[libFuzzer] reimplement the way we do -only_ascii to allow more 'const' in function declarations. Add a test for -only_ascii. NFC intended

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@270900 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Fuzzer/FuzzerInternal.h

@@ -122,6 +122,7 @@ void ComputeSHA1(const uint8_t *Data, size_t Len, uint8_t *Out);
 // 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);
 
 int NumberOfCpuCores();
 int GetPid();
@@ -408,13 +409,13 @@ private:
   void MutateAndTestOne();
   void ReportNewCoverage(const Unit &U);
   bool RunOne(const Unit &U) { return RunOne(U.data(), U.size()); }
-  void RunOneAndUpdateCorpus(uint8_t *Data, size_t Size);
+  void RunOneAndUpdateCorpus(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");
   void PrintStatusForNewUnit(const Unit &U);
   void ShuffleCorpus(UnitVector *V);
-  void TryDetectingAMemoryLeak(uint8_t *Data, size_t Size);
+  void TryDetectingAMemoryLeak(const uint8_t *Data, size_t Size);
   void CheckForMemoryLeaks();
 
   // Updates the probability distribution for the units in the corpus.
@@ -437,7 +438,7 @@ private:
   void DumpCurrentUnit(const char *Prefix);
   void DeathCallback();
 
-  uint8_t *CurrentUnitData = nullptr;
+  const uint8_t *CurrentUnitData = nullptr;
   size_t CurrentUnitSize = 0;
   bool InOOMState = false;
 

lib/Fuzzer/FuzzerLoop.cpp

@@ -457,11 +457,9 @@ bool Fuzzer::RunOne(const uint8_t *Data, size_t Size) {
   return Res;
 }
 
-void Fuzzer::RunOneAndUpdateCorpus(uint8_t *Data, size_t Size) {
+void Fuzzer::RunOneAndUpdateCorpus(const uint8_t *Data, size_t Size) {
   if (TotalNumberOfRuns >= Options.MaxNumberOfRuns)
     return;
-  if (Options.OnlyASCII)
-    ToASCII(Data, Size);
   if (RunOne(Data, Size))
     ReportNewCoverage({Data, Data + Size});
 }
@@ -525,13 +523,14 @@ std::string Fuzzer::Coverage::DebugString() const {
 }
 
 void Fuzzer::WriteToOutputCorpus(const Unit &U) {
+  if (Options.OnlyASCII)
+    assert(IsASCII(U));
   if (Options.OutputCorpus.empty())
     return;
   std::string Path = DirPlusFile(Options.OutputCorpus, Hash(U));
   WriteToFile(U, Path);
   if (Options.Verbosity >= 2)
     Printf("Written to %s\n", Path.c_str());
-  assert(!Options.OnlyASCII || IsASCII(U));
 }
 
 void Fuzzer::WriteUnitToFileWithPrefix(const Unit &U, const char *Prefix) {
@@ -659,7 +658,7 @@ void Fuzzer::CheckForMemoryLeaks() {
 
 // Tries detecting a memory leak on the particular input that we have just
 // executed before calling this function.
-void Fuzzer::TryDetectingAMemoryLeak(uint8_t *Data, size_t Size) {
+void Fuzzer::TryDetectingAMemoryLeak(const uint8_t *Data, size_t Size) {
   if (!HasMoreMallocsThanFrees) return;  // mallocs==frees, a leak is unlikely.
   if (!Options.DetectLeaks) return;
   if (!&__lsan_enable || !&__lsan_disable || !__lsan_do_recoverable_leak_check)
@@ -710,6 +709,8 @@ void Fuzzer::MutateAndTestOne() {
     assert(NewSize <= Options.MaxLen &&
            "Mutator return overisized unit");
     Size = NewSize;
+    if (Options.OnlyASCII)
+      ToASCII(MutateInPlaceHere.data(), Size);
     if (i == 0)
       StartTraceRecording();
     RunOneAndUpdateCorpus(MutateInPlaceHere.data(), Size);

lib/Fuzzer/FuzzerTraceState.cpp

@@ -175,7 +175,7 @@ static bool RecordingMemcmp = false;
 class TraceState {
  public:
    TraceState(MutationDispatcher &MD, const Fuzzer::FuzzingOptions &Options,
-              uint8_t **CurrentUnitData, size_t *CurrentUnitSize)
+              const uint8_t **CurrentUnitData, size_t *CurrentUnitSize)
        : MD(MD), Options(Options), CurrentUnitData(CurrentUnitData),
          CurrentUnitSize(CurrentUnitSize) {
      // Current trace collection is not thread-friendly and it probably
@@ -293,7 +293,7 @@ class TraceState {
   size_t LastDfsanLabel = 0;
   MutationDispatcher &MD;
   const Fuzzer::FuzzingOptions &Options;
-  uint8_t **CurrentUnitData;
+  const uint8_t **CurrentUnitData;
   size_t *CurrentUnitSize;
   std::map<Word, size_t> AutoDictUnitCounts;
   size_t AutoDictAdds = 0;

lib/Fuzzer/FuzzerUtil.cpp

@@ -164,9 +164,11 @@ bool ToASCII(uint8_t *Data, size_t Size) {
   return Changed;
 }
 
-bool IsASCII(const Unit &U) {
-  for (auto X : U)
-    if (!(isprint(X) || isspace(X))) return false;
+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;
 }
 

lib/Fuzzer/test/fuzzer.test

@@ -5,6 +5,9 @@ RUN: LLVMFuzzer-SimpleTest 2>&1 | FileCheck %s
 RUN: not LLVMFuzzer-NullDerefTest %S/hi.txt 2>&1 | FileCheck %s --check-prefix=SingleInput
 SingleInput-NOT: Test unit written to ./crash-
 
+# only_ascii mode. Will perform some minimal self-validation.
+RUN: LLVMFuzzer-SimpleTest -only_ascii=1 2>&1
+
 RUN: LLVMFuzzer-SimpleCmpTest -max_total_time=1 2>&1 | FileCheck %s --check-prefix=MaxTotalTime
 MaxTotalTime: Done {{.*}} runs in {{.}} second(s)