summaryrefslogtreecommitdiff
path: root/lib/fuzzer
diff options
context:
space:
mode:
authorDimitry Andric <dim@FreeBSD.org>2019-08-20 20:51:06 +0000
committerDimitry Andric <dim@FreeBSD.org>2019-08-20 20:51:06 +0000
commit8f3cadc28cb2bb9e8f9d69eeaaea1f57f2f7b2ab (patch)
tree05a2b6ec297fe6283d9557c791445d1daf88dcd0 /lib/fuzzer
parent63714eb5809e39666dec2454c354195e76f916ba (diff)
Notes
Diffstat (limited to 'lib/fuzzer')
-rw-r--r--lib/fuzzer/FuzzerBuiltins.h7
-rw-r--r--lib/fuzzer/FuzzerBuiltinsMsvc.h9
-rw-r--r--lib/fuzzer/FuzzerCommand.h7
-rw-r--r--lib/fuzzer/FuzzerCorpus.h17
-rw-r--r--lib/fuzzer/FuzzerCrossOver.cpp7
-rw-r--r--lib/fuzzer/FuzzerDataFlowTrace.cpp270
-rw-r--r--lib/fuzzer/FuzzerDataFlowTrace.h89
-rw-r--r--lib/fuzzer/FuzzerDefs.h41
-rw-r--r--lib/fuzzer/FuzzerDictionary.h7
-rw-r--r--lib/fuzzer/FuzzerDriver.cpp171
-rw-r--r--lib/fuzzer/FuzzerExtFunctions.def12
-rw-r--r--lib/fuzzer/FuzzerExtFunctions.h7
-rw-r--r--lib/fuzzer/FuzzerExtFunctionsDlsym.cpp7
-rw-r--r--lib/fuzzer/FuzzerExtFunctionsWeak.cpp7
-rw-r--r--lib/fuzzer/FuzzerExtFunctionsWindows.cpp9
-rw-r--r--lib/fuzzer/FuzzerExtraCounters.cpp7
-rw-r--r--lib/fuzzer/FuzzerFlags.def46
-rw-r--r--lib/fuzzer/FuzzerFork.cpp409
-rw-r--r--lib/fuzzer/FuzzerFork.h24
-rw-r--r--lib/fuzzer/FuzzerIO.cpp44
-rw-r--r--lib/fuzzer/FuzzerIO.h29
-rw-r--r--lib/fuzzer/FuzzerIOPosix.cpp47
-rw-r--r--lib/fuzzer/FuzzerIOWindows.cpp94
-rw-r--r--lib/fuzzer/FuzzerInterface.h28
-rw-r--r--lib/fuzzer/FuzzerInternal.h21
-rw-r--r--lib/fuzzer/FuzzerLoop.cpp162
-rw-r--r--lib/fuzzer/FuzzerMain.cpp9
-rw-r--r--lib/fuzzer/FuzzerMerge.cpp224
-rw-r--r--lib/fuzzer/FuzzerMerge.h28
-rw-r--r--lib/fuzzer/FuzzerMutate.cpp30
-rw-r--r--lib/fuzzer/FuzzerMutate.h13
-rw-r--r--lib/fuzzer/FuzzerOptions.h18
-rw-r--r--lib/fuzzer/FuzzerRandom.h18
-rw-r--r--lib/fuzzer/FuzzerSHA1.cpp10
-rw-r--r--lib/fuzzer/FuzzerSHA1.h7
-rw-r--r--lib/fuzzer/FuzzerShmem.h69
-rw-r--r--lib/fuzzer/FuzzerShmemFuchsia.cpp38
-rw-r--r--lib/fuzzer/FuzzerShmemPosix.cpp108
-rw-r--r--lib/fuzzer/FuzzerShmemWindows.cpp64
-rw-r--r--lib/fuzzer/FuzzerTracePC.cpp343
-rw-r--r--lib/fuzzer/FuzzerTracePC.h102
-rw-r--r--lib/fuzzer/FuzzerUtil.cpp7
-rw-r--r--lib/fuzzer/FuzzerUtil.h23
-rw-r--r--lib/fuzzer/FuzzerUtilDarwin.cpp7
-rw-r--r--lib/fuzzer/FuzzerUtilFuchsia.cpp143
-rw-r--r--lib/fuzzer/FuzzerUtilLinux.cpp15
-rw-r--r--lib/fuzzer/FuzzerUtilPosix.cpp34
-rw-r--r--lib/fuzzer/FuzzerUtilWindows.cpp17
-rw-r--r--lib/fuzzer/FuzzerValueBitMap.h13
-rw-r--r--lib/fuzzer/utils/FuzzedDataProvider.h245
50 files changed, 2078 insertions, 1085 deletions
diff --git a/lib/fuzzer/FuzzerBuiltins.h b/lib/fuzzer/FuzzerBuiltins.h
index a80938d9a54d7..5f1ccef8a9cad 100644
--- a/lib/fuzzer/FuzzerBuiltins.h
+++ b/lib/fuzzer/FuzzerBuiltins.h
@@ -1,9 +1,8 @@
//===- FuzzerBuiltins.h - Internal header for builtins ----------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerBuiltinsMsvc.h b/lib/fuzzer/FuzzerBuiltinsMsvc.h
index 67dd57ff9a933..82709cfe7b404 100644
--- a/lib/fuzzer/FuzzerBuiltinsMsvc.h
+++ b/lib/fuzzer/FuzzerBuiltinsMsvc.h
@@ -1,9 +1,8 @@
//===- FuzzerBuiltinsMSVC.h - Internal header for builtins ------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -25,7 +24,7 @@
// __builtin_return_address() cannot be compiled with MSVC. Use the equivalent
// from <intrin.h>
-#define GET_CALLER_PC() reinterpret_cast<uintptr_t>(_ReturnAddress())
+#define GET_CALLER_PC() _ReturnAddress()
namespace fuzzer {
diff --git a/lib/fuzzer/FuzzerCommand.h b/lib/fuzzer/FuzzerCommand.h
index 9d258a228f89c..87308864af531 100644
--- a/lib/fuzzer/FuzzerCommand.h
+++ b/lib/fuzzer/FuzzerCommand.h
@@ -1,9 +1,8 @@
//===- FuzzerCommand.h - Interface representing a process -------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
diff --git a/lib/fuzzer/FuzzerCorpus.h b/lib/fuzzer/FuzzerCorpus.h
index f844c07c75726..6a95ef3a8e643 100644
--- a/lib/fuzzer/FuzzerCorpus.h
+++ b/lib/fuzzer/FuzzerCorpus.h
@@ -1,9 +1,8 @@
//===- FuzzerCorpus.h - Internal header for the Fuzzer ----------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -86,9 +85,10 @@ class InputCorpus {
bool empty() const { return Inputs.empty(); }
const Unit &operator[] (size_t Idx) const { return Inputs[Idx]->U; }
- void AddToCorpus(const Unit &U, size_t NumFeatures, bool MayDeleteFile,
- bool HasFocusFunction, const Vector<uint32_t> &FeatureSet,
- const DataFlowTrace &DFT, const InputInfo *BaseII) {
+ InputInfo *AddToCorpus(const Unit &U, size_t NumFeatures, bool MayDeleteFile,
+ bool HasFocusFunction,
+ 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);
@@ -114,6 +114,7 @@ class InputCorpus {
UpdateCorpusDistribution();
PrintCorpus();
// ValidateFeatureSet();
+ return &II;
}
// Debug-only
@@ -170,7 +171,7 @@ class InputCorpus {
InputInfo &II = *Inputs[ChooseUnitIdxToMutate(Rand)];
assert(!II.U.empty());
return II;
- };
+ }
// Returns an index of random unit from the corpus to mutate.
size_t ChooseUnitIdxToMutate(Random &Rand) {
diff --git a/lib/fuzzer/FuzzerCrossOver.cpp b/lib/fuzzer/FuzzerCrossOver.cpp
index 8b0fd7d529a81..83d9f8d47cb18 100644
--- a/lib/fuzzer/FuzzerCrossOver.cpp
+++ b/lib/fuzzer/FuzzerCrossOver.cpp
@@ -1,9 +1,8 @@
//===- FuzzerCrossOver.cpp - Cross over two test inputs -------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerDataFlowTrace.cpp b/lib/fuzzer/FuzzerDataFlowTrace.cpp
index 764f3e49fd2d3..99ff918f7c6c6 100644
--- a/lib/fuzzer/FuzzerDataFlowTrace.cpp
+++ b/lib/fuzzer/FuzzerDataFlowTrace.cpp
@@ -1,91 +1,281 @@
//===- FuzzerDataFlowTrace.cpp - DataFlowTrace ---*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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);
+ }
+}
-void DataFlowTrace::Init(const std::string &DirPath,
- const std::string &FocusFunction) {
- if (DirPath.empty()) return;
- const char *kFunctionsTxt = "functions.txt";
+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 FocusFuncIdx = SIZE_MAX;
size_t NumFunctions = 0;
while (std::getline(IF, L, '\n')) {
+ FunctionNames.push_back(L);
NumFunctions++;
- if (FocusFunction == L)
+ 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;
+ 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;
- auto ParseError = [&](const char *Err) {
- Printf("DataFlowTrace: parse error: %s\n File: %s\n Line: %s\n", Err,
- Name.c_str(), L.c_str());
- };
+ 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 SpacePos = L.find(' ');
- if (SpacePos == std::string::npos)
- return ParseError("no space in the trace line");
- if (L.empty() || L[0] != 'F')
- return ParseError("the trace line doesn't start with 'F'");
- size_t N = std::atol(L.c_str() + 1);
- if (N >= NumFunctions)
- return ParseError("N is greater than the number of functions");
- if (N == FocusFuncIdx) {
+ size_t FunctionNum = 0;
+ std::string DFTString;
+ if (ParseDFTLine(L, &FunctionNum, &DFTString) &&
+ FunctionNum == FocusFuncIdx) {
NumTracesWithFocusFunction++;
- const char *Beg = L.c_str() + SpacePos + 1;
- const char *End = L.c_str() + L.size();
- assert(Beg < End);
- size_t Len = End - Beg;
- Vector<uint8_t> V(Len);
- for (size_t I = 0; I < Len; I++) {
- if (Beg[I] != '0' && Beg[I] != '1')
- ParseError("the trace should contain only 0 or 1");
- V[I] = Beg[I] == '1';
- }
- Traces[Name] = V;
+
+ 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 && Len <= 16)
- Printf("%s => |%s|\n", Name.c_str(), L.c_str() + SpacePos + 1);
- break; // No need to parse the following lines.
+ 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.
}
}
}
- assert(NumTraceFiles == Files.size() - 1);
Printf("INFO: DataFlowTrace: %zd trace files, %zd functions, "
"%zd traces with focus function\n",
NumTraceFiles, NumFunctions, NumTracesWithFocusFunction);
+ return NumTraceFiles > 0;
}
-} // namespace fuzzer
+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());
+ static char DFSanEnv[] = "DFSAN_OPTIONS=fast16labels=1: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
diff --git a/lib/fuzzer/FuzzerDataFlowTrace.h b/lib/fuzzer/FuzzerDataFlowTrace.h
index ad4faeab7b2fe..d6e3de30a4ef4 100644
--- a/lib/fuzzer/FuzzerDataFlowTrace.h
+++ b/lib/fuzzer/FuzzerDataFlowTrace.h
@@ -1,9 +1,8 @@
//===- FuzzerDataFlowTrace.h - Internal header for the Fuzzer ---*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -30,15 +29,93 @@
#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 Init(const std::string &DirPath, const std::string &FocusFunction);
+ 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);
@@ -50,6 +127,8 @@ class DataFlowTrace {
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
diff --git a/lib/fuzzer/FuzzerDefs.h b/lib/fuzzer/FuzzerDefs.h
index c3dccbcd86f5c..320b37d5f8e3a 100644
--- a/lib/fuzzer/FuzzerDefs.h
+++ b/lib/fuzzer/FuzzerDefs.h
@@ -1,9 +1,8 @@
//===- FuzzerDefs.h - Internal header for the Fuzzer ------------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -120,31 +119,39 @@
# define ALWAYS_INLINE
#endif // __clang__
-#define ATTRIBUTE_NO_SANITIZE_ADDRESS __attribute__((no_sanitize_address))
-
-#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
+#if LIBFUZZER_WINDOWS
+#define ATTRIBUTE_NO_SANITIZE_ADDRESS
#else
-# define ATTRIBUTE_NO_SANITIZE_ALL
+#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 \
- __attribute__((tls_model("initial-exec"))) thread_local static
+#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
namespace fuzzer {
diff --git a/lib/fuzzer/FuzzerDictionary.h b/lib/fuzzer/FuzzerDictionary.h
index 0d9d91bcd2f15..301c5d9afecf8 100644
--- a/lib/fuzzer/FuzzerDictionary.h
+++ b/lib/fuzzer/FuzzerDictionary.h
@@ -1,9 +1,8 @@
//===- FuzzerDictionary.h - Internal header for the Fuzzer ------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
diff --git a/lib/fuzzer/FuzzerDriver.cpp b/lib/fuzzer/FuzzerDriver.cpp
index ff2a639ac4a37..54c7ff0795857 100644
--- a/lib/fuzzer/FuzzerDriver.cpp
+++ b/lib/fuzzer/FuzzerDriver.cpp
@@ -1,9 +1,8 @@
//===- FuzzerDriver.cpp - FuzzerDriver function and flags -----------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
//
//===----------------------------------------------------------------------===//
// FuzzerDriver and flag parsing.
@@ -11,12 +10,13 @@
#include "FuzzerCommand.h"
#include "FuzzerCorpus.h"
+#include "FuzzerFork.h"
#include "FuzzerIO.h"
#include "FuzzerInterface.h"
#include "FuzzerInternal.h"
+#include "FuzzerMerge.h"
#include "FuzzerMutate.h"
#include "FuzzerRandom.h"
-#include "FuzzerShmem.h"
#include "FuzzerTracePC.h"
#include <algorithm>
#include <atomic>
@@ -26,10 +26,16 @@
#include <mutex>
#include <string>
#include <thread>
+#include <fstream>
// This function should be present in the libFuzzer so that the client
// binary can test for its existence.
+#if LIBFUZZER_MSVC
+extern "C" void __libfuzzer_is_present() {}
+#pragma comment(linker, "/include:__libfuzzer_is_present")
+#else
extern "C" __attribute__((used)) void __libfuzzer_is_present() {}
+#endif // LIBFUZZER_MSVC
namespace fuzzer {
@@ -176,7 +182,8 @@ static bool ParseOneFlag(const char *Param) {
}
// We don't use any library to minimize dependencies.
-static void ParseFlags(const Vector<std::string> &Args) {
+static void ParseFlags(const Vector<std::string> &Args,
+ const ExternalFunctions *EF) {
for (size_t F = 0; F < kNumFlags; F++) {
if (FlagDescriptions[F].IntFlag)
*FlagDescriptions[F].IntFlag = FlagDescriptions[F].Default;
@@ -186,6 +193,11 @@ static void ParseFlags(const Vector<std::string> &Args) {
if (FlagDescriptions[F].StrFlag)
*FlagDescriptions[F].StrFlag = nullptr;
}
+
+ // Disable len_control by default, if LLVMFuzzerCustomMutator is used.
+ if (EF->LLVMFuzzerCustomMutator)
+ Flags.len_control = 0;
+
Inputs = new Vector<std::string>;
for (size_t A = 1; A < Args.size(); A++) {
if (ParseOneFlag(Args[A].c_str())) {
@@ -316,10 +328,8 @@ int CleanseCrashInput(const Vector<std::string> &Args,
assert(Cmd.hasArgument(InputFilePath));
Cmd.removeArgument(InputFilePath);
- auto LogFilePath = DirPlusFile(
- TmpDir(), "libFuzzerTemp." + std::to_string(GetPid()) + ".txt");
- auto TmpFilePath = DirPlusFile(
- TmpDir(), "libFuzzerTemp." + std::to_string(GetPid()) + ".repro");
+ auto LogFilePath = TempPath(".txt");
+ auto TmpFilePath = TempPath(".repro");
Cmd.addArgument(TmpFilePath);
Cmd.setOutputFile(LogFilePath);
Cmd.combineOutAndErr();
@@ -379,8 +389,7 @@ int MinimizeCrashInput(const Vector<std::string> &Args,
BaseCmd.addFlag("max_total_time", "600");
}
- auto LogFilePath = DirPlusFile(
- TmpDir(), "libFuzzerTemp." + std::to_string(GetPid()) + ".txt");
+ auto LogFilePath = TempPath(".txt");
BaseCmd.setOutputFile(LogFilePath);
BaseCmd.combineOutAndErr();
@@ -464,6 +473,34 @@ int MinimizeCrashInputInternalStep(Fuzzer *F, InputCorpus *Corpus) {
return 0;
}
+void Merge(Fuzzer *F, FuzzingOptions &Options, const Vector<std::string> &Args,
+ const Vector<std::string> &Corpora, const char *CFPathOrNull) {
+ if (Corpora.size() < 2) {
+ Printf("INFO: Merge requires two or more corpus dirs\n");
+ exit(0);
+ }
+
+ Vector<SizedFile> OldCorpus, NewCorpus;
+ GetSizedFilesFromDir(Corpora[0], &OldCorpus);
+ for (size_t i = 1; i < Corpora.size(); i++)
+ GetSizedFilesFromDir(Corpora[i], &NewCorpus);
+ std::sort(OldCorpus.begin(), OldCorpus.end());
+ std::sort(NewCorpus.begin(), NewCorpus.end());
+
+ std::string CFPath = CFPathOrNull ? CFPathOrNull : TempPath(".txt");
+ Vector<std::string> NewFiles;
+ Set<uint32_t> NewFeatures, NewCov;
+ CrashResistantMerge(Args, OldCorpus, NewCorpus, &NewFiles, {}, &NewFeatures,
+ {}, &NewCov, CFPath, true);
+ for (auto &Path : NewFiles)
+ F->WriteToOutputCorpus(FileToVector(Path, Options.MaxLen));
+ // We are done, delete the control file if it was a temporary one.
+ if (!Flags.merge_control_file)
+ RemoveFile(CFPath);
+
+ exit(0);
+}
+
int AnalyzeDictionary(Fuzzer *F, const Vector<Unit>& Dict,
UnitVector& Corpus) {
Printf("Started dictionary minimization (up to %d tests)\n",
@@ -530,6 +567,45 @@ int AnalyzeDictionary(Fuzzer *F, const Vector<Unit>& Dict,
return 0;
}
+Vector<std::string> ParseSeedInuts(const char *seed_inputs) {
+ // Parse -seed_inputs=file1,file2,... or -seed_inputs=@seed_inputs_file
+ Vector<std::string> Files;
+ if (!seed_inputs) return Files;
+ std::string SeedInputs;
+ if (Flags.seed_inputs[0] == '@')
+ SeedInputs = FileToString(Flags.seed_inputs + 1); // File contains list.
+ else
+ SeedInputs = Flags.seed_inputs; // seed_inputs contains the list.
+ if (SeedInputs.empty()) {
+ Printf("seed_inputs is empty or @file does not exist.\n");
+ exit(1);
+ }
+ // Parse SeedInputs.
+ size_t comma_pos = 0;
+ while ((comma_pos = SeedInputs.find_last_of(',')) != std::string::npos) {
+ Files.push_back(SeedInputs.substr(comma_pos + 1));
+ SeedInputs = SeedInputs.substr(0, comma_pos);
+ }
+ Files.push_back(SeedInputs);
+ return Files;
+}
+
+static Vector<SizedFile> ReadCorpora(const Vector<std::string> &CorpusDirs,
+ const Vector<std::string> &ExtraSeedFiles) {
+ Vector<SizedFile> SizedFiles;
+ size_t LastNumFiles = 0;
+ for (auto &Dir : CorpusDirs) {
+ GetSizedFilesFromDir(Dir, &SizedFiles);
+ Printf("INFO: % 8zd files found in %s\n", SizedFiles.size() - LastNumFiles,
+ Dir.c_str());
+ LastNumFiles = SizedFiles.size();
+ }
+ for (auto &File : ExtraSeedFiles)
+ if (auto Size = FileSize(File))
+ SizedFiles.push_back({File, Size});
+ return SizedFiles;
+}
+
int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
using namespace fuzzer;
assert(argc && argv && "Argument pointers cannot be nullptr");
@@ -546,7 +622,7 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
Printf("ERROR: argv[0] has been modified in LLVMFuzzerInitialize\n");
exit(1);
}
- ParseFlags(Args);
+ ParseFlags(Args, EF);
if (Flags.help) {
PrintHelp();
return 0;
@@ -573,6 +649,9 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
Options.UnitTimeoutSec = Flags.timeout;
Options.ErrorExitCode = Flags.error_exitcode;
Options.TimeoutExitCode = Flags.timeout_exitcode;
+ Options.IgnoreTimeouts = Flags.ignore_timeouts;
+ Options.IgnoreOOMs = Flags.ignore_ooms;
+ Options.IgnoreCrashes = Flags.ignore_crashes;
Options.MaxTotalTimeSec = Flags.max_total_time;
Options.DoCrossOver = Flags.cross_over;
Options.MutateDepth = Flags.mutate_depth;
@@ -609,15 +688,14 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
return 1;
if (Flags.verbosity > 0 && !Dictionary.empty())
Printf("Dictionary: %zd entries\n", Dictionary.size());
- bool DoPlainRun = AllInputsAreFiles();
+ bool RunIndividualFiles = AllInputsAreFiles();
Options.SaveArtifacts =
- !DoPlainRun || Flags.minimize_crash_internal_step;
+ !RunIndividualFiles || Flags.minimize_crash_internal_step;
Options.PrintNewCovPcs = Flags.print_pcs;
Options.PrintNewCovFuncs = Flags.print_funcs;
Options.PrintFinalStats = Flags.print_final_stats;
Options.PrintCorpusStats = Flags.print_corpus_stats;
Options.PrintCoverage = Flags.print_coverage;
- Options.DumpCoverage = Flags.dump_coverage;
if (Flags.exit_on_src_pos)
Options.ExitOnSrcPos = Flags.exit_on_src_pos;
if (Flags.exit_on_item)
@@ -626,6 +704,13 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
Options.FocusFunction = Flags.focus_function;
if (Flags.data_flow_trace)
Options.DataFlowTrace = Flags.data_flow_trace;
+ if (Flags.features_dir)
+ Options.FeaturesDir = Flags.features_dir;
+ if (Flags.collect_data_flow)
+ Options.CollectDataFlow = Flags.collect_data_flow;
+ Options.LazyCounters = Flags.lazy_counters;
+ if (Flags.stop_file)
+ Options.StopFile = Flags.stop_file;
unsigned Seed = Flags.seed;
// Initialize Seed.
@@ -635,6 +720,15 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
if (Flags.verbosity)
Printf("INFO: Seed: %u\n", Seed);
+ if (Flags.collect_data_flow && !Flags.fork && !Flags.merge) {
+ if (RunIndividualFiles)
+ return CollectDataFlow(Flags.collect_data_flow, Flags.data_flow_trace,
+ ReadCorpora({}, *Inputs));
+ else
+ return CollectDataFlow(Flags.collect_data_flow, Flags.data_flow_trace,
+ ReadCorpora(*Inputs, {}));
+ }
+
Random Rand(Seed);
auto *MD = new MutationDispatcher(Rand, Options);
auto *Corpus = new InputCorpus(Options.OutputCorpus);
@@ -669,35 +763,7 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
if (Flags.cleanse_crash)
return CleanseCrashInput(Args, Options);
-#if 0 // deprecated, to be removed.
- if (auto Name = Flags.run_equivalence_server) {
- SMR.Destroy(Name);
- if (!SMR.Create(Name)) {
- Printf("ERROR: can't create shared memory region\n");
- return 1;
- }
- Printf("INFO: EQUIVALENCE SERVER UP\n");
- while (true) {
- SMR.WaitClient();
- size_t Size = SMR.ReadByteArraySize();
- SMR.WriteByteArray(nullptr, 0);
- const Unit tmp(SMR.GetByteArray(), SMR.GetByteArray() + Size);
- F->ExecuteCallback(tmp.data(), tmp.size());
- SMR.PostServer();
- }
- return 0;
- }
-
- if (auto Name = Flags.use_equivalence_server) {
- if (!SMR.Open(Name)) {
- Printf("ERROR: can't open shared memory region\n");
- return 1;
- }
- Printf("INFO: EQUIVALENCE CLIENT UP\n");
- }
-#endif
-
- if (DoPlainRun) {
+ if (RunIndividualFiles) {
Options.SaveArtifacts = false;
int Runs = std::max(1, Flags.runs);
Printf("%s: Running %zd inputs %d time(s) each.\n", ProgName->c_str(),
@@ -719,13 +785,11 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
exit(0);
}
- if (Flags.merge) {
- F->CrashResistantMerge(Args, *Inputs,
- Flags.load_coverage_summary,
- Flags.save_coverage_summary,
- Flags.merge_control_file);
- exit(0);
- }
+ if (Flags.fork)
+ FuzzWithFork(F->GetMD().GetRand(), Options, Args, *Inputs, Flags.fork);
+
+ if (Flags.merge)
+ Merge(F, Options, Args, *Inputs, Flags.merge_control_file);
if (Flags.merge_inner) {
const size_t kDefaultMaxMergeLen = 1 << 20;
@@ -757,7 +821,8 @@ int FuzzerDriver(int *argc, char ***argv, UserCallback Callback) {
exit(0);
}
- F->Loop(*Inputs);
+ auto CorporaFiles = ReadCorpora(*Inputs, ParseSeedInuts(Flags.seed_inputs));
+ F->Loop(CorporaFiles);
if (Flags.verbosity)
Printf("Done %zd runs in %zd second(s)\n", F->getTotalNumberOfRuns(),
diff --git a/lib/fuzzer/FuzzerExtFunctions.def b/lib/fuzzer/FuzzerExtFunctions.def
index 8bfffdde56d47..41fa0fd2b7482 100644
--- a/lib/fuzzer/FuzzerExtFunctions.def
+++ b/lib/fuzzer/FuzzerExtFunctions.def
@@ -1,9 +1,8 @@
//===- FuzzerExtFunctions.def - External functions --------------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -29,7 +28,7 @@ EXT_FUNC(LLVMFuzzerCustomCrossOver, size_t,
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, bool, (), true);
+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 *)),
@@ -44,8 +43,7 @@ EXT_FUNC(__sanitizer_get_module_and_offset_for_pc, int,
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(__sanitizer_dump_coverage, void, (const uintptr_t *, uintptr_t),
- 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);
diff --git a/lib/fuzzer/FuzzerExtFunctions.h b/lib/fuzzer/FuzzerExtFunctions.h
index 2672a385478d1..c88aac4e67c25 100644
--- a/lib/fuzzer/FuzzerExtFunctions.h
+++ b/lib/fuzzer/FuzzerExtFunctions.h
@@ -1,9 +1,8 @@
//===- FuzzerExtFunctions.h - Interface to external functions ---*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerExtFunctionsDlsym.cpp b/lib/fuzzer/FuzzerExtFunctionsDlsym.cpp
index 06bddd5de38f3..dcd7134594870 100644
--- a/lib/fuzzer/FuzzerExtFunctionsDlsym.cpp
+++ b/lib/fuzzer/FuzzerExtFunctionsDlsym.cpp
@@ -1,9 +1,8 @@
//===- FuzzerExtFunctionsDlsym.cpp - Interface to external functions ------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
diff --git a/lib/fuzzer/FuzzerExtFunctionsWeak.cpp b/lib/fuzzer/FuzzerExtFunctionsWeak.cpp
index 6a6ef4932f416..ea5b87bd5196f 100644
--- a/lib/fuzzer/FuzzerExtFunctionsWeak.cpp
+++ b/lib/fuzzer/FuzzerExtFunctionsWeak.cpp
@@ -1,9 +1,8 @@
//===- FuzzerExtFunctionsWeak.cpp - Interface to external functions -------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
diff --git a/lib/fuzzer/FuzzerExtFunctionsWindows.cpp b/lib/fuzzer/FuzzerExtFunctionsWindows.cpp
index b01871439ca06..55efe8f80e908 100644
--- a/lib/fuzzer/FuzzerExtFunctionsWindows.cpp
+++ b/lib/fuzzer/FuzzerExtFunctionsWindows.cpp
@@ -1,9 +1,8 @@
//=== FuzzerExtWindows.cpp - Interface to external functions --------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -50,7 +49,7 @@ extern "C" {
Printf("ERROR: Function \"%s\" not defined.\n", #NAME); \
exit(1); \
} \
- EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG;
+ EXTERNAL_FUNC(NAME, NAME##Def) RETURN_TYPE NAME FUNC_SIG
#include "FuzzerExtFunctions.def"
diff --git a/lib/fuzzer/FuzzerExtraCounters.cpp b/lib/fuzzer/FuzzerExtraCounters.cpp
index c99cd89be293a..3f38f4fb70c12 100644
--- a/lib/fuzzer/FuzzerExtraCounters.cpp
+++ b/lib/fuzzer/FuzzerExtraCounters.cpp
@@ -1,9 +1,8 @@
//===- FuzzerExtraCounters.cpp - Extra coverage counters ------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerFlags.def b/lib/fuzzer/FuzzerFlags.def
index 91281c979c50a..a11cfe4405f31 100644
--- a/lib/fuzzer/FuzzerFlags.def
+++ b/lib/fuzzer/FuzzerFlags.def
@@ -1,9 +1,8 @@
//===- FuzzerFlags.def - Run-time flags -------------------------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -20,7 +19,10 @@ FUZZER_FLAG_INT(max_len, 0, "Maximum length of the test input. "
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.")
+ "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-seperated list.")
FUZZER_FLAG_INT(cross_over, 1, "If 1, cross over inputs.")
FUZZER_FLAG_INT(mutate_depth, 5,
"Apply this number of consecutive mutations to each input.")
@@ -35,27 +37,26 @@ FUZZER_FLAG_INT(
"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, 77, "When libFuzzer reports a timeout "
+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.")
-FUZZER_FLAG_STRING(save_coverage_summary, "Experimental:"
- " save coverage summary to a given file."
- " Used with -merge=1")
-FUZZER_FLAG_STRING(load_coverage_summary, "Experimental:"
- " load coverage summary from a given file."
- " Treat this coverage as belonging to the first corpus. "
- " Used with -merge=1")
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."
@@ -68,6 +69,10 @@ FUZZER_FLAG_INT(cleanse_crash, 0, "If 1, tries to cleanse the provided"
" 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_INT(use_counters, 1, "Use coverage counters")
FUZZER_FLAG_INT(use_memmem, 1,
"Use hints from intercepting memmem, strstr, etc")
@@ -107,9 +112,7 @@ 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."
- " If 1, dump coverage information as a"
- " .sancov file 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.")
@@ -120,6 +123,9 @@ 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(lazy_counters, 0, "If 1, a performance optimization is"
+ "enabled for the 8bit inline counters. "
+ "Requires that libFuzzer successfully installs its SEGV handler")
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.")
@@ -146,10 +152,12 @@ 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")
+ "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_DEPRECATED_FLAG(run_equivalence_server)
-FUZZER_DEPRECATED_FLAG(use_equivalence_server)
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")
diff --git a/lib/fuzzer/FuzzerFork.cpp b/lib/fuzzer/FuzzerFork.cpp
new file mode 100644
index 0000000000000..95ed36551463b
--- /dev/null
+++ b/lib/fuzzer/FuzzerFork.cpp
@@ -0,0 +1,409 @@
+//===- 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(".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];
+
+ 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
diff --git a/lib/fuzzer/FuzzerFork.h b/lib/fuzzer/FuzzerFork.h
new file mode 100644
index 0000000000000..b29a43e13fbc5
--- /dev/null
+++ b/lib/fuzzer/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
diff --git a/lib/fuzzer/FuzzerIO.cpp b/lib/fuzzer/FuzzerIO.cpp
index c4c31e82471ff..7e5ba30a2e7dd 100644
--- a/lib/fuzzer/FuzzerIO.cpp
+++ b/lib/fuzzer/FuzzerIO.cpp
@@ -1,17 +1,17 @@
//===- FuzzerIO.cpp - IO utils. -------------------------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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 "FuzzerIO.h"
#include "FuzzerDefs.h"
#include "FuzzerExtFunctions.h"
+#include "FuzzerIO.h"
+#include "FuzzerUtil.h"
#include <algorithm>
#include <cstdarg>
#include <fstream>
@@ -61,10 +61,19 @@ void CopyFileToErr(const std::string &Path) {
}
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(), "w");
+ FILE *Out = fopen(Path.c_str(), "wb");
if (!Out) return;
- fwrite(U.data(), sizeof(U[0]), U.size(), Out);
+ fwrite(Data, sizeof(Data[0]), Size, Out);
fclose(Out);
}
@@ -126,4 +135,25 @@ void Printf(const char *Fmt, ...) {
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);
+}
+
+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 *Extension) {
+ return DirPlusFile(TmpDir(),
+ "libFuzzerTemp." + std::to_string(GetPid()) + Extension);
+}
+
} // namespace fuzzer
diff --git a/lib/fuzzer/FuzzerIO.h b/lib/fuzzer/FuzzerIO.h
index b4a68190e7808..fe0d7b451758d 100644
--- a/lib/fuzzer/FuzzerIO.h
+++ b/lib/fuzzer/FuzzerIO.h
@@ -1,9 +1,8 @@
//===- FuzzerIO.h - Internal header for IO utils ----------------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -25,6 +24,9 @@ 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 ReadDirToVectorOfUnits(const char *Path, Vector<Unit> *V,
@@ -40,6 +42,8 @@ std::string DirName(const std::string &FileName);
// Returns path to a TmpDir.
std::string TmpDir();
+std::string TempPath(const char *Extension);
+
bool IsInterestingCoverageFile(const std::string &FileName);
void DupAndCloseStderr();
@@ -47,6 +51,7 @@ 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);
@@ -58,6 +63,16 @@ size_t FileSize(const std::string &Path);
void ListFilesInDirRecursive(const std::string &Dir, long *Epoch,
Vector<std::string> *V, bool TopDir);
+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;
@@ -77,11 +92,17 @@ int CloseFile(int Fd);
int DuplicateFile(int Fd);
void RemoveFile(const std::string &Path);
+void RenameFile(const std::string &OldPath, const std::string &NewPath);
void DiscardOutput(int Fd);
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
diff --git a/lib/fuzzer/FuzzerIOPosix.cpp b/lib/fuzzer/FuzzerIOPosix.cpp
index 401b4cbbf74fa..cfd69bbc8111a 100644
--- a/lib/fuzzer/FuzzerIOPosix.cpp
+++ b/lib/fuzzer/FuzzerIOPosix.cpp
@@ -1,9 +1,8 @@
//===- FuzzerIOPosix.cpp - IO utils for Posix. ----------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -79,6 +78,28 @@ void ListFilesInDirRecursive(const std::string &Dir, long *Epoch,
*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 '/';
}
@@ -99,6 +120,10 @@ 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());
+}
+
void DiscardOutput(int Fd) {
FILE* Temp = fopen("/dev/null", "w");
if (!Temp)
@@ -137,11 +162,23 @@ bool IsInterestingCoverageFile(const std::string &FileName) {
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
diff --git a/lib/fuzzer/FuzzerIOWindows.cpp b/lib/fuzzer/FuzzerIOWindows.cpp
index 75dcaf72a9402..510afebef738b 100644
--- a/lib/fuzzer/FuzzerIOWindows.cpp
+++ b/lib/fuzzer/FuzzerIOWindows.cpp
@@ -1,9 +1,8 @@
//===- FuzzerIOWindows.cpp - IO utils for Windows. ------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -72,6 +71,11 @@ bool IsFile(const std::string &Path) {
return IsFile(Path, Att);
}
+static bool IsDir(DWORD FileAttrs) {
+ if (FileAttrs == INVALID_FILE_ATTRIBUTES) return false;
+ return FileAttrs & FILE_ATTRIBUTE_DIRECTORY;
+}
+
std::string Basename(const std::string &Path) {
size_t Pos = Path.find_last_of("/\\");
if (Pos == std::string::npos) return Path;
@@ -82,8 +86,10 @@ std::string Basename(const std::string &Path) {
size_t FileSize(const std::string &Path) {
WIN32_FILE_ATTRIBUTE_DATA attr;
if (!GetFileAttributesExA(Path.c_str(), GetFileExInfoStandard, &attr)) {
- Printf("GetFileAttributesExA() failed for \"%s\" (Error code: %lu).\n",
- Path.c_str(), GetLastError());
+ 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;
@@ -141,6 +147,58 @@ void ListFilesInDirRecursive(const std::string &Dir, long *Epoch,
*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 '\\';
}
@@ -161,6 +219,10 @@ 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());
+}
+
void DiscardOutput(int Fd) {
FILE* Temp = fopen("nul", "w");
if (!Temp)
@@ -334,8 +396,24 @@ bool IsInterestingCoverageFile(const std::string &FileName) {
}
void RawPrint(const char *Str) {
- // Not tested, may or may not work. Fix if needed.
- Printf("%s", 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
diff --git a/lib/fuzzer/FuzzerInterface.h b/lib/fuzzer/FuzzerInterface.h
index 0f7effb2ab6a2..4f62822eac051 100644
--- a/lib/fuzzer/FuzzerInterface.h
+++ b/lib/fuzzer/FuzzerInterface.h
@@ -1,9 +1,8 @@
//===- FuzzerInterface.h - Interface header for the Fuzzer ------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -26,25 +25,32 @@
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.
-__attribute__((visibility("default"))) int
+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.
-__attribute__((visibility("default"))) int LLVMFuzzerInitialize(int *argc,
- char ***argv);
+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.
-__attribute__((visibility("default"))) size_t
+FUZZER_INTERFACE_VISIBILITY size_t
LLVMFuzzerCustomMutator(uint8_t *Data, size_t Size, size_t MaxSize,
unsigned int Seed);
@@ -52,7 +58,7 @@ LLVMFuzzerCustomMutator(uint8_t *Data, size_t Size, size_t MaxSize,
// 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.
-__attribute__((visibility("default"))) size_t
+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);
@@ -61,9 +67,11 @@ LLVMFuzzerCustomCrossOver(const uint8_t *Data1, size_t Size1,
// 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.
-__attribute__((visibility("default"))) size_t
+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
diff --git a/lib/fuzzer/FuzzerInternal.h b/lib/fuzzer/FuzzerInternal.h
index a7fdc89cba5f9..f2a4c437de386 100644
--- a/lib/fuzzer/FuzzerInternal.h
+++ b/lib/fuzzer/FuzzerInternal.h
@@ -1,9 +1,8 @@
//===- FuzzerInternal.h - Internal header for the Fuzzer --------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -36,8 +35,8 @@ public:
Fuzzer(UserCallback CB, InputCorpus &Corpus, MutationDispatcher &MD,
FuzzingOptions Options);
~Fuzzer();
- void Loop(const Vector<std::string> &CorpusDirs);
- void ReadAndExecuteSeedCorpora(const Vector<std::string> &CorpusDirs);
+ void Loop(Vector<SizedFile> &CorporaFiles);
+ void ReadAndExecuteSeedCorpora(Vector<SizedFile> &CorporaFiles);
void MinimizeCrashLoop(const Unit &U);
void RereadOutputCorpus(size_t MaxSize);
@@ -72,11 +71,6 @@ public:
// Merge Corpora[1:] into Corpora[0].
void Merge(const Vector<std::string> &Corpora);
- void CrashResistantMerge(const Vector<std::string> &Args,
- const Vector<std::string> &Corpora,
- const char *CoverageSummaryInputPathOrNull,
- const char *CoverageSummaryOutputPathOrNull,
- const char *MergeControlFilePathOrNull);
void CrashResistantMergeInternalStep(const std::string &ControlFilePath);
MutationDispatcher &GetMD() { return MD; }
void PrintFinalStats();
@@ -90,20 +84,19 @@ public:
bool DuringInitialCorpusExecution);
void HandleMalloc(size_t Size);
- void AnnounceOutput(const uint8_t *Data, size_t Size);
+ static void MaybeExitGracefully();
+ std::string WriteToOutputCorpus(const Unit &U);
private:
void AlarmCallback();
void CrashCallback();
void ExitCallback();
- void MaybeExitGracefully();
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 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);
void PrintStatusForNewUnit(const Unit &U, const char *Text);
diff --git a/lib/fuzzer/FuzzerLoop.cpp b/lib/fuzzer/FuzzerLoop.cpp
index a32a307230040..f773f9a13398e 100644
--- a/lib/fuzzer/FuzzerLoop.cpp
+++ b/lib/fuzzer/FuzzerLoop.cpp
@@ -1,9 +1,8 @@
//===- FuzzerLoop.cpp - Fuzzer's main loop --------------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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's main loop.
@@ -14,7 +13,6 @@
#include "FuzzerInternal.h"
#include "FuzzerMutate.h"
#include "FuzzerRandom.h"
-#include "FuzzerShmem.h"
#include "FuzzerTracePC.h"
#include <algorithm>
#include <cstring>
@@ -41,8 +39,6 @@ static const size_t kMaxUnitSizeToPrint = 256;
thread_local bool Fuzzer::IsMyThread;
-SharedMemoryRegion SMR;
-
bool RunningUserCallback = false;
// Only one Fuzzer per process.
@@ -135,7 +131,7 @@ void Fuzzer::HandleMalloc(size_t Size) {
DumpCurrentUnit("oom-");
Printf("SUMMARY: libFuzzer: out-of-memory\n");
PrintFinalStats();
- _Exit(Options.ErrorExitCode); // Stop right now.
+ _Exit(Options.OOMExitCode); // Stop right now.
}
Fuzzer::Fuzzer(UserCallback CB, InputCorpus &Corpus, MutationDispatcher &MD,
@@ -157,12 +153,10 @@ Fuzzer::Fuzzer(UserCallback CB, InputCorpus &Corpus, MutationDispatcher &MD,
if (!Options.OutputCorpus.empty() && Options.ReloadIntervalSec)
EpochOfLastReadOfOutputCorpus = GetEpoch(Options.OutputCorpus);
MaxInputLen = MaxMutationLen = Options.MaxLen;
- TmpMaxMutationLen = Max(size_t(4), Corpus.MaxInputSize());
+ TmpMaxMutationLen = 0; // Will be set once we load the corpus.
AllocateCurrentUnitData();
CurrentUnitSize = 0;
memset(BaseSha1, 0, sizeof(BaseSha1));
- TPC.SetFocusFunction(Options.FocusFunction);
- DFT.Init(Options.DataFlowTrace, Options.FocusFunction);
}
Fuzzer::~Fuzzer() {}
@@ -231,8 +225,9 @@ void Fuzzer::StaticFileSizeExceedCallback() {
}
void Fuzzer::CrashCallback() {
- if (EF->__sanitizer_acquire_crash_state)
- EF->__sanitizer_acquire_crash_state();
+ if (EF->__sanitizer_acquire_crash_state &&
+ !EF->__sanitizer_acquire_crash_state())
+ return;
Printf("==%lu== ERROR: libFuzzer: deadly signal\n", GetPid());
PrintStackTrace();
Printf("NOTE: libFuzzer has rudimentary signal handlers.\n"
@@ -259,16 +254,20 @@ void Fuzzer::ExitCallback() {
}
void Fuzzer::MaybeExitGracefully() {
- if (!GracefulExitRequested) return;
+ if (!F->GracefulExitRequested) return;
Printf("==%lu== INFO: libFuzzer: exiting as requested\n", GetPid());
- PrintFinalStats();
+ RmDirRecursive(TempPath(".dir"));
+ F->PrintFinalStats();
_Exit(0);
}
void Fuzzer::InterruptCallback() {
Printf("==%lu== libFuzzer: run interrupted; exiting\n", GetPid());
PrintFinalStats();
- _Exit(0); // Stop right now, don't perform any at-exit actions.
+ ScopedDisableMsanInterceptorChecks S; // RmDirRecursive may call opendir().
+ RmDirRecursive(TempPath(".dir"));
+ // Stop right now, don't perform any at-exit actions.
+ _Exit(Options.InterruptExitCode);
}
NO_SANITIZE_MEMORY
@@ -317,7 +316,7 @@ void Fuzzer::RssLimitCallback() {
DumpCurrentUnit("oom-");
Printf("SUMMARY: libFuzzer: out-of-memory\n");
PrintFinalStats();
- _Exit(Options.ErrorExitCode); // Stop right now.
+ _Exit(Options.OOMExitCode); // Stop right now.
}
void Fuzzer::PrintStats(const char *Where, const char *End, size_t Units) {
@@ -355,8 +354,6 @@ void Fuzzer::PrintStats(const char *Where, const char *End, size_t Units) {
void Fuzzer::PrintFinalStats() {
if (Options.PrintCoverage)
TPC.PrintCoverage();
- if (Options.DumpCoverage)
- TPC.DumpCoverage();
if (Options.PrintCorpusStats)
Corpus.PrintStats();
if (!Options.PrintFinalStats)
@@ -388,10 +385,10 @@ void Fuzzer::SetMaxMutationLen(size_t MaxMutationLen) {
void Fuzzer::CheckExitOnSrcPosOrItem() {
if (!Options.ExitOnSrcPos.empty()) {
static auto *PCsSet = new Set<uintptr_t>;
- auto HandlePC = [&](uintptr_t PC) {
- if (!PCsSet->insert(PC).second)
+ auto HandlePC = [&](const TracePC::PCTableEntry *TE) {
+ if (!PCsSet->insert(TE->PC).second)
return;
- std::string Descr = DescribePC("%F %L", PC + 1);
+ std::string Descr = DescribePC("%F %L", TE->PC + 1);
if (Descr.find(Options.ExitOnSrcPos) != std::string::npos) {
Printf("INFO: found line matching '%s', exiting.\n",
Options.ExitOnSrcPos.c_str());
@@ -447,6 +444,23 @@ void Fuzzer::PrintPulseAndReportSlowInput(const uint8_t *Data, size_t Size) {
}
}
+static void WriteFeatureSetToFile(const std::string &FeaturesDir,
+ const std::string &FileName,
+ const Vector<uint32_t> &FeatureSet) {
+ if (FeaturesDir.empty() || FeatureSet.empty()) return;
+ WriteToFile(reinterpret_cast<const uint8_t *>(FeatureSet.data()),
+ FeatureSet.size() * sizeof(FeatureSet[0]),
+ DirPlusFile(FeaturesDir, FileName));
+}
+
+static void RenameFeatureSetFile(const std::string &FeaturesDir,
+ const std::string &OldFile,
+ const std::string &NewFile) {
+ if (FeaturesDir.empty()) return;
+ RenameFile(DirPlusFile(FeaturesDir, OldFile),
+ DirPlusFile(FeaturesDir, NewFile));
+}
+
bool Fuzzer::RunOne(const uint8_t *Data, size_t Size, bool MayDeleteFile,
InputInfo *II, bool *FoundUniqFeatures) {
if (!Size)
@@ -471,15 +485,21 @@ bool Fuzzer::RunOne(const uint8_t *Data, size_t Size, bool MayDeleteFile,
size_t NumNewFeatures = Corpus.NumFeatureUpdates() - NumUpdatesBefore;
if (NumNewFeatures) {
TPC.UpdateObservedPCs();
- Corpus.AddToCorpus({Data, Data + Size}, NumNewFeatures, MayDeleteFile,
- TPC.ObservedFocusFunction(), UniqFeatureSetTmp, DFT, II);
+ auto NewII = Corpus.AddToCorpus({Data, Data + Size}, NumNewFeatures,
+ MayDeleteFile, TPC.ObservedFocusFunction(),
+ UniqFeatureSetTmp, DFT, II);
+ WriteFeatureSetToFile(Options.FeaturesDir, Sha1ToString(NewII->Sha1),
+ NewII->UniqFeatureSet);
return true;
}
if (II && FoundUniqFeaturesOfII &&
II->DataFlowTraceForFocusFunction.empty() &&
FoundUniqFeaturesOfII == II->UniqFeatureSet.size() &&
II->U.size() > Size) {
+ auto OldFeaturesFile = Sha1ToString(II->Sha1);
Corpus.Replace(II, {Data, Data + Size});
+ RenameFeatureSetFile(Options.FeaturesDir, OldFeaturesFile,
+ Sha1ToString(II->Sha1));
return true;
}
return false;
@@ -513,14 +533,14 @@ void Fuzzer::ExecuteCallback(const uint8_t *Data, size_t Size) {
TPC.RecordInitialStack();
TotalNumberOfRuns++;
assert(InFuzzingThread());
- if (SMR.IsClient())
- SMR.WriteByteArray(Data, Size);
// We copy the contents of Unit into a separate heap buffer
// so that we reliably find buffer overflows in it.
uint8_t *DataCopy = new uint8_t[Size];
memcpy(DataCopy, Data, Size);
if (EF->__msan_unpoison)
EF->__msan_unpoison(DataCopy, Size);
+ if (EF->__msan_unpoison_param)
+ EF->__msan_unpoison_param(2);
if (CurrentUnitData && CurrentUnitData != Data)
memcpy(CurrentUnitData, Data, Size);
CurrentUnitSize = Size;
@@ -543,15 +563,16 @@ void Fuzzer::ExecuteCallback(const uint8_t *Data, size_t Size) {
delete[] DataCopy;
}
-void Fuzzer::WriteToOutputCorpus(const Unit &U) {
+std::string Fuzzer::WriteToOutputCorpus(const Unit &U) {
if (Options.OnlyASCII)
assert(IsASCII(U));
if (Options.OutputCorpus.empty())
- return;
+ return "";
std::string Path = DirPlusFile(Options.OutputCorpus, Hash(U));
WriteToFile(U, Path);
if (Options.Verbosity >= 2)
Printf("Written %zd bytes to %s\n", U.size(), Path.c_str());
+ return Path;
}
void Fuzzer::WriteUnitToFileWithPrefix(const Unit &U, const char *Prefix) {
@@ -637,6 +658,8 @@ void Fuzzer::MutateAndTestOne() {
MD.StartMutationSequence();
auto &II = Corpus.ChooseUnitToMutate(MD.GetRand());
+ if (Options.DoCrossOver)
+ MD.SetCrossOverWith(&Corpus.ChooseUnitToMutate(MD.GetRand()).U);
const auto &U = II.U;
memcpy(BaseSha1, II.Sha1, sizeof(BaseSha1));
assert(CurrentUnitData);
@@ -659,7 +682,9 @@ void Fuzzer::MutateAndTestOne() {
Size <= CurrentMaxMutationLen)
NewSize = MD.MutateWithMask(CurrentUnitData, Size, Size,
II.DataFlowTraceForFocusFunction);
- else
+
+ // If MutateWithMask either failed or wasn't called, call default Mutate.
+ if (!NewSize)
NewSize = MD.Mutate(CurrentUnitData, Size, CurrentMaxMutationLen);
assert(NewSize > 0 && "Mutator returned empty unit");
assert(NewSize <= CurrentMaxMutationLen && "Mutator return oversized unit");
@@ -676,7 +701,7 @@ void Fuzzer::MutateAndTestOne() {
break; // We will mutate this input more in the next rounds.
}
if (Options.ReduceDepth && !FoundUniqFeatures)
- break;
+ break;
}
}
@@ -695,21 +720,13 @@ void Fuzzer::PurgeAllocator() {
LastAllocatorPurgeAttemptTime = system_clock::now();
}
-void Fuzzer::ReadAndExecuteSeedCorpora(const Vector<std::string> &CorpusDirs) {
+void Fuzzer::ReadAndExecuteSeedCorpora(Vector<SizedFile> &CorporaFiles) {
const size_t kMaxSaneLen = 1 << 20;
const size_t kMinDefaultLen = 4096;
- Vector<SizedFile> SizedFiles;
size_t MaxSize = 0;
size_t MinSize = -1;
size_t TotalSize = 0;
- size_t LastNumFiles = 0;
- for (auto &Dir : CorpusDirs) {
- GetSizedFilesFromDir(Dir, &SizedFiles);
- Printf("INFO: % 8zd files found in %s\n", SizedFiles.size() - LastNumFiles,
- Dir.c_str());
- LastNumFiles = SizedFiles.size();
- }
- for (auto &File : SizedFiles) {
+ for (auto &File : CorporaFiles) {
MaxSize = Max(File.Size, MaxSize);
MinSize = Min(File.Size, MinSize);
TotalSize += File.Size;
@@ -722,24 +739,28 @@ void Fuzzer::ReadAndExecuteSeedCorpora(const Vector<std::string> &CorpusDirs) {
uint8_t dummy = 0;
ExecuteCallback(&dummy, 0);
- if (SizedFiles.empty()) {
+ // Protect lazy counters here, after the once-init code has been executed.
+ if (Options.LazyCounters)
+ TPC.ProtectLazyCounters();
+
+ if (CorporaFiles.empty()) {
Printf("INFO: A corpus is not provided, starting from an empty corpus\n");
Unit U({'\n'}); // Valid ASCII input.
RunOne(U.data(), U.size());
} else {
Printf("INFO: seed corpus: files: %zd min: %zdb max: %zdb total: %zdb"
" rss: %zdMb\n",
- SizedFiles.size(), MinSize, MaxSize, TotalSize, GetPeakRSSMb());
+ CorporaFiles.size(), MinSize, MaxSize, TotalSize, GetPeakRSSMb());
if (Options.ShuffleAtStartUp)
- std::shuffle(SizedFiles.begin(), SizedFiles.end(), MD.GetRand());
+ std::shuffle(CorporaFiles.begin(), CorporaFiles.end(), MD.GetRand());
if (Options.PreferSmall) {
- std::stable_sort(SizedFiles.begin(), SizedFiles.end());
- assert(SizedFiles.front().Size <= SizedFiles.back().Size);
+ std::stable_sort(CorporaFiles.begin(), CorporaFiles.end());
+ assert(CorporaFiles.front().Size <= CorporaFiles.back().Size);
}
// Load and execute inputs one by one.
- for (auto &SF : SizedFiles) {
+ for (auto &SF : CorporaFiles) {
auto U = FileToVector(SF.File, MaxInputLen, /*ExitOnError=*/false);
assert(U.size() <= MaxInputLen);
RunOne(U.data(), U.size());
@@ -764,16 +785,25 @@ void Fuzzer::ReadAndExecuteSeedCorpora(const Vector<std::string> &CorpusDirs) {
}
}
-void Fuzzer::Loop(const Vector<std::string> &CorpusDirs) {
- ReadAndExecuteSeedCorpora(CorpusDirs);
+void Fuzzer::Loop(Vector<SizedFile> &CorporaFiles) {
+ auto FocusFunctionOrAuto = Options.FocusFunction;
+ DFT.Init(Options.DataFlowTrace, &FocusFunctionOrAuto, CorporaFiles,
+ MD.GetRand());
+ TPC.SetFocusFunction(FocusFunctionOrAuto);
+ ReadAndExecuteSeedCorpora(CorporaFiles);
DFT.Clear(); // No need for DFT any more.
TPC.SetPrintNewPCs(Options.PrintNewCovPcs);
TPC.SetPrintNewFuncs(Options.PrintNewCovFuncs);
system_clock::time_point LastCorpusReload = system_clock::now();
- if (Options.DoCrossOver)
- MD.SetCorpus(&Corpus);
+
+ TmpMaxMutationLen =
+ Min(MaxMutationLen, Max(size_t(4), Corpus.MaxInputSize()));
+
while (true) {
auto Now = system_clock::now();
+ if (!Options.StopFile.empty() &&
+ !FileToVector(Options.StopFile, 1, false).empty())
+ break;
if (duration_cast<seconds>(Now - LastCorpusReload).count() >=
Options.ReloadIntervalSec) {
RereadOutputCorpus(MaxInputLen);
@@ -824,44 +854,14 @@ void Fuzzer::MinimizeCrashLoop(const Unit &U) {
}
}
-void Fuzzer::AnnounceOutput(const uint8_t *Data, size_t Size) {
- if (SMR.IsServer()) {
- SMR.WriteByteArray(Data, Size);
- } else if (SMR.IsClient()) {
- SMR.PostClient();
- SMR.WaitServer();
- size_t OtherSize = SMR.ReadByteArraySize();
- uint8_t *OtherData = SMR.GetByteArray();
- if (Size != OtherSize || memcmp(Data, OtherData, Size) != 0) {
- size_t i = 0;
- for (i = 0; i < Min(Size, OtherSize); i++)
- if (Data[i] != OtherData[i])
- break;
- Printf("==%lu== ERROR: libFuzzer: equivalence-mismatch. Sizes: %zd %zd; "
- "offset %zd\n",
- GetPid(), Size, OtherSize, i);
- DumpCurrentUnit("mismatch-");
- Printf("SUMMARY: libFuzzer: equivalence-mismatch\n");
- PrintFinalStats();
- _Exit(Options.ErrorExitCode);
- }
- }
-}
-
} // namespace fuzzer
extern "C" {
-__attribute__((visibility("default"))) size_t
+ATTRIBUTE_INTERFACE size_t
LLVMFuzzerMutate(uint8_t *Data, size_t Size, size_t MaxSize) {
assert(fuzzer::F);
return fuzzer::F->GetMD().DefaultMutate(Data, Size, MaxSize);
}
-// Experimental
-__attribute__((visibility("default"))) void
-LLVMFuzzerAnnounceOutput(const uint8_t *Data, size_t Size) {
- assert(fuzzer::F);
- fuzzer::F->AnnounceOutput(Data, Size);
-}
} // extern "C"
diff --git a/lib/fuzzer/FuzzerMain.cpp b/lib/fuzzer/FuzzerMain.cpp
index f2c8e9c7bb111..771a34aed3167 100644
--- a/lib/fuzzer/FuzzerMain.cpp
+++ b/lib/fuzzer/FuzzerMain.cpp
@@ -1,9 +1,8 @@
//===- FuzzerMain.cpp - main() function and flags -------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
//
//===----------------------------------------------------------------------===//
// main() and flags.
@@ -16,6 +15,6 @@ extern "C" {
int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size);
} // extern "C"
-__attribute__((visibility("default"))) int main(int argc, char **argv) {
+ATTRIBUTE_INTERFACE int main(int argc, char **argv) {
return fuzzer::FuzzerDriver(&argc, &argv, LLVMFuzzerTestOneInput);
}
diff --git a/lib/fuzzer/FuzzerMerge.cpp b/lib/fuzzer/FuzzerMerge.cpp
index 5f3052a39c162..75b2b5d59b9c0 100644
--- a/lib/fuzzer/FuzzerMerge.cpp
+++ b/lib/fuzzer/FuzzerMerge.cpp
@@ -1,9 +1,8 @@
//===- FuzzerMerge.cpp - merging corpora ----------------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -43,10 +42,12 @@ void Merger::ParseOrExit(std::istream &IS, bool ParseCoverage) {
// file1
// file2 # One file name per line.
// STARTED 0 123 # FileID, file size
-// DONE 0 1 4 6 8 # FileID COV1 COV2 ...
-// STARTED 1 456 # If DONE is missing, the input crashed while processing.
+// 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
-// DONE 2 8 9
+// FT 2 8 9
+// COV 2 11 12
bool Merger::Parse(std::istream &IS, bool ParseCoverage) {
LastFailure.clear();
std::string Line;
@@ -71,11 +72,12 @@ bool Merger::Parse(std::istream &IS, bool ParseCoverage) {
if (!std::getline(IS, Files[i].Name, '\n'))
return false;
- // Parse STARTED and DONE lines.
+ // 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;
@@ -90,19 +92,25 @@ bool Merger::Parse(std::istream &IS, bool ParseCoverage) {
LastSeenStartMarker = ExpectedStartMarker;
assert(ExpectedStartMarker < Files.size());
ExpectedStartMarker++;
- } else if (Marker == "DONE") {
- // DONE FILE_ID COV1 COV2 COV3 ...
+ } 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 >> std::hex >> N)
+ 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;
}
@@ -121,21 +129,21 @@ size_t Merger::ApproximateMemoryConsumption() const {
return Res;
}
-// Decides which files need to be merged (add thost to NewFiles).
+// 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);
+ 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());
}
- size_t InitialNumFeatures = AllFeatures.size();
-
// 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;
@@ -161,22 +169,20 @@ size_t Merger::Merge(const Set<uint32_t> &InitialFeatures,
auto &Cur = Files[i].Features;
// Printf("%s -> sz %zd ft %zd\n", Files[i].Name.c_str(),
// Files[i].Size, Cur.size());
- size_t OldSize = AllFeatures.size();
- AllFeatures.insert(Cur.begin(), Cur.end());
- if (AllFeatures.size() > OldSize)
+ 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 AllFeatures.size() - InitialNumFeatures;
-}
-
-void Merger::PrintSummary(std::ostream &OS) {
- for (auto &File : Files) {
- OS << std::hex;
- OS << File.Name << " size: " << File.Size << " features: ";
- for (auto Feature : File.Features)
- OS << " " << Feature;
- OS << "\n";
- }
+ return NewFeatures->size();
}
Set<uint32_t> Merger::AllFeatures() const {
@@ -186,25 +192,6 @@ Set<uint32_t> Merger::AllFeatures() const {
return S;
}
-Set<uint32_t> Merger::ParseSummary(std::istream &IS) {
- std::string Line, Tmp;
- Set<uint32_t> Res;
- while (std::getline(IS, Line, '\n')) {
- size_t N;
- std::istringstream ISS1(Line);
- ISS1 >> Tmp; // Name
- ISS1 >> Tmp; // size:
- assert(Tmp == "size:" && "Corrupt summary file");
- ISS1 >> std::hex;
- ISS1 >> N; // File Size
- ISS1 >> Tmp; // features:
- assert(Tmp == "features:" && "Corrupt summary file");
- while (ISS1 >> std::hex >> N)
- Res.insert(N);
- }
- return Res;
-}
-
// 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());
@@ -223,8 +210,9 @@ void Fuzzer::CrashResistantMergeInternalStep(const std::string &CFPath) {
std::ofstream OF(CFPath, std::ofstream::out | std::ofstream::app);
Set<size_t> AllFeatures;
+ Set<const TracePC::PCTableEntry *> AllPCs;
for (size_t i = M.FirstNotProcessedFile; i < M.Files.size(); i++) {
- MaybeExitGracefully();
+ Fuzzer::MaybeExitGracefully();
auto U = FileToVector(M.Files[i].Name);
if (U.size() > MaxInputLen) {
U.resize(MaxInputLen);
@@ -232,7 +220,7 @@ void Fuzzer::CrashResistantMergeInternalStep(const std::string &CFPath) {
}
std::ostringstream StartedLine;
// Write the pre-run marker.
- OF << "STARTED " << std::dec << i << " " << U.size() << "\n";
+ OF << "STARTED " << i << " " << U.size() << "\n";
OF.flush(); // Flush is important since Command::Execute may crash.
// Run.
TPC.ResetMaps();
@@ -247,26 +235,36 @@ void Fuzzer::CrashResistantMergeInternalStep(const std::string &CFPath) {
if (AllFeatures.insert(Feature).second)
UniqFeatures.insert(Feature);
});
+ TPC.UpdateObservedPCs();
// Show stats.
if (!(TotalNumberOfRuns & (TotalNumberOfRuns - 1)))
PrintStats("pulse ");
// Write the post-run marker and the coverage.
- OF << "DONE " << i;
+ OF << "FT " << i;
for (size_t F : UniqFeatures)
- OF << " " << std::hex << F;
+ 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();
}
+ PrintStats("DONE ");
}
static void WriteNewControlFile(const std::string &CFPath,
- const Vector<SizedFile> &AllFiles,
- size_t NumFilesInFirstCorpus) {
+ const Vector<SizedFile> &OldCorpus,
+ const Vector<SizedFile> &NewCorpus) {
RemoveFile(CFPath);
std::ofstream ControlFile(CFPath);
- ControlFile << AllFiles.size() << "\n";
- ControlFile << NumFilesInFirstCorpus << "\n";
- for (auto &SF: AllFiles)
+ ControlFile << (OldCorpus.size() + NewCorpus.size()) << "\n";
+ ControlFile << OldCorpus.size() << "\n";
+ for (auto &SF: OldCorpus)
+ ControlFile << SF.File << "\n";
+ for (auto &SF: NewCorpus)
ControlFile << SF.File << "\n";
if (!ControlFile) {
Printf("MERGE-OUTER: failed to write to the control file: %s\n",
@@ -275,116 +273,90 @@ static void WriteNewControlFile(const std::string &CFPath,
}
}
-// Outer process. Does not call the target code and thus sohuld not fail.
-void Fuzzer::CrashResistantMerge(const Vector<std::string> &Args,
- const Vector<std::string> &Corpora,
- const char *CoverageSummaryInputPathOrNull,
- const char *CoverageSummaryOutputPathOrNull,
- const char *MergeControlFilePathOrNull) {
- if (Corpora.size() <= 1) {
- Printf("Merge requires two or more corpus dirs\n");
- return;
- }
- auto CFPath =
- MergeControlFilePathOrNull
- ? MergeControlFilePathOrNull
- : DirPlusFile(TmpDir(),
- "libFuzzerTemp." + std::to_string(GetPid()) + ".txt");
-
+// 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;
- if (MergeControlFilePathOrNull && FileSize(MergeControlFilePathOrNull)) {
- Printf("MERGE-OUTER: non-empty control file provided: '%s'\n",
- MergeControlFilePathOrNull);
+ if (FileSize(CFPath)) {
+ VPrintf(V, "MERGE-OUTER: non-empty control file provided: '%s'\n",
+ CFPath.c_str());
Merger M;
- std::ifstream IF(MergeControlFilePathOrNull);
+ std::ifstream IF(CFPath);
if (M.Parse(IF, /*ParseCoverage=*/false)) {
- Printf("MERGE-OUTER: control file ok, %zd files total,"
+ 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())
- Printf("MERGE-OUTER: '%s' will be skipped as unlucky "
+ 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()) {
- Printf("MERGE-OUTER: nothing to do, merge has been completed before\n");
+ VPrintf(
+ V, "MERGE-OUTER: nothing to do, merge has been completed before\n");
exit(0);
}
NumAttempts = M.Files.size() - M.FirstNotProcessedFile;
} else {
- Printf("MERGE-OUTER: bad control file, will overwrite it\n");
+ 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.
- Vector<SizedFile> AllFiles;
- GetSizedFilesFromDir(Corpora[0], &AllFiles);
- size_t NumFilesInFirstCorpus = AllFiles.size();
- std::sort(AllFiles.begin(), AllFiles.end());
- for (size_t i = 1; i < Corpora.size(); i++)
- GetSizedFilesFromDir(Corpora[i], &AllFiles);
- std::sort(AllFiles.begin() + NumFilesInFirstCorpus, AllFiles.end());
- Printf("MERGE-OUTER: %zd files, %zd in the initial corpus\n",
- AllFiles.size(), NumFilesInFirstCorpus);
- WriteNewControlFile(CFPath, AllFiles, NumFilesInFirstCorpus);
- NumAttempts = AllFiles.size();
+ NumAttempts = OldCorpus.size() + NewCorpus.size();
+ VPrintf(V, "MERGE-OUTER: %zd files, %zd in the initial corpus\n",
+ NumAttempts, OldCorpus.size());
+ WriteNewControlFile(CFPath, OldCorpus, NewCorpus);
}
// Execute the inner process until it passes.
// Every inner process should execute at least one input.
Command BaseCmd(Args);
BaseCmd.removeFlag("merge");
- bool Success = false;
+ BaseCmd.removeFlag("fork");
+ BaseCmd.removeFlag("collect_data_flow");
for (size_t Attempt = 1; Attempt <= NumAttempts; Attempt++) {
- MaybeExitGracefully();
- Printf("MERGE-OUTER: attempt %zd\n", 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) {
- Printf("MERGE-OUTER: succesfull in %zd attempt(s)\n", Attempt);
- Success = true;
+ VPrintf(V, "MERGE-OUTER: succesfull in %zd attempt(s)\n", Attempt);
break;
}
}
- if (!Success) {
- Printf("MERGE-OUTER: zero succesfull attempts, exiting\n");
- exit(1);
- }
// Read the control file and do the merge.
Merger M;
std::ifstream IF(CFPath);
IF.seekg(0, IF.end);
- Printf("MERGE-OUTER: the control file has %zd bytes\n", (size_t)IF.tellg());
+ 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();
- Printf("MERGE-OUTER: consumed %zdMb (%zdMb rss) to parse the control file\n",
- M.ApproximateMemoryConsumption() >> 20, GetPeakRSSMb());
- if (CoverageSummaryOutputPathOrNull) {
- Printf("MERGE-OUTER: writing coverage summary for %zd files to %s\n",
- M.Files.size(), CoverageSummaryOutputPathOrNull);
- std::ofstream SummaryOut(CoverageSummaryOutputPathOrNull);
- M.PrintSummary(SummaryOut);
- }
- Vector<std::string> NewFiles;
- Set<uint32_t> InitialFeatures;
- if (CoverageSummaryInputPathOrNull) {
- std::ifstream SummaryIn(CoverageSummaryInputPathOrNull);
- InitialFeatures = M.ParseSummary(SummaryIn);
- Printf("MERGE-OUTER: coverage summary loaded from %s, %zd features found\n",
- CoverageSummaryInputPathOrNull, InitialFeatures.size());
- }
- size_t NumNewFeatures = M.Merge(InitialFeatures, &NewFiles);
- Printf("MERGE-OUTER: %zd new files with %zd new features added\n",
- NewFiles.size(), NumNewFeatures);
- for (auto &F: NewFiles)
- WriteToOutputCorpus(FileToVector(F, MaxInputLen));
- // We are done, delete the control file if it was a temporary one.
- if (!MergeControlFilePathOrNull)
- RemoveFile(CFPath);
+ VPrintf(V,
+ "MERGE-OUTER: consumed %zdMb (%zdMb rss) to parse the control file\n",
+ M.ApproximateMemoryConsumption() >> 20, GetPeakRSSMb());
+ 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
diff --git a/lib/fuzzer/FuzzerMerge.h b/lib/fuzzer/FuzzerMerge.h
index e54885a1ebaef..c14dd589e62dd 100644
--- a/lib/fuzzer/FuzzerMerge.h
+++ b/lib/fuzzer/FuzzerMerge.h
@@ -1,9 +1,8 @@
//===- FuzzerMerge.h - merging corpa ----------------------------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -52,7 +51,7 @@ namespace fuzzer {
struct MergeFileInfo {
std::string Name;
size_t Size = 0;
- Vector<uint32_t> Features;
+ Vector<uint32_t> Features, Cov;
};
struct Merger {
@@ -64,17 +63,24 @@ struct Merger {
bool Parse(std::istream &IS, bool ParseCoverage);
bool Parse(const std::string &Str, bool ParseCoverage);
void ParseOrExit(std::istream &IS, bool ParseCoverage);
- void PrintSummary(std::ostream &OS);
- Set<uint32_t> ParseSummary(std::istream &IS);
- size_t Merge(const Set<uint32_t> &InitialFeatures,
+ 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 Merge(Vector<std::string> *NewFiles) {
- return Merge(Set<uint32_t>{}, 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
diff --git a/lib/fuzzer/FuzzerMutate.cpp b/lib/fuzzer/FuzzerMutate.cpp
index 142b2b0b00143..29541eac5dc60 100644
--- a/lib/fuzzer/FuzzerMutate.cpp
+++ b/lib/fuzzer/FuzzerMutate.cpp
@@ -1,20 +1,19 @@
//===- FuzzerMutate.cpp - Mutate a test input -----------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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 "FuzzerMutate.h"
-#include "FuzzerCorpus.h"
#include "FuzzerDefs.h"
#include "FuzzerExtFunctions.h"
#include "FuzzerIO.h"
+#include "FuzzerMutate.h"
#include "FuzzerOptions.h"
+#include "FuzzerTracePC.h"
namespace fuzzer {
@@ -73,10 +72,10 @@ size_t MutationDispatcher::Mutate_Custom(uint8_t *Data, size_t Size,
size_t MutationDispatcher::Mutate_CustomCrossOver(uint8_t *Data, size_t Size,
size_t MaxSize) {
- if (!Corpus || Corpus->size() < 2 || Size == 0)
+ if (Size == 0)
return 0;
- size_t Idx = Rand(Corpus->size());
- const Unit &Other = (*Corpus)[Idx];
+ if (!CrossOverWith) return 0;
+ const Unit &Other = *CrossOverWith;
if (Other.empty())
return 0;
CustomCrossOverInPlaceHere.resize(MaxSize);
@@ -422,9 +421,9 @@ size_t MutationDispatcher::Mutate_ChangeBinaryInteger(uint8_t *Data,
size_t MutationDispatcher::Mutate_CrossOver(uint8_t *Data, size_t Size,
size_t MaxSize) {
if (Size > MaxSize) return 0;
- if (!Corpus || Corpus->size() < 2 || Size == 0) return 0;
- size_t Idx = Rand(Corpus->size());
- const Unit &O = (*Corpus)[Idx];
+ if (Size == 0) return 0;
+ if (!CrossOverWith) return 0;
+ const Unit &O = *CrossOverWith;
if (O.empty()) return 0;
MutateInPlaceHere.resize(MaxSize);
auto &U = MutateInPlaceHere;
@@ -530,7 +529,7 @@ size_t MutationDispatcher::MutateImpl(uint8_t *Data, size_t Size,
size_t MutationDispatcher::MutateWithMask(uint8_t *Data, size_t Size,
size_t MaxSize,
const Vector<uint8_t> &Mask) {
- assert(Size <= Mask.size());
+ size_t MaskedSize = std::min(Size, Mask.size());
// * Copy the worthy bytes into a temporary array T
// * Mutate T
// * Copy T back.
@@ -539,16 +538,17 @@ size_t MutationDispatcher::MutateWithMask(uint8_t *Data, size_t Size,
if (T.size() < Size)
T.resize(Size);
size_t OneBits = 0;
- for (size_t I = 0; I < Size; I++)
+ 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 < Size; I++)
+ for (size_t I = 0, J = 0; I < MaskedSize; I++)
if (Mask[I])
Data[I] = T[J++];
return Size;
diff --git a/lib/fuzzer/FuzzerMutate.h b/lib/fuzzer/FuzzerMutate.h
index a51c7fb44d468..6cbce80276248 100644
--- a/lib/fuzzer/FuzzerMutate.h
+++ b/lib/fuzzer/FuzzerMutate.h
@@ -1,9 +1,8 @@
//===- FuzzerMutate.h - Internal header for the Fuzzer ----------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -64,7 +63,7 @@ public:
/// 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 some other element of the corpus.
+ /// CrossOver Data with CrossOverWith.
size_t Mutate_CrossOver(uint8_t *Data, size_t Size, size_t MaxSize);
/// Applies one of the configured mutations.
@@ -89,7 +88,7 @@ public:
void PrintRecommendedDictionary();
- void SetCorpus(const InputCorpus *Corpus) { this->Corpus = Corpus; }
+ void SetCrossOverWith(const Unit *U) { CrossOverWith = U; }
Random &GetRand() { return Rand; }
@@ -140,7 +139,7 @@ public:
DictionaryEntry CmpDictionaryEntriesDeque[kCmpDictionaryEntriesDequeSize];
size_t CmpDictionaryEntriesDequeIdx = 0;
- const InputCorpus *Corpus = nullptr;
+ const Unit *CrossOverWith = nullptr;
Vector<uint8_t> MutateInPlaceHere;
Vector<uint8_t> MutateWithMaskTemp;
// CustomCrossOver needs its own buffer as a custom implementation may call
diff --git a/lib/fuzzer/FuzzerOptions.h b/lib/fuzzer/FuzzerOptions.h
index ab90df82a63d1..ad3df015bc778 100644
--- a/lib/fuzzer/FuzzerOptions.h
+++ b/lib/fuzzer/FuzzerOptions.h
@@ -1,8 +1,7 @@
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -20,8 +19,13 @@ struct FuzzingOptions {
size_t MaxLen = 0;
size_t LenControl = 1000;
int UnitTimeoutSec = 300;
- int TimeoutExitCode = 77;
+ 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;
@@ -47,6 +51,9 @@ struct FuzzingOptions {
std::string ExitOnItem;
std::string FocusFunction;
std::string DataFlowTrace;
+ std::string CollectDataFlow;
+ std::string FeaturesDir;
+ std::string StopFile;
bool SaveArtifacts = true;
bool PrintNEW = true; // Print a status line when new units are found;
bool PrintNewCovPcs = false;
@@ -68,6 +75,7 @@ struct FuzzingOptions {
bool HandleXfsz = false;
bool HandleUsr1 = false;
bool HandleUsr2 = false;
+ bool LazyCounters = false;
};
} // namespace fuzzer
diff --git a/lib/fuzzer/FuzzerRandom.h b/lib/fuzzer/FuzzerRandom.h
index 8a1aa3ef5fdc1..659283eee2074 100644
--- a/lib/fuzzer/FuzzerRandom.h
+++ b/lib/fuzzer/FuzzerRandom.h
@@ -1,9 +1,8 @@
//===- FuzzerRandom.h - Internal header for the Fuzzer ----------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -15,12 +14,17 @@
#include <random>
namespace fuzzer {
-class Random : public std::mt19937 {
+class Random : public std::minstd_rand {
public:
- Random(unsigned int seed) : std::mt19937(seed) {}
- result_type operator()() { return this->std::mt19937::operator()(); }
+ 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);
diff --git a/lib/fuzzer/FuzzerSHA1.cpp b/lib/fuzzer/FuzzerSHA1.cpp
index d2f8e811bbf8b..43e5e78cd7877 100644
--- a/lib/fuzzer/FuzzerSHA1.cpp
+++ b/lib/fuzzer/FuzzerSHA1.cpp
@@ -1,9 +1,8 @@
//===- FuzzerSHA1.h - Private copy of the SHA1 implementation ---*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -32,7 +31,8 @@ namespace { // Added for LibFuzzer
#ifdef __BIG_ENDIAN__
# define SHA_BIG_ENDIAN
-#elif defined __LITTLE_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__
diff --git a/lib/fuzzer/FuzzerSHA1.h b/lib/fuzzer/FuzzerSHA1.h
index 3b5e6e807f420..05cbacda87d07 100644
--- a/lib/fuzzer/FuzzerSHA1.h
+++ b/lib/fuzzer/FuzzerSHA1.h
@@ -1,9 +1,8 @@
//===- FuzzerSHA1.h - Internal header for the SHA1 utils --------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerShmem.h b/lib/fuzzer/FuzzerShmem.h
deleted file mode 100644
index 53568e0acb69c..0000000000000
--- a/lib/fuzzer/FuzzerShmem.h
+++ /dev/null
@@ -1,69 +0,0 @@
-//===- FuzzerShmem.h - shared memory interface ------------------*- C++ -* ===//
-//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
-//
-//===----------------------------------------------------------------------===//
-// SharedMemoryRegion
-//===----------------------------------------------------------------------===//
-
-#ifndef LLVM_FUZZER_SHMEM_H
-#define LLVM_FUZZER_SHMEM_H
-
-#include <algorithm>
-#include <cstring>
-#include <string>
-
-#include "FuzzerDefs.h"
-
-namespace fuzzer {
-
-class SharedMemoryRegion {
- public:
- bool Create(const char *Name);
- bool Open(const char *Name);
- bool Destroy(const char *Name);
- uint8_t *GetData() { return Data; }
- void PostServer() {Post(0);}
- void WaitServer() {Wait(0);}
- void PostClient() {Post(1);}
- void WaitClient() {Wait(1);}
-
- size_t WriteByteArray(const uint8_t *Bytes, size_t N) {
- assert(N <= kShmemSize - sizeof(N));
- memcpy(GetData(), &N, sizeof(N));
- memcpy(GetData() + sizeof(N), Bytes, N);
- assert(N == ReadByteArraySize());
- return N;
- }
- size_t ReadByteArraySize() {
- size_t Res;
- memcpy(&Res, GetData(), sizeof(Res));
- return Res;
- }
- uint8_t *GetByteArray() { return GetData() + sizeof(size_t); }
-
- bool IsServer() const { return Data && IAmServer; }
- bool IsClient() const { return Data && !IAmServer; }
-
-private:
-
- static const size_t kShmemSize = 1 << 22;
- bool IAmServer;
- std::string Path(const char *Name);
- std::string SemName(const char *Name, int Idx);
- void Post(int Idx);
- void Wait(int Idx);
-
- bool Map(int fd);
- uint8_t *Data = nullptr;
- void *Semaphore[2];
-};
-
-extern SharedMemoryRegion SMR;
-
-} // namespace fuzzer
-
-#endif // LLVM_FUZZER_SHMEM_H
diff --git a/lib/fuzzer/FuzzerShmemFuchsia.cpp b/lib/fuzzer/FuzzerShmemFuchsia.cpp
deleted file mode 100644
index e9ce50c2ac8a9..0000000000000
--- a/lib/fuzzer/FuzzerShmemFuchsia.cpp
+++ /dev/null
@@ -1,38 +0,0 @@
-//===- FuzzerShmemPosix.cpp - Posix shared memory ---------------*- C++ -* ===//
-//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
-//
-//===----------------------------------------------------------------------===//
-// SharedMemoryRegion. For Fuchsia, this is just stubs as equivalence servers
-// are not currently supported.
-//===----------------------------------------------------------------------===//
-#include "FuzzerDefs.h"
-
-#if LIBFUZZER_FUCHSIA
-
-#include "FuzzerShmem.h"
-
-namespace fuzzer {
-
-bool SharedMemoryRegion::Create(const char *Name) {
- return false;
-}
-
-bool SharedMemoryRegion::Open(const char *Name) {
- return false;
-}
-
-bool SharedMemoryRegion::Destroy(const char *Name) {
- return false;
-}
-
-void SharedMemoryRegion::Post(int Idx) {}
-
-void SharedMemoryRegion::Wait(int Idx) {}
-
-} // namespace fuzzer
-
-#endif // LIBFUZZER_FUCHSIA
diff --git a/lib/fuzzer/FuzzerShmemPosix.cpp b/lib/fuzzer/FuzzerShmemPosix.cpp
deleted file mode 100644
index 41a93f61004b7..0000000000000
--- a/lib/fuzzer/FuzzerShmemPosix.cpp
+++ /dev/null
@@ -1,108 +0,0 @@
-//===- FuzzerShmemPosix.cpp - Posix shared memory ---------------*- C++ -* ===//
-//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
-//
-//===----------------------------------------------------------------------===//
-// SharedMemoryRegion
-//===----------------------------------------------------------------------===//
-#include "FuzzerDefs.h"
-#if LIBFUZZER_POSIX
-
-#include "FuzzerIO.h"
-#include "FuzzerShmem.h"
-
-#include <errno.h>
-#include <fcntl.h>
-#include <semaphore.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <sys/mman.h>
-#include <sys/stat.h>
-#include <sys/types.h>
-#include <unistd.h>
-
-namespace fuzzer {
-
-std::string SharedMemoryRegion::Path(const char *Name) {
- return DirPlusFile(TmpDir(), Name);
-}
-
-std::string SharedMemoryRegion::SemName(const char *Name, int Idx) {
- std::string Res(Name);
- // When passing a name without a leading <slash> character to
- // sem_open, the behaviour is unspecified in POSIX. Add a leading
- // <slash> character for the name if there is no such one.
- if (!Res.empty() && Res[0] != '/')
- Res.insert(Res.begin(), '/');
- return Res + (char)('0' + Idx);
-}
-
-bool SharedMemoryRegion::Map(int fd) {
- Data =
- (uint8_t *)mmap(0, kShmemSize, PROT_WRITE | PROT_READ, MAP_SHARED, fd, 0);
- if (Data == (uint8_t*)-1)
- return false;
- return true;
-}
-
-bool SharedMemoryRegion::Create(const char *Name) {
- int fd = open(Path(Name).c_str(), O_CREAT | O_RDWR, 0777);
- if (fd < 0) return false;
- if (ftruncate(fd, kShmemSize) < 0) return false;
- if (!Map(fd))
- return false;
- for (int i = 0; i < 2; i++) {
- sem_unlink(SemName(Name, i).c_str());
- Semaphore[i] = sem_open(SemName(Name, i).c_str(), O_CREAT, 0644, 0);
- if (Semaphore[i] == SEM_FAILED)
- return false;
- }
- IAmServer = true;
- return true;
-}
-
-bool SharedMemoryRegion::Open(const char *Name) {
- int fd = open(Path(Name).c_str(), O_RDWR);
- if (fd < 0) return false;
- struct stat stat_res;
- if (0 != fstat(fd, &stat_res))
- return false;
- assert(stat_res.st_size == kShmemSize);
- if (!Map(fd))
- return false;
- for (int i = 0; i < 2; i++) {
- Semaphore[i] = sem_open(SemName(Name, i).c_str(), 0);
- if (Semaphore[i] == SEM_FAILED)
- return false;
- }
- IAmServer = false;
- return true;
-}
-
-bool SharedMemoryRegion::Destroy(const char *Name) {
- return 0 == unlink(Path(Name).c_str());
-}
-
-void SharedMemoryRegion::Post(int Idx) {
- assert(Idx == 0 || Idx == 1);
- sem_post((sem_t*)Semaphore[Idx]);
-}
-
-void SharedMemoryRegion::Wait(int Idx) {
- assert(Idx == 0 || Idx == 1);
- for (int i = 0; i < 10 && sem_wait((sem_t*)Semaphore[Idx]); i++) {
- // sem_wait may fail if interrupted by a signal.
- sleep(i);
- if (i)
- Printf("%s: sem_wait[%d] failed %s\n", i < 9 ? "WARNING" : "ERROR", i,
- strerror(errno));
- if (i == 9) abort();
- }
-}
-
-} // namespace fuzzer
-
-#endif // LIBFUZZER_POSIX
diff --git a/lib/fuzzer/FuzzerShmemWindows.cpp b/lib/fuzzer/FuzzerShmemWindows.cpp
deleted file mode 100644
index d330ebf4fd07a..0000000000000
--- a/lib/fuzzer/FuzzerShmemWindows.cpp
+++ /dev/null
@@ -1,64 +0,0 @@
-//===- FuzzerShmemWindows.cpp - Posix shared memory -------------*- C++ -* ===//
-//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
-//
-//===----------------------------------------------------------------------===//
-// SharedMemoryRegion
-//===----------------------------------------------------------------------===//
-#include "FuzzerDefs.h"
-#if LIBFUZZER_WINDOWS
-
-#include "FuzzerIO.h"
-#include "FuzzerShmem.h"
-
-#include <fcntl.h>
-#include <stdio.h>
-#include <sys/stat.h>
-#include <sys/types.h>
-
-namespace fuzzer {
-
-std::string SharedMemoryRegion::Path(const char *Name) {
- return DirPlusFile(TmpDir(), Name);
-}
-
-std::string SharedMemoryRegion::SemName(const char *Name, int Idx) {
- std::string Res(Name);
- return Res + (char)('0' + Idx);
-}
-
-bool SharedMemoryRegion::Map(int fd) {
- assert(0 && "UNIMPLEMENTED");
- return false;
-}
-
-bool SharedMemoryRegion::Create(const char *Name) {
- assert(0 && "UNIMPLEMENTED");
- return false;
-}
-
-bool SharedMemoryRegion::Open(const char *Name) {
- assert(0 && "UNIMPLEMENTED");
- return false;
-}
-
-bool SharedMemoryRegion::Destroy(const char *Name) {
- assert(0 && "UNIMPLEMENTED");
- return false;
-}
-
-void SharedMemoryRegion::Post(int Idx) {
- assert(0 && "UNIMPLEMENTED");
-}
-
-void SharedMemoryRegion::Wait(int Idx) {
- Semaphore[1] = nullptr;
- assert(0 && "UNIMPLEMENTED");
-}
-
-} // namespace fuzzer
-
-#endif // LIBFUZZER_WINDOWS
diff --git a/lib/fuzzer/FuzzerTracePC.cpp b/lib/fuzzer/FuzzerTracePC.cpp
index 80b33105bb22d..4a1308de55044 100644
--- a/lib/fuzzer/FuzzerTracePC.cpp
+++ b/lib/fuzzer/FuzzerTracePC.cpp
@@ -1,9 +1,8 @@
//===- FuzzerTracePC.cpp - PC tracing--------------------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -24,15 +23,6 @@
#include "FuzzerValueBitMap.h"
#include <set>
-// The coverage counters and PCs.
-// These are declared as global variables named "__sancov_*" to simplify
-// experiments with inlined instrumentation.
-alignas(64) ATTRIBUTE_INTERFACE
-uint8_t __sancov_trace_pc_guard_8bit_counters[fuzzer::TracePC::kNumPCs];
-
-ATTRIBUTE_INTERFACE
-uintptr_t __sancov_trace_pc_pcs[fuzzer::TracePC::kNumPCs];
-
// Used by -fsanitize-coverage=stack-depth to track stack depth
ATTRIBUTES_INTERFACE_TLS_INITIAL_EXEC uintptr_t __sancov_lowest_stack;
@@ -40,33 +30,80 @@ namespace fuzzer {
TracePC TPC;
-uint8_t *TracePC::Counters() const {
- return __sancov_trace_pc_guard_8bit_counters;
-}
-
-uintptr_t *TracePC::PCs() const {
- return __sancov_trace_pc_pcs;
-}
-
size_t TracePC::GetTotalPCCoverage() {
- if (ObservedPCs.size())
- return ObservedPCs.size();
- size_t Res = 0;
- for (size_t i = 1, N = GetNumPCs(); i < N; i++)
- if (PCs()[i])
- Res++;
- return Res;
+ return ObservedPCs.size();
}
void TracePC::HandleInline8bitCountersInit(uint8_t *Start, uint8_t *Stop) {
if (Start == Stop) return;
- if (NumModulesWithInline8bitCounters &&
- ModuleCounters[NumModulesWithInline8bitCounters-1].Start == Start) return;
- assert(NumModulesWithInline8bitCounters <
- sizeof(ModuleCounters) / sizeof(ModuleCounters[0]));
- ModuleCounters[NumModulesWithInline8bitCounters++] = {Start, Stop};
- NumInline8bitCounters += Stop - Start;
+ 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();
+}
+
+// Mark all full page counter regions as PROT_NONE and set Enabled=false.
+// The first time the instrumented code hits such a protected/disabled
+// counter region we should catch a SEGV and call UnprotectLazyCounters,
+// which will mark the page as PROT_READ|PROT_WRITE and set Enabled=true.
+//
+// Whenever other functions iterate over the counters they should ignore
+// regions with Enabled=false.
+void TracePC::ProtectLazyCounters() {
+ size_t NumPagesProtected = 0;
+ IterateCounterRegions([&](Module::Region &R) {
+ if (!R.OneFullPage) return;
+ if (Mprotect(R.Start, R.Stop - R.Start, false)) {
+ R.Enabled = false;
+ NumPagesProtected++;
+ }
+ });
+ if (NumPagesProtected)
+ Printf("INFO: %zd pages of counters where protected;"
+ " libFuzzer's SEGV handler must be installed\n",
+ NumPagesProtected);
+}
+
+bool TracePC::UnprotectLazyCounters(void *CounterPtr) {
+ // Printf("UnprotectLazyCounters: %p\n", CounterPtr);
+ if (!CounterPtr)
+ return false;
+ bool Done = false;
+ uint8_t *Addr = reinterpret_cast<uint8_t *>(CounterPtr);
+ IterateCounterRegions([&](Module::Region &R) {
+ if (!R.OneFullPage || R.Enabled || Done) return;
+ if (Addr >= R.Start && Addr < R.Stop)
+ if (Mprotect(R.Start, R.Stop - R.Start, true)) {
+ R.Enabled = true;
+ Done = true;
+ }
+ });
+ return Done;
}
void TracePC::HandlePCsInit(const uintptr_t *Start, const uintptr_t *Stop) {
@@ -78,38 +115,13 @@ void TracePC::HandlePCsInit(const uintptr_t *Start, const uintptr_t *Stop) {
NumPCsInPCTables += E - B;
}
-void TracePC::HandleInit(uint32_t *Start, uint32_t *Stop) {
- if (Start == Stop || *Start) return;
- assert(NumModules < sizeof(Modules) / sizeof(Modules[0]));
- for (uint32_t *P = Start; P < Stop; P++) {
- NumGuards++;
- if (NumGuards == kNumPCs) {
- RawPrint(
- "WARNING: The binary has too many instrumented PCs.\n"
- " You may want to reduce the size of the binary\n"
- " for more efficient fuzzing and precise coverage data\n");
- }
- *P = NumGuards % kNumPCs;
- }
- Modules[NumModules].Start = Start;
- Modules[NumModules].Stop = Stop;
- NumModules++;
-}
-
void TracePC::PrintModuleInfo() {
- if (NumGuards) {
- Printf("INFO: Loaded %zd modules (%zd guards): ", NumModules, NumGuards);
- for (size_t i = 0; i < NumModules; i++)
- Printf("%zd [%p, %p), ", Modules[i].Stop - Modules[i].Start,
- Modules[i].Start, Modules[i].Stop);
- Printf("\n");
- }
- if (NumModulesWithInline8bitCounters) {
+ if (NumModules) {
Printf("INFO: Loaded %zd modules (%zd inline 8-bit counters): ",
- NumModulesWithInline8bitCounters, NumInline8bitCounters);
- for (size_t i = 0; i < NumModulesWithInline8bitCounters; i++)
- Printf("%zd [%p, %p), ", ModuleCounters[i].Stop - ModuleCounters[i].Start,
- ModuleCounters[i].Start, ModuleCounters[i].Stop);
+ 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) {
@@ -121,8 +133,7 @@ void TracePC::PrintModuleInfo() {
}
Printf("\n");
- if ((NumGuards && NumGuards != NumPCsInPCTables) ||
- (NumInline8bitCounters && NumInline8bitCounters != NumPCsInPCTables)) {
+ 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"
@@ -163,7 +174,7 @@ inline ALWAYS_INLINE uintptr_t GetPreviousInstructionPc(uintptr_t PC) {
/// \return the address of the next instruction.
/// Note: the logic is copied from `sanitizer_common/sanitizer_stacktrace.cc`
-inline ALWAYS_INLINE uintptr_t GetNextInstructionPc(uintptr_t PC) {
+ALWAYS_INLINE uintptr_t TracePC::GetNextInstructionPc(uintptr_t PC) {
#if defined(__mips__)
return PC + 8;
#elif defined(__powerpc__) || defined(__sparc__) || defined(__arm__) || \
@@ -176,41 +187,34 @@ inline ALWAYS_INLINE uintptr_t GetNextInstructionPc(uintptr_t PC) {
void TracePC::UpdateObservedPCs() {
Vector<uintptr_t> CoveredFuncs;
- auto ObservePC = [&](uintptr_t PC) {
- if (ObservedPCs.insert(PC).second && DoPrintNewPCs) {
- PrintPC("\tNEW_PC: %p %F %L", "\tNEW_PC: %p", GetNextInstructionPc(PC));
+ 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 (TE.PCFlags & 1)
- if (++ObservedFuncs[TE.PC] == 1 && NumPrintNewFuncs)
- CoveredFuncs.push_back(TE.PC);
- ObservePC(TE.PC);
+ 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 < NumModulesWithInline8bitCounters; i++) {
- uint8_t *Beg = ModuleCounters[i].Start;
- size_t Size = ModuleCounters[i].Stop - Beg;
- assert(Size ==
- (size_t)(ModulePCTable[i].Stop - ModulePCTable[i].Start));
- for (size_t j = 0; j < Size; j++)
- if (Beg[j])
- Observe(ModulePCTable[i].Start[j]);
- }
- } else if (NumGuards == NumPCsInPCTables) {
- size_t GuardIdx = 1;
for (size_t i = 0; i < NumModules; i++) {
- uint32_t *Beg = Modules[i].Start;
- size_t Size = Modules[i].Stop - Beg;
- assert(Size ==
+ auto &M = Modules[i];
+ assert(M.Size() ==
(size_t)(ModulePCTable[i].Stop - ModulePCTable[i].Start));
- for (size_t j = 0; j < Size; j++, GuardIdx++)
- if (Counters()[GuardIdx])
- Observe(ModulePCTable[i].Start[j]);
+ 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)]);
+ }
}
}
}
@@ -223,6 +227,27 @@ void TracePC::UpdateObservedPCs() {
}
}
+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++?
@@ -242,47 +267,38 @@ void TracePC::IterateCoveredFunctions(CallBack CB) {
auto ModuleName = GetModuleName(M.Start->PC);
for (auto NextFE = M.Start; NextFE < M.Stop; ) {
auto FE = NextFE;
- assert((FE->PCFlags & 1) && "Not a function entry point");
+ assert(PcIsFuncEntry(FE) && "Not a function entry point");
do {
NextFE++;
- } while (NextFE < M.Stop && !(NextFE->PCFlags & 1));
- if (ObservedFuncs.count(FE->PC))
- CB(FE, NextFE, ObservedFuncs[FE->PC]);
+ } 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(FocusFunction.first > NumModulesWithInline8bitCounters);
+ assert(!FocusFunctionCounterPtr);
if (FuncName.empty())
return;
- for (size_t M = 0; M < NumModulesWithInline8bitCounters; M++) {
+ 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 (!(PCTE.Start[I].PCFlags & 1)) continue; // not a function entry.
+ 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());
- FocusFunction = {M, I};
+ FocusFunctionCounterPtr = Modules[M].Start() + I;
return;
}
}
}
bool TracePC::ObservedFocusFunction() {
- size_t I = FocusFunction.first;
- size_t J = FocusFunction.second;
- if (I >= NumModulesWithInline8bitCounters)
- return false;
- auto &MC = ModuleCounters[I];
- size_t Size = MC.Stop - MC.Start;
- if (J >= Size)
- return false;
- return MC.Start[J] != 0;
+ return FocusFunctionCounterPtr && *FocusFunctionCounterPtr;
}
void TracePC::PrintCoverage() {
@@ -306,32 +322,24 @@ void TracePC::PrintCoverage() {
if (FunctionStr.find("in ") == 0)
FunctionStr = FunctionStr.substr(3);
std::string LineStr = DescribePC("%l", VisualizePC);
- size_t Line = std::stoul(LineStr);
size_t NumEdges = Last - First;
Vector<uintptr_t> UncoveredPCs;
for (auto TE = First; TE < Last; TE++)
- if (!ObservedPCs.count(TE->PC))
+ if (!ObservedPCs.count(TE))
UncoveredPCs.push_back(TE->PC);
- Printf("COVERED_FUNC: hits: %zd", Counter);
+ Printf("%sCOVERED_FUNC: hits: %zd", Counter ? "" : "UN", Counter);
Printf(" edges: %zd/%zd", NumEdges - UncoveredPCs.size(), NumEdges);
- Printf(" %s %s:%zd\n", FunctionStr.c_str(), FileStr.c_str(), Line);
- for (auto PC: UncoveredPCs)
- Printf(" UNCOVERED_PC: %s\n",
- DescribePC("%s:%l", GetNextInstructionPc(PC)).c_str());
+ 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);
}
-void TracePC::DumpCoverage() {
- if (EF->__sanitizer_dump_coverage) {
- Vector<uintptr_t> PCsCopy(GetNumPCs());
- for (size_t i = 0; i < GetNumPCs(); i++)
- PCsCopy[i] = PCs()[i] ? GetPreviousInstructionPc(PCs()[i]) : 0;
- EF->__sanitizer_dump_coverage(PCsCopy.data(), PCsCopy.size());
- }
-}
-
// Value profile.
// We keep track of various values that affect control flow.
// These values are inserted into a bit-set-based hash map.
@@ -361,11 +369,16 @@ void TracePC::AddValueForMemcmp(void *caller_pc, const void *s1, const void *s2,
Hash ^= (T << 8) | B2[i];
}
size_t I = 0;
- for (; I < Len; I++)
- if (B1[I] != B2[I] || (StopAtZero && B1[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));
}
@@ -400,11 +413,10 @@ static size_t InternalStrnlen2(const char *S1, const char *S2) {
}
void TracePC::ClearInlineCounters() {
- for (size_t i = 0; i < NumModulesWithInline8bitCounters; i++) {
- uint8_t *Beg = ModuleCounters[i].Start;
- size_t Size = ModuleCounters[i].Stop - Beg;
- memset(Beg, 0, Size);
- }
+ IterateCounterRegions([](const Module::Region &R){
+ if (R.Enabled)
+ memset(R.Start, 0, R.Stop - R.Start);
+ });
}
ATTRIBUTE_NO_SANITIZE_ALL
@@ -417,16 +429,25 @@ 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) {
- uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
- uint32_t Idx = *Guard;
- __sancov_trace_pc_pcs[Idx] = PC;
- __sancov_trace_pc_guard_8bit_counters[Idx]++;
+ fuzzer::WarnAboutDeprecatedInstrumentation(
+ "-fsanitize-coverage=trace-pc-guard");
}
// Best-effort support for -fsanitize-coverage=trace-pc, which is available
@@ -434,15 +455,13 @@ void __sanitizer_cov_trace_pc_guard(uint32_t *Guard) {
ATTRIBUTE_INTERFACE
ATTRIBUTE_NO_SANITIZE_ALL
void __sanitizer_cov_trace_pc() {
- uintptr_t PC = reinterpret_cast<uintptr_t>(GET_CALLER_PC());
- uintptr_t Idx = PC & (((uintptr_t)1 << fuzzer::TracePC::kTracePcBits) - 1);
- __sancov_trace_pc_pcs[Idx] = PC;
- __sancov_trace_pc_guard_8bit_counters[Idx]++;
+ fuzzer::WarnAboutDeprecatedInstrumentation("-fsanitize-coverage=trace-pc");
}
ATTRIBUTE_INTERFACE
void __sanitizer_cov_trace_pc_guard_init(uint32_t *Start, uint32_t *Stop) {
- fuzzer::TPC.HandleInit(Start, Stop);
+ fuzzer::WarnAboutDeprecatedInstrumentation(
+ "-fsanitize-coverage=trace-pc-guard");
}
ATTRIBUTE_INTERFACE
@@ -537,24 +556,44 @@ void __sanitizer_cov_trace_switch(uint64_t Val, uint64_t *Cases) {
uint64_t N = Cases[0];
uint64_t ValSizeInBits = Cases[1];
uint64_t *Vals = Cases + 2;
- // Skip the most common and the most boring case.
- if (Vals[N - 1] < 256 && Val < 256)
+ // 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 Token = 0;
+ 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++) {
- Token = Val ^ Vals[i];
- if (Val < Vals[i])
+ if (Val < Vals[i]) {
+ Larger = Vals[i];
break;
+ }
+ if (Val > Vals[i]) Smaller = Vals[i];
}
- if (ValSizeInBits == 16)
- fuzzer::TPC.HandleCmp(PC + i, static_cast<uint16_t>(Token), (uint16_t)(0));
- else if (ValSizeInBits == 32)
- fuzzer::TPC.HandleCmp(PC + i, static_cast<uint32_t>(Token), (uint32_t)(0));
- else
- fuzzer::TPC.HandleCmp(PC + i, Token, (uint64_t)(0));
+ // 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
diff --git a/lib/fuzzer/FuzzerTracePC.h b/lib/fuzzer/FuzzerTracePC.h
index 46d6c24887fda..4f5ebeb047a16 100644
--- a/lib/fuzzer/FuzzerTracePC.h
+++ b/lib/fuzzer/FuzzerTracePC.h
@@ -1,9 +1,8 @@
//===- FuzzerTracePC.h - Internal header for the Fuzzer ---------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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
@@ -70,11 +69,6 @@ struct MemMemTable {
class TracePC {
public:
- static const size_t kNumPCs = 1 << 21;
- // How many bits of PC are used from __sanitizer_cov_trace_pc.
- static const size_t kTracePcBits = 18;
-
- void HandleInit(uint32_t *Start, uint32_t *Stop);
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);
@@ -89,8 +83,6 @@ class TracePC {
void ResetMaps() {
ValueProfileMap.Reset();
- if (NumModules)
- memset(Counters(), 0, GetNumPCs());
ClearExtraCounters();
ClearInlineCounters();
}
@@ -103,7 +95,6 @@ class TracePC {
void PrintModuleInfo();
void PrintCoverage();
- void DumpCoverage();
template<class CallBack>
void IterateCoveredFunctions(CallBack CB);
@@ -116,14 +107,6 @@ class TracePC {
TableOfRecentCompares<Word, 32> TORCW;
MemMemTable<1024> MMT;
- size_t GetNumPCs() const {
- return NumGuards == 0 ? (1 << kTracePcBits) : Min(kNumPCs, NumGuards + 1);
- }
- uintptr_t GetPC(size_t Idx) {
- assert(Idx < GetNumPCs());
- return PCs()[Idx];
- }
-
void RecordInitialStack();
uintptr_t GetMaxStackOffset() const;
@@ -136,39 +119,63 @@ class TracePC {
void SetFocusFunction(const std::string &FuncName);
bool ObservedFocusFunction();
+ void ProtectLazyCounters();
+ bool UnprotectLazyCounters(void *CounterPtr);
+
+ 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 {
- uint32_t *Start, *Stop;
+ 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 NumGuards; // linker-initialized.
-
- struct { uint8_t *Start, *Stop; } ModuleCounters[4096];
- size_t NumModulesWithInline8bitCounters; // linker-initialized.
size_t NumInline8bitCounters;
- struct PCTableEntry {
- uintptr_t PC, PCFlags;
- };
+ 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;
- uint8_t *Counters() const;
- uintptr_t *PCs() const;
-
- Set<uintptr_t> ObservedPCs;
+ Set<const PCTableEntry*> ObservedPCs;
std::unordered_map<uintptr_t, uintptr_t> ObservedFuncs; // PC => Counter.
- std::pair<size_t, size_t> FocusFunction = {-1, -1}; // Module and PC IDs.
+ uint8_t *FocusFunctionCounterPtr = nullptr;
ValueBitMap ValueProfileMap;
uintptr_t InitialStack;
@@ -177,7 +184,7 @@ private:
template <class Callback>
// void Callback(size_t FirstFeature, size_t Idx, uint8_t Value);
ATTRIBUTE_NO_SANITIZE_ALL
-void ForEachNonZeroByte(const uint8_t *Begin, const uint8_t *End,
+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);
@@ -199,6 +206,7 @@ void ForEachNonZeroByte(const uint8_t *Begin, const uint8_t *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].
@@ -229,10 +237,8 @@ unsigned CounterToFeature(T Counter) {
template <class Callback> // void Callback(size_t Feature)
ATTRIBUTE_NO_SANITIZE_ADDRESS
-__attribute__((noinline))
+ATTRIBUTE_NOINLINE
void TracePC::CollectFeatures(Callback HandleFeature) const {
- uint8_t *Counters = this->Counters();
- size_t N = GetNumPCs();
auto Handle8bitCounter = [&](size_t FirstFeature,
size_t Idx, uint8_t Counter) {
if (UseCounters)
@@ -243,22 +249,18 @@ void TracePC::CollectFeatures(Callback HandleFeature) const {
size_t FirstFeature = 0;
- if (!NumInline8bitCounters) {
- ForEachNonZeroByte(Counters, Counters + N, FirstFeature, Handle8bitCounter);
- FirstFeature += N * 8;
- }
-
- if (NumInline8bitCounters) {
- for (size_t i = 0; i < NumModulesWithInline8bitCounters; i++) {
- ForEachNonZeroByte(ModuleCounters[i].Start, ModuleCounters[i].Stop,
- FirstFeature, Handle8bitCounter);
- FirstFeature += 8 * (ModuleCounters[i].Stop - ModuleCounters[i].Start);
+ 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);
}
}
- ForEachNonZeroByte(ExtraCountersBegin(), ExtraCountersEnd(), FirstFeature,
- Handle8bitCounter);
- FirstFeature += (ExtraCountersEnd() - ExtraCountersBegin()) * 8;
+ FirstFeature +=
+ 8 * ForEachNonZeroByte(ExtraCountersBegin(), ExtraCountersEnd(),
+ FirstFeature, Handle8bitCounter);
if (UseValueProfileMask) {
ValueProfileMap.ForEach([&](size_t Idx) {
diff --git a/lib/fuzzer/FuzzerUtil.cpp b/lib/fuzzer/FuzzerUtil.cpp
index 6286f9a718adb..7aa84a1faad72 100644
--- a/lib/fuzzer/FuzzerUtil.cpp
+++ b/lib/fuzzer/FuzzerUtil.cpp
@@ -1,9 +1,8 @@
//===- FuzzerUtil.cpp - Misc utils ----------------------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerUtil.h b/lib/fuzzer/FuzzerUtil.h
index d2f1d5de4267d..0a127911df3cc 100644
--- a/lib/fuzzer/FuzzerUtil.h
+++ b/lib/fuzzer/FuzzerUtil.h
@@ -1,9 +1,8 @@
//===- FuzzerUtil.h - Internal header for the Fuzzer Utils ------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -53,6 +52,8 @@ void SetSignalHandler(const FuzzingOptions& Options);
void SleepSeconds(int Seconds);
+bool Mprotect(void *Ptr, size_t Size, bool AllowReadWrite);
+
unsigned long GetPid();
size_t GetPeakRSSMb();
@@ -88,6 +89,20 @@ 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);
+}
+
} // namespace fuzzer
#endif // LLVM_FUZZER_UTIL_H
diff --git a/lib/fuzzer/FuzzerUtilDarwin.cpp b/lib/fuzzer/FuzzerUtilDarwin.cpp
index 4bfbc11a50cf8..171db23570c45 100644
--- a/lib/fuzzer/FuzzerUtilDarwin.cpp
+++ b/lib/fuzzer/FuzzerUtilDarwin.cpp
@@ -1,9 +1,8 @@
//===- FuzzerUtilDarwin.cpp - Misc utils ----------------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
diff --git a/lib/fuzzer/FuzzerUtilFuchsia.cpp b/lib/fuzzer/FuzzerUtilFuchsia.cpp
index cd48fefef3526..1f04b33c154e4 100644
--- a/lib/fuzzer/FuzzerUtilFuchsia.cpp
+++ b/lib/fuzzer/FuzzerUtilFuchsia.cpp
@@ -1,9 +1,8 @@
//===- FuzzerUtilFuchsia.cpp - Misc utils for Fuchsia. --------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -14,6 +13,7 @@
#include "FuzzerInternal.h"
#include "FuzzerUtil.h"
+#include <cassert>
#include <cerrno>
#include <cinttypes>
#include <cstdint>
@@ -30,7 +30,7 @@
#include <zircon/syscalls.h>
#include <zircon/syscalls/debug.h>
#include <zircon/syscalls/exception.h>
-#include <zircon/syscalls/port.h>
+#include <zircon/syscalls/object.h>
#include <zircon/types.h>
namespace fuzzer {
@@ -49,10 +49,6 @@ void CrashTrampolineAsm() __asm__("CrashTrampolineAsm");
namespace {
-// A magic value for the Zircon exception port, chosen to spell 'FUZZING'
-// when interpreted as a byte sequence on little-endian platforms.
-const uint64_t kFuzzingCrash = 0x474e495a5a5546;
-
// Helper function to handle Zircon syscall failures.
void ExitOnErr(zx_status_t Status, const char *Syscall) {
if (Status != ZX_OK) {
@@ -218,76 +214,101 @@ void CrashHandler(zx_handle_t *Event) {
zx_handle_t Handle = ZX_HANDLE_INVALID;
};
- // Create and bind the exception port. We need to claim to be a "debugger" so
- // the kernel will allow us to modify and resume dying threads (see below).
- // Once the port is set, we can signal the main thread to continue and wait
+ // 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 Port;
- ExitOnErr(_zx_port_create(0, &Port.Handle), "_zx_port_create");
+ ScopedHandle Channel;
zx_handle_t Self = _zx_process_self();
-
- ExitOnErr(_zx_task_bind_exception_port(Self, Port.Handle, kFuzzingCrash,
- ZX_EXCEPTION_PORT_DEBUGGER),
- "_zx_task_bind_exception_port");
+ 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");
- zx_port_packet_t Packet;
- ExitOnErr(_zx_port_wait(Port.Handle, ZX_TIME_INFINITE, &Packet),
- "_zx_port_wait");
-
- // 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_object_get_child(Self, Packet.exception.tid,
- ZX_RIGHT_SAME_RIGHTS, &Thread.Handle),
- "_zx_object_get_child");
-
- 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.
+ // 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 - (128 + sizeof(GeneralRegisters))) &
- -(uintptr_t)16;
- __unsanitized_memcpy(reinterpret_cast<void *>(StackPtr), &GeneralRegisters,
- sizeof(GeneralRegisters));
- GeneralRegisters.rsp = StackPtr;
- GeneralRegisters.rip = reinterpret_cast<zx_vaddr_t>(CrashTrampolineAsm);
+ uintptr_t StackPtr =
+ (GeneralRegisters.rsp - (128 + sizeof(GeneralRegisters))) &
+ -(uintptr_t)16;
+ __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 - sizeof(GeneralRegisters)) & -(uintptr_t)16;
- __unsanitized_memcpy(reinterpret_cast<void *>(StackPtr), &GeneralRegisters,
- sizeof(GeneralRegisters));
- GeneralRegisters.sp = StackPtr;
- GeneralRegisters.pc = reinterpret_cast<zx_vaddr_t>(CrashTrampolineAsm);
+ uintptr_t StackPtr =
+ (GeneralRegisters.sp - sizeof(GeneralRegisters)) & -(uintptr_t)16;
+ __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");
-
- ExitOnErr(_zx_task_resume_from_exception(Thread.Handle, Port.Handle, 0),
- "_zx_task_resume_from_exception");
+ // 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
+bool Mprotect(void *Ptr, size_t Size, bool AllowReadWrite) {
+ return false; // UNIMPLEMENTED
+}
+
// Platform specific functions.
void SetSignalHandler(const FuzzingOptions &Options) {
// Set up alarm handler if needed.
diff --git a/lib/fuzzer/FuzzerUtilLinux.cpp b/lib/fuzzer/FuzzerUtilLinux.cpp
index c103fd230b0c6..d5a15d19f2a95 100644
--- a/lib/fuzzer/FuzzerUtilLinux.cpp
+++ b/lib/fuzzer/FuzzerUtilLinux.cpp
@@ -1,9 +1,8 @@
//===- FuzzerUtilLinux.cpp - Misc utils for Linux. ------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -14,12 +13,18 @@
#include "FuzzerCommand.h"
#include <stdlib.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+
namespace fuzzer {
int ExecuteCommand(const Command &Cmd) {
std::string CmdLine = Cmd.toString();
- return system(CmdLine.c_str());
+ int exit_code = system(CmdLine.c_str());
+ if (WIFEXITED(exit_code))
+ return WEXITSTATUS(exit_code);
+ return exit_code;
}
} // namespace fuzzer
diff --git a/lib/fuzzer/FuzzerUtilPosix.cpp b/lib/fuzzer/FuzzerUtilPosix.cpp
index bc64d3293702f..110785d874137 100644
--- a/lib/fuzzer/FuzzerUtilPosix.cpp
+++ b/lib/fuzzer/FuzzerUtilPosix.cpp
@@ -1,9 +1,8 @@
//===- FuzzerUtilPosix.cpp - Misc utils for Posix. ------------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -12,6 +11,7 @@
#if LIBFUZZER_POSIX
#include "FuzzerIO.h"
#include "FuzzerInternal.h"
+#include "FuzzerTracePC.h"
#include <cassert>
#include <chrono>
#include <cstring>
@@ -19,6 +19,7 @@
#include <iomanip>
#include <signal.h>
#include <stdio.h>
+#include <sys/mman.h>
#include <sys/resource.h>
#include <sys/syscall.h>
#include <sys/time.h>
@@ -32,6 +33,16 @@ 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 (TPC.UnprotectLazyCounters(si->si_addr)) return;
+ if (upstream_segv_handler)
+ return upstream_segv_handler(sig, si, ucontext);
+ Fuzzer::StaticCrashSignalCallback();
+}
+
static void CrashHandler(int, siginfo_t *, void *) {
Fuzzer::StaticCrashSignalCallback();
}
@@ -56,8 +67,11 @@ static void SetSigaction(int signum,
exit(1);
}
if (sigact.sa_flags & SA_SIGINFO) {
- if (sigact.sa_sigaction)
- return;
+ 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)
@@ -65,6 +79,7 @@ static void SetSigaction(int signum,
}
sigact = {};
+ sigact.sa_flags = SA_SIGINFO;
sigact.sa_sigaction = callback;
if (sigaction(signum, &sigact, 0)) {
Printf("libFuzzer: sigaction failed with %d\n", errno);
@@ -83,6 +98,11 @@ void SetTimer(int Seconds) {
SetSigaction(SIGALRM, AlarmHandler);
}
+bool Mprotect(void *Ptr, size_t Size, bool AllowReadWrite) {
+ return 0 == mprotect(Ptr, Size,
+ AllowReadWrite ? (PROT_READ | PROT_WRITE) : PROT_NONE);
+}
+
void SetSignalHandler(const FuzzingOptions& Options) {
if (Options.UnitTimeoutSec > 0)
SetTimer(Options.UnitTimeoutSec / 2 + 1);
@@ -91,7 +111,7 @@ void SetSignalHandler(const FuzzingOptions& Options) {
if (Options.HandleTerm)
SetSigaction(SIGTERM, InterruptHandler);
if (Options.HandleSegv)
- SetSigaction(SIGSEGV, CrashHandler);
+ SetSigaction(SIGSEGV, SegvHandler);
if (Options.HandleBus)
SetSigaction(SIGBUS, CrashHandler);
if (Options.HandleAbrt)
diff --git a/lib/fuzzer/FuzzerUtilWindows.cpp b/lib/fuzzer/FuzzerUtilWindows.cpp
index 393b4768be7ed..074e1eb423094 100644
--- a/lib/fuzzer/FuzzerUtilWindows.cpp
+++ b/lib/fuzzer/FuzzerUtilWindows.cpp
@@ -1,9 +1,8 @@
//===- FuzzerUtilWindows.cpp - Misc utils for Windows. --------------------===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -86,11 +85,11 @@ void CALLBACK AlarmHandler(PVOID, BOOLEAN) {
class TimerQ {
HANDLE TimerQueue;
public:
- TimerQ() : TimerQueue(NULL) {};
+ TimerQ() : TimerQueue(NULL) {}
~TimerQ() {
if (TimerQueue)
DeleteTimerQueueEx(TimerQueue, NULL);
- };
+ }
void SetTimer(int Seconds) {
if (!TimerQueue) {
TimerQueue = CreateTimerQueue();
@@ -105,13 +104,17 @@ class TimerQ {
Printf("libFuzzer: CreateTimerQueueTimer failed.\n");
exit(1);
}
- };
+ }
};
static TimerQ Timer;
static void CrashHandler(int) { Fuzzer::StaticCrashSignalCallback(); }
+bool Mprotect(void *Ptr, size_t Size, bool AllowReadWrite) {
+ return false; // UNIMPLEMENTED
+}
+
void SetSignalHandler(const FuzzingOptions& Options) {
HandlerOpt = &Options;
diff --git a/lib/fuzzer/FuzzerValueBitMap.h b/lib/fuzzer/FuzzerValueBitMap.h
index 13d7cbd95dd73..bc039f1df27f3 100644
--- a/lib/fuzzer/FuzzerValueBitMap.h
+++ b/lib/fuzzer/FuzzerValueBitMap.h
@@ -1,9 +1,8 @@
//===- FuzzerValueBitMap.h - INTERNAL - Bit map -----------------*- C++ -* ===//
//
-// The LLVM Compiler Infrastructure
-//
-// This file is distributed under the University of Illinois Open Source
-// License. See LICENSE.TXT for details.
+// 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.
@@ -35,7 +34,7 @@ struct ValueBitMap {
uintptr_t WordIdx = Idx / kBitsInWord;
uintptr_t BitIdx = Idx % kBitsInWord;
uintptr_t Old = Map[WordIdx];
- uintptr_t New = Old | (1UL << BitIdx);
+ uintptr_t New = Old | (1ULL << BitIdx);
Map[WordIdx] = New;
return New != Old;
}
@@ -49,7 +48,7 @@ struct ValueBitMap {
assert(Idx < kMapSizeInBits);
uintptr_t WordIdx = Idx / kBitsInWord;
uintptr_t BitIdx = Idx % kBitsInWord;
- return Map[WordIdx] & (1UL << BitIdx);
+ return Map[WordIdx] & (1ULL << BitIdx);
}
size_t SizeInBits() const { return kMapSizeInBits; }
@@ -65,7 +64,7 @@ struct ValueBitMap {
}
private:
- uintptr_t Map[kMapSizeInWords] __attribute__((aligned(512)));
+ ATTRIBUTE_ALIGNED(512) uintptr_t Map[kMapSizeInWords];
};
} // namespace fuzzer
diff --git a/lib/fuzzer/utils/FuzzedDataProvider.h b/lib/fuzzer/utils/FuzzedDataProvider.h
new file mode 100644
index 0000000000000..1b5b4bb012696
--- /dev/null
+++ b/lib/fuzzer/utils/FuzzedDataProvider.h
@@ -0,0 +1,245 @@
+//===- FuzzedDataProvider.h - Utility header for fuzz targets ---*- 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
+//
+//===----------------------------------------------------------------------===//
+// A single header library providing an utility class to break up an array of
+// bytes. Whenever run on the same input, provides the same output, as long as
+// its methods are called in the same order, with the same arguments.
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_
+#define LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_
+
+#include <limits.h>
+#include <stddef.h>
+#include <stdint.h>
+
+#include <algorithm>
+#include <cstring>
+#include <initializer_list>
+#include <string>
+#include <type_traits>
+#include <utility>
+#include <vector>
+
+class FuzzedDataProvider {
+public:
+ // |data| is an array of length |size| that the FuzzedDataProvider wraps to
+ // provide more granular access. |data| must outlive the FuzzedDataProvider.
+ FuzzedDataProvider(const uint8_t *data, size_t size)
+ : data_ptr_(data), remaining_bytes_(size) {}
+ ~FuzzedDataProvider() = default;
+
+ // Returns a std::vector containing |num_bytes| of input data. If fewer than
+ // |num_bytes| of data remain, returns a shorter std::vector containing all
+ // of the data that's left. Can be used with any byte sized type, such as
+ // char, unsigned char, uint8_t, etc.
+ template <typename T> std::vector<T> ConsumeBytes(size_t num_bytes) {
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ return ConsumeBytes<T>(num_bytes, num_bytes);
+ }
+
+ // Similar to |ConsumeBytes|, but also appends the terminator value at the end
+ // of the resulting vector. Useful, when a mutable null-terminated C-string is
+ // needed, for example. But that is a rare case. Better avoid it, if possible,
+ // and prefer using |ConsumeBytes| or |ConsumeBytesAsString| methods.
+ template <typename T>
+ std::vector<T> ConsumeBytesWithTerminator(size_t num_bytes,
+ T terminator = 0) {
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ std::vector<T> result = ConsumeBytes<T>(num_bytes + 1, num_bytes);
+ result.back() = terminator;
+ return result;
+ }
+
+ // Returns a std::string containing |num_bytes| of input data. Using this and
+ // |.c_str()| on the resulting string is the best way to get an immutable
+ // null-terminated C string. If fewer than |num_bytes| of data remain, returns
+ // a shorter std::string containing all of the data that's left.
+ std::string ConsumeBytesAsString(size_t num_bytes) {
+ static_assert(sizeof(std::string::value_type) == sizeof(uint8_t),
+ "ConsumeBytesAsString cannot convert the data to a string.");
+
+ num_bytes = std::min(num_bytes, remaining_bytes_);
+ std::string result(
+ reinterpret_cast<const std::string::value_type *>(data_ptr_),
+ num_bytes);
+ Advance(num_bytes);
+ return result;
+ }
+
+ // Returns a number in the range [min, max] by consuming bytes from the
+ // input data. The value might not be uniformly distributed in the given
+ // range. If there's no input data left, always returns |min|. |min| must
+ // be less than or equal to |max|.
+ template <typename T> T ConsumeIntegralInRange(T min, T max) {
+ static_assert(std::is_integral<T>::value, "An integral type is required.");
+ static_assert(sizeof(T) <= sizeof(uint64_t), "Unsupported integral type.");
+
+ if (min > max)
+ abort();
+
+ // Use the biggest type possible to hold the range and the result.
+ uint64_t range = static_cast<uint64_t>(max) - min;
+ uint64_t result = 0;
+ size_t offset = 0;
+
+ while (offset < sizeof(T) * CHAR_BIT && (range >> offset) > 0 &&
+ remaining_bytes_ != 0) {
+ // Pull bytes off the end of the seed data. Experimentally, this seems to
+ // allow the fuzzer to more easily explore the input space. This makes
+ // sense, since it works by modifying inputs that caused new code to run,
+ // and this data is often used to encode length of data read by
+ // |ConsumeBytes|. Separating out read lengths makes it easier modify the
+ // contents of the data that is actually read.
+ --remaining_bytes_;
+ result = (result << CHAR_BIT) | data_ptr_[remaining_bytes_];
+ offset += CHAR_BIT;
+ }
+
+ // Avoid division by 0, in case |range + 1| results in overflow.
+ if (range != std::numeric_limits<decltype(range)>::max())
+ result = result % (range + 1);
+
+ return static_cast<T>(min + result);
+ }
+
+ // Returns a std::string of length from 0 to |max_length|. When it runs out of
+ // input data, returns what remains of the input. Designed to be more stable
+ // with respect to a fuzzer inserting characters than just picking a random
+ // length and then consuming that many bytes with |ConsumeBytes|.
+ std::string ConsumeRandomLengthString(size_t max_length) {
+ // Reads bytes from the start of |data_ptr_|. Maps "\\" to "\", and maps "\"
+ // followed by anything else to the end of the string. As a result of this
+ // logic, a fuzzer can insert characters into the string, and the string
+ // will be lengthened to include those new characters, resulting in a more
+ // stable fuzzer than picking the length of a string independently from
+ // picking its contents.
+ std::string result;
+
+ // Reserve the anticipated capaticity to prevent several reallocations.
+ result.reserve(std::min(max_length, remaining_bytes_));
+ for (size_t i = 0; i < max_length && remaining_bytes_ != 0; ++i) {
+ char next = ConvertUnsignedToSigned<char>(data_ptr_[0]);
+ Advance(1);
+ if (next == '\\' && remaining_bytes_ != 0) {
+ next = ConvertUnsignedToSigned<char>(data_ptr_[0]);
+ Advance(1);
+ if (next != '\\')
+ break;
+ }
+ result += next;
+ }
+
+ result.shrink_to_fit();
+ return result;
+ }
+
+ // Returns a std::vector containing all remaining bytes of the input data.
+ template <typename T> std::vector<T> ConsumeRemainingBytes() {
+ return ConsumeBytes<T>(remaining_bytes_);
+ }
+
+ // Prefer using |ConsumeRemainingBytes| unless you actually need a std::string
+ // object.
+ // Returns a std::vector containing all remaining bytes of the input data.
+ std::string ConsumeRemainingBytesAsString() {
+ return ConsumeBytesAsString(remaining_bytes_);
+ }
+
+ // Returns a number in the range [Type's min, Type's max]. The value might
+ // not be uniformly distributed in the given range. If there's no input data
+ // left, always returns |min|.
+ template <typename T> T ConsumeIntegral() {
+ return ConsumeIntegralInRange(std::numeric_limits<T>::min(),
+ std::numeric_limits<T>::max());
+ }
+
+ // Reads one byte and returns a bool, or false when no data remains.
+ bool ConsumeBool() { return 1 & ConsumeIntegral<uint8_t>(); }
+
+ // Returns a copy of a value selected from a fixed-size |array|.
+ template <typename T, size_t size>
+ T PickValueInArray(const T (&array)[size]) {
+ static_assert(size > 0, "The array must be non empty.");
+ return array[ConsumeIntegralInRange<size_t>(0, size - 1)];
+ }
+
+ template <typename T>
+ T PickValueInArray(std::initializer_list<const T> list) {
+ // static_assert(list.size() > 0, "The array must be non empty.");
+ return *(list.begin() + ConsumeIntegralInRange<size_t>(0, list.size() - 1));
+ }
+
+ // Return an enum value. The enum must start at 0 and be contiguous. It must
+ // also contain |kMaxValue| aliased to its largest (inclusive) value. Such as:
+ // enum class Foo { SomeValue, OtherValue, kMaxValue = OtherValue };
+ template <typename T> T ConsumeEnum() {
+ static_assert(std::is_enum<T>::value, "|T| must be an enum type.");
+ return static_cast<T>(ConsumeIntegralInRange<uint32_t>(
+ 0, static_cast<uint32_t>(T::kMaxValue)));
+ }
+
+ // Reports the remaining bytes available for fuzzed input.
+ size_t remaining_bytes() { return remaining_bytes_; }
+
+private:
+ FuzzedDataProvider(const FuzzedDataProvider &) = delete;
+ FuzzedDataProvider &operator=(const FuzzedDataProvider &) = delete;
+
+ void Advance(size_t num_bytes) {
+ if (num_bytes > remaining_bytes_)
+ abort();
+
+ data_ptr_ += num_bytes;
+ remaining_bytes_ -= num_bytes;
+ }
+
+ template <typename T>
+ std::vector<T> ConsumeBytes(size_t size, size_t num_bytes_to_consume) {
+ static_assert(sizeof(T) == sizeof(uint8_t), "Incompatible data type.");
+
+ // The point of using the size-based constructor below is to increase the
+ // odds of having a vector object with capacity being equal to the length.
+ // That part is always implementation specific, but at least both libc++ and
+ // libstdc++ allocate the requested number of bytes in that constructor,
+ // which seems to be a natural choice for other implementations as well.
+ // To increase the odds even more, we also call |shrink_to_fit| below.
+ std::vector<T> result(size);
+ std::memcpy(result.data(), data_ptr_, num_bytes_to_consume);
+ Advance(num_bytes_to_consume);
+
+ // Even though |shrink_to_fit| is also implementation specific, we expect it
+ // to provide an additional assurance in case vector's constructor allocated
+ // a buffer which is larger than the actual amount of data we put inside it.
+ result.shrink_to_fit();
+ return result;
+ }
+
+ template <typename TS, typename TU> TS ConvertUnsignedToSigned(TU value) {
+ static_assert(sizeof(TS) == sizeof(TU), "Incompatible data types.");
+ static_assert(!std::numeric_limits<TU>::is_signed,
+ "Source type must be unsigned.");
+
+ // TODO(Dor1s): change to `if constexpr` once C++17 becomes mainstream.
+ if (std::numeric_limits<TS>::is_modulo)
+ return static_cast<TS>(value);
+
+ // Avoid using implementation-defined unsigned to signer conversions.
+ // To learn more, see https://stackoverflow.com/questions/13150449.
+ if (value <= std::numeric_limits<TS>::max())
+ return static_cast<TS>(value);
+ else {
+ constexpr auto TS_min = std::numeric_limits<TS>::min();
+ return TS_min + static_cast<char>(value - TS_min);
+ }
+ }
+
+ const uint8_t *data_ptr_;
+ size_t remaining_bytes_;
+};
+
+#endif // LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_