tools/llvm-mc-disassemble-fuzzer/llvm-mc-disassemble-fuzzer.cpp

   1 //===-- llvm-mc-disassemble-fuzzer.cpp - Fuzzer for the MC layer ----------===//
   2 //
   3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   4 // See https://llvm.org/LICENSE.txt for license information.
   5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   6 //
   7 //===----------------------------------------------------------------------===//
   8 //
   9 //===----------------------------------------------------------------------===//
  10
  11 #include "llvm-c/Disassembler.h"
  12 #include "llvm-c/Target.h"
  13 #include "llvm/MC/SubtargetFeature.h"
  14 #include "llvm/Support/CommandLine.h"
  15 #include "llvm/Support/raw_ostream.h"
  16
  17 using namespace llvm;
  18
  19 const unsigned AssemblyTextBufSize = 80;
  20
  21 static cl::opt<std::string>
  22     TripleName("triple", cl::desc("Target triple to assemble for, "
  23                                   "see -version for available targets"));
  24
  25 static cl::opt<std::string>
  26     MCPU("mcpu",
  27          cl::desc("Target a specific cpu type (-mcpu=help for details)"),
  28          cl::value_desc("cpu-name"), cl::init(""));
  29
  30 // This is useful for variable-length instruction sets.
  31 static cl::opt<unsigned> InsnLimit(
  32     "insn-limit",
  33     cl::desc("Limit the number of instructions to process (0 for no limit)"),
  34     cl::value_desc("count"), cl::init(0));
  35
  36 static cl::list<std::string>
  37     MAttrs("mattr", cl::CommaSeparated,
  38            cl::desc("Target specific attributes (-mattr=help for details)"),
  39            cl::value_desc("a1,+a2,-a3,..."));
  40 // The feature string derived from -mattr's values.
  41 std::string FeaturesStr;
  42
  43 static cl::list<std::string>
  44     FuzzerArgs("fuzzer-args", cl::Positional,
  45                cl::desc("Options to pass to the fuzzer"), cl::ZeroOrMore,
  46                cl::PositionalEatsArgs);
  47 static std::vector<char *> ModifiedArgv;
  48
  49 int DisassembleOneInput(const uint8_t *Data, size_t Size) {
  50   char AssemblyText[AssemblyTextBufSize];
  51
  52   std::vector<uint8_t> DataCopy(Data, Data + Size);
  53
  54   LLVMDisasmContextRef Ctx = LLVMCreateDisasmCPUFeatures(
  55       TripleName.c_str(), MCPU.c_str(), FeaturesStr.c_str(), nullptr, 0,
  56       nullptr, nullptr);
  57   assert(Ctx);
  58   uint8_t *p = DataCopy.data();
  59   unsigned Consumed;
  60   unsigned InstructionsProcessed = 0;
  61   do {
  62     Consumed = LLVMDisasmInstruction(Ctx, p, Size, 0, AssemblyText,
  63                                      AssemblyTextBufSize);
  64     Size -= Consumed;
  65     p += Consumed;
  66
  67     InstructionsProcessed ++;
  68     if (InsnLimit != 0 && InstructionsProcessed < InsnLimit)
  69       break;
  70   } while (Consumed != 0);
  71   LLVMDisasmDispose(Ctx);
  72   return 0;
  73 }
  74
  75 extern "C" int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size) {
  76   return DisassembleOneInput(Data, Size);
  77 }
  78
  79 extern "C" LLVM_ATTRIBUTE_USED int LLVMFuzzerInitialize(int *argc,
  80                                                         char ***argv) {
  81   // The command line is unusual compared to other fuzzers due to the need to
  82   // specify the target. Options like -triple, -mcpu, and -mattr work like
  83   // their counterparts in llvm-mc, while -fuzzer-args collects options for the
  84   // fuzzer itself.
  85   //
  86   // Examples:
  87   //
  88   // Fuzz the big-endian MIPS32R6 disassembler using 100,000 inputs of up to
  89   // 4-bytes each and use the contents of ./corpus as the test corpus:
  90   //   llvm-mc-fuzzer -triple mips-linux-gnu -mcpu=mips32r6 -disassemble \
  91   //       -fuzzer-args -max_len=4 -runs=100000 ./corpus
  92   //
  93   // Infinitely fuzz the little-endian MIPS64R2 disassembler with the MSA
  94   // feature enabled using up to 64-byte inputs:
  95   //   llvm-mc-fuzzer -triple mipsel-linux-gnu -mcpu=mips64r2 -mattr=msa \
  96   //       -disassemble -fuzzer-args ./corpus
  97   //
  98   // If your aim is to find instructions that are not tested, then it is
  99   // advisable to constrain the maximum input size to a single instruction
 100   // using -max_len as in the first example. This results in a test corpus of
 101   // individual instructions that test unique paths. Without this constraint,
 102   // there will be considerable redundancy in the corpus.
 103
 104   char **OriginalArgv = *argv;
 105
 106   LLVMInitializeAllTargetInfos();
 107   LLVMInitializeAllTargetMCs();
 108   LLVMInitializeAllDisassemblers();
 109
 110   cl::ParseCommandLineOptions(*argc, OriginalArgv);
 111
 112   // Rebuild the argv without the arguments llvm-mc-fuzzer consumed so that
 113   // the driver can parse its arguments.
 114   //
 115   // FuzzerArgs cannot provide the non-const pointer that OriginalArgv needs.
 116   // Re-use the strings from OriginalArgv instead of copying FuzzerArg to a
 117   // non-const buffer to avoid the need to clean up when the fuzzer terminates.
 118   ModifiedArgv.push_back(OriginalArgv[0]);
 119   for (const auto &FuzzerArg : FuzzerArgs) {
 120     for (int i = 1; i < *argc; ++i) {
 121       if (FuzzerArg == OriginalArgv[i])
 122         ModifiedArgv.push_back(OriginalArgv[i]);
 123     }
 124   }
 125   *argc = ModifiedArgv.size();
 126   *argv = ModifiedArgv.data();
 127
 128   // Package up features to be passed to target/subtarget
 129   // We have to pass it via a global since the callback doesn't
 130   // permit any user data.
 131   if (MAttrs.size()) {
 132     SubtargetFeatures Features;
 133     for (unsigned i = 0; i != MAttrs.size(); ++i)
 134       Features.AddFeature(MAttrs[i]);
 135     FeaturesStr = Features.getString();
 136   }
 137
 138   if (TripleName.empty())
 139     TripleName = sys::getDefaultTargetTriple();
 140
 141   return 0;
 142 }