mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2025-11-04 05:17:07 +00:00
7cb77e188f
Implement minor library linking optimization. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@6181 91177308-0d34-0410-b5e6-96231b3b80d8
407 lines
15 KiB
C++
407 lines
15 KiB
C++
//===- gccld.cpp - LLVM 'ld' compatible linker ----------------------------===//
|
|
//
|
|
// This utility is intended to be compatible with GCC, and follows standard
|
|
// system 'ld' conventions. As such, the default output file is ./a.out.
|
|
// Additionally, this program outputs a shell script that is used to invoke LLI
|
|
// to execute the program. In this manner, the generated executable (a.out for
|
|
// example), is directly executable, whereas the bytecode file actually lives in
|
|
// the a.out.bc file generated by this program. Also, Force is on by default.
|
|
//
|
|
// Note that if someone (or a script) deletes the executable program generated,
|
|
// the .bc file will be left around. Considering that this is a temporary hack,
|
|
// I'm not to worried about this.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "llvm/Transforms/Utils/Linker.h"
|
|
#include "llvm/Module.h"
|
|
#include "llvm/PassManager.h"
|
|
#include "llvm/Bytecode/Reader.h"
|
|
#include "llvm/Bytecode/WriteBytecodePass.h"
|
|
#include "llvm/Target/TargetData.h"
|
|
#include "llvm/Transforms/IPO.h"
|
|
#include "llvm/Transforms/Scalar.h"
|
|
#include "Support/CommandLine.h"
|
|
#include "Support/Signals.h"
|
|
#include <fstream>
|
|
#include <memory>
|
|
#include <set>
|
|
#include <algorithm>
|
|
#include <sys/types.h> // For FileExists
|
|
#include <sys/stat.h>
|
|
|
|
namespace {
|
|
cl::list<std::string>
|
|
InputFilenames(cl::Positional, cl::desc("<input bytecode files>"),
|
|
cl::OneOrMore);
|
|
|
|
cl::opt<std::string>
|
|
OutputFilename("o", cl::desc("Override output filename"), cl::init("a.out"),
|
|
cl::value_desc("filename"));
|
|
|
|
cl::opt<bool>
|
|
Verbose("v", cl::desc("Print information about actions taken"));
|
|
|
|
cl::list<std::string>
|
|
LibPaths("L", cl::desc("Specify a library search path"), cl::Prefix,
|
|
cl::value_desc("directory"));
|
|
|
|
cl::list<std::string>
|
|
Libraries("l", cl::desc("Specify libraries to link to"), cl::Prefix,
|
|
cl::value_desc("library prefix"));
|
|
|
|
cl::opt<bool>
|
|
Strip("s", cl::desc("Strip symbol info from executable"));
|
|
|
|
cl::opt<bool>
|
|
NoInternalize("disable-internalize",
|
|
cl::desc("Do not mark all symbols as internal"));
|
|
|
|
cl::opt<bool>
|
|
LinkAsLibrary("link-as-library", cl::desc("Link the .bc files together as a"
|
|
" library, not an executable"));
|
|
|
|
// Compatibility options that are ignored, but support by LD
|
|
cl::opt<std::string>
|
|
CO3("soname", cl::Hidden, cl::desc("Compatibility option: ignored"));
|
|
cl::opt<std::string>
|
|
CO4("version-script", cl::Hidden, cl::desc("Compatibility option: ignored"));
|
|
cl::opt<bool>
|
|
CO5("eh-frame-hdr", cl::Hidden, cl::desc("Compatibility option: ignored"));
|
|
}
|
|
|
|
// FileExists - Return true if the specified string is an openable file...
|
|
static inline bool FileExists(const std::string &FN) {
|
|
struct stat StatBuf;
|
|
return stat(FN.c_str(), &StatBuf) != -1;
|
|
}
|
|
|
|
|
|
// LoadObject - Read the specified "object file", which should not search the
|
|
// library path to find it.
|
|
static inline std::auto_ptr<Module> LoadObject(std::string FN,
|
|
std::string &OutErrorMessage) {
|
|
if (Verbose) std::cerr << "Loading '" << FN << "'\n";
|
|
if (!FileExists(FN)) {
|
|
// Attempt to load from the LLVM_LIB_SEARCH_PATH directory... if we would
|
|
// otherwise fail. This is used to locate objects like crtend.o.
|
|
//
|
|
char *SearchPath = getenv("LLVM_LIB_SEARCH_PATH");
|
|
if (SearchPath && FileExists(std::string(SearchPath)+"/"+FN))
|
|
FN = std::string(SearchPath)+"/"+FN;
|
|
else {
|
|
OutErrorMessage = "could not find input file '" + FN + "'!";
|
|
return std::auto_ptr<Module>();
|
|
}
|
|
}
|
|
|
|
std::string ErrorMessage;
|
|
Module *Result = ParseBytecodeFile(FN, &ErrorMessage);
|
|
if (Result) return std::auto_ptr<Module>(Result);
|
|
|
|
OutErrorMessage = "Bytecode file '" + FN + "' corrupt!";
|
|
if (ErrorMessage.size()) OutErrorMessage += ": " + ErrorMessage;
|
|
return std::auto_ptr<Module>();
|
|
}
|
|
|
|
|
|
static Module *LoadSingleLibraryObject(const std::string &Filename) {
|
|
std::string ErrorMessage;
|
|
std::auto_ptr<Module> M = LoadObject(Filename, ErrorMessage);
|
|
if (M.get() == 0 && Verbose) {
|
|
std::cerr << "Error loading '" + Filename + "'";
|
|
if (!ErrorMessage.empty()) std::cerr << ": " << ErrorMessage;
|
|
std::cerr << "\n";
|
|
}
|
|
|
|
return M.release();
|
|
}
|
|
|
|
|
|
// LoadLibraryFromDirectory - This looks for a .a, .so, or .bc file in a
|
|
// particular directory. It returns true if no library is found, otherwise it
|
|
// puts the loaded modules into the Objects list, and sets isArchive to true if
|
|
// a .a file was loaded.
|
|
//
|
|
static inline bool LoadLibraryFromDirectory(const std::string &LibName,
|
|
const std::string &Directory,
|
|
std::vector<Module*> &Objects,
|
|
bool &isArchive) {
|
|
if (FileExists(Directory + "lib" + LibName + ".a")) {
|
|
std::string ErrorMessage;
|
|
if (Verbose) std::cerr << " Loading '" << Directory << "lib"
|
|
<< LibName << ".a'\n";
|
|
if (!ReadArchiveFile(Directory + "lib" + LibName + ".a", Objects,
|
|
&ErrorMessage)) { // Read the archive file
|
|
isArchive = true;
|
|
return false; // Success!
|
|
}
|
|
|
|
if (Verbose) {
|
|
std::cerr << " Error loading archive '" + Directory +"lib"+LibName+".a'";
|
|
if (!ErrorMessage.empty()) std::cerr << ": " << ErrorMessage;
|
|
std::cerr << "\n";
|
|
}
|
|
}
|
|
|
|
if (FileExists(Directory + "lib" + LibName + ".so"))
|
|
if (Module *M = LoadSingleLibraryObject(Directory + "lib" + LibName+".so")){
|
|
isArchive = false;
|
|
Objects.push_back(M);
|
|
return false;
|
|
}
|
|
|
|
if (FileExists(Directory + "lib" + LibName + ".bc"))
|
|
if (Module *M = LoadSingleLibraryObject(Directory + "lib" + LibName+".bc")){
|
|
isArchive = false;
|
|
Objects.push_back(M);
|
|
return false;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
// LoadLibrary - This searches for a .a, .so, or .bc file which provides the
|
|
// LLVM bytecode for the library. It returns true if no library is found,
|
|
// otherwise it puts the loaded modules into the Objects list, and sets
|
|
// isArchive to true if a .a file was loaded.
|
|
//
|
|
static inline bool LoadLibrary(const std::string &LibName,
|
|
std::vector<Module*> &Objects, bool &isArchive,
|
|
std::string &ErrorMessage) {
|
|
std::string Directory;
|
|
unsigned NextLibPathIdx = 0;
|
|
|
|
while (1) {
|
|
// Try loading from the current directory...
|
|
if (Verbose) std::cerr << " Looking in directory '" << Directory << "'\n";
|
|
if (!LoadLibraryFromDirectory(LibName, Directory, Objects, isArchive))
|
|
return false;
|
|
|
|
if (NextLibPathIdx == LibPaths.size()) break;
|
|
Directory = LibPaths[NextLibPathIdx++]+"/";
|
|
}
|
|
|
|
ErrorMessage = "error linking library '-l" + LibName+ "': library not found!";
|
|
return true;
|
|
}
|
|
|
|
static void GetAllDefinedSymbols(Module *M,
|
|
std::set<std::string> &DefinedSymbols) {
|
|
for (Module::iterator I = M->begin(), E = M->end(); I != E; ++I)
|
|
if (I->hasName() && !I->isExternal() && !I->hasInternalLinkage())
|
|
DefinedSymbols.insert(I->getName());
|
|
for (Module::giterator I = M->gbegin(), E = M->gend(); I != E; ++I)
|
|
if (I->hasName() && !I->isExternal() && !I->hasInternalLinkage())
|
|
DefinedSymbols.insert(I->getName());
|
|
}
|
|
|
|
// GetAllUndefinedSymbols - This calculates the set of undefined symbols that
|
|
// still exist in an LLVM module. This is a bit tricky because there may be two
|
|
// symbols with the same name, but different LLVM types that will be resolved to
|
|
// each other, but aren't currently (thus we need to treat it as resolved).
|
|
//
|
|
static void GetAllUndefinedSymbols(Module *M,
|
|
std::set<std::string> &UndefinedSymbols) {
|
|
std::set<std::string> DefinedSymbols;
|
|
UndefinedSymbols.clear(); // Start out empty
|
|
|
|
for (Module::iterator I = M->begin(), E = M->end(); I != E; ++I)
|
|
if (I->hasName()) {
|
|
if (I->isExternal())
|
|
UndefinedSymbols.insert(I->getName());
|
|
else if (!I->hasInternalLinkage())
|
|
DefinedSymbols.insert(I->getName());
|
|
}
|
|
for (Module::giterator I = M->gbegin(), E = M->gend(); I != E; ++I)
|
|
if (I->hasName()) {
|
|
if (I->isExternal())
|
|
UndefinedSymbols.insert(I->getName());
|
|
else if (!I->hasInternalLinkage())
|
|
DefinedSymbols.insert(I->getName());
|
|
}
|
|
|
|
// Prune out any defined symbols from the undefined symbols set...
|
|
for (std::set<std::string>::iterator I = UndefinedSymbols.begin();
|
|
I != UndefinedSymbols.end(); )
|
|
if (DefinedSymbols.count(*I))
|
|
UndefinedSymbols.erase(I++); // This symbol really is defined!
|
|
else
|
|
++I; // Keep this symbol in the undefined symbols list
|
|
}
|
|
|
|
|
|
static bool LinkLibrary(Module *M, const std::string &LibName,
|
|
std::string &ErrorMessage) {
|
|
std::set<std::string> UndefinedSymbols;
|
|
GetAllUndefinedSymbols(M, UndefinedSymbols);
|
|
if (UndefinedSymbols.empty()) {
|
|
if (Verbose) std::cerr << " No symbols undefined, don't link library!\n";
|
|
return false; // No need to link anything in!
|
|
}
|
|
|
|
std::vector<Module*> Objects;
|
|
bool isArchive;
|
|
if (LoadLibrary(LibName, Objects, isArchive, ErrorMessage)) return true;
|
|
|
|
// Figure out which symbols are defined by all of the modules in the .a file
|
|
std::vector<std::set<std::string> > DefinedSymbols;
|
|
DefinedSymbols.resize(Objects.size());
|
|
for (unsigned i = 0; i != Objects.size(); ++i)
|
|
GetAllDefinedSymbols(Objects[i], DefinedSymbols[i]);
|
|
|
|
bool Linked = true;
|
|
while (Linked) { // While we are linking in object files, loop.
|
|
Linked = false;
|
|
|
|
for (unsigned i = 0; i != Objects.size(); ++i) {
|
|
// Consider whether we need to link in this module... we only need to
|
|
// link it in if it defines some symbol which is so far undefined.
|
|
//
|
|
const std::set<std::string> &DefSymbols = DefinedSymbols[i];
|
|
|
|
bool ObjectRequired = false;
|
|
for (std::set<std::string>::iterator I = UndefinedSymbols.begin(),
|
|
E = UndefinedSymbols.end(); I != E; ++I)
|
|
if (DefSymbols.count(*I)) {
|
|
if (Verbose)
|
|
std::cerr << " Found object providing symbol '" << *I << "'...\n";
|
|
ObjectRequired = true;
|
|
break;
|
|
}
|
|
|
|
// We DO need to link this object into the program...
|
|
if (ObjectRequired) {
|
|
if (LinkModules(M, Objects[i], &ErrorMessage))
|
|
return true; // Couldn't link in the right object file...
|
|
|
|
// Since we have linked in this object, delete it from the list of
|
|
// objects to consider in this archive file.
|
|
std::swap(Objects[i], Objects.back());
|
|
std::swap(DefinedSymbols[i], DefinedSymbols.back());
|
|
Objects.pop_back();
|
|
DefinedSymbols.pop_back();
|
|
--i; // Do not skip an entry
|
|
|
|
// The undefined symbols set should have shrunk.
|
|
GetAllUndefinedSymbols(M, UndefinedSymbols);
|
|
Linked = true; // We have linked something in!
|
|
}
|
|
}
|
|
}
|
|
|
|
return false;
|
|
}
|
|
|
|
static int PrintAndReturn(const char *progname, const std::string &Message,
|
|
const std::string &Extra = "") {
|
|
std::cerr << progname << Extra << ": " << Message << "\n";
|
|
return 1;
|
|
}
|
|
|
|
|
|
int main(int argc, char **argv) {
|
|
cl::ParseCommandLineOptions(argc, argv, " llvm linker for GCC\n");
|
|
|
|
std::string ErrorMessage;
|
|
std::auto_ptr<Module> Composite(LoadObject(InputFilenames[0], ErrorMessage));
|
|
if (Composite.get() == 0)
|
|
return PrintAndReturn(argv[0], ErrorMessage);
|
|
|
|
// If the user specied an extra search path in their environment, respect it.
|
|
if (char *SearchPath = getenv("LLVM_LIB_SEARCH_PATH"))
|
|
LibPaths.push_back(SearchPath);
|
|
|
|
for (unsigned i = 1; i < InputFilenames.size(); ++i) {
|
|
std::auto_ptr<Module> M(LoadObject(InputFilenames[i], ErrorMessage));
|
|
if (M.get() == 0)
|
|
return PrintAndReturn(argv[0], ErrorMessage);
|
|
|
|
if (Verbose) std::cerr << "Linking in '" << InputFilenames[i] << "'\n";
|
|
|
|
if (LinkModules(Composite.get(), M.get(), &ErrorMessage))
|
|
return PrintAndReturn(argv[0], ErrorMessage,
|
|
": error linking in '" + InputFilenames[i] + "'");
|
|
}
|
|
|
|
// Remove any consecutive duplicates of the same library...
|
|
Libraries.erase(std::unique(Libraries.begin(), Libraries.end()),
|
|
Libraries.end());
|
|
|
|
// Link in all of the libraries next...
|
|
for (unsigned i = 0; i != Libraries.size(); ++i) {
|
|
if (Verbose) std::cerr << "Linking in library: -l" << Libraries[i] << "\n";
|
|
if (LinkLibrary(Composite.get(), Libraries[i], ErrorMessage))
|
|
return PrintAndReturn(argv[0], ErrorMessage);
|
|
}
|
|
|
|
// In addition to just linking the input from GCC, we also want to spiff it up
|
|
// a little bit. Do this now.
|
|
//
|
|
PassManager Passes;
|
|
|
|
// Add an appropriate TargetData instance for this module...
|
|
Passes.add(new TargetData("gccas", Composite.get()));
|
|
|
|
// Linking modules together can lead to duplicated global constants, only keep
|
|
// one copy of each constant...
|
|
//
|
|
Passes.add(createConstantMergePass());
|
|
|
|
// If the -s command line option was specified, strip the symbols out of the
|
|
// resulting program to make it smaller. -s is a GCC option that we are
|
|
// supporting.
|
|
//
|
|
if (Strip)
|
|
Passes.add(createSymbolStrippingPass());
|
|
|
|
// Often if the programmer does not specify proper prototypes for the
|
|
// functions they are calling, they end up calling a vararg version of the
|
|
// function that does not get a body filled in (the real function has typed
|
|
// arguments). This pass merges the two functions.
|
|
//
|
|
Passes.add(createFunctionResolvingPass());
|
|
|
|
if (!NoInternalize) {
|
|
// Now that composite has been compiled, scan through the module, looking
|
|
// for a main function. If main is defined, mark all other functions
|
|
// internal.
|
|
//
|
|
Passes.add(createInternalizePass());
|
|
}
|
|
|
|
// Now that we have optimized the program, discard unreachable functions...
|
|
//
|
|
Passes.add(createGlobalDCEPass());
|
|
|
|
// Add the pass that writes bytecode to the output file...
|
|
std::string RealBytecodeOutput = OutputFilename;
|
|
if (!LinkAsLibrary) RealBytecodeOutput += ".bc";
|
|
std::ofstream Out(RealBytecodeOutput.c_str());
|
|
if (!Out.good())
|
|
return PrintAndReturn(argv[0], "error opening '" + RealBytecodeOutput +
|
|
"' for writing!");
|
|
Passes.add(new WriteBytecodePass(&Out)); // Write bytecode to file...
|
|
|
|
// Make sure that the Out file gets unlink'd from the disk if we get a SIGINT
|
|
RemoveFileOnSignal(RealBytecodeOutput);
|
|
|
|
// Run our queue of passes all at once now, efficiently.
|
|
Passes.run(*Composite.get());
|
|
Out.close();
|
|
|
|
if (!LinkAsLibrary) {
|
|
// Output the script to start the program...
|
|
std::ofstream Out2(OutputFilename.c_str());
|
|
if (!Out2.good())
|
|
return PrintAndReturn(argv[0], "error opening '" + OutputFilename +
|
|
"' for writing!");
|
|
Out2 << "#!/bin/sh\nlli -q -abort-on-exception $0.bc $*\n";
|
|
Out2.close();
|
|
|
|
// Make the script executable...
|
|
chmod(OutputFilename.c_str(), 0755);
|
|
}
|
|
|
|
return 0;
|
|
}
|