[libFuzzer] refactoring in preparation for -reduce_inputs; NFC intended

llvm-svn: 307857

5 files changed, 40 insertions, 32 deletions

diff --git a/llvm/lib/Fuzzer/FuzzerDriver.cpp b/llvm/lib/Fuzzer/FuzzerDriver.cpp
index 0453a7f443b..f88b5b7c2bb 100644
--- a/llvm/lib/Fuzzer/FuzzerDriver.cpp
+++ b/llvm/lib/Fuzzer/FuzzerDriver.cpp
@@ -265,7 +265,7 @@ int RunOneTest(Fuzzer *F, const char *InputFilePath, size_t MaxLen) {
   Unit U = FileToVector(InputFilePath);
   if (MaxLen && MaxLen < U.size())
     U.resize(MaxLen);
-  F->RunOne(U.data(), U.size());
+  F->ExecuteCallback(U.data(), U.size());
   F->TryDetectingAMemoryLeak(U.data(), U.size(), true);
   return 0;
 }
@@ -572,6 +572,7 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
   Options.UseCmp = Flags.use_cmp;
   Options.UseValueProfile = Flags.use_value_profile;
   Options.Shrink = Flags.shrink;
+  Options.ReduceInputs = Flags.reduce_inputs;
   Options.ShuffleAtStartUp = Flags.shuffle;
   Options.PreferSmall = Flags.prefer_small;
   Options.ReloadIntervalSec = Flags.reload;
@@ -657,7 +658,7 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
       size_t Size = SMR.ReadByteArraySize();
       SMR.WriteByteArray(nullptr, 0);
       const Unit tmp(SMR.GetByteArray(), SMR.GetByteArray() + Size);
-      F->RunOne(tmp.data(), tmp.size());
+      F->ExecuteCallback(tmp.data(), tmp.size());
       SMR.PostServer();
     }
     return 0;
diff --git a/llvm/lib/Fuzzer/FuzzerFlags.def b/llvm/lib/Fuzzer/FuzzerFlags.def
index 7ff196c8fa9..5e70cbad3cf 100644
--- a/llvm/lib/Fuzzer/FuzzerFlags.def
+++ b/llvm/lib/Fuzzer/FuzzerFlags.def
@@ -65,7 +65,9 @@ FUZZER_FLAG_INT(use_memmem, 1,
 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 elements.")
+FUZZER_FLAG_INT(shrink, 0, "Experimental. Try to shrink corpus inputs.")
+FUZZER_FLAG_INT(reduce_inputs, 0, "Experimental. "
+  "Try to reduce the size of inputs wile 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.")
diff --git a/llvm/lib/Fuzzer/FuzzerInternal.h b/llvm/lib/Fuzzer/FuzzerInternal.h
index 5f184c2316e..573923fddc8 100644
--- a/llvm/lib/Fuzzer/FuzzerInternal.h
+++ b/llvm/lib/Fuzzer/FuzzerInternal.h
@@ -65,7 +65,7 @@ public:
   static void StaticFileSizeExceedCallback();
 
   void ExecuteCallback(const uint8_t *Data, size_t Size);
-  size_t RunOne(const uint8_t *Data, size_t Size);
+  bool RunOne(const uint8_t *Data, size_t Size, bool MayDeleteFile = false);
 
   // Merge Corpora[1:] into Corpora[0].
   void Merge(const std::vector<std::string> &Corpora);
@@ -95,7 +95,7 @@ private:
   void InterruptCallback();
   void MutateAndTestOne();
   void ReportNewCoverage(InputInfo *II, const Unit &U);
-  size_t RunOne(const Unit &U) { return RunOne(U.data(), U.size()); }
+  void PrintPulseAndReportSlowInput(const uint8_t *Data, size_t Size);
   void WriteToOutputCorpus(const Unit &U);
   void WriteUnitToFileWithPrefix(const Unit &U, const char *Prefix);
   void PrintStats(const char *Where, const char *End = "\n", size_t Units = 0);
@@ -142,6 +142,8 @@ private:
   size_t MaxInputLen = 0;
   size_t MaxMutationLen = 0;
 
+  std::vector<size_t> FeatureSetTmp;
+
   // Need to know our own thread.
   static thread_local bool IsMyThread;
 };
diff --git a/llvm/lib/Fuzzer/FuzzerLoop.cpp b/llvm/lib/Fuzzer/FuzzerLoop.cpp
index fbf18357ede..88ec844ab44 100644
--- a/llvm/lib/Fuzzer/FuzzerLoop.cpp
+++ b/llvm/lib/Fuzzer/FuzzerLoop.cpp
@@ -348,11 +348,8 @@ void Fuzzer::RereadOutputCorpus(size_t MaxSize) {
     if (U.size() > MaxSize)
       U.resize(MaxSize);
     if (!Corpus.HasUnit(U)) {
-      if (size_t NumFeatures = RunOne(U)) {
-        CheckExitOnSrcPosOrItem();
-        Corpus.AddToCorpus(U, NumFeatures);
+      if (RunOne(U.data(), U.size()))
         Reloaded = true;
-      }
     }
   }
   if (Reloaded)
@@ -377,10 +374,7 @@ void Fuzzer::ShuffleAndMinimize(UnitVector *InitialCorpus) {
   ExecuteCallback(&dummy, 0);
 
   for (const auto &U : *InitialCorpus) {
-    if (size_t NumFeatures = RunOne(U)) {
-      CheckExitOnSrcPosOrItem();
-      Corpus.AddToCorpus(U, NumFeatures);
-    }
+    RunOne(U.data(), U.size());
     TryDetectingAMemoryLeak(U.data(), U.size(),
                             /*DuringInitialCorpusExecution*/ true);
   }
@@ -392,18 +386,7 @@ void Fuzzer::ShuffleAndMinimize(UnitVector *InitialCorpus) {
   }
 }
 
-size_t Fuzzer::RunOne(const uint8_t *Data, size_t Size) {
-  if (!Size) return 0;
-  TotalNumberOfRuns++;
-
-  ExecuteCallback(Data, Size);
-
-  size_t NumUpdatesBefore = Corpus.NumFeatureUpdates();
-  TPC.CollectFeatures([&](size_t Feature) {
-    Corpus.AddFeature(Feature, Size, Options.Shrink);
-  });
-  size_t NumUpdatesAfter = Corpus.NumFeatureUpdates();
-
+void Fuzzer::PrintPulseAndReportSlowInput(const uint8_t *Data, size_t Size) {
   auto TimeOfUnit =
       duration_cast<seconds>(UnitStopTime - UnitStartTime).count();
   if (!(TotalNumberOfRuns & (TotalNumberOfRuns - 1)) &&
@@ -415,7 +398,27 @@ size_t Fuzzer::RunOne(const uint8_t *Data, size_t Size) {
     Printf("Slowest unit: %zd s:\n", TimeOfLongestUnitInSeconds);
     WriteUnitToFileWithPrefix({Data, Data + Size}, "slow-unit-");
   }
-  return NumUpdatesAfter - NumUpdatesBefore;
+}
+
+bool Fuzzer::RunOne(const uint8_t *Data, size_t Size, bool MayDeleteFile) {
+  if (!Size) return false;
+
+  ExecuteCallback(Data, Size);
+
+  FeatureSetTmp.clear();
+  size_t NumUpdatesBefore = Corpus.NumFeatureUpdates();
+  TPC.CollectFeatures([&](size_t Feature) {
+    Corpus.AddFeature(Feature, Size, Options.Shrink);
+    if (Options.ReduceInputs)
+      FeatureSetTmp.push_back(Feature);
+  });
+  PrintPulseAndReportSlowInput(Data, Size);
+  size_t NumNewFeatures = Corpus.NumFeatureUpdates() - NumUpdatesBefore;
+  if (NumNewFeatures) {
+    CheckExitOnSrcPosOrItem();
+    Corpus.AddToCorpus({Data, Data + Size}, NumNewFeatures, MayDeleteFile);
+  }
+  return NumNewFeatures > 0;
 }
 
 size_t Fuzzer::GetCurrentUnitInFuzzingThead(const uint8_t **Data) const {
@@ -443,6 +446,7 @@ static bool LooseMemeq(const uint8_t *A, const uint8_t *B, size_t Size) {
 }
 
 void Fuzzer::ExecuteCallback(const uint8_t *Data, size_t Size) {
+  TotalNumberOfRuns++;
   assert(InFuzzingThread());
   if (SMR.IsClient())
     SMR.WriteByteArray(Data, Size);
@@ -595,12 +599,9 @@ void Fuzzer::MutateAndTestOne() {
     if (i == 0)
       StartTraceRecording();
     II.NumExecutedMutations++;
-    if (size_t NumFeatures = RunOne(CurrentUnitData, Size)) {
-      Corpus.AddToCorpus({CurrentUnitData, CurrentUnitData + Size}, NumFeatures,
-                         /*MayDeleteFile=*/true);
+    if (RunOne(CurrentUnitData, Size, /*MayDeleteFile=*/true))
       ReportNewCoverage(&II, {CurrentUnitData, CurrentUnitData + Size});
-      CheckExitOnSrcPosOrItem();
-    }
+
     StopTraceRecording();
     TryDetectingAMemoryLeak(CurrentUnitData, Size,
                             /*DuringInitialCorpusExecution*/ false);
@@ -638,7 +639,8 @@ void Fuzzer::MinimizeCrashLoop(const Unit &U) {
     for (int i = 0; i < Options.MutateDepth; i++) {
       size_t NewSize = MD.Mutate(CurrentUnitData, U.size(), MaxMutationLen);
       assert(NewSize > 0 && NewSize <= MaxMutationLen);
-      RunOne(CurrentUnitData, NewSize);
+      ExecuteCallback(CurrentUnitData, NewSize);
+      PrintPulseAndReportSlowInput(CurrentUnitData, NewSize);
       TryDetectingAMemoryLeak(CurrentUnitData, NewSize,
                               /*DuringInitialCorpusExecution*/ false);
     }
diff --git a/llvm/lib/Fuzzer/FuzzerOptions.h b/llvm/lib/Fuzzer/FuzzerOptions.h
index b1366789be0..9500235e2b1 100644
--- a/llvm/lib/Fuzzer/FuzzerOptions.h
+++ b/llvm/lib/Fuzzer/FuzzerOptions.h
@@ -32,6 +32,7 @@ struct FuzzingOptions {
   bool UseCmp = false;
   bool UseValueProfile = false;
   bool Shrink = false;
+  bool ReduceInputs = false;
   int ReloadIntervalSec = 1;
   bool ShuffleAtStartUp = true;
   bool PreferSmall = true;