path: root/tools/gold/gold-plugin.cpp

//===-- gold-plugin.cpp - Plugin to gold for Link Time Optimization  ------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This is a gold plugin for LLVM. It provides an LLVM implementation of the
// interface described in http://gcc.gnu.org/wiki/whopr/driver .
//
//===----------------------------------------------------------------------===//

#include "llvm/Config/config.h" // plugin-api.h requires HAVE_STDINT_H
#include "plugin-api.h"
#include "llvm-c/lto.h"
#include "llvm/ADT/OwningPtr.h"
#include "llvm/Support/Errno.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/Program.h"
#include "llvm/Support/ToolOutputFile.h"
#include "llvm/Support/system_error.h"
#include <cerrno>
#include <cstdlib>
#include <cstring>
#include <fstream>
#include <list>
#include <vector>

// Support Windows/MinGW crazyness.
#ifdef _WIN32
# include <io.h>
# define lseek _lseek
# define read _read
#endif

using namespace llvm;

namespace {
  ld_plugin_status discard_message(int level, const char *format, ...) {
    // Die loudly. Recent versions of Gold pass ld_plugin_message as the first
    // callback in the transfer vector. This should never be called.
    abort();
  }

  ld_plugin_add_symbols add_symbols = NULL;
  ld_plugin_get_symbols get_symbols = NULL;
  ld_plugin_add_input_file add_input_file = NULL;
  ld_plugin_add_input_library add_input_library = NULL;
  ld_plugin_set_extra_library_path set_extra_library_path = NULL;
  ld_plugin_get_view get_view = NULL;
  ld_plugin_message message = discard_message;
  // @LOCALMOD-BEGIN
  // REL, DYN, or EXEC
  ld_plugin_output_file_type linker_output;

  // Callback for getting link soname from gold
  ld_plugin_get_output_soname get_output_soname = NULL;

  // Callback for getting needed libraries from gold
  ld_plugin_get_needed get_needed = NULL;

  // Callback for getting number of needed library from gold
  ld_plugin_get_num_needed get_num_needed = NULL;

  // Callback for getting the number of --wrap'd symbols.
  ld_plugin_get_num_wrapped get_num_wrapped = NULL;

  // Callback for getting the name of a wrapped symbol.
  ld_plugin_get_wrapped get_wrapped = NULL;
  // @LOCALMOD-END

  int api_version = 0;
  int gold_version = 0;

  struct claimed_file {
    void *handle;
    std::vector<ld_plugin_symbol> syms;
    bool is_linked_in; // @LOCALMOD
  };

  lto_codegen_model output_type = LTO_CODEGEN_PIC_MODEL_STATIC;
  std::string output_name = "";
  std::list<claimed_file> Modules;

  // @LOCALMOD-BEGIN
  std::vector<std::string> DepLibs;
  // @LOCALMOD-END

  std::vector<sys::Path> Cleanup;
  lto_code_gen_t code_gen = NULL;
}

namespace options {
  enum generate_bc { BC_NO, BC_ALSO, BC_ONLY };
  static bool generate_api_file = false;
  static bool gather_then_link = true; // @LOCALMOD
  static generate_bc generate_bc_file = BC_NO;
  static std::string bc_path;
  static std::string obj_path;
  static std::string extra_library_path;
  static std::string triple;
  static std::string mcpu;
  // Additional options to pass into the code generator.
  // Note: This array will contain all plugin options which are not claimed
  // as plugin exclusive to pass to the code generator.
  // For example, "generate-api-file" and "as"options are for the plugin
  // use only and will not be passed.
  static std::vector<std::string> extra;

  static void process_plugin_option(const char* opt_)
  {
    if (opt_ == NULL)
      return;
    llvm::StringRef opt = opt_;

    if (opt == "generate-api-file") {
      generate_api_file = true;
    } else if (opt.startswith("mcpu=")) {
      mcpu = opt.substr(strlen("mcpu="));
    } else if (opt.startswith("extra-library-path=")) {
      extra_library_path = opt.substr(strlen("extra_library_path="));
    } else if (opt.startswith("mtriple=")) {
      triple = opt.substr(strlen("mtriple="));
    } else if (opt.startswith("obj-path=")) {
      obj_path = opt.substr(strlen("obj-path="));
      // @LOCALMOD-BEGIN
    } else if (opt == "no-gather-then-link") {
      gather_then_link = false;
      // @LOCALMOD-END
    } else if (opt == "emit-llvm") {
      generate_bc_file = BC_ONLY;
    } else if (opt == "also-emit-llvm") {
      generate_bc_file = BC_ALSO;
    } else if (opt.startswith("also-emit-llvm=")) {
      llvm::StringRef path = opt.substr(strlen("also-emit-llvm="));
      generate_bc_file = BC_ALSO;
      if (!bc_path.empty()) {
        (*message)(LDPL_WARNING, "Path to the output IL file specified twice. "
                   "Discarding %s", opt_);
      } else {
        bc_path = path;
      }
    } else {
      // Save this option to pass to the code generator.
      extra.push_back(opt);
    }
  }
}

// @LOCALMOD-BEGIN
static const char *get_basename(const char *path) {
  if (path == NULL)
    return NULL;
  const char *slash = strrchr(path, '/');
  if (slash)
    return slash + 1;

  return path;
}
// @LOCALMOD-END

static ld_plugin_status claim_file_hook(const ld_plugin_input_file *file,
                                        int *claimed);
static ld_plugin_status all_symbols_read_hook(void);
static ld_plugin_status cleanup_hook(void);

extern "C" ld_plugin_status onload(ld_plugin_tv *tv);
ld_plugin_status onload(ld_plugin_tv *tv) {
  // We're given a pointer to the first transfer vector. We read through them
  // until we find one where tv_tag == LDPT_NULL. The REGISTER_* tagged values
  // contain pointers to functions that we need to call to register our own
  // hooks. The others are addresses of functions we can use to call into gold
  // for services.

  bool registeredClaimFile = false;

  for (; tv->tv_tag != LDPT_NULL; ++tv) {
    switch (tv->tv_tag) {
      case LDPT_API_VERSION:
        api_version = tv->tv_u.tv_val;
        break;
      case LDPT_GOLD_VERSION:  // major * 100 + minor
        gold_version = tv->tv_u.tv_val;
        break;
      case LDPT_OUTPUT_NAME:
        output_name = tv->tv_u.tv_string;
        break;
      case LDPT_LINKER_OUTPUT:
        // @LOCALMOD-BEGIN
        linker_output =
          static_cast<ld_plugin_output_file_type>(tv->tv_u.tv_val);
        // @LOCALMOD-END
        switch (tv->tv_u.tv_val) {
          case LDPO_REL:  // .o
          case LDPO_DYN:  // .so
          // FIXME: Replace 3 with LDPO_PIE once that is in a released binutils.
          case 3: // position independent executable
            output_type = LTO_CODEGEN_PIC_MODEL_DYNAMIC;
            break;
          case LDPO_EXEC:  // .exe
            output_type = LTO_CODEGEN_PIC_MODEL_STATIC;
            break;
          default:
            (*message)(LDPL_ERROR, "Unknown output file type %d",
                       tv->tv_u.tv_val);
            return LDPS_ERR;
        }
        // TODO: add an option to disable PIC.
        //output_type = LTO_CODEGEN_PIC_MODEL_DYNAMIC_NO_PIC;
        break;
      case LDPT_OPTION:
        options::process_plugin_option(tv->tv_u.tv_string);
        break;
      case LDPT_REGISTER_CLAIM_FILE_HOOK: {
        ld_plugin_register_claim_file callback;
        callback = tv->tv_u.tv_register_claim_file;

        if ((*callback)(claim_file_hook) != LDPS_OK)
          return LDPS_ERR;

        registeredClaimFile = true;
      } break;
      case LDPT_REGISTER_ALL_SYMBOLS_READ_HOOK: {
        ld_plugin_register_all_symbols_read callback;
        callback = tv->tv_u.tv_register_all_symbols_read;

        if ((*callback)(all_symbols_read_hook) != LDPS_OK)
          return LDPS_ERR;

        code_gen = lto_codegen_create();
      } break;
      case LDPT_REGISTER_CLEANUP_HOOK: {
        ld_plugin_register_cleanup callback;
        callback = tv->tv_u.tv_register_cleanup;

        if ((*callback)(cleanup_hook) != LDPS_OK)
          return LDPS_ERR;
      } break;
      case LDPT_ADD_SYMBOLS:
        add_symbols = tv->tv_u.tv_add_symbols;
        break;
      case LDPT_GET_SYMBOLS:
        get_symbols = tv->tv_u.tv_get_symbols;
        break;
      case LDPT_ADD_INPUT_FILE:
        add_input_file = tv->tv_u.tv_add_input_file;
        break;
      case LDPT_ADD_INPUT_LIBRARY:
        add_input_library = tv->tv_u.tv_add_input_file;
        break;
      case LDPT_SET_EXTRA_LIBRARY_PATH:
        set_extra_library_path = tv->tv_u.tv_set_extra_library_path;
        break;
      case LDPT_GET_VIEW:
        get_view = tv->tv_u.tv_get_view;
      // @LOCALMOD-BEGIN
      case LDPT_GET_OUTPUT_SONAME:
        get_output_soname = tv->tv_u.tv_get_output_soname;
        break;
      case LDPT_GET_NEEDED:
        get_needed = tv->tv_u.tv_get_needed;
        break;
      case LDPT_GET_NUM_NEEDED:
        get_num_needed = tv->tv_u.tv_get_num_needed;
        break;
      case LDPT_GET_WRAPPED:
        get_wrapped = tv->tv_u.tv_get_wrapped;
        break;
      case LDPT_GET_NUM_WRAPPED:
        get_num_wrapped = tv->tv_u.tv_get_num_wrapped;
        break;
      // @LOCALMOD-END
      case LDPT_MESSAGE:
        message = tv->tv_u.tv_message;
        break;
      default:
        break;
    }
  }

  if (!registeredClaimFile) {
    (*message)(LDPL_ERROR, "register_claim_file not passed to LLVMgold.");
    return LDPS_ERR;
  }
  if (!add_symbols) {
    (*message)(LDPL_ERROR, "add_symbols not passed to LLVMgold.");
    return LDPS_ERR;
  }

  // @LOCALMOD-BEGIN
  // Parse extra command-line options
  // Although lto_codegen provides a way to parse command-line arguments,
  // we need the arguments to be parsed and applied before LTOModules are
  // even created. In particular, this is needed because the
  // "-add-nacl-read-tp-dependency" flag affects how modules are created.
  if (!options::extra.empty()) {
    for (std::vector<std::string>::iterator it = options::extra.begin();
         it != options::extra.end(); ++it) {
      lto_add_command_line_option((*it).c_str());
    }
    lto_parse_command_line_options();
    // We clear the options so that they don't get parsed again in
    // lto_codegen_debug_options.
    options::extra.clear();
  }
  // @LOCALMOD-END

  return LDPS_OK;
}

/// claim_file_hook - called by gold to see whether this file is one that
/// our plugin can handle. We'll try to open it and register all the symbols
/// with add_symbol if possible.
static ld_plugin_status claim_file_hook(const ld_plugin_input_file *file,
                                        int *claimed) {
  lto_module_t M;
  const void *view;
  OwningPtr<MemoryBuffer> buffer;
  if (get_view) {
    if (get_view(file->handle, &view) != LDPS_OK) {
      (*message)(LDPL_ERROR, "Failed to get a view of %s", file->name);
      return LDPS_ERR;
    }
  } else {
    int64_t offset = 0;
    // Gold has found what might be IR part-way inside of a file, such as
    // an .a archive.
    if (file->offset) {
      offset = file->offset;
    }
    if (error_code ec =
        MemoryBuffer::getOpenFile(file->fd, file->name, buffer, file->filesize,
                                  -1, offset, false)) {
      (*message)(LDPL_ERROR, ec.message().c_str());
      return LDPS_ERR;
    }
    view = buffer->getBufferStart();
  }

  if (!lto_module_is_object_file_in_memory(view, file->filesize))
    return LDPS_OK;

  M = lto_module_create_from_memory(view, file->filesize);
  if (!M) {
    if (const char* msg = lto_get_error_message()) {
      (*message)(LDPL_ERROR,
                 "LLVM gold plugin has failed to create LTO module: %s",
                 msg);
      return LDPS_ERR;
    }
    return LDPS_OK;
  }

  *claimed = 1;
  Modules.resize(Modules.size() + 1);
  claimed_file &cf = Modules.back();

  if (!options::triple.empty())
    lto_module_set_target_triple(M, options::triple.c_str());

  cf.handle = file->handle;
  unsigned sym_count = lto_module_get_num_symbols(M);
  cf.syms.reserve(sym_count);

  for (unsigned i = 0; i != sym_count; ++i) {
    lto_symbol_attributes attrs = lto_module_get_symbol_attribute(M, i);
    if ((attrs & LTO_SYMBOL_SCOPE_MASK) == LTO_SYMBOL_SCOPE_INTERNAL)
      continue;

    cf.syms.push_back(ld_plugin_symbol());
    ld_plugin_symbol &sym = cf.syms.back();
    sym.name = const_cast<char *>(lto_module_get_symbol_name(M, i));
    sym.name = strdup(sym.name);
    // @LOCALMOD-BEGIN
    // Localmods have disabled the use of the 'version' field for passing
    // version information to Gold. Instead, the version is now transmitted as
    // part of the 'name' field, which has the form "sym@VER" or "sym@@VER".
    // This is nicer because it communicates one extra bit of information (@@
    // marks the default version), and allows us to access the real symbol
    // name in all_symbols_read.

    // These fields are set by Gold to communicate the updated version info
    // to the plugin. They are used in all_symbols_read_hook().
    // Initialize them for predictability.
    sym.version = NULL;
    sym.is_default = false;
    sym.dynfile = NULL;
    // @LOCALMOD-END

    int scope = attrs & LTO_SYMBOL_SCOPE_MASK;
    switch (scope) {
      case LTO_SYMBOL_SCOPE_HIDDEN:
        sym.visibility = LDPV_HIDDEN;
        break;
      case LTO_SYMBOL_SCOPE_PROTECTED:
        sym.visibility = LDPV_PROTECTED;
        break;
      case 0: // extern
      case LTO_SYMBOL_SCOPE_DEFAULT:
        sym.visibility = LDPV_DEFAULT;
        break;
      default:
        (*message)(LDPL_ERROR, "Unknown scope attribute: %d", scope);
        return LDPS_ERR;
    }

    int definition = attrs & LTO_SYMBOL_DEFINITION_MASK;
    sym.comdat_key = NULL;
    switch (definition) {
      case LTO_SYMBOL_DEFINITION_REGULAR:
        sym.def = LDPK_DEF;
        break;
      case LTO_SYMBOL_DEFINITION_UNDEFINED:
        sym.def = LDPK_UNDEF;
        break;
      case LTO_SYMBOL_DEFINITION_TENTATIVE:
        sym.def = LDPK_COMMON;
        break;
      case LTO_SYMBOL_DEFINITION_WEAK:
        sym.comdat_key = sym.name;
        sym.def = LDPK_WEAKDEF;
        break;
      case LTO_SYMBOL_DEFINITION_WEAKUNDEF:
        sym.def = LDPK_WEAKUNDEF;
        break;
      default:
        (*message)(LDPL_ERROR, "Unknown definition attribute: %d", definition);
        return LDPS_ERR;
    }

    sym.size = 0;

    sym.resolution = LDPR_UNKNOWN;
  }

  cf.syms.reserve(cf.syms.size());
  // @LOCALMOD-BEGIN
  bool is_shared =
    (lto_module_get_output_format(M) == LTO_OUTPUT_FORMAT_SHARED);
  const char* soname = lto_module_get_soname(M);
  if (soname[0] == '\0')
    soname = NULL;
  // @LOCALMOD-END

  if (!cf.syms.empty()) {
    if ((*add_symbols)(cf.handle, cf.syms.size(), &cf.syms[0],
                       is_shared, soname) != LDPS_OK) { // @LOCALMOD
      (*message)(LDPL_ERROR, "Unable to add symbols!");
      return LDPS_ERR;
    }
  }

  // @LOCALMOD-BEGIN
  // Do not merge the module if it's a PSO.
  // If the PSO's soname is set, add it to DepLibs.
  cf.is_linked_in = false;
  if (code_gen) {
    if (is_shared) {
      if (soname && strlen(soname) > 0) {
        DepLibs.push_back(soname);
      }
    } else {
      if (options::gather_then_link) {
        lto_codegen_gather_module_for_link(code_gen, M);
      } else {
        lto_codegen_add_module(code_gen, M);
      }
      cf.is_linked_in = true;
    }
  }

  // With gather_then_link, the modules are disposed when linking.
  if (!options::gather_then_link)
    lto_module_dispose(M);
  // @LOCALMOD-END

  return LDPS_OK;
}

/// all_symbols_read_hook - gold informs us that all symbols have been read.
/// At this point, we use get_symbols to see if any of our definitions have
/// been overridden by a native object file. Then, perform optimization and
/// codegen.
static ld_plugin_status all_symbols_read_hook(void) {
  std::ofstream api_file;
  assert(code_gen);

  // @LOCALMOD-BEGIN
  if (options::gather_then_link) {
    lto_codegen_link_gathered_modules_and_dispose(code_gen);
  }
  // @LOCALMOD-END

  if (options::generate_api_file) {
    api_file.open("apifile.txt", std::ofstream::out | std::ofstream::trunc);
    if (!api_file.is_open()) {
      (*message)(LDPL_FATAL, "Unable to open apifile.txt for writing.");
      abort();
    }
  }

  for (std::list<claimed_file>::iterator I = Modules.begin(),
         E = Modules.end(); I != E; ++I) {
    if (I->syms.empty())
      continue;
    (*get_symbols)(I->handle, I->syms.size(), &I->syms[0]);
    for (unsigned i = 0, e = I->syms.size(); i != e; i++) {
      // @LOCALMOD-BEGIN
      // Don't process the symbols inside a dynamic object.
      if (!I->is_linked_in)
        continue;
      // @LOCALMOD-END

      if (I->syms[i].resolution == LDPR_PREVAILING_DEF) {
        // @LOCALMOD-BEGIN
        // Set the symbol version in the module.
        if (linker_output != LDPO_REL && I->syms[i].version) {
          // NOTE: This may change the name of the symbol, so it must happen
          // before the call to lto_codegen_add_must_preserve_symbols() below.
          I->syms[i].name = const_cast<char *>(
            lto_codegen_set_symbol_def_version(code_gen, I->syms[i].name,
                                               I->syms[i].version,
                                               I->syms[i].is_default));
        }
        lto_codegen_add_must_preserve_symbol(code_gen, I->syms[i].name);
        // @LOCALMOD-END

        if (options::generate_api_file)
          api_file << I->syms[i].name << "\n";
      }
      // @LOCALMOD-BEGIN
      else if (linker_output != LDPO_REL &&
               (I->syms[i].resolution == LDPR_RESOLVED_DYN ||
                I->syms[i].resolution == LDPR_UNDEF)) {
        // This symbol is provided by an external object.
        // Set the version and source dynamic file for it.
        const char *ver = I->syms[i].version;
        const char *dynfile = I->syms[i].dynfile;
        dynfile = get_basename(dynfile);
        // NOTE: This may change the name of the symbol.
        I->syms[i].name = const_cast<char *>(
          lto_codegen_set_symbol_needed(code_gen, I->syms[i].name,
                                        ver ? ver : "",
                                        dynfile ? dynfile : ""));
      }
      // @LOCALMOD-END
    }
  }

  if (options::generate_api_file)
    api_file.close();

  lto_codegen_set_pic_model(code_gen, output_type);
  lto_codegen_set_debug_model(code_gen, LTO_DEBUG_MODEL_DWARF);
  if (!options::mcpu.empty())
    lto_codegen_set_cpu(code_gen, options::mcpu.c_str());

  // @LOCALMOD-BEGIN (COMMENT)
  // "extra" will always be empty below, because we process the extra
  // options earlier, at the end of onload().
  // @LOCALMOD-END

  // Pass through extra options to the code generator.
  if (!options::extra.empty()) {
    for (std::vector<std::string>::iterator it = options::extra.begin();
         it != options::extra.end(); ++it) {
      lto_codegen_debug_options(code_gen, (*it).c_str());
    }
  }

  // @LOCALMOD-BEGIN
  // Store the linker output format into the bitcode.
  lto_output_format format;
  switch (linker_output) {
    case LDPO_REL:
      format = LTO_OUTPUT_FORMAT_OBJECT;
      break;
    case LDPO_DYN:
      format = LTO_OUTPUT_FORMAT_SHARED;
      break;
    case LDPO_EXEC:
      format = LTO_OUTPUT_FORMAT_EXEC;
      break;
    default:
      (*message)(LDPL_FATAL, "Unknown linker output format (gold-plugin)");
      abort();
      break;
  }
  lto_codegen_set_merged_module_output_format(code_gen, format);
  // @LOCALMOD-END

  // @LOCALMOD-BEGIN
  // For -shared linking, store the soname into the bitcode.
  if (linker_output == LDPO_DYN) {
    const char *soname = (*get_output_soname)();
    lto_codegen_set_merged_module_soname(code_gen, soname);
  }
  // @LOCALMOD-END

  // @LOCALMOD-BEGIN
  // Add the needed libraries to the bitcode.
  unsigned int num_needed = (*get_num_needed)();
  for (unsigned i=0; i < num_needed; ++i) {
    const char *soname = (*get_needed)(i);
    soname = get_basename(soname);
    lto_codegen_add_merged_module_library_dep(code_gen, soname);
  }
  for (std::vector<std::string>::iterator I = DepLibs.begin(),
           E = DepLibs.end(); I != E; ++I) {
    lto_codegen_add_merged_module_library_dep(code_gen, I->c_str());
  }
  // @LOCALMOD-END

  // @LOCALMOD-BEGIN
  // Perform symbol wrapping.
  unsigned int num_wrapped = (*get_num_wrapped)();
  for (unsigned i=0; i < num_wrapped; ++i) {
    const char *sym = (*get_wrapped)(i);
    lto_codegen_wrap_symbol_in_merged_module(code_gen, sym);
  }
  // @LOCALMOD-END
  if (options::generate_bc_file != options::BC_NO) {
    std::string path;
    if (options::generate_bc_file == options::BC_ONLY)
      path = output_name;
    else if (!options::bc_path.empty())
      path = options::bc_path;
    else
      path = output_name + ".bc";
    bool err = lto_codegen_write_merged_modules(code_gen, path.c_str());
    if (err)
      (*message)(LDPL_FATAL, "Failed to write the output file.");
    if (options::generate_bc_file == options::BC_ONLY)
      exit(0);
  }
  const char *objPath;
  if (lto_codegen_compile_to_file(code_gen, &objPath)) {
    (*message)(LDPL_ERROR, "Could not produce a combined object file\n");
  }

  lto_codegen_dispose(code_gen);
  for (std::list<claimed_file>::iterator I = Modules.begin(),
         E = Modules.end(); I != E; ++I) {
    for (unsigned i = 0; i != I->syms.size(); ++i) {
      ld_plugin_symbol &sym = I->syms[i];
      free(sym.name);
    }
  }

  if ((*add_input_file)(objPath) != LDPS_OK) {
    (*message)(LDPL_ERROR, "Unable to add .o file to the link.");
    (*message)(LDPL_ERROR, "File left behind in: %s", objPath);
    return LDPS_ERR;
  }

  if (!options::extra_library_path.empty() &&
      set_extra_library_path(options::extra_library_path.c_str()) != LDPS_OK) {
    (*message)(LDPL_ERROR, "Unable to set the extra library path.");
    return LDPS_ERR;
  }

  if (options::obj_path.empty())
    Cleanup.push_back(sys::Path(objPath));

  return LDPS_OK;
}

static ld_plugin_status cleanup_hook(void) {
  std::string ErrMsg;

  for (int i = 0, e = Cleanup.size(); i != e; ++i)
    if (Cleanup[i].eraseFromDisk(false, &ErrMsg))
      (*message)(LDPL_ERROR, "Failed to delete '%s': %s", Cleanup[i].c_str(),
                 ErrMsg.c_str());

  return LDPS_OK;
}