| 1 | //===- BugDriver.cpp - Top-Level BugPoint class implementation ------------===// |
| 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 | // This class contains all of the shared state and information that is used by |
| 10 | // the BugPoint tool to track down errors in optimizations. This class is the |
| 11 | // main driver class that invokes all sub-functionality. |
| 12 | // |
| 13 | //===----------------------------------------------------------------------===// |
| 14 | |
| 15 | #include "BugDriver.h" |
| 16 | #include "ToolRunner.h" |
| 17 | #include "llvm/IR/Module.h" |
| 18 | #include "llvm/IR/Verifier.h" |
| 19 | #include "llvm/IRReader/IRReader.h" |
| 20 | #include "llvm/Linker/Linker.h" |
| 21 | #include "llvm/Pass.h" |
| 22 | #include "llvm/Support/CommandLine.h" |
| 23 | #include "llvm/Support/FileUtilities.h" |
| 24 | #include "llvm/Support/SourceMgr.h" |
| 25 | #include "llvm/Support/raw_ostream.h" |
| 26 | #include "llvm/TargetParser/Host.h" |
| 27 | #include <memory> |
| 28 | using namespace llvm; |
| 29 | |
| 30 | namespace llvm { |
| 31 | Triple TargetTriple; |
| 32 | } |
| 33 | |
| 34 | DiscardTemp::~DiscardTemp() { |
| 35 | if (SaveTemps) { |
| 36 | if (Error E = File.keep()) |
| 37 | errs() << "Failed to keep temp file " << toString(E: std::move(E)) << '\n'; |
| 38 | return; |
| 39 | } |
| 40 | if (Error E = File.discard()) |
| 41 | errs() << "Failed to delete temp file " << toString(E: std::move(E)) << '\n'; |
| 42 | } |
| 43 | |
| 44 | // Anonymous namespace to define command line options for debugging. |
| 45 | // |
| 46 | namespace { |
| 47 | // Output - The user can specify a file containing the expected output of the |
| 48 | // program. If this filename is set, it is used as the reference diff source, |
| 49 | // otherwise the raw input run through an interpreter is used as the reference |
| 50 | // source. |
| 51 | // |
| 52 | cl::opt<std::string> OutputFile("output" , |
| 53 | cl::desc("Specify a reference program output " |
| 54 | "(for miscompilation detection)" )); |
| 55 | } |
| 56 | |
| 57 | /// If we reduce or update the program somehow, call this method to update |
| 58 | /// bugdriver with it. This deletes the old module and sets the specified one |
| 59 | /// as the current program. |
| 60 | void BugDriver::setNewProgram(std::unique_ptr<Module> M) { |
| 61 | Program = std::move(M); |
| 62 | } |
| 63 | |
| 64 | /// getPassesString - Turn a list of passes into a string which indicates the |
| 65 | /// command line options that must be passed to add the passes. |
| 66 | /// |
| 67 | std::string llvm::getPassesString(const std::vector<std::string> &Passes) { |
| 68 | std::string Result; |
| 69 | for (unsigned i = 0, e = Passes.size(); i != e; ++i) { |
| 70 | if (i) |
| 71 | Result += " " ; |
| 72 | Result += "-" ; |
| 73 | Result += Passes[i]; |
| 74 | } |
| 75 | return Result; |
| 76 | } |
| 77 | |
| 78 | BugDriver::BugDriver(const char *toolname, bool find_bugs, unsigned timeout, |
| 79 | unsigned memlimit, bool use_valgrind, LLVMContext &ctxt) |
| 80 | : Context(ctxt), ToolName(toolname), ReferenceOutputFile(OutputFile), |
| 81 | Program(nullptr), Interpreter(nullptr), SafeInterpreter(nullptr), |
| 82 | cc(nullptr), run_find_bugs(find_bugs), Timeout(timeout), |
| 83 | MemoryLimit(memlimit), UseValgrind(use_valgrind) {} |
| 84 | |
| 85 | BugDriver::~BugDriver() { |
| 86 | if (Interpreter != SafeInterpreter) |
| 87 | delete Interpreter; |
| 88 | delete SafeInterpreter; |
| 89 | delete cc; |
| 90 | } |
| 91 | |
| 92 | std::unique_ptr<Module> llvm::parseInputFile(StringRef Filename, |
| 93 | LLVMContext &Ctxt) { |
| 94 | SMDiagnostic Err; |
| 95 | std::unique_ptr<Module> Result = parseIRFile(Filename, Err, Context&: Ctxt); |
| 96 | if (!Result) { |
| 97 | Err.print(ProgName: "bugpoint" , S&: errs()); |
| 98 | return Result; |
| 99 | } |
| 100 | |
| 101 | if (verifyModule(M: *Result, OS: &errs())) { |
| 102 | errs() << "bugpoint: " << Filename << ": error: input module is broken!\n" ; |
| 103 | return std::unique_ptr<Module>(); |
| 104 | } |
| 105 | |
| 106 | // If we don't have an override triple, use the first one to configure |
| 107 | // bugpoint, or use the host triple if none provided. |
| 108 | if (TargetTriple.getTriple().empty()) { |
| 109 | Triple TheTriple(Result->getTargetTriple()); |
| 110 | |
| 111 | if (TheTriple.getTriple().empty()) |
| 112 | TheTriple.setTriple(sys::getDefaultTargetTriple()); |
| 113 | |
| 114 | TargetTriple.setTriple(TheTriple.getTriple()); |
| 115 | } |
| 116 | |
| 117 | // override the triple |
| 118 | Result->setTargetTriple(TargetTriple); |
| 119 | return Result; |
| 120 | } |
| 121 | |
| 122 | std::unique_ptr<Module> BugDriver::swapProgramIn(std::unique_ptr<Module> M) { |
| 123 | std::unique_ptr<Module> OldProgram = std::move(Program); |
| 124 | Program = std::move(M); |
| 125 | return OldProgram; |
| 126 | } |
| 127 | |
| 128 | // This method takes the specified list of LLVM input files, attempts to load |
| 129 | // them, either as assembly or bitcode, then link them together. It returns |
| 130 | // true on failure (if, for example, an input bitcode file could not be |
| 131 | // parsed), and false on success. |
| 132 | // |
| 133 | bool BugDriver::addSources(const std::vector<std::string> &Filenames) { |
| 134 | assert(!Program && "Cannot call addSources multiple times!" ); |
| 135 | assert(!Filenames.empty() && "Must specify at least on input filename!" ); |
| 136 | |
| 137 | // Load the first input file. |
| 138 | Program = parseInputFile(Filename: Filenames[0], Ctxt&: Context); |
| 139 | if (!Program) |
| 140 | return true; |
| 141 | |
| 142 | outs() << "Read input file : '" << Filenames[0] << "'\n" ; |
| 143 | |
| 144 | for (unsigned i = 1, e = Filenames.size(); i != e; ++i) { |
| 145 | std::unique_ptr<Module> M = parseInputFile(Filename: Filenames[i], Ctxt&: Context); |
| 146 | if (!M) |
| 147 | return true; |
| 148 | |
| 149 | outs() << "Linking in input file: '" << Filenames[i] << "'\n" ; |
| 150 | if (Linker::linkModules(Dest&: *Program, Src: std::move(M))) |
| 151 | return true; |
| 152 | } |
| 153 | |
| 154 | outs() << "*** All input ok\n" ; |
| 155 | |
| 156 | // All input files read successfully! |
| 157 | return false; |
| 158 | } |
| 159 | |
| 160 | /// run - The top level method that is invoked after all of the instance |
| 161 | /// variables are set up from command line arguments. |
| 162 | /// |
| 163 | Error BugDriver::run() { |
| 164 | if (run_find_bugs) { |
| 165 | // Rearrange the passes and apply them to the program. Repeat this process |
| 166 | // until the user kills the program or we find a bug. |
| 167 | return runManyPasses(AllPasses: PassesToRun); |
| 168 | } |
| 169 | |
| 170 | // If we're not running as a child, the first thing that we must do is |
| 171 | // determine what the problem is. Does the optimization series crash the |
| 172 | // compiler, or does it produce illegal code? We make the top-level |
| 173 | // decision by trying to run all of the passes on the input program, |
| 174 | // which should generate a bitcode file. If it does generate a bitcode |
| 175 | // file, then we know the compiler didn't crash, so try to diagnose a |
| 176 | // miscompilation. |
| 177 | if (!PassesToRun.empty()) { |
| 178 | outs() << "Running selected passes on program to test for crash: " ; |
| 179 | if (runPasses(M&: *Program, PassesToRun)) |
| 180 | return debugOptimizerCrash(); |
| 181 | } |
| 182 | |
| 183 | // Set up the execution environment, selecting a method to run LLVM bitcode. |
| 184 | if (Error E = initializeExecutionEnvironment()) |
| 185 | return E; |
| 186 | |
| 187 | // Test to see if we have a code generator crash. |
| 188 | outs() << "Running the code generator to test for a crash: " ; |
| 189 | if (Error E = compileProgram(M&: *Program)) { |
| 190 | outs() << toString(E: std::move(E)); |
| 191 | return debugCodeGeneratorCrash(); |
| 192 | } |
| 193 | outs() << '\n'; |
| 194 | |
| 195 | // Run the raw input to see where we are coming from. If a reference output |
| 196 | // was specified, make sure that the raw output matches it. If not, it's a |
| 197 | // problem in the front-end or the code generator. |
| 198 | // |
| 199 | bool CreatedOutput = false; |
| 200 | if (ReferenceOutputFile.empty()) { |
| 201 | outs() << "Generating reference output from raw program: " ; |
| 202 | if (Error E = createReferenceFile(M&: *Program)) { |
| 203 | errs() << toString(E: std::move(E)); |
| 204 | return debugCodeGeneratorCrash(); |
| 205 | } |
| 206 | CreatedOutput = true; |
| 207 | } |
| 208 | |
| 209 | // Make sure the reference output file gets deleted on exit from this |
| 210 | // function, if appropriate. |
| 211 | std::string ROF(ReferenceOutputFile); |
| 212 | FileRemover RemoverInstance(ROF, CreatedOutput && !SaveTemps); |
| 213 | |
| 214 | // Diff the output of the raw program against the reference output. If it |
| 215 | // matches, then we assume there is a miscompilation bug and try to |
| 216 | // diagnose it. |
| 217 | outs() << "*** Checking the code generator...\n" ; |
| 218 | Expected<bool> Diff = diffProgram(Program: *Program, BitcodeFile: "" , SharedObj: "" , RemoveBitcode: false); |
| 219 | if (Error E = Diff.takeError()) { |
| 220 | errs() << toString(E: std::move(E)); |
| 221 | return debugCodeGeneratorCrash(); |
| 222 | } |
| 223 | if (!*Diff) { |
| 224 | outs() << "\n*** Output matches: Debugging miscompilation!\n" ; |
| 225 | if (Error E = debugMiscompilation()) { |
| 226 | errs() << toString(E: std::move(E)); |
| 227 | return debugCodeGeneratorCrash(); |
| 228 | } |
| 229 | return Error::success(); |
| 230 | } |
| 231 | |
| 232 | outs() << "\n*** Input program does not match reference diff!\n" ; |
| 233 | outs() << "Debugging code generator problem!\n" ; |
| 234 | if (Error E = debugCodeGenerator()) { |
| 235 | errs() << toString(E: std::move(E)); |
| 236 | return debugCodeGeneratorCrash(); |
| 237 | } |
| 238 | return Error::success(); |
| 239 | } |
| 240 | |
| 241 | void llvm::PrintFunctionList(const std::vector<Function *> &Funcs) { |
| 242 | unsigned NumPrint = Funcs.size(); |
| 243 | if (NumPrint > 10) |
| 244 | NumPrint = 10; |
| 245 | for (unsigned i = 0; i != NumPrint; ++i) |
| 246 | outs() << " " << Funcs[i]->getName(); |
| 247 | if (NumPrint < Funcs.size()) |
| 248 | outs() << "... <" << Funcs.size() << " total>" ; |
| 249 | outs().flush(); |
| 250 | } |
| 251 | |
| 252 | void llvm::PrintGlobalVariableList(const std::vector<GlobalVariable *> &GVs) { |
| 253 | unsigned NumPrint = GVs.size(); |
| 254 | if (NumPrint > 10) |
| 255 | NumPrint = 10; |
| 256 | for (unsigned i = 0; i != NumPrint; ++i) |
| 257 | outs() << " " << GVs[i]->getName(); |
| 258 | if (NumPrint < GVs.size()) |
| 259 | outs() << "... <" << GVs.size() << " total>" ; |
| 260 | outs().flush(); |
| 261 | } |
| 262 | |