tools/bugpoint/BugDriver.cpp

   1 //===- BugDriver.cpp - Top-Level BugPoint class implementation ------------===//
   2 //
   3 //                     The LLVM Compiler Infrastructure
   4 //
   5 // This file is distributed under the University of Illinois Open Source
   6 // License. See LICENSE.TXT for details.
   7 //
   8 //===----------------------------------------------------------------------===//
   9 //
  10 // This class contains all of the shared state and information that is used by
  11 // the BugPoint tool to track down errors in optimizations.  This class is the
  12 // main driver class that invokes all sub-functionality.
  13 //
  14 //===----------------------------------------------------------------------===//
  15
  16 #include "BugDriver.h"
  17 #include "ToolRunner.h"
  18 #include "llvm/IR/Module.h"
  19 #include "llvm/IR/Verifier.h"
  20 #include "llvm/IRReader/IRReader.h"
  21 #include "llvm/Linker/Linker.h"
  22 #include "llvm/Pass.h"
  23 #include "llvm/Support/CommandLine.h"
  24 #include "llvm/Support/FileUtilities.h"
  25 #include "llvm/Support/Host.h"
  26 #include "llvm/Support/SourceMgr.h"
  27 #include "llvm/Support/raw_ostream.h"
  28 #include <memory>
  29 using namespace llvm;
  30
  31 namespace llvm {
  32 Triple TargetTriple;
  33 }
  34
  35 // Anonymous namespace to define command line options for debugging.
  36 //
  37 namespace {
  38 // Output - The user can specify a file containing the expected output of the
  39 // program.  If this filename is set, it is used as the reference diff source,
  40 // otherwise the raw input run through an interpreter is used as the reference
  41 // source.
  42 //
  43 cl::opt<std::string> OutputFile("output",
  44                                 cl::desc("Specify a reference program output "
  45                                          "(for miscompilation detection)"));
  46 }
  47
  48 /// setNewProgram - If we reduce or update the program somehow, call this method
  49 /// to update bugdriver with it.  This deletes the old module and sets the
  50 /// specified one as the current program.
  51 void BugDriver::setNewProgram(Module *M) {
  52   delete Program;
  53   Program = M;
  54 }
  55
  56 /// getPassesString - Turn a list of passes into a string which indicates the
  57 /// command line options that must be passed to add the passes.
  58 ///
  59 std::string llvm::getPassesString(const std::vector<std::string> &Passes) {
  60   std::string Result;
  61   for (unsigned i = 0, e = Passes.size(); i != e; ++i) {
  62     if (i)
  63       Result += " ";
  64     Result += "-";
  65     Result += Passes[i];
  66   }
  67   return Result;
  68 }
  69
  70 BugDriver::BugDriver(const char *toolname, bool find_bugs, unsigned timeout,
  71                      unsigned memlimit, bool use_valgrind, LLVMContext &ctxt)
  72     : Context(ctxt), ToolName(toolname), ReferenceOutputFile(OutputFile),
  73       Program(nullptr), Interpreter(nullptr), SafeInterpreter(nullptr),
  74       cc(nullptr), run_find_bugs(find_bugs), Timeout(timeout),
  75       MemoryLimit(memlimit), UseValgrind(use_valgrind) {}
  76
  77 BugDriver::~BugDriver() {
  78   delete Program;
  79   if (Interpreter != SafeInterpreter)
  80     delete Interpreter;
  81   delete SafeInterpreter;
  82   delete cc;
  83 }
  84
  85 std::unique_ptr<Module> llvm::parseInputFile(StringRef Filename,
  86                                              LLVMContext &Ctxt) {
  87   SMDiagnostic Err;
  88   std::unique_ptr<Module> Result = parseIRFile(Filename, Err, Ctxt);
  89   if (!Result) {
  90     Err.print("bugpoint", errs());
  91     return Result;
  92   }
  93
  94   if (verifyModule(*Result, &errs())) {
  95     errs() << "bugpoint: " << Filename << ": error: input module is broken!\n";
  96     return std::unique_ptr<Module>();
  97   }
  98
  99   // If we don't have an override triple, use the first one to configure
 100   // bugpoint, or use the host triple if none provided.
 101   if (TargetTriple.getTriple().empty()) {
 102     Triple TheTriple(Result->getTargetTriple());
 103
 104     if (TheTriple.getTriple().empty())
 105       TheTriple.setTriple(sys::getDefaultTargetTriple());
 106
 107     TargetTriple.setTriple(TheTriple.getTriple());
 108   }
 109
 110   Result->setTargetTriple(TargetTriple.getTriple()); // override the triple
 111   return Result;
 112 }
 113
 114 // This method takes the specified list of LLVM input files, attempts to load
 115 // them, either as assembly or bitcode, then link them together. It returns
 116 // true on failure (if, for example, an input bitcode file could not be
 117 // parsed), and false on success.
 118 //
 119 bool BugDriver::addSources(const std::vector<std::string> &Filenames) {
 120   assert(!Program && "Cannot call addSources multiple times!");
 121   assert(!Filenames.empty() && "Must specify at least on input filename!");
 122
 123   // Load the first input file.
 124   Program = parseInputFile(Filenames[0], Context).release();
 125   if (!Program)
 126     return true;
 127
 128   outs() << "Read input file      : '" << Filenames[0] << "'\n";
 129
 130   for (unsigned i = 1, e = Filenames.size(); i != e; ++i) {
 131     std::unique_ptr<Module> M = parseInputFile(Filenames[i], Context);
 132     if (!M.get())
 133       return true;
 134
 135     outs() << "Linking in input file: '" << Filenames[i] << "'\n";
 136     if (Linker::linkModules(*Program, std::move(M)))
 137       return true;
 138   }
 139
 140   outs() << "*** All input ok\n";
 141
 142   // All input files read successfully!
 143   return false;
 144 }
 145
 146 /// run - The top level method that is invoked after all of the instance
 147 /// variables are set up from command line arguments.
 148 ///
 149 Error BugDriver::run() {
 150   if (run_find_bugs) {
 151     // Rearrange the passes and apply them to the program. Repeat this process
 152     // until the user kills the program or we find a bug.
 153     return runManyPasses(PassesToRun);
 154   }
 155
 156   // If we're not running as a child, the first thing that we must do is
 157   // determine what the problem is. Does the optimization series crash the
 158   // compiler, or does it produce illegal code?  We make the top-level
 159   // decision by trying to run all of the passes on the input program,
 160   // which should generate a bitcode file.  If it does generate a bitcode
 161   // file, then we know the compiler didn't crash, so try to diagnose a
 162   // miscompilation.
 163   if (!PassesToRun.empty()) {
 164     outs() << "Running selected passes on program to test for crash: ";
 165     if (runPasses(Program, PassesToRun))
 166       return debugOptimizerCrash();
 167   }
 168
 169   // Set up the execution environment, selecting a method to run LLVM bitcode.
 170   if (Error E = initializeExecutionEnvironment())
 171     return E;
 172
 173   // Test to see if we have a code generator crash.
 174   outs() << "Running the code generator to test for a crash: ";
 175   if (Error E = compileProgram(Program)) {
 176     outs() << toString(std::move(E));
 177     return debugCodeGeneratorCrash();
 178   }
 179   outs() << '\n';
 180
 181   // Run the raw input to see where we are coming from.  If a reference output
 182   // was specified, make sure that the raw output matches it.  If not, it's a
 183   // problem in the front-end or the code generator.
 184   //
 185   bool CreatedOutput = false;
 186   if (ReferenceOutputFile.empty()) {
 187     outs() << "Generating reference output from raw program: ";
 188     if (Error E = createReferenceFile(Program)) {
 189       errs() << toString(std::move(E));
 190       return debugCodeGeneratorCrash();
 191     }
 192     CreatedOutput = true;
 193   }
 194
 195   // Make sure the reference output file gets deleted on exit from this
 196   // function, if appropriate.
 197   std::string ROF(ReferenceOutputFile);
 198   FileRemover RemoverInstance(ROF, CreatedOutput && !SaveTemps);
 199
 200   // Diff the output of the raw program against the reference output.  If it
 201   // matches, then we assume there is a miscompilation bug and try to
 202   // diagnose it.
 203   outs() << "*** Checking the code generator...\n";
 204   Expected<bool> Diff = diffProgram(Program, "", "", false);
 205   if (Error E = Diff.takeError()) {
 206     errs() << toString(std::move(E));
 207     return debugCodeGeneratorCrash();
 208   }
 209   if (!*Diff) {
 210     outs() << "\n*** Output matches: Debugging miscompilation!\n";
 211     if (Error E = debugMiscompilation()) {
 212       errs() << toString(std::move(E));
 213       return debugCodeGeneratorCrash();
 214     }
 215     return Error::success();
 216   }
 217
 218   outs() << "\n*** Input program does not match reference diff!\n";
 219   outs() << "Debugging code generator problem!\n";
 220   if (Error E = debugCodeGenerator()) {
 221     errs() << toString(std::move(E));
 222     return debugCodeGeneratorCrash();
 223   }
 224   return Error::success();
 225 }
 226
 227 void llvm::PrintFunctionList(const std::vector<Function *> &Funcs) {
 228   unsigned NumPrint = Funcs.size();
 229   if (NumPrint > 10)
 230     NumPrint = 10;
 231   for (unsigned i = 0; i != NumPrint; ++i)
 232     outs() << " " << Funcs[i]->getName();
 233   if (NumPrint < Funcs.size())
 234     outs() << "... <" << Funcs.size() << " total>";
 235   outs().flush();
 236 }
 237
 238 void llvm::PrintGlobalVariableList(const std::vector<GlobalVariable *> &GVs) {
 239   unsigned NumPrint = GVs.size();
 240   if (NumPrint > 10)
 241     NumPrint = 10;
 242   for (unsigned i = 0; i != NumPrint; ++i)
 243     outs() << " " << GVs[i]->getName();
 244   if (NumPrint < GVs.size())
 245     outs() << "... <" << GVs.size() << " total>";
 246   outs().flush();
 247 }