lld/COFF/Chunks.cpp

//===- Chunks.cpp ---------------------------------------------------------===//
//
//                             The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//

#include "Chunks.h"
#include "InputFiles.h"
#include "Writer.h"
#include "llvm/ADT/Hashing.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/Object/COFF.h"
#include "llvm/Support/COFF.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/Endian.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>

using namespace llvm;
using namespace llvm::object;
using namespace llvm::support::endian;
using namespace llvm::COFF;
using llvm::support::ulittle32_t;

namespace lld {
namespace coff {

SectionChunk::SectionChunk(ObjectFile *F, const coff_section *H)
    : Chunk(SectionKind), Ptr(this), File(F), Header(H),
      Relocs(File->getCOFFObj()->getRelocations(Header)),
      NumRelocs(std::distance(Relocs.begin(), Relocs.end())) {
  // Initialize SectionName.
  File->getCOFFObj()->getSectionName(Header, SectionName);

  // Bit [20:24] contains section alignment. Both 0 and 1 mean alignment 1.
  unsigned Shift = (Header->Characteristics >> 20) & 0xF;
  if (Shift > 0)
    Align = uint32_t(1) << (Shift - 1);

  // COMDAT sections are not GC root. Non-text sections are not
  // subject of garbage collection (thus they are root).
  Root = !isCOMDAT() && !(Header->Characteristics & IMAGE_SCN_CNT_CODE);
}

static void add16(uint8_t *P, int16_t V) { write16le(P, read16le(P) + V); }
static void add32(uint8_t *P, int32_t V) { write32le(P, read32le(P) + V); }
static void add64(uint8_t *P, int64_t V) { write64le(P, read64le(P) + V); }

void SectionChunk::applyRelX64(uint8_t *Off, uint16_t Type, uint64_t S,
                               uint64_t P) {
  switch (Type) {
  case IMAGE_REL_AMD64_ADDR32:   add32(Off, S + Config->ImageBase); break;
  case IMAGE_REL_AMD64_ADDR64:   add64(Off, S + Config->ImageBase); break;
  case IMAGE_REL_AMD64_ADDR32NB: add32(Off, S); break;
  case IMAGE_REL_AMD64_REL32:    add32(Off, S - P - 4); break;
  case IMAGE_REL_AMD64_REL32_1:  add32(Off, S - P - 5); break;
  case IMAGE_REL_AMD64_REL32_2:  add32(Off, S - P - 6); break;
  case IMAGE_REL_AMD64_REL32_3:  add32(Off, S - P - 7); break;
  case IMAGE_REL_AMD64_REL32_4:  add32(Off, S - P - 8); break;
  case IMAGE_REL_AMD64_REL32_5:  add32(Off, S - P - 9); break;
  case IMAGE_REL_AMD64_SECTION:  add16(Off, Out->SectionIndex); break;
  case IMAGE_REL_AMD64_SECREL:   add32(Off, S - Out->getRVA()); break;
  default:
    llvm::report_fatal_error("Unsupported relocation type");
  }
}

void SectionChunk::applyRelX86(uint8_t *Off, uint16_t Type, uint64_t S,
                               uint64_t P) {
  switch (Type) {
  case IMAGE_REL_I386_ABSOLUTE: break;
  case IMAGE_REL_I386_DIR32:    add32(Off, S + Config->ImageBase); break;
  case IMAGE_REL_I386_DIR32NB:  add32(Off, S); break;
  case IMAGE_REL_I386_REL32:    add32(Off, S - P - 4); break;
  case IMAGE_REL_I386_SECTION:  add16(Off, Out->SectionIndex); break;
  case IMAGE_REL_I386_SECREL:   add32(Off, S - Out->getRVA()); break;
  default:
    llvm::report_fatal_error("Unsupported relocation type");
  }
}

void SectionChunk::writeTo(uint8_t *Buf) {
  if (!hasData())
    return;
  // Copy section contents from source object file to output file.
  ArrayRef<uint8_t> A = getContents();
  memcpy(Buf + FileOff, A.data(), A.size());

  // Apply relocations.
  for (const coff_relocation &Rel : Relocs) {
    uint8_t *Off = Buf + FileOff + Rel.VirtualAddress;
    SymbolBody *Body = File->getSymbolBody(Rel.SymbolTableIndex)->repl();
    uint64_t S = cast<Defined>(Body)->getRVA();
    uint64_t P = RVA + Rel.VirtualAddress;
    switch (Config->MachineType) {
    case IMAGE_FILE_MACHINE_AMD64:
      applyRelX64(Off, Rel.Type, S, P);
      break;
    case IMAGE_FILE_MACHINE_I386:
      applyRelX86(Off, Rel.Type, S, P);
      break;
    default:
      llvm_unreachable("unknown machine type");
    }
  }
}

void SectionChunk::addAssociative(SectionChunk *Child) {
  AssocChildren.push_back(Child);
  // Associative sections are live if their parent COMDATs are live,
  // and vice versa, so they are not considered live by themselves.
  Child->Root = false;
}

static uint8_t getBaserelType(const coff_relocation &Rel) {
  switch (Config->MachineType) {
  case IMAGE_FILE_MACHINE_AMD64:
    if (Rel.Type == IMAGE_REL_AMD64_ADDR64)
      return IMAGE_REL_BASED_DIR64;
    return IMAGE_REL_BASED_ABSOLUTE;
  case IMAGE_FILE_MACHINE_I386:
    if (Rel.Type == IMAGE_REL_I386_DIR32)
      return IMAGE_REL_BASED_HIGHLOW;
    return IMAGE_REL_BASED_ABSOLUTE;
  default:
    llvm_unreachable("unknown machine type");
  }
}

// Windows-specific.
// Collect all locations that contain absolute addresses, which need to be
// fixed by the loader if load-time relocation is needed.
// Only called when base relocation is enabled.
void SectionChunk::getBaserels(std::vector<Baserel> *Res) {
  for (const coff_relocation &Rel : Relocs) {
    uint8_t Ty = getBaserelType(Rel);
    if (Ty == IMAGE_REL_BASED_ABSOLUTE)
      continue;
    SymbolBody *Body = File->getSymbolBody(Rel.SymbolTableIndex)->repl();
    if (isa<DefinedAbsolute>(Body))
      continue;
    Res->emplace_back(RVA + Rel.VirtualAddress, Ty);
  }
}

bool SectionChunk::hasData() const {
  return !(Header->Characteristics & IMAGE_SCN_CNT_UNINITIALIZED_DATA);
}

uint32_t SectionChunk::getPermissions() const {
  return Header->Characteristics & PermMask;
}

bool SectionChunk::isCOMDAT() const {
  return Header->Characteristics & IMAGE_SCN_LNK_COMDAT;
}

void SectionChunk::printDiscardedMessage() const {
  if (this == Ptr) {
    // Removed by dead-stripping.
    llvm::dbgs() << "Discarded " << Sym->getName() << "\n";
  } else {
    // Removed by ICF.
    llvm::dbgs() << "Replaced " << Sym->getName() << "\n";
  }
}

StringRef SectionChunk::getDebugName() {
  return Sym->getName();
}

uint64_t SectionChunk::getHash() const {
  ArrayRef<uint8_t> A = getContents();
  return hash_combine(getPermissions(),
                      llvm::hash_value(SectionName),
                      NumRelocs,
                      uint32_t(Header->SizeOfRawData),
                      std::distance(Relocs.end(), Relocs.begin()),
                      hash_combine_range(A.data(), A.data() + A.size()));
}

// Returns true if this and a given chunk are identical COMDAT sections.
bool SectionChunk::equals(const SectionChunk *X) const {
  // Compare headers
  if (getPermissions() != X->getPermissions())
    return false;
  if (SectionName != X->SectionName)
    return false;
  if (Header->SizeOfRawData != X->Header->SizeOfRawData)
    return false;
  if (NumRelocs != X->NumRelocs)
    return false;

  // Compare data
  if (getContents() != X->getContents())
    return false;

  // Compare associative sections
  if (AssocChildren.size() != X->AssocChildren.size())
    return false;
  for (size_t I = 0, E = AssocChildren.size(); I != E; ++I)
    if (AssocChildren[I]->Ptr != X->AssocChildren[I]->Ptr)
      return false;

  // Compare relocations
  auto Eq = [&](const coff_relocation &R1, const coff_relocation &R2) {
    if (R1.Type != R2.Type)
      return false;
    if (R1.VirtualAddress != R2.VirtualAddress)
      return false;
    SymbolBody *B1 = File->getSymbolBody(R1.SymbolTableIndex)->repl();
    SymbolBody *B2 = X->File->getSymbolBody(R2.SymbolTableIndex)->repl();
    if (B1 == B2)
      return true;
    auto *D1 = dyn_cast<DefinedRegular>(B1);
    auto *D2 = dyn_cast<DefinedRegular>(B2);
    return (D1 && D2 &&
            D1->getValue() == D2->getValue() &&
            D1->getChunk() == D2->getChunk());
  };
  return std::equal(Relocs.begin(), Relocs.end(), X->Relocs.begin(), Eq);
}

ArrayRef<uint8_t> SectionChunk::getContents() const {
  ArrayRef<uint8_t> A;
  File->getCOFFObj()->getSectionContents(Header, A);
  return A;
}

void SectionChunk::replaceWith(SectionChunk *Other) {
  Ptr = Other->Ptr;
  Live = false;
}

CommonChunk::CommonChunk(const COFFSymbolRef S) : Sym(S) {
  // Common symbols are aligned on natural boundaries up to 32 bytes.
  // This is what MSVC link.exe does.
  Align = std::min(uint64_t(32), NextPowerOf2(Sym.getValue()));
}

uint32_t CommonChunk::getPermissions() const {
  return IMAGE_SCN_CNT_UNINITIALIZED_DATA | IMAGE_SCN_MEM_READ |
         IMAGE_SCN_MEM_WRITE;
}

void StringChunk::writeTo(uint8_t *Buf) {
  memcpy(Buf + FileOff, Str.data(), Str.size());
}

ImportThunkChunkX64::ImportThunkChunkX64(Defined *S) : ImpSymbol(S) {
  // Intel Optimization Manual says that all branch targets
  // should be 16-byte aligned. MSVC linker does this too.
  if (Config->MachineType == AMD64)
      Align = 16;
}

void ImportThunkChunkX64::writeTo(uint8_t *Buf) {
  memcpy(Buf + FileOff, ImportThunkX86, sizeof(ImportThunkX86));
  // The first two bytes is a JMP instruction. Fill its operand.
  write32le(Buf + FileOff + 2, ImpSymbol->getRVA() - RVA - getSize());
}

void ImportThunkChunkX86::getBaserels(std::vector<Baserel> *Res) {
  Res->emplace_back(getRVA() + 2);
}

void ImportThunkChunkX86::writeTo(uint8_t *Buf) {
  memcpy(Buf + FileOff, ImportThunkX86, sizeof(ImportThunkX86));
  // The first two bytes is a JMP instruction. Fill its operand.
  write32le(Buf + FileOff + 2, ImpSymbol->getRVA() + Config->ImageBase);
}

void LocalImportChunk::getBaserels(std::vector<Baserel> *Res) {
  Res->emplace_back(getRVA());
}

size_t LocalImportChunk::getSize() const {
  return Config->is64() ? 8 : 4;
}

void LocalImportChunk::writeTo(uint8_t *Buf) {
  if (Config->is64()) {
    write64le(Buf + FileOff, Sym->getRVA() + Config->ImageBase);
  } else {
    write32le(Buf + FileOff, Sym->getRVA() + Config->ImageBase);
  }
}

void SEHTableChunk::writeTo(uint8_t *Buf) {
  ulittle32_t *Begin = reinterpret_cast<ulittle32_t *>(Buf + FileOff);
  size_t Cnt = 0;
  for (Defined *D : Syms)
    Begin[Cnt++] = D->getRVA();
  std::sort(Begin, Begin + Cnt);
}

// Windows-specific.
// This class represents a block in .reloc section.
BaserelChunk::BaserelChunk(uint32_t Page, Baserel *Begin, Baserel *End) {
  // Block header consists of 4 byte page RVA and 4 byte block size.
  // Each entry is 2 byte. Last entry may be padding.
  Data.resize(RoundUpToAlignment((End - Begin) * 2 + 8, 4));
  uint8_t *P = Data.data();
  write32le(P, Page);
  write32le(P + 4, Data.size());
  P += 8;
  for (Baserel *I = Begin; I != End; ++I) {
    write16le(P, (I->Type << 12) | (I->RVA - Page));
    P += 2;
  }
}

void BaserelChunk::writeTo(uint8_t *Buf) {
  memcpy(Buf + FileOff, Data.data(), Data.size());
}

uint8_t Baserel::getDefaultType() {
  switch (Config->MachineType) {
  case IMAGE_FILE_MACHINE_AMD64:
    return IMAGE_REL_BASED_DIR64;
  case IMAGE_FILE_MACHINE_I386:
    return IMAGE_REL_BASED_HIGHLOW;
  default:
    llvm_unreachable("unknown machine type");
  }
}

} // namespace coff
} // namespace lld