinttypes>
+  #include <cstdint>
+  #include <fcntl.h>
+  #include <lib/fdio/fdio.h>
+  #include <lib/fdio/spawn.h>
+  #include <string>
+  #include <sys/select.h>
+  #include <thread>
+  #include <unistd.h>
+  #include <zircon/errors.h>
+  #include <zircon/process.h>
+  #include <zircon/sanitizer.h>
+  #include <zircon/status.h>
+  #include <zircon/syscalls.h>
+  #include <zircon/syscalls/debug.h>
+  #include <zircon/syscalls/exception.h>
+  #include <zircon/syscalls/object.h>
+  #include <zircon/types.h>
+
+  #include <vector>
+
+namespace fuzzer {
+
+// Given that Fuchsia doesn't have the POSIX signals that libFuzzer was written
+// around, the general approach is to spin up dedicated threads to watch for
+// each requested condition (alarm, interrupt, crash).  Of these, the crash
+// handler is the most involved, as it requires resuming the crashed thread in
+// order to invoke the sanitizers to get the needed state.
+
+// Forward declaration of assembly trampoline needed to resume crashed threads.
+// This appears to have external linkage to  C++, which is why it's not in the
+// anonymous namespace.  The assembly definition inside MakeTrampoline()
+// actually defines the symbol with internal linkage only.
+void CrashTrampolineAsm() __asm__("CrashTrampolineAsm");
+
+namespace {
+
+// Helper function to handle Zircon syscall failures.
+void ExitOnErr(zx_status_t Status, const char *Syscall) {
+
+  if (Status != ZX_OK) {
+
+    Printf("libFuzzer: %s failed: %s\n", Syscall,
+           _zx_status_get_string(Status));
+    exit(1);
+
+  }
+
+}
+
+void AlarmHandler(int Seconds) {
+
+  while (true) {
+
+    SleepSeconds(Seconds);
+    Fuzzer::StaticAlarmCallback();
+
+  }
+
+}
+
+void InterruptHandler() {
+
+  fd_set readfds;
+  // Ctrl-C sends ETX in Zircon.
+  do {
+
+    FD_ZERO(&readfds);
+    FD_SET(STDIN_FILENO, &readfds);
+    select(STDIN_FILENO + 1, &readfds, nullptr, nullptr, nullptr);
+
+  } while (!FD_ISSET(STDIN_FILENO, &readfds) || getchar() != 0x03);
+
+  Fuzzer::StaticInterruptCallback();
+
+}
+
+  // CFAOffset is used to reference the stack pointer before entering the
+  // trampoline (Stack Pointer + CFAOffset = prev Stack Pointer). Before jumping
+  // to the trampoline we copy all the registers onto the stack. We need to make
+  // sure that the new stack has enough space to store all the registers.
+  //
+  // The trampoline holds CFI information regarding the registers stored in the
+  // stack, which is then used by the unwinder to restore them.
+  #if defined(__x86_64__)
+// In x86_64 the crashing function might also be using the red zone (128 bytes
+// on top of their rsp).
+constexpr size_t CFAOffset = 128 + sizeof(zx_thread_state_general_regs_t);
+  #elif defined(__aarch64__)
+// In aarch64 we need to always have the stack pointer aligned to 16 bytes, so
+// we make sure that we are keeping that same alignment.
+constexpr size_t CFAOffset =
+    (sizeof(zx_thread_state_general_regs_t) + 15) & -(uintptr_t)16;
+  #endif
+
+  // For the crash handler, we need to call Fuzzer::StaticCrashSignalCallback
+  // without POSIX signal handlers.  To achieve this, we use an assembly
+  // function to add the necessary CFI unwinding information and a C function to
+  // bridge from that back into C++.
+
+  // FIXME: This works as a short-term solution, but this code really shouldn't
+  // be architecture dependent. A better long term solution is to implement
+  // remote unwinding and expose the necessary APIs through sanitizer_common
+  // and/or ASAN to allow the exception handling thread to gather the crash
+  // state directly.
+  //
+  // Alternatively, Fuchsia may in future actually implement basic signal
+  // handling for the machine trap signals.
+  #if defined(__x86_64__)
+    #define FOREACH_REGISTER(OP_REG, OP_NUM) \
+      OP_REG(rax)                            \
+      OP_REG(rbx)                            \
+      OP_REG(rcx)                            \
+      OP_REG(rdx)                            \
+      OP_REG(rsi)                            \
+      OP_REG(rdi)                            \
+      OP_REG(rbp)                            \
+      OP_REG(rsp)                            \
+      OP_REG(r8)                             \
+      OP_REG(r9)                             \
+      OP_REG(r10)                            \
+      OP_REG(r11)                            \
+      OP_REG(r12)                            \
+      OP_REG(r13)                            \
+      OP_REG(r14)                            \
+      OP_REG(r15)                            \
+      OP_REG(rip)
+
+  #elif defined(__aarch64__)
+    #define FOREACH_REGISTER(OP_REG, OP_NUM) \
+      OP_NUM(0)                              \
+      OP_NUM(1)                              \
+      OP_NUM(2)                              \
+      OP_NUM(3)                              \
+      OP_NUM(4)                              \
+      OP_NUM(5)                              \
+      OP_NUM(6)                              \
+      OP_NUM(7)                              \
+      OP_NUM(8)                              \
+      OP_NUM(9)                              \
+      OP_NUM(10)                             \
+      OP_NUM(11)                             \
+      OP_NUM(12)                             \
+      OP_NUM(13)                             \
+      OP_NUM(14)                             \
+      OP_NUM(15)                             \
+      OP_NUM(16)                             \
+      OP_NUM(17)                             \
+      OP_NUM(18)                             \
+      OP_NUM(19)                             \
+      OP_NUM(20)                             \
+      OP_NUM(21)                             \
+      OP_NUM(22)                             \
+      OP_NUM(23)                             \
+      OP_NUM(24)                             \
+      OP_NUM(25)                             \
+      OP_NUM(26)                             \
+      OP_NUM(27)                             \
+      OP_NUM(28)                             \
+      OP_NUM(29)                             \
+      OP_REG(sp)
+
+  #else
+    #error "Unsupported architecture for fuzzing on Fuchsia"
+  #endif
+
+  // Produces a CFI directive for the named or numbered register.
+  // The value used refers to an assembler immediate operand with the same name
+  // as the register (see ASM_OPERAND_REG).
+  #define CFI_OFFSET_REG(reg) ".cfi_offset " #reg ", %c[" #reg "]\n"
+  #define CFI_OFFSET_NUM(num) CFI_OFFSET_REG(x##num)
+
+  // Produces an assembler immediate operand for the named or numbered register.
+  // This operand contains the offset of the register relative to the CFA.
+  #define ASM_OPERAND_REG(reg) \
+    [reg] "i"(offsetof(zx_thread_state_general_regs_t, reg) - CFAOffset),
+  #define ASM_OPERAND_NUM(num) \
+    [x##num] "i"(offsetof(zx_thread_state_general_regs_t, r[num]) - CFAOffset),
+
+// Trampoline to bridge from the assembly below to the static C++ crash
+// callback.
+__attribute__((noreturn)) static void StaticCrashHandler() {
+
+  Fuzzer::StaticCrashSignalCallback();
+  for (;;) {
+
+    _Exit(1);
+
+  }
+
+}
+
+// Creates the trampoline with the necessary CFI information to unwind through
+// to the crashing call stack:
+//  * Defining the CFA so that it points to the stack pointer at the point
+//    of crash.
+//  * Storing all registers at the point of crash in the stack and refer to them
+//    via CFI information (relative to the CFA).
+//  * Setting the return column so the unwinder knows how to continue unwinding.
+//  * (x86_64) making sure rsp is aligned before calling StaticCrashHandler.
+//  * Calling StaticCrashHandler that will trigger the unwinder.
+//
+// The __attribute__((used)) is necessary because the function
+// is never called; it's just a container around the assembly to allow it to
+// use operands for compile-time computed constants.
+__attribute__((used)) void MakeTrampoline() {
+
+  __asm__(".cfi_endproc\n"
+    ".pushsection .text.CrashTrampolineAsm\n"
+    ".type CrashTrampolineAsm,STT_FUNC\n"
+"CrashTrampolineAsm:\n"
+    ".cfi_startproc simple\n"
+    ".cfi_signal_frame\n"
+  #if defined(__x86_64__)
+    ".cfi_return_column rip\n"
+    ".cfi_def_cfa rsp, %c[CFAOffset]\n"
+    FOREACH_REGISTER(CFI_OFFSET_REG, CFI_OFFSET_NUM)
+    "mov %%rsp, %%rbp\n"
+    ".cfi_def_cfa_register rbp\n"
+    "andq $-16, %%rsp\n"
+    "call %c[StaticCrashHandler]\n"
+    "ud2\n"
+  #elif defined(__aarch64__)
+    ".cfi_return_column 33\n"
+    ".cfi_def_cfa sp, %c[CFAOffset]\n"
+    FOREACH_REGISTER(CFI_OFFSET_REG, CFI_OFFSET_NUM)
+    ".cfi_offset 33, %c[pc]\n"
+    ".cfi_offset 30, %c[lr]\n"
+    "bl %c[StaticCrashHandler]\n"
+    "brk 1\n"
+  #else
+    #error "Unsupported architecture for fuzzing on Fuchsia"
+  #endif
+    ".cfi_endproc\n"
+    ".size CrashTrampolineAsm, . - CrashTrampolineAsm\n"
+    ".popsection\n"
+    ".cfi_startproc\n"
+    : // No outputs
+    : FOREACH_REGISTER(ASM_OPERAND_REG, ASM_OPERAND_NUM)
+  #if defined(__aarch64__)
+      ASM_OPERAND_REG(pc)
+      ASM_OPERAND_REG(lr)
+  #endif
+      [StaticCrashHandler] "i" (StaticCrashHandler),
+      [CFAOffset] "i" (CFAOffset));
+
+}
+
+void CrashHandler(zx_handle_t *Event) {
+
+  // This structure is used to ensure we close handles to objects we create in
+  // this handler.
+  struct ScopedHandle {
+
+    ~ScopedHandle() {
+
+      _zx_handle_close(Handle);
+
+    }
+
+    zx_handle_t Handle = ZX_HANDLE_INVALID;
+
+  };
+
+  // Create the exception channel.  We need to claim to be a "debugger" so the
+  // kernel will allow us to modify and resume dying threads (see below). Once
+  // the channel is set, we can signal the main thread to continue and wait
+  // for the exception to arrive.
+  ScopedHandle Channel;
+  zx_handle_t  Self = _zx_process_self();
+  ExitOnErr(_zx_task_create_exception_channel(
+                Self, ZX_EXCEPTION_CHANNEL_DEBUGGER, &Channel.Handle),
+            "_zx_task_create_exception_channel");
+
+  ExitOnErr(_zx_object_signal(*Event, 0, ZX_USER_SIGNAL_0),
+            "_zx_object_signal");
+
+  // This thread lives as long as the process in order to keep handling
+  // crashes.  In practice, the first crashed thread to reach the end of the
+  // StaticCrashHandler will end the process.
+  while (true) {
+
+    ExitOnErr(_zx_object_wait_one(Channel.Handle, ZX_CHANNEL_READABLE,
+                                  ZX_TIME_INFINITE, nullptr),
+              "_zx_object_wait_one");
+
+    zx_exception_info_t ExceptionInfo;
+    ScopedHandle        Exception;
+    ExitOnErr(
+        _zx_channel_read(Channel.Handle, 0, &ExceptionInfo, &Exception.Handle,
+                         sizeof(ExceptionInfo), 1, nullptr, nullptr),
+        "_zx_channel_read");
+
+    // Ignore informational synthetic exceptions.
+    if (ZX_EXCP_THREAD_STARTING == ExceptionInfo.type ||
+        ZX_EXCP_THREAD_EXITING == ExceptionInfo.type ||
+        ZX_EXCP_PROCESS_STARTING == ExceptionInfo.type) {
+
+      continue;
+
+    }
+
+    // At this point, we want to get the state of the crashing thread, but
+    // libFuzzer and the sanitizers assume this will happen from that same
+    // thread via a POSIX signal handler. "Resurrecting" the thread in the
+    // middle of the appropriate callback is as simple as forcibly setting the
+    // instruction pointer/program counter, provided we NEVER EVER return from
+    // that function (since otherwise our stack will not be valid).
+    ScopedHandle Thread;
+    ExitOnErr(_zx_exception_get_thread(Exception.Handle, &Thread.Handle),
+              "_zx_exception_get_thread");
+
+    zx_thread_state_general_regs_t GeneralRegisters;
+    ExitOnErr(
+        _zx_thread_read_state(Thread.Handle, ZX_THREAD_STATE_GENERAL_REGS,
+                              &GeneralRegisters, sizeof(GeneralRegisters)),
+        "_zx_thread_read_state");
+
+    // To unwind properly, we need to push the crashing thread's register state
+    // onto the stack and jump into a trampoline with CFI instructions on how
+    // to restore it.
+  #if defined(__x86_64__)
+    uintptr_t StackPtr = GeneralRegisters.rsp - CFAOffset;
+    __unsanitized_memcpy(reinterpret_cast<void *>(StackPtr), &GeneralRegisters,
+                         sizeof(GeneralRegisters));
+    GeneralRegisters.rsp = StackPtr;
+    GeneralRegisters.rip = reinterpret_cast<zx_vaddr_t>(CrashTrampolineAsm);
+
+  #elif defined(__aarch64__)
+    uintptr_t StackPtr = GeneralRegisters.sp - CFAOffset;
+    __unsanitized_memcpy(reinterpret_cast<void *>(StackPtr), &GeneralRegisters,
+                         sizeof(GeneralRegisters));
+    GeneralRegisters.sp = StackPtr;
+    GeneralRegisters.pc = reinterpret_cast<zx_vaddr_t>(CrashTrampolineAsm);
+
+  #else
+    #error "Unsupported architecture for fuzzing on Fuchsia"
+  #endif
+
+    // Now force the crashing thread's state.
+    ExitOnErr(
+        _zx_thread_write_state(Thread.Handle, ZX_THREAD_STATE_GENERAL_REGS,
+                               &GeneralRegisters, sizeof(GeneralRegisters)),
+        "_zx_thread_write_state");
+
+    // Set the exception to HANDLED so it resumes the thread on close.
+    uint32_t ExceptionState = ZX_EXCEPTION_STATE_HANDLED;
+    ExitOnErr(_zx_object_set_property(Exception.Handle, ZX_PROP_EXCEPTION_STATE,
+                                      &ExceptionState, sizeof(ExceptionState)),
+              "zx_object_set_property");
+
+  }
+
+}
+
+}  // namespace
+
+// Platform specific functions.
+void SetSignalHandler(const FuzzingOptions &Options) {
+
+  // Make sure information from libFuzzer and the sanitizers are easy to
+  // reassemble. `__sanitizer_log_write` has the added benefit of ensuring the
+  // DSO map is always available for the symbolizer.
+  // A uint64_t fits in 20 chars, so 64 is plenty.
+  char Buf[64];
+  memset(Buf, 0, sizeof(Buf));
+  snprintf(Buf, sizeof(Buf), "==%lu== INFO: libFuzzer starting.\n", GetPid());
+  if (EF->__sanitizer_log_write) __sanitizer_log_write(Buf, sizeof(Buf));
+  Printf("%s", Buf);
+
+  // Set up alarm handler if needed.
+  if (Options.HandleAlrm && Options.UnitTimeoutSec > 0) {
+
+    std::thread T(AlarmHandler, Options.UnitTimeoutSec / 2 + 1);
+    T.detach();
+
+  }
+
+  // Set up interrupt handler if needed.
+  if (Options.HandleInt || Options.HandleTerm) {
+
+    std::thread T(InterruptHandler);
+    T.detach();
+
+  }
+
+  // Early exit if no crash handler needed.
+  if (!Options.HandleSegv && !Options.HandleBus && !Options.HandleIll &&
+      !Options.HandleFpe && !Options.HandleAbrt)
+    return;
+
+  // Set up the crash handler and wait until it is ready before proceeding.
+  zx_handle_t Event;
+  ExitOnErr(_zx_event_create(0, &Event), "_zx_event_create");
+
+  std::thread T(CrashHandler, &Event);
+  zx_status_t Status =
+      _zx_object_wait_one(Event, ZX_USER_SIGNAL_0, ZX_TIME_INFINITE, nullptr);
+  _zx_handle_close(Event);
+  ExitOnErr(Status, "_zx_object_wait_one");
+
+  T.detach();
+
+}
+
+void SleepSeconds(int Seconds) {
+
+  _zx_nanosleep(_zx_deadline_after(ZX_SEC(Seconds)));
+
+}
+
+unsigned long GetPid() {
+
+  zx_status_t            rc;
+  zx_info_handle_basic_t Info;
+  if ((rc = _zx_object_get_info(_zx_process_self(), ZX_INFO_HANDLE_BASIC, &Info,
+                                sizeof(Info), NULL, NULL)) != ZX_OK) {
+
+    Printf("libFuzzer: unable to get info about self: %s\n",
+           _zx_status_get_string(rc));
+    exit(1);
+
+  }
+
+  return Info.koid;
+
+}
+
+size_t GetPeakRSSMb() {
+
+  zx_status_t          rc;
+  zx_info_task_stats_t Info;
+  if ((rc = _zx_object_get_info(_zx_process_self(), ZX_INFO_TASK_STATS, &Info,
+                                sizeof(Info), NULL, NULL)) != ZX_OK) {
+
+    Printf("libFuzzer: unable to get info about self: %s\n",
+           _zx_status_get_string(rc));
+    exit(1);
+
+  }
+
+  return (Info.mem_private_bytes + Info.mem_shared_bytes) >> 20;
+
+}
+
+template <typename Fn>
+class RunOnDestruction {
+
+ public:
+  explicit RunOnDestruction(Fn fn) : fn_(fn) {
+
+  }
+
+  ~RunOnDestruction() {
+
+    fn_();
+
+  }
+
+ private:
+  Fn fn_;
+
+};
+
+template <typename Fn>
+RunOnDestruction<Fn> at_scope_exit(Fn fn) {
+
+  return RunOnDestruction<Fn>(fn);
+
+}
+
+static fdio_spawn_action_t clone_fd_action(int localFd, int targetFd) {
+
+  return {
+
+      .action = FDIO_SPAWN_ACTION_CLONE_FD,
+      .fd =
+          {
+
+              .local_fd = localFd,
+              .target_fd = targetFd,
+
+          },
+
+  };
+
+}
+
+int ExecuteCommand(const Command &Cmd) {
+
+  zx_status_t rc;
+
+  // Convert arguments to C array
+  auto   Args = Cmd.getArguments();
+  size_t Argc = Args.size();
+  assert(Argc != 0);
+  std::unique_ptr<const char *[]> Argv(new const char *[Argc + 1]);
+  for (size_t i = 0; i < Argc; ++i)
+    Argv[i] = Args[i].c_str();
+  Argv[Argc] = nullptr;
+
+  // Determine output.  On Fuchsia, the fuzzer is typically run as a component
+  // that lacks a mutable working directory. Fortunately, when this is the case
+  // a mutable output directory must be specified using "-artifact_prefix=...",
+  // so write the log file(s) there.
+  // However, we don't want to apply this logic for absolute paths.
+  int  FdOut = STDOUT_FILENO;
+  bool discardStdout = false;
+  bool discardStderr = false;
+
+  if (Cmd.hasOutputFile()) {
+
+    std::string Path = Cmd.getOutputFile();
+    if (Path == getDevNull()) {
+
+      // On Fuchsia, there's no "/dev/null" like-file, so we
+      // just don't copy the FDs into the spawned process.
+      discardStdout = true;
+
+    } else {
+
+      bool IsAbsolutePath = Path.length() > 1 && Path[0] == '/';
+      if (!IsAbsolutePath && Cmd.hasFlag("artifact_prefix"))
+        Path = Cmd.getFlagValue("artifact_prefix") + "/" + Path;
+
+      FdOut = open(Path.c_str(), O_WRONLY | O_CREAT | O_TRUNC, 0);
+      if (FdOut == -1) {
+
+        Printf("libFuzzer: failed to open %s: %s\n", Path.c_str(),
+               strerror(errno));
+        return ZX_ERR_IO;
+
+      }
+
+    }
+
+  }
+
+  auto CloseFdOut = at_scope_exit([FdOut]() {
+
+    if (FdOut != STDOUT_FILENO) close(FdOut);
+
+  });
+
+  // Determine stderr
+  int FdErr = STDERR_FILENO;
+  if (Cmd.isOutAndErrCombined()) {
+
+    FdErr = FdOut;
+    if (discardStdout) discardStderr = true;
+
+  }
+
+  // Clone the file descriptors into the new process
+  std::vector<fdio_spawn_action_t> SpawnActions;
+  SpawnActions.push_back(clone_fd_action(STDIN_FILENO, STDIN_FILENO));
+
+  if (!discardStdout)
+    SpawnActions.push_back(clone_fd_action(FdOut, STDOUT_FILENO));
+  if (!discardStderr)
+    SpawnActions.push_back(clone_fd_action(FdErr, STDERR_FILENO));
+
+  // Start the process.
+  char        ErrorMsg[FDIO_SPAWN_ERR_MSG_MAX_LENGTH];
+  zx_handle_t ProcessHandle = ZX_HANDLE_INVALID;
+  rc = fdio_spawn_etc(ZX_HANDLE_INVALID,
+                      FDIO_SPAWN_CLONE_ALL & (~FDIO_SPAWN_CLONE_STDIO), Argv[0],
+                      Argv.get(), nullptr, SpawnActions.size(),
+                      SpawnActions.data(), &ProcessHandle, ErrorMsg);
+
+  if (rc != ZX_OK) {
+
+    Printf("libFuzzer: failed to launch '%s': %s, %s\n", Argv[0], ErrorMsg,
+           _zx_status_get_string(rc));
+    return rc;
+
+  }
+
+  auto CloseHandle = at_scope_exit([&]() { _zx_handle_close(ProcessHandle); });
+
+  // Now join the process and return the exit status.
+  if ((rc = _zx_object_wait_one(ProcessHandle, ZX_PROCESS_TERMINATED,
+                                ZX_TIME_INFINITE, nullptr)) != ZX_OK) {
+
+    Printf("libFuzzer: failed to join '%s': %s\n", Argv[0],
+           _zx_status_get_string(rc));
+    return rc;
+
+  }
+
+  zx_info_process_t Info;
+  if ((rc = _zx_object_get_info(ProcessHandle, ZX_INFO_PROCESS, &Info,
+                                sizeof(Info), nullptr, nullptr)) != ZX_OK) {
+
+    Printf("libFuzzer: unable to get return code from '%s': %s\n", Argv[0],
+           _zx_status_get_string(rc));
+    return rc;
+
+  }
+
+  return Info.return_code;
+
+}
+
+bool ExecuteCommand(const Command &BaseCmd, std::string *CmdOutput) {
+
+  auto    LogFilePath = TempPath("SimPopenOut", ".txt");
+  Command Cmd(BaseCmd);
+  Cmd.setOutputFile(LogFilePath);
+  int Ret = ExecuteCommand(Cmd);
+  *CmdOutput = FileToString(LogFilePath);
+  RemoveFile(LogFilePath);
+  return Ret == 0;
+
+}
+
+const void *SearchMemory(const void *Data, size_t DataLen, const void *Patt,
+                         size_t PattLen) {
+
+  return memmem(Data, DataLen, Patt, PattLen);
+
+}
+
+// In fuchsia, accessing /dev/null is not supported. There's nothing
+// similar to a file that discards everything that is written to it.
+// The way of doing something similar in fuchsia is by using
+// fdio_null_create and binding that to a file descriptor.
+void DiscardOutput(int Fd) {
+
+  fdio_t *fdio_null = fdio_null_create();
+  if (fdio_null == nullptr) return;
+  int nullfd = fdio_bind_to_fd(fdio_null, -1, 0);
+  if (nullfd < 0) return;
+  dup2(nullfd, Fd);
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_FUCHSIA
+

custom_mutators/libfuzzer/FuzzerUtilLinux.cpp

@@ -0,0 +1,43 @@
+//===- FuzzerUtilLinux.cpp - Misc utils for Linux. ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Misc utils for Linux.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_LINUX || LIBFUZZER_NETBSD || LIBFUZZER_FREEBSD || \
+    LIBFUZZER_OPENBSD || LIBFUZZER_EMSCRIPTEN
+  #include "FuzzerCommand.h"
+
+  #include <stdlib.h>
+  #include <sys/types.h>
+  #include <sys/wait.h>
+  #include <unistd.h>
+
+namespace fuzzer {
+
+int ExecuteCommand(const Command &Cmd) {
+
+  std::string CmdLine = Cmd.toString();
+  int         exit_code = system(CmdLine.c_str());
+  if (WIFEXITED(exit_code)) return WEXITSTATUS(exit_code);
+  return exit_code;
+
+}
+
+void DiscardOutput(int Fd) {
+
+  FILE *Temp = fopen("/dev/null", "w");
+  if (!Temp) return;
+  dup2(fileno(Temp), Fd);
+  fclose(Temp);
+
+}
+
+}  // namespace fuzzer
+
+#endif
+

custom_mutators/libfuzzer/FuzzerUtilPosix.cpp

@@ -0,0 +1,239 @@
+//===- FuzzerUtilPosix.cpp - Misc utils for Posix. ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Misc utils implementation using Posix API.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_POSIX
+  #include "FuzzerIO.h"
+  #include "FuzzerInternal.h"
+  #include "FuzzerTracePC.h"
+  #include <cassert>
+  #include <chrono>
+  #include <cstring>
+  #include <errno.h>
+  #include <iomanip>
+  #include <signal.h>
+  #include <stdio.h>
+  #include <sys/mman.h>
+  #include <sys/resource.h>
+  #include <sys/syscall.h>
+  #include <sys/time.h>
+  #include <sys/types.h>
+  #include <thread>
+  #include <unistd.h>
+
+namespace fuzzer {
+
+static void AlarmHandler(int, siginfo_t *, void *) {
+
+  Fuzzer::StaticAlarmCallback();
+
+}
+
+static void (*upstream_segv_handler)(int, siginfo_t *, void *);
+
+static void SegvHandler(int sig, siginfo_t *si, void *ucontext) {
+
+  assert(si->si_signo == SIGSEGV);
+  if (upstream_segv_handler) return upstream_segv_handler(sig, si, ucontext);
+  Fuzzer::StaticCrashSignalCallback();
+
+}
+
+static void CrashHandler(int, siginfo_t *, void *) {
+
+  Fuzzer::StaticCrashSignalCallback();
+
+}
+
+static void InterruptHandler(int, siginfo_t *, void *) {
+
+  Fuzzer::StaticInterruptCallback();
+
+}
+
+static void GracefulExitHandler(int, siginfo_t *, void *) {
+
+  Fuzzer::StaticGracefulExitCallback();
+
+}
+
+static void FileSizeExceedHandler(int, siginfo_t *, void *) {
+
+  Fuzzer::StaticFileSizeExceedCallback();
+
+}
+
+static void SetSigaction(int signum,
+                         void (*callback)(int, siginfo_t *, void *)) {
+
+  struct sigaction sigact = {};
+  if (sigaction(signum, nullptr, &sigact)) {
+
+    Printf("libFuzzer: sigaction failed with %d\n", errno);
+    exit(1);
+
+  }
+
+  if (sigact.sa_flags & SA_SIGINFO) {
+
+    if (sigact.sa_sigaction) {
+
+      if (signum != SIGSEGV) return;
+      upstream_segv_handler = sigact.sa_sigaction;
+
+    }
+
+  } else {
+
+    if (sigact.sa_handler != SIG_DFL && sigact.sa_handler != SIG_IGN &&
+        sigact.sa_handler != SIG_ERR)
+      return;
+
+  }
+
+  sigact = {};
+  sigact.sa_flags = SA_SIGINFO;
+  sigact.sa_sigaction = callback;
+  if (sigaction(signum, &sigact, 0)) {
+
+    Printf("libFuzzer: sigaction failed with %d\n", errno);
+    exit(1);
+
+  }
+
+}
+
+// Return true on success, false otherwise.
+bool ExecuteCommand(const Command &Cmd, std::string *CmdOutput) {
+
+  FILE *Pipe = popen(Cmd.toString().c_str(), "r");
+  if (!Pipe) return false;
+
+  if (CmdOutput) {
+
+    char TmpBuffer[128];
+    while (fgets(TmpBuffer, sizeof(TmpBuffer), Pipe))
+      CmdOutput->append(TmpBuffer);
+
+  }
+
+  return pclose(Pipe) == 0;
+
+}
+
+void SetTimer(int Seconds) {
+
+  struct itimerval T {
+
+    {Seconds, 0}, {
+
+      Seconds, 0
+
+    }
+
+  };
+
+  if (setitimer(ITIMER_REAL, &T, nullptr)) {
+
+    Printf("libFuzzer: setitimer failed with %d\n", errno);
+    exit(1);
+
+  }
+
+  SetSigaction(SIGALRM, AlarmHandler);
+
+}
+
+void SetSignalHandler(const FuzzingOptions &Options) {
+
+  // setitimer is not implemented in emscripten.
+  if (Options.HandleAlrm && Options.UnitTimeoutSec > 0 && !LIBFUZZER_EMSCRIPTEN)
+    SetTimer(Options.UnitTimeoutSec / 2 + 1);
+  if (Options.HandleInt) SetSigaction(SIGINT, InterruptHandler);
+  if (Options.HandleTerm) SetSigaction(SIGTERM, InterruptHandler);
+  if (Options.HandleSegv) SetSigaction(SIGSEGV, SegvHandler);
+  if (Options.HandleBus) SetSigaction(SIGBUS, CrashHandler);
+  if (Options.HandleAbrt) SetSigaction(SIGABRT, CrashHandler);
+  if (Options.HandleIll) SetSigaction(SIGILL, CrashHandler);
+  if (Options.HandleFpe) SetSigaction(SIGFPE, CrashHandler);
+  if (Options.HandleXfsz) SetSigaction(SIGXFSZ, FileSizeExceedHandler);
+  if (Options.HandleUsr1) SetSigaction(SIGUSR1, GracefulExitHandler);
+  if (Options.HandleUsr2) SetSigaction(SIGUSR2, GracefulExitHandler);
+
+}
+
+void SleepSeconds(int Seconds) {
+
+  sleep(Seconds);  // Use C API to avoid coverage from instrumented libc++.
+
+}
+
+unsigned long GetPid() {
+
+  return (unsigned long)getpid();
+
+}
+
+size_t GetPeakRSSMb() {
+
+  struct rusage usage;
+  if (getrusage(RUSAGE_SELF, &usage)) return 0;
+  if (LIBFUZZER_LINUX || LIBFUZZER_FREEBSD || LIBFUZZER_NETBSD ||
+      LIBFUZZER_OPENBSD || LIBFUZZER_EMSCRIPTEN) {
+
+    // ru_maxrss is in KiB
+    return usage.ru_maxrss >> 10;
+
+  } else if (LIBFUZZER_APPLE) {
+
+    // ru_maxrss is in bytes
+    return usage.ru_maxrss >> 20;
+
+  }
+
+  assert(0 && "GetPeakRSSMb() is not implemented for your platform");
+  return 0;
+
+}
+
+FILE *OpenProcessPipe(const char *Command, const char *Mode) {
+
+  return popen(Command, Mode);
+
+}
+
+int CloseProcessPipe(FILE *F) {
+
+  return pclose(F);
+
+}
+
+const void *SearchMemory(const void *Data, size_t DataLen, const void *Patt,
+                         size_t PattLen) {
+
+  return memmem(Data, DataLen, Patt, PattLen);
+
+}
+
+std::string DisassembleCmd(const std::string &FileName) {
+
+  return "objdump -d " + FileName;
+
+}
+
+std::string SearchRegexCmd(const std::string &Regex) {
+
+  return "grep '" + Regex + "'";
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_POSIX
+

custom_mutators/libfuzzer/FuzzerUtilWindows.cpp

@@ -0,0 +1,279 @@
+//===- FuzzerUtilWindows.cpp - Misc utils for Windows. --------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// Misc utils implementation for Windows.
+//===----------------------------------------------------------------------===//
+#include "FuzzerPlatform.h"
+#if LIBFUZZER_WINDOWS
+  #include "FuzzerCommand.h"
+  #include "FuzzerIO.h"
+  #include "FuzzerInternal.h"
+  #include <cassert>
+  #include <chrono>
+  #include <cstring>
+  #include <errno.h>
+  #include <io.h>
+  #include <iomanip>
+  #include <signal.h>
+  #include <stdio.h>
+  #include <sys/types.h>
+  #include <windows.h>
+
+  // This must be included after windows.h.
+  #include <psapi.h>
+
+namespace fuzzer {
+
+static const FuzzingOptions *HandlerOpt = nullptr;
+
+static LONG CALLBACK ExceptionHandler(PEXCEPTION_POINTERS ExceptionInfo) {
+
+  switch (ExceptionInfo->ExceptionRecord->ExceptionCode) {
+
+    case EXCEPTION_ACCESS_VIOLATION:
+    case EXCEPTION_ARRAY_BOUNDS_EXCEEDED:
+    case EXCEPTION_STACK_OVERFLOW:
+      if (HandlerOpt->HandleSegv) Fuzzer::StaticCrashSignalCallback();
+      break;
+    case EXCEPTION_DATATYPE_MISALIGNMENT:
+    case EXCEPTION_IN_PAGE_ERROR:
+      if (HandlerOpt->HandleBus) Fuzzer::StaticCrashSignalCallback();
+      break;
+    case EXCEPTION_ILLEGAL_INSTRUCTION:
+    case EXCEPTION_PRIV_INSTRUCTION:
+      if (HandlerOpt->HandleIll) Fuzzer::StaticCrashSignalCallback();
+      break;
+    case EXCEPTION_FLT_DENORMAL_OPERAND:
+    case EXCEPTION_FLT_DIVIDE_BY_ZERO:
+    case EXCEPTION_FLT_INEXACT_RESULT:
+    case EXCEPTION_FLT_INVALID_OPERATION:
+    case EXCEPTION_FLT_OVERFLOW:
+    case EXCEPTION_FLT_STACK_CHECK:
+    case EXCEPTION_FLT_UNDERFLOW:
+    case EXCEPTION_INT_DIVIDE_BY_ZERO:
+    case EXCEPTION_INT_OVERFLOW:
+      if (HandlerOpt->HandleFpe) Fuzzer::StaticCrashSignalCallback();
+      break;
+      // TODO: handle (Options.HandleXfsz)
+
+  }
+
+  return EXCEPTION_CONTINUE_SEARCH;
+
+}
+
+BOOL WINAPI CtrlHandler(DWORD dwCtrlType) {
+
+  switch (dwCtrlType) {
+
+    case CTRL_C_EVENT:
+      if (HandlerOpt->HandleInt) Fuzzer::StaticInterruptCallback();
+      return TRUE;
+    case CTRL_BREAK_EVENT:
+      if (HandlerOpt->HandleTerm) Fuzzer::StaticInterruptCallback();
+      return TRUE;
+
+  }
+
+  return FALSE;
+
+}
+
+void CALLBACK AlarmHandler(PVOID, BOOLEAN) {
+
+  Fuzzer::StaticAlarmCallback();
+
+}
+
+class TimerQ {
+
+  HANDLE TimerQueue;
+
+ public:
+  TimerQ() : TimerQueue(NULL) {
+
+  }
+
+  ~TimerQ() {
+
+    if (TimerQueue) DeleteTimerQueueEx(TimerQueue, NULL);
+
+  }
+
+  void SetTimer(int Seconds) {
+
+    if (!TimerQueue) {
+
+      TimerQueue = CreateTimerQueue();
+      if (!TimerQueue) {
+
+        Printf("libFuzzer: CreateTimerQueue failed.\n");
+        exit(1);
+
+      }
+
+    }
+
+    HANDLE Timer;
+    if (!CreateTimerQueueTimer(&Timer, TimerQueue, AlarmHandler, NULL,
+                               Seconds * 1000, Seconds * 1000, 0)) {
+
+      Printf("libFuzzer: CreateTimerQueueTimer failed.\n");
+      exit(1);
+
+    }
+
+  }
+
+};
+
+static TimerQ Timer;
+
+static void CrashHandler(int) {
+
+  Fuzzer::StaticCrashSignalCallback();
+
+}
+
+void SetSignalHandler(const FuzzingOptions &Options) {
+
+  HandlerOpt = &Options;
+
+  if (Options.HandleAlrm && Options.UnitTimeoutSec > 0)
+    Timer.SetTimer(Options.UnitTimeoutSec / 2 + 1);
+
+  if (Options.HandleInt || Options.HandleTerm)
+    if (!SetConsoleCtrlHandler(CtrlHandler, TRUE)) {
+
+      DWORD LastError = GetLastError();
+      Printf("libFuzzer: SetConsoleCtrlHandler failed (Error code: %lu).\n",
+             LastError);
+      exit(1);
+
+    }
+
+  if (Options.HandleSegv || Options.HandleBus || Options.HandleIll ||
+      Options.HandleFpe)
+    SetUnhandledExceptionFilter(ExceptionHandler);
+
+  if (Options.HandleAbrt)
+    if (SIG_ERR == signal(SIGABRT, CrashHandler)) {
+
+      Printf("libFuzzer: signal failed with %d\n", errno);
+      exit(1);
+
+    }
+
+}
+
+void SleepSeconds(int Seconds) {
+
+  Sleep(Seconds * 1000);
+
+}
+
+unsigned long GetPid() {
+
+  return GetCurrentProcessId();
+
+}
+
+size_t GetPeakRSSMb() {
+
+  PROCESS_MEMORY_COUNTERS info;
+  if (!GetProcessMemoryInfo(GetCurrentProcess(), &info, sizeof(info))) return 0;
+  return info.PeakWorkingSetSize >> 20;
+
+}
+
+FILE *OpenProcessPipe(const char *Command, const char *Mode) {
+
+  return _popen(Command, Mode);
+
+}
+
+int CloseProcessPipe(FILE *F) {
+
+  return _pclose(F);
+
+}
+
+int ExecuteCommand(const Command &Cmd) {
+
+  std::string CmdLine = Cmd.toString();
+  return system(CmdLine.c_str());
+
+}
+
+bool ExecuteCommand(const Command &Cmd, std::string *CmdOutput) {
+
+  FILE *Pipe = _popen(Cmd.toString().c_str(), "r");
+  if (!Pipe) return false;
+
+  if (CmdOutput) {
+
+    char TmpBuffer[128];
+    while (fgets(TmpBuffer, sizeof(TmpBuffer), Pipe))
+      CmdOutput->append(TmpBuffer);
+
+  }
+
+  return _pclose(Pipe) == 0;
+
+}
+
+const void *SearchMemory(const void *Data, size_t DataLen, const void *Patt,
+                         size_t PattLen) {
+
+  // TODO: make this implementation more efficient.
+  const char *Cdata = (const char *)Data;
+  const char *Cpatt = (const char *)Patt;
+
+  if (!Data || !Patt || DataLen == 0 || PattLen == 0 || DataLen < PattLen)
+    return NULL;
+
+  if (PattLen == 1) return memchr(Data, *Cpatt, DataLen);
+
+  const char *End = Cdata + DataLen - PattLen + 1;
+
+  for (const char *It = Cdata; It < End; ++It)
+    if (It[0] == Cpatt[0] && memcmp(It, Cpatt, PattLen) == 0) return It;
+
+  return NULL;
+
+}
+
+std::string DisassembleCmd(const std::string &FileName) {
+
+  Vector<std::string> command_vector;
+  command_vector.push_back("dumpbin /summary > nul");
+  if (ExecuteCommand(Command(command_vector)) == 0)
+    return "dumpbin /disasm " + FileName;
+  Printf("libFuzzer: couldn't find tool to disassemble (dumpbin)\n");
+  exit(1);
+
+}
+
+std::string SearchRegexCmd(const std::string &Regex) {
+
+  return "findstr /r \"" + Regex + "\"";
+
+}
+
+void DiscardOutput(int Fd) {
+
+  FILE *Temp = fopen("nul", "w");
+  if (!Temp) return;
+  _dup2(_fileno(Temp), Fd);
+  fclose(Temp);
+
+}
+
+}  // namespace fuzzer
+
+#endif  // LIBFUZZER_WINDOWS
+

custom_mutators/libfuzzer/FuzzerValueBitMap.h

@@ -0,0 +1,73 @@
+//===- FuzzerValueBitMap.h - INTERNAL - Bit map -----------------*- C++ -* ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// ValueBitMap.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_VALUE_BIT_MAP_H
+#define LLVM_FUZZER_VALUE_BIT_MAP_H
+
+#include "FuzzerPlatform.h"
+#include <cstdint>
+
+namespace fuzzer {
+
+// A bit map containing kMapSizeInWords bits.
+struct ValueBitMap {
+  static const size_t kMapSizeInBits = 1 << 16;
+  static const size_t kMapPrimeMod = 65371;  // Largest Prime < kMapSizeInBits;
+  static const size_t kBitsInWord = (sizeof(uintptr_t) * 8);
+  static const size_t kMapSizeInWords = kMapSizeInBits / kBitsInWord;
+ public:
+
+  // Clears all bits.
+  void Reset() { memset(Map, 0, sizeof(Map)); }
+
+  // Computes a hash function of Value and sets the corresponding bit.
+  // Returns true if the bit was changed from 0 to 1.
+  ATTRIBUTE_NO_SANITIZE_ALL
+  inline bool AddValue(uintptr_t Value) {
+    uintptr_t Idx = Value % kMapSizeInBits;
+    uintptr_t WordIdx = Idx / kBitsInWord;
+    uintptr_t BitIdx = Idx % kBitsInWord;
+    uintptr_t Old = Map[WordIdx];
+    uintptr_t New = Old | (1ULL << BitIdx);
+    Map[WordIdx] = New;
+    return New != Old;
+  }
+
+  ATTRIBUTE_NO_SANITIZE_ALL
+  inline bool AddValueModPrime(uintptr_t Value) {
+    return AddValue(Value % kMapPrimeMod);
+  }
+
+  inline bool Get(uintptr_t Idx) {
+    assert(Idx < kMapSizeInBits);
+    uintptr_t WordIdx = Idx / kBitsInWord;
+    uintptr_t BitIdx = Idx % kBitsInWord;
+    return Map[WordIdx] & (1ULL << BitIdx);
+  }
+
+  size_t SizeInBits() const { return kMapSizeInBits; }
+
+  template <class Callback>
+  ATTRIBUTE_NO_SANITIZE_ALL
+  void ForEach(Callback CB) const {
+    for (size_t i = 0; i < kMapSizeInWords; i++)
+      if (uintptr_t M = Map[i])
+        for (size_t j = 0; j < sizeof(M) * 8; j++)
+          if (M & ((uintptr_t)1 << j))
+            CB(i * sizeof(M) * 8 + j);
+  }
+
+ private:
+  ATTRIBUTE_ALIGNED(512) uintptr_t Map[kMapSizeInWords];
+};
+
+}  // namespace fuzzer
+
+#endif  // LLVM_FUZZER_VALUE_BIT_MAP_H

custom_mutators/libfuzzer/Makefile

@@ -0,0 +1,81 @@
+
+#CFLAGS = -O3 -funroll-loops -fPIC -fpermissive -std=c++11
+CFLAGS = -g -O0 -fPIC -fpermissive -std=c++11
+CC := clang++
+
+all: libfuzzer-mutator.so
+
+FuzzerCrossOver.o:	FuzzerCrossOver.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerDataFlowTrace.o:	FuzzerDataFlowTrace.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerDriver.o:	FuzzerDriver.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerExtFunctionsDlsym.o:	FuzzerExtFunctionsDlsym.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerExtFunctionsWeak.o:	FuzzerExtFunctionsWeak.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerExtFunctionsWindows.o:	FuzzerExtFunctionsWindows.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerExtraCounters.o:	FuzzerExtraCounters.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerFork.o:	FuzzerFork.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerIO.o:	FuzzerIO.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerIOPosix.o:	FuzzerIOPosix.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerIOWindows.o:	FuzzerIOWindows.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerLoop.o:	FuzzerLoop.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerMerge.o:	FuzzerMerge.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerMutate.o:	FuzzerMutate.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerSHA1.o:	FuzzerSHA1.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerTracePC.o:	FuzzerTracePC.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerUtil.o:	FuzzerUtil.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerUtilDarwin.o:	FuzzerUtilDarwin.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerUtilFuchsia.o:	FuzzerUtilFuchsia.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerUtilLinux.o:	FuzzerUtilLinux.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerUtilPosix.o:	FuzzerUtilPosix.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+FuzzerUtilWindows.o:	FuzzerUtilWindows.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+libfuzzer.o:	libfuzzer.cpp
+	$(CC) $(CFLAGS) -I../../include -I. -c $^
+
+libfuzzer-mutator.so:	FuzzerCrossOver.o FuzzerDataFlowTrace.o FuzzerDriver.o FuzzerExtFunctionsDlsym.o FuzzerExtFunctionsWeak.o FuzzerExtFunctionsWindows.o FuzzerExtraCounters.o FuzzerFork.o FuzzerIO.o FuzzerIOPosix.o FuzzerIOWindows.o FuzzerLoop.o FuzzerMerge.o FuzzerMutate.o FuzzerSHA1.o FuzzerTracePC.o FuzzerUtil.o FuzzerUtilDarwin.o FuzzerUtilFuchsia.o FuzzerUtilLinux.o FuzzerUtilPosix.o FuzzerUtilWindows.o libfuzzer.o
+	$(CC) $(CFLAGS) -I../../include -I. -shared -o libfuzzer-mutator.so *.o
+
+clean:
+	rm -f *.o *~ *.so core

custom_mutators/libfuzzer/README.md

@@ -0,0 +1,24 @@
+# custum mutator: libfuzzer LLVMFuzzerMutate()
+
+This uses the libfuzzer LLVMFuzzerMutate() function in llvm 12.
+
+just type `make` to build
+
+```AFL_CUSTOM_MUTATOR_LIBRARY=custom_mutators/libfuzzer/libfuzzer-mutator.so afl-fuzz ...```
+
+Note that is is currently simple and is missing two features:
+  * Splicing ("Crossover")
+  * Dictionary support
+
+To update the source, all that is needed is that FuzzerDriver.cpp has to receive
+```
+#include "libfuzzer.inc"
+```
+before the closing namespace bracket.
+
+It is also libfuzzer.inc where the configuration of the libfuzzer mutations
+are done.
+
+> Original repository: https://github.com/llvm/llvm-project
+> Path: compiler-rt/lib/fuzzer/*.{h|cpp}
+> Source commit: d4b88ac1658d681e143482336cac27c6a74b8b24

custom_mutators/libfuzzer/libfuzzer.cpp

@@ -0,0 +1,147 @@
+#include <stdio.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+//#include "config.h"
+//#include "debug.h"
+#include "afl-fuzz.h"
+
+afl_state_t *afl_struct;
+
+extern "C" size_t LLVMFuzzerMutate(uint8_t *Data, size_t Size, size_t MaxSize);
+extern "C" int    LLVMFuzzerRunDriver(int *argc, char ***argv,
+                                      int (*UserCb)(const uint8_t *Data,
+                                                 size_t         Size));
+extern "C" void   LLVMFuzzerMyInit(int (*UserCb)(const uint8_t *Data,
+                                               size_t         Size),
+                                   unsigned int Seed);
+
+typedef struct my_mutator {
+
+  afl_state_t *afl;
+  u8 *         mutator_buf;
+  unsigned int seed;
+  unsigned int extras_cnt, a_extras_cnt;
+
+} my_mutator_t;
+
+extern "C" int dummy(const uint8_t *Data, size_t Size) {
+
+  (void)(Data);
+  (void)(Size);
+  fprintf(stderr, "dummy() called\n");
+  return 0;
+
+}
+
+extern "C" my_mutator_t *afl_custom_init(afl_state_t *afl, unsigned int seed) {
+
+  my_mutator_t *data = (my_mutator_t *)calloc(1, sizeof(my_mutator_t));
+  if (!data) {
+
+    perror("afl_custom_init alloc");
+    return NULL;
+
+  }
+
+  if ((data->mutator_buf = (u8 *)malloc(MAX_FILE)) == NULL) {
+
+    perror("mutator_buf alloc");
+    return NULL;
+
+  }
+
+  data->afl = afl;
+  data->seed = seed;
+  afl_struct = afl;
+
+  /*
+    char **argv;
+    argv = (char**)malloc(sizeof(size_t) * 2);
+    argv[0] = (char*)"foo";
+    argv[1] = NULL;
+    int eins = 1;
+    LLVMFuzzerRunDriver(&eins, &argv, dummy);
+  */
+
+  LLVMFuzzerMyInit(dummy, seed);
+
+  return data;
+
+}
+
+/* When a new queue entry is added we check if there are new dictionary
+   entries to add to honggfuzz structure */
+#if ß
+extern "C" void afl_custom_queue_new_entry(my_mutator_t * data,
+                                           const uint8_t *filename_new_queue,
+                                           const uint8_t *filename_orig_queue) {
+
+  while (data->extras_cnt < afl_struct->extras_cnt) {
+
+    /*
+        memcpy(run.global->mutate.dictionary[run.global->mutate.dictionaryCnt].val,
+               afl_struct->extras[data->extras_cnt].data,
+               afl_struct->extras[data->extras_cnt].len);
+        run.global->mutate.dictionary[run.global->mutate.dictionaryCnt].len =
+            afl_struct->extras[data->extras_cnt].len;
+        run.global->mutate.dictionaryCnt++;
+    */
+    data->extras_cnt++;
+
+  }
+
+  while (data->a_extras_cnt < afl_struct->a_extras_cnt) {
+
+    /*
+        memcpy(run.global->mutate.dictionary[run.global->mutate.dictionaryCnt].val,
+               afl_struct->a_extras[data->a_extras_cnt].data,
+               afl_struct->a_extras[data->a_extras_cnt].len);
+        run.global->mutate.dictionary[run.global->mutate.dictionaryCnt].len =
+            afl_struct->a_extras[data->a_extras_cnt].len;
+        run.global->mutate.dictionaryCnt++;
+        data->a_extras_cnt++;
+    */
+
+  }
+
+}
+
+#endif
+/* we could set only_printable if is_ascii is set ... let's see
+uint8_t afl_custom_queue_get(void *data, const uint8_t *filename) {
+
+  //run.global->cfg.only_printable = ...
+
+}
+
+*/
+
+/* here we run the honggfuzz mutator, which is really good */
+
+extern "C" size_t afl_custom_fuzz(my_mutator_t *data, uint8_t *buf,
+                                  size_t buf_size, u8 **out_buf,
+                                  uint8_t *add_buf, size_t add_buf_size,
+                                  size_t max_size) {
+
+  memcpy(data->mutator_buf, buf, buf_size);
+  size_t ret = LLVMFuzzerMutate(data->mutator_buf, buf_size, max_size);
+
+  /* return size of mutated data */
+  *out_buf = data->mutator_buf;
+  return ret;
+
+}
+
+/**
+ * Deinitialize everything
+ *
+ * @param data The data ptr from afl_custom_init
+ */
+extern "C" void afl_custom_deinit(my_mutator_t *data) {
+
+  free(data->mutator_buf);
+  free(data);
+
+}
+

custom_mutators/libfuzzer/libfuzzer.inc

@@ -0,0 +1,36 @@
+
+
+extern "C" ATTRIBUTE_INTERFACE void
+LLVMFuzzerMyInit(int (*Callback)(const uint8_t *Data, size_t Size), unsigned int Seed) {
+  Random Rand(Seed);
+  FuzzingOptions Options;
+  Options.Verbosity = 3;
+  Options.MaxLen = 1024000;
+  Options.LenControl = true;
+  Options.DoCrossOver = false;
+  Options.MutateDepth = 6;
+  Options.UseCounters = false;
+  Options.UseMemmem = false;
+  Options.UseCmp = false;
+  Options.UseValueProfile = false;
+  Options.Shrink = false;
+  Options.ReduceInputs = false;
+  Options.PreferSmall = false;
+  Options.ReloadIntervalSec = 0;
+  Options.OnlyASCII = false;
+  Options.DetectLeaks = false;
+  Options.PurgeAllocatorIntervalSec = 0;
+  Options.TraceMalloc = false;
+  Options.RssLimitMb = 100;
+  Options.MallocLimitMb = 100;
+  Options.MaxNumberOfRuns = 0;
+  Options.ReportSlowUnits = false;
+  Options.Entropic = false;
+  
+  struct EntropicOptions Entropic;
+  Entropic.Enabled = Options.Entropic;
+  EF = new ExternalFunctions();
+  auto *MD = new MutationDispatcher(Rand, Options);
+  auto *Corpus = new InputCorpus(Options.OutputCorpus, Entropic);
+  auto *F = new Fuzzer(Callback, *Corpus, *MD, Options);
+}

custom_mutators/symcc/symcc.c

@@ -12,7 +12,8 @@ afl_state_t *afl_struct;
 #ifdef DEBUG
   #define DBG(x...) fprintf(stderr, x)
 #else
-  #define DBG(x...) {}
+  #define DBG(x...) \
+    {}
 #endif
 
 typedef struct my_mutator {
@@ -177,8 +178,8 @@ size_t afl_custom_fuzz(my_mutator_t *data, uint8_t *buf, size_t buf_size,
                        size_t max_size) {
 
   struct dirent **nl;
-  int32_t i, done = 0, items = scandir(data->out_dir, &nl, NULL, NULL);
-  size_t  size = 0;
+  int32_t         i, done = 0, items = scandir(data->out_dir, &nl, NULL, NULL);
+  size_t          size = 0;
 
   if (items <= 0) return 0;
 

docs/Changelog.md

@@ -33,6 +33,8 @@ sending a mail to <afl-users+subscribe@googlegroups.com>.
     - LTO autodict now also collects interesting cmp comparisons,
       std::string compare + find + ==, bcmp
   - added a new custom mutator: symcc -> https://github.com/eurecom-s3/symcc/
+  - added a new custom mutator: libfuzzer that integrates libfuzzer mutations
+  - Our afl++ Grammar-Mutator is now better integrated into custom_mutators/
 
 
 ### Version ++2.68c (release)

include/afl-prealloc.h

@@ -60,7 +60,7 @@ typedef enum prealloc_status {
                                                                                \
     if ((prealloc_counter) >= (prealloc_size)) {                               \
                                                                                \
-      el_ptr = (void *)malloc(sizeof(*el_ptr));                                \
+      el_ptr = (element_t *)malloc(sizeof(*el_ptr));                           \
       if (!el_ptr) { FATAL("error in list.h -> out of memory for element!"); } \
       el_ptr->pre_status = PRE_STATUS_MALLOC;                                  \
                                                                                \

include/alloc-inl.h

@@ -668,7 +668,7 @@ static inline void *afl_realloc(void **buf, size_t size_needed) {
   if (likely(*buf)) {
 
     /* the size is always stored at buf - 1*size_t */
-    new_buf = afl_alloc_bufptr(*buf);
+    new_buf = (struct afl_alloc_buf *)afl_alloc_bufptr(*buf);
     current_size = new_buf->complete_size;
 
   }
@@ -694,7 +694,7 @@ static inline void *afl_realloc(void **buf, size_t size_needed) {
   }
 
   /* alloc */
-  new_buf = realloc(new_buf, next_size);
+  new_buf = (struct afl_alloc_buf *)realloc(new_buf, next_size);
   if (unlikely(!new_buf)) {
 
     *buf = NULL;

include/list.h

@@ -81,6 +81,7 @@ static inline void list_append(list_t *list, void *el) {
   }
 
   element_t *el_box = NULL;
+
   PRE_ALLOC(el_box, list->element_prealloc_buf, LIST_PREALLOC_SIZE,
             list->element_prealloc_count);
   if (!el_box) { FATAL("failed to allocate list element"); }

src/afl-cc.c

@@ -1508,9 +1508,9 @@ int main(int argc, char **argv, char **envp) {
 
   if (debug) {
 
-    SAYF(cMGN "[D]" cRST " cd \"%s\";", getthecwd());
+    SAYF(cMGN "[D]" cRST " cd '%s';", getthecwd());
     for (i = 0; i < argc; i++)
-      SAYF(" \"%s\"", argv[i]);
+      SAYF(" '%s'", argv[i]);
     SAYF("\n");
 
   }
@@ -1536,9 +1536,9 @@ int main(int argc, char **argv, char **envp) {
 
   if (debug) {
 
-    SAYF(cMGN "[D]" cRST " cd \"%s\";", getthecwd());
+    SAYF(cMGN "[D]" cRST " cd '%s';", getthecwd());
     for (i = 0; i < cc_par_cnt; i++)
-      SAYF(" \"%s\"", cc_params[i]);
+      SAYF(" '%s'", cc_params[i]);
     SAYF("\n");
 
   }