d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge lld trunk r338150, and resolve conflicts.

Browse Source
This commit is contained in:
Dimitry Andric 2018-07-31 17:18:35 +00:00
commit c85947bf32
125 changed files with 9180 additions and 4482 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
contrib/llvm/tools/lld/COFF/CMakeLists.txt
View File

@ -18,7 +18,6 @@ add_lld_library(lldCOFF
MarkLive.cpp
MinGW.cpp
PDB.cpp
Strings.cpp
SymbolTable.cpp
Symbols.cpp
Writer.cpp

175
contrib/llvm/tools/lld/COFF/Chunks.cpp
View File

@ -31,8 +31,7 @@ namespace coff {
SectionChunk::SectionChunk(ObjFile *F, const coff_section *H)
: Chunk(SectionKind), Repl(this), Header(H), File(F),
Relocs(File->getCOFFObj()->getRelocations(Header)),
NumRelocs(std::distance(Relocs.begin(), Relocs.end())) {
Relocs(File->getCOFFObj()->getRelocations(Header)) {
// Initialize SectionName.
File->getCOFFObj()->getSectionName(Header, SectionName);
@ -51,13 +50,21 @@ static void add64(uint8_t *P, int64_t V) { write64le(P, read64le(P) + V); }
static void or16(uint8_t *P, uint16_t V) { write16le(P, read16le(P) | V); }
static void or32(uint8_t *P, uint32_t V) { write32le(P, read32le(P) | V); }
// Verify that given sections are appropriate targets for SECREL
// relocations. This check is relaxed because unfortunately debug
// sections have section-relative relocations against absolute symbols.
static bool checkSecRel(const SectionChunk *Sec, OutputSection *OS) {
if (OS)
return true;
if (Sec->isCodeView())
return false;
fatal("SECREL relocation cannot be applied to absolute symbols");
}
static void applySecRel(const SectionChunk *Sec, uint8_t *Off,
OutputSection *OS, uint64_t S) {
if (!OS) {
if (Sec->isCodeView())
return;
fatal("SECREL relocation cannot be applied to absolute symbols");
}
if (!checkSecRel(Sec, OS))
return;
uint64_t SecRel = S - OS->getRVA();
if (SecRel > UINT32_MAX) {
error("overflow in SECREL relocation in section: " + Sec->getSectionName());
@ -67,10 +74,13 @@ static void applySecRel(const SectionChunk *Sec, uint8_t *Off,
}
static void applySecIdx(uint8_t *Off, OutputSection *OS) {
// If we have no output section, this must be an absolute symbol. Use the
// sentinel absolute symbol section index.
uint16_t SecIdx = OS ? OS->SectionIndex : DefinedAbsolute::OutputSectionIndex;
add16(Off, SecIdx);
// Absolute symbol doesn't have section index, but section index relocation
// against absolute symbol should be resolved to one plus the last output
// section index. This is required for compatibility with MSVC.
if (OS)
add16(Off, OS->SectionIndex);
else
add16(Off, DefinedAbsolute::NumOutputSections + 1);
}
void SectionChunk::applyRelX64(uint8_t *Off, uint16_t Type, OutputSection *OS,
@ -88,7 +98,8 @@ void SectionChunk::applyRelX64(uint8_t *Off, uint16_t Type, OutputSection *OS,
case IMAGE_REL_AMD64_SECTION: applySecIdx(Off, OS); break;
case IMAGE_REL_AMD64_SECREL: applySecRel(this, Off, OS, S); break;
default:
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type));
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type) + " in " +
toString(File));
}
}
@ -102,7 +113,8 @@ void SectionChunk::applyRelX86(uint8_t *Off, uint16_t Type, OutputSection *OS,
case IMAGE_REL_I386_SECTION: applySecIdx(Off, OS); break;
case IMAGE_REL_I386_SECREL: applySecRel(this, Off, OS, S); break;
default:
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type));
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type) + " in " +
toString(File));
}
}
@ -112,11 +124,10 @@ static void applyMOV(uint8_t *Off, uint16_t V) {
}
static uint16_t readMOV(uint8_t *Off) {
uint16_t Opcode1 = read16le(Off);
uint16_t Opcode2 = read16le(Off + 2);
uint16_t Imm = (Opcode2 & 0x00ff) | ((Opcode2 >> 4) & 0x0700);
Imm |= ((Opcode1 << 1) & 0x0800) | ((Opcode1 & 0x000f) << 12);
return Imm;
uint16_t Op1 = read16le(Off);
uint16_t Op2 = read16le(Off + 2);
return (Op2 & 0x00ff) | ((Op2 >> 4) & 0x0700) | ((Op1 << 1) & 0x0800) |
((Op1 & 0x000f) << 12);
}
void applyMOV32T(uint8_t *Off, uint32_t V) {
@ -153,7 +164,7 @@ void SectionChunk::applyRelARM(uint8_t *Off, uint16_t Type, OutputSection *OS,
uint64_t S, uint64_t P) const {
// Pointer to thumb code must have the LSB set.
uint64_t SX = S;
if (OS && (OS->getPermissions() & IMAGE_SCN_MEM_EXECUTE))
if (OS && (OS->Header.Characteristics & IMAGE_SCN_MEM_EXECUTE))
SX |= 1;
switch (Type) {
case IMAGE_REL_ARM_ADDR32: add32(Off, SX + Config->ImageBase); break;
@ -165,18 +176,19 @@ void SectionChunk::applyRelARM(uint8_t *Off, uint16_t Type, OutputSection *OS,
case IMAGE_REL_ARM_SECTION: applySecIdx(Off, OS); break;
case IMAGE_REL_ARM_SECREL: applySecRel(this, Off, OS, S); break;
default:
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type));
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type) + " in " +
toString(File));
}
}
// Interpret the existing immediate value as a byte offset to the
// target symbol, then update the instruction with the immediate as
// the page offset from the current instruction to the target.
static void applyArm64Addr(uint8_t *Off, uint64_t S, uint64_t P) {
static void applyArm64Addr(uint8_t *Off, uint64_t S, uint64_t P, int Shift) {
uint32_t Orig = read32le(Off);
uint64_t Imm = ((Orig >> 29) & 0x3) | ((Orig >> 3) & 0x1FFFFC);
S += Imm;
Imm = (S >> 12) - (P >> 12);
Imm = (S >> Shift) - (P >> Shift);
uint32_t ImmLo = (Imm & 0x3) << 29;
uint32_t ImmHi = (Imm & 0x1FFFFC) << 3;
uint64_t Mask = (0x3 << 29) | (0x1FFFFC << 3);
@ -213,19 +225,70 @@ static void applyArm64Ldr(uint8_t *Off, uint64_t Imm) {
applyArm64Imm(Off, Imm >> Size, Size);
}
static void applySecRelLow12A(const SectionChunk *Sec, uint8_t *Off,
OutputSection *OS, uint64_t S) {
if (checkSecRel(Sec, OS))
applyArm64Imm(Off, (S - OS->getRVA()) & 0xfff, 0);
}
static void applySecRelHigh12A(const SectionChunk *Sec, uint8_t *Off,
OutputSection *OS, uint64_t S) {
if (!checkSecRel(Sec, OS))
return;
uint64_t SecRel = (S - OS->getRVA()) >> 12;
if (0xfff < SecRel) {
error("overflow in SECREL_HIGH12A relocation in section: " +
Sec->getSectionName());
return;
}
applyArm64Imm(Off, SecRel & 0xfff, 0);
}
static void applySecRelLdr(const SectionChunk *Sec, uint8_t *Off,
OutputSection *OS, uint64_t S) {
if (checkSecRel(Sec, OS))
applyArm64Ldr(Off, (S - OS->getRVA()) & 0xfff);
}
static void applyArm64Branch26(uint8_t *Off, int64_t V) {
if (!isInt<28>(V))
fatal("relocation out of range");
or32(Off, (V & 0x0FFFFFFC) >> 2);
}
static void applyArm64Branch19(uint8_t *Off, int64_t V) {
if (!isInt<21>(V))
fatal("relocation out of range");
or32(Off, (V & 0x001FFFFC) << 3);
}
static void applyArm64Branch14(uint8_t *Off, int64_t V) {
if (!isInt<16>(V))
fatal("relocation out of range");
or32(Off, (V & 0x0000FFFC) << 3);
}
void SectionChunk::applyRelARM64(uint8_t *Off, uint16_t Type, OutputSection *OS,
uint64_t S, uint64_t P) const {
switch (Type) {
case IMAGE_REL_ARM64_PAGEBASE_REL21: applyArm64Addr(Off, S, P); break;
case IMAGE_REL_ARM64_PAGEBASE_REL21: applyArm64Addr(Off, S, P, 12); break;
case IMAGE_REL_ARM64_REL21: applyArm64Addr(Off, S, P, 0); break;
case IMAGE_REL_ARM64_PAGEOFFSET_12A: applyArm64Imm(Off, S & 0xfff, 0); break;
case IMAGE_REL_ARM64_PAGEOFFSET_12L: applyArm64Ldr(Off, S & 0xfff); break;
case IMAGE_REL_ARM64_BRANCH26: or32(Off, ((S - P) & 0x0FFFFFFC) >> 2); break;
case IMAGE_REL_ARM64_BRANCH26: applyArm64Branch26(Off, S - P); break;
case IMAGE_REL_ARM64_BRANCH19: applyArm64Branch19(Off, S - P); break;
case IMAGE_REL_ARM64_BRANCH14: applyArm64Branch14(Off, S - P); break;
case IMAGE_REL_ARM64_ADDR32: add32(Off, S + Config->ImageBase); break;
case IMAGE_REL_ARM64_ADDR32NB: add32(Off, S); break;
case IMAGE_REL_ARM64_ADDR64: add64(Off, S + Config->ImageBase); break;
case IMAGE_REL_ARM64_SECREL: applySecRel(this, Off, OS, S); break;
case IMAGE_REL_ARM64_SECREL_LOW12A: applySecRelLow12A(this, Off, OS, S); break;
case IMAGE_REL_ARM64_SECREL_HIGH12A: applySecRelHigh12A(this, Off, OS, S); break;
case IMAGE_REL_ARM64_SECREL_LOW12L: applySecRelLdr(this, Off, OS, S); break;
case IMAGE_REL_ARM64_SECTION: applySecIdx(Off, OS); break;
default:
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type));
fatal("unsupported relocation type 0x" + Twine::utohexstr(Type) + " in " +
toString(File));
}
}
@ -234,7 +297,8 @@ void SectionChunk::writeTo(uint8_t *Buf) const {
return;
// Copy section contents from source object file to output file.
ArrayRef<uint8_t> A = getContents();
memcpy(Buf + OutputSectionOff, A.data(), A.size());
if (!A.empty())
memcpy(Buf + OutputSectionOff, A.data(), A.size());
// Apply relocations.
size_t InputSize = getSize();
@ -350,8 +414,8 @@ bool SectionChunk::hasData() const {
return !(Header->Characteristics & IMAGE_SCN_CNT_UNINITIALIZED_DATA);
}
uint32_t SectionChunk::getPermissions() const {
return Header->Characteristics & PermMask;
uint32_t SectionChunk::getOutputCharacteristics() const {
return Header->Characteristics & (PermMask | TypeMask);
}
bool SectionChunk::isCOMDAT() const {
@ -378,6 +442,7 @@ ArrayRef<uint8_t> SectionChunk::getContents() const {
}
void SectionChunk::replace(SectionChunk *Other) {
Alignment = std::max(Alignment, Other->Alignment);
Other->Repl = Repl;
Other->Live = false;
}
@ -388,7 +453,7 @@ CommonChunk::CommonChunk(const COFFSymbolRef S) : Sym(S) {
Alignment = std::min(uint64_t(32), PowerOf2Ceil(Sym.getValue()));
}
uint32_t CommonChunk::getPermissions() const {
uint32_t CommonChunk::getOutputCharacteristics() const {
return IMAGE_SCN_CNT_UNINITIALIZED_DATA | IMAGE_SCN_MEM_READ |
IMAGE_SCN_MEM_WRITE;
}
@ -433,7 +498,7 @@ void ImportThunkChunkARM::writeTo(uint8_t *Buf) const {
void ImportThunkChunkARM64::writeTo(uint8_t *Buf) const {
int64_t Off = ImpSymbol->getRVA() & 0xfff;
memcpy(Buf + OutputSectionOff, ImportThunkARM64, sizeof(ImportThunkARM64));
applyArm64Addr(Buf + OutputSectionOff, ImpSymbol->getRVA(), RVA);
applyArm64Addr(Buf + OutputSectionOff, ImpSymbol->getRVA(), RVA, 12);
applyArm64Ldr(Buf + OutputSectionOff + 4, Off);
}
@ -453,12 +518,14 @@ void LocalImportChunk::writeTo(uint8_t *Buf) const {
}
}
void SEHTableChunk::writeTo(uint8_t *Buf) const {
void RVATableChunk::writeTo(uint8_t *Buf) const {
ulittle32_t *Begin = reinterpret_cast<ulittle32_t *>(Buf + OutputSectionOff);
size_t Cnt = 0;
for (Defined *D : Syms)
Begin[Cnt++] = D->getRVA();
for (const ChunkAndOffset &CO : Syms)
Begin[Cnt++] = CO.InputChunk->getRVA() + CO.Offset;
std::sort(Begin, Begin + Cnt);
assert(std::unique(Begin, Begin + Cnt) == Begin + Cnt &&
"RVA tables should be de-duplicated");
}
// Windows-specific. This class represents a block in .reloc section.
@ -531,5 +598,47 @@ uint8_t Baserel::getDefaultType() {
}
}
std::map<uint32_t, MergeChunk *> MergeChunk::Instances;
MergeChunk::MergeChunk(uint32_t Alignment)
: Builder(StringTableBuilder::RAW, Alignment) {
this->Alignment = Alignment;
}
void MergeChunk::addSection(SectionChunk *C) {
auto *&MC = Instances[C->Alignment];
if (!MC)
MC = make<MergeChunk>(C->Alignment);
MC->Sections.push_back(C);
}
void MergeChunk::finalizeContents() {
for (SectionChunk *C : Sections)
if (C->isLive())
Builder.add(toStringRef(C->getContents()));
Builder.finalize();
for (SectionChunk *C : Sections) {
if (!C->isLive())
continue;
size_t Off = Builder.getOffset(toStringRef(C->getContents()));
C->setOutputSection(Out);
C->setRVA(RVA + Off);
C->OutputSectionOff = OutputSectionOff + Off;
}
}
uint32_t MergeChunk::getOutputCharacteristics() const {
return IMAGE_SCN_MEM_READ | IMAGE_SCN_CNT_INITIALIZED_DATA;
}
size_t MergeChunk::getSize() const {
return Builder.getSize();
}
void MergeChunk::writeTo(uint8_t *Buf) const {
Builder.write(Buf + OutputSectionOff);
}
} // namespace coff
} // namespace lld

92
contrib/llvm/tools/lld/COFF/Chunks.h
View File

@ -16,6 +16,7 @@
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/iterator.h"
#include "llvm/ADT/iterator_range.h"
#include "llvm/MC/StringTableBuilder.h"
#include "llvm/Object/COFF.h"
#include <utility>
#include <vector>
@ -37,9 +38,11 @@ class ObjFile;
class OutputSection;
class Symbol;
// Mask for section types (code, data, bss, disacardable, etc.)
// and permissions (writable, readable or executable).
const uint32_t PermMask = 0xFF0000F0;
// Mask for permissions (discardable, writable, readable, executable, etc).
const uint32_t PermMask = 0xFE000000;
// Mask for section types (code, data, bss).
const uint32_t TypeMask = 0x000000E0;
// A Chunk represents a chunk of data that will occupy space in the
// output (if the resolver chose that). It may or may not be backed by
@ -60,6 +63,10 @@ class Chunk {
// before calling this function.
virtual void writeTo(uint8_t *Buf) const {}
// Called by the writer after an RVA is assigned, but before calling
// getSize().
virtual void finalizeContents() {}
// The writer sets and uses the addresses.
uint64_t getRVA() const { return RVA; }
void setRVA(uint64_t V) { RVA = V; }
@ -70,7 +77,7 @@ class Chunk {
virtual bool hasData() const { return true; }
// Returns readable/writable/executable bits.
virtual uint32_t getPermissions() const { return 0; }
virtual uint32_t getOutputCharacteristics() const { return 0; }
// Returns the section name if this is a section chunk.
// It is illegal to call this function on non-section chunks.
@ -137,7 +144,7 @@ class SectionChunk final : public Chunk {
ArrayRef<uint8_t> getContents() const;
void writeTo(uint8_t *Buf) const override;
bool hasData() const override;
uint32_t getPermissions() const override;
uint32_t getOutputCharacteristics() const override;
StringRef getSectionName() const override { return SectionName; }
void getBaserels(std::vector<Baserel> *Res) override;
bool isCOMDAT() const;
@ -208,11 +215,11 @@ class SectionChunk final : public Chunk {
// The COMDAT leader symbol if this is a COMDAT chunk.
DefinedRegular *Sym = nullptr;
ArrayRef<coff_relocation> Relocs;
private:
StringRef SectionName;
std::vector<SectionChunk *> AssocChildren;
llvm::iterator_range<const coff_relocation *> Relocs;
size_t NumRelocs;
// Used by the garbage collector.
bool Live;
@ -222,13 +229,40 @@ class SectionChunk final : public Chunk {
uint32_t Class[2] = {0, 0};
};
// This class is used to implement an lld-specific feature (not implemented in
// MSVC) that minimizes the output size by finding string literals sharing tail
// parts and merging them.
//
// If string tail merging is enabled and a section is identified as containing a
// string literal, it is added to a MergeChunk with an appropriate alignment.
// The MergeChunk then tail merges the strings using the StringTableBuilder
// class and assigns RVAs and section offsets to each of the member chunks based
// on the offsets assigned by the StringTableBuilder.
class MergeChunk : public Chunk {
public:
MergeChunk(uint32_t Alignment);
static void addSection(SectionChunk *C);
void finalizeContents() override;
uint32_t getOutputCharacteristics() const override;
StringRef getSectionName() const override { return ".rdata"; }
size_t getSize() const override;
void writeTo(uint8_t *Buf) const override;
static std::map<uint32_t, MergeChunk *> Instances;
std::vector<SectionChunk *> Sections;
private:
llvm::StringTableBuilder Builder;
};
// A chunk for common symbols. Common chunks don't have actual data.
class CommonChunk : public Chunk {
public:
CommonChunk(const COFFSymbolRef Sym);
size_t getSize() const override { return Sym.getValue(); }
bool hasData() const override { return false; }
uint32_t getPermissions() const override;
uint32_t getOutputCharacteristics() const override;
StringRef getSectionName() const override { return ".bss"; }
private:
@ -320,17 +354,41 @@ class LocalImportChunk : public Chunk {
Defined *Sym;
};
// Windows-specific.
// A chunk for SEH table which contains RVAs of safe exception handler
// functions. x86-only.
class SEHTableChunk : public Chunk {
// Duplicate RVAs are not allowed in RVA tables, so unique symbols by chunk and
// offset into the chunk. Order does not matter as the RVA table will be sorted
// later.
struct ChunkAndOffset {
Chunk *InputChunk;
uint32_t Offset;
struct DenseMapInfo {
static ChunkAndOffset getEmptyKey() {
return {llvm::DenseMapInfo<Chunk *>::getEmptyKey(), 0};
}
static ChunkAndOffset getTombstoneKey() {
return {llvm::DenseMapInfo<Chunk *>::getTombstoneKey(), 0};
}
static unsigned getHashValue(const ChunkAndOffset &CO) {
return llvm::DenseMapInfo<std::pair<Chunk *, uint32_t>>::getHashValue(
{CO.InputChunk, CO.Offset});
}
static bool isEqual(const ChunkAndOffset &LHS, const ChunkAndOffset &RHS) {
return LHS.InputChunk == RHS.InputChunk && LHS.Offset == RHS.Offset;
}
};
};
using SymbolRVASet = llvm::DenseSet<ChunkAndOffset>;
// Table which contains symbol RVAs. Used for /safeseh and /guard:cf.
class RVATableChunk : public Chunk {
public:
explicit SEHTableChunk(std::set<Defined *> S) : Syms(std::move(S)) {}
explicit RVATableChunk(SymbolRVASet S) : Syms(std::move(S)) {}
size_t getSize() const override { return Syms.size() * 4; }
void writeTo(uint8_t *Buf) const override;
private:
std::set<Defined *> Syms;
SymbolRVASet Syms;
};
// Windows-specific.
@ -362,4 +420,10 @@ void applyBranch24T(uint8_t *Off, int32_t V);
} // namespace coff
} // namespace lld
namespace llvm {
template <>
struct DenseMapInfo<lld::coff::ChunkAndOffset>
: lld::coff::ChunkAndOffset::DenseMapInfo {};
}
#endif

29
contrib/llvm/tools/lld/COFF/Config.h
View File

@ -10,6 +10,7 @@
#ifndef LLD_COFF_CONFIG_H
#define LLD_COFF_CONFIG_H
#include "llvm/ADT/StringMap.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/Object/COFF.h"
#include "llvm/Support/CachePruning.h"
@ -71,6 +72,12 @@ enum class DebugType {
Fixup = 0x4, /// Relocation Table
};
enum class GuardCFLevel {
Off,
NoLongJmp, // Emit gfids but no longjmp tables
Full, // Enable all protections.
};
// Global configuration.
struct Configuration {
enum ManifestKind { SideBySide, Embed, No };
@ -85,13 +92,19 @@ struct Configuration {
std::string ImportName;
bool DoGC = true;
bool DoICF = true;
bool TailMerge;
bool Relocatable = true;
bool Force = false;
bool Debug = false;
bool DebugDwarf = false;
bool DebugGHashes = false;
bool DebugSymtab = false;
bool ShowTiming = false;
unsigned DebugTypes = static_cast<unsigned>(DebugType::None);
std::vector<std::string> NatvisFiles;
llvm::SmallString<128> PDBAltPath;
llvm::SmallString<128> PDBPath;
llvm::SmallString<128> PDBSourcePath;
std::vector<llvm::StringRef> Argv;
// Symbols in this set are considered as live by the garbage collector.
@ -110,15 +123,18 @@ struct Configuration {
bool SaveTemps = false;
// /guard:cf
GuardCFLevel GuardCF = GuardCFLevel::Off;
// Used for SafeSEH.
Symbol *SEHTable = nullptr;
Symbol *SEHCount = nullptr;
// Used for /opt:lldlto=N
unsigned LTOOptLevel = 2;
unsigned LTOO = 2;
// Used for /opt:lldltojobs=N
unsigned LTOJobs = 0;
unsigned ThinLTOJobs = 0;
// Used for /opt:lldltopartitions=N
unsigned LTOPartitions = 1;
@ -152,6 +168,9 @@ struct Configuration {
// Used for /alternatename.
std::map<StringRef, StringRef> AlternateNames;
// Used for /order.
llvm::StringMap<int> Order;
// Used for /lldmap.
std::string MapFile;
@ -164,7 +183,7 @@ struct Configuration {
uint32_t MinorImageVersion = 0;
uint32_t MajorOSVersion = 6;
uint32_t MinorOSVersion = 0;
bool CanExitEarly = false;
uint32_t Timestamp = 0;
bool DynamicBase = true;
bool AllowBind = true;
bool NxCompat = true;
@ -174,8 +193,12 @@ struct Configuration {
bool HighEntropyVA = false;
bool AppContainer = false;
bool MinGW = false;
bool WarnMissingOrderSymbol = true;
bool WarnLocallyDefinedImported = true;
bool Incremental = true;
bool IntegrityCheck = false;
bool KillAt = false;
bool Repro = false;
};
extern Configuration *Config;

2
contrib/llvm/tools/lld/COFF/DLL.cpp
View File

@ -18,8 +18,8 @@
//
//===----------------------------------------------------------------------===//
#include "Chunks.h"
#include "DLL.h"
#include "Chunks.h"
#include "llvm/Object/COFF.h"
#include "llvm/Support/Endian.h"
#include "llvm/Support/Path.h"

5
contrib/llvm/tools/lld/COFF/DLL.h
View File

@ -76,6 +76,11 @@ class EdataContents {
public:
EdataContents();
std::vector<Chunk *> Chunks;
uint64_t getRVA() { return Chunks[0]->getRVA(); }
uint64_t getSize() {
return Chunks.back()->getRVA() + Chunks.back()->getSize() - getRVA();
}
};
} // namespace coff

429
contrib/llvm/tools/lld/COFF/Driver.cpp
View File

@ -9,14 +9,18 @@
#include "Driver.h"
#include "Config.h"
#include "ICF.h"
#include "InputFiles.h"
#include "MarkLive.h"
#include "MinGW.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "Writer.h"
#include "lld/Common/Args.h"
#include "lld/Common/Driver.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Timer.h"
#include "lld/Common/Version.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/StringSwitch.h"
@ -35,9 +39,8 @@
#include "llvm/Support/raw_ostream.h"
#include "llvm/ToolDrivers/llvm-lib/LibDriver.h"
#include <algorithm>
#include <memory>
#include <future>
#include <memory>
using namespace llvm;
using namespace llvm::object;
@ -47,20 +50,20 @@ using llvm::sys::Process;
namespace lld {
namespace coff {
static Timer InputFileTimer("Input File Reading", Timer::root());
Configuration *Config;
LinkerDriver *Driver;
bool link(ArrayRef<const char *> Args, bool CanExitEarly, raw_ostream &Diag) {
errorHandler().LogName = Args[0];
errorHandler().LogName = sys::path::filename(Args[0]);
errorHandler().ErrorOS = &Diag;
errorHandler().ColorDiagnostics = Diag.has_colors();
errorHandler().ErrorLimitExceededMsg =
"too many errors emitted, stopping now"
" (use /ERRORLIMIT:0 to see all errors)";
" (use /errorlimit:0 to see all errors)";
errorHandler().ExitEarly = CanExitEarly;
Config = make<Configuration>();
Config->Argv = {Args.begin(), Args.end()};
Config->CanExitEarly = CanExitEarly;
Symtab = make<SymbolTable>();
@ -72,6 +75,9 @@ bool link(ArrayRef<const char *> Args, bool CanExitEarly, raw_ostream &Diag) {
exitLld(errorCount() ? 1 : 0);
freeArena();
ObjFile::Instances.clear();
ImportFile::Instances.clear();
BitcodeFile::Instances.clear();
return !errorCount();
}
@ -93,7 +99,7 @@ typedef std::pair<std::unique_ptr<MemoryBuffer>, std::error_code> MBErrPair;
// Create a std::future that opens and maps a file using the best strategy for
// the host platform.
static std::future<MBErrPair> createFutureForFile(std::string Path) {
#if LLVM_ON_WIN32
#if _WIN32
// On Windows, file I/O is relatively slow so it is best to do this
// asynchronously.
auto Strategy = std::launch::async;
@ -101,7 +107,9 @@ static std::future<MBErrPair> createFutureForFile(std::string Path) {
auto Strategy = std::launch::deferred;
#endif
return std::async(Strategy, [=]() {
auto MBOrErr = MemoryBuffer::getFile(Path);
auto MBOrErr = MemoryBuffer::getFile(Path,
/*FileSize*/ -1,
/*RequiresNullTerminator*/ false);
if (!MBOrErr)
return MBErrPair{nullptr, MBOrErr.getError()};
return MBErrPair{std::move(*MBOrErr), std::error_code()};
@ -120,40 +128,47 @@ MemoryBufferRef LinkerDriver::takeBuffer(std::unique_ptr<MemoryBuffer> MB) {
void LinkerDriver::addBuffer(std::unique_ptr<MemoryBuffer> MB,
bool WholeArchive) {
StringRef Filename = MB->getBufferIdentifier();
MemoryBufferRef MBRef = takeBuffer(std::move(MB));
FilePaths.push_back(MBRef.getBufferIdentifier());
FilePaths.push_back(Filename);
// File type is detected by contents, not by file extension.
switch (identify_magic(MBRef.getBuffer())) {
case file_magic::windows_resource:
Resources.push_back(MBRef);
break;
case file_magic::archive:
if (WholeArchive) {
std::unique_ptr<Archive> File =
CHECK(Archive::create(MBRef),
MBRef.getBufferIdentifier() + ": failed to parse archive");
CHECK(Archive::create(MBRef), Filename + ": failed to parse archive");
for (MemoryBufferRef M : getArchiveMembers(File.get()))
addArchiveBuffer(M, "<whole-archive>", MBRef.getBufferIdentifier());
addArchiveBuffer(M, "<whole-archive>", Filename);
return;
}
Symtab->addFile(make<ArchiveFile>(MBRef));
break;
case file_magic::bitcode:
Symtab->addFile(make<BitcodeFile>(MBRef));
break;
case file_magic::coff_cl_gl_object:
error(MBRef.getBufferIdentifier() + ": is not a native COFF file. "
"Recompile without /GL");
break;
default:
case file_magic::coff_object:
case file_magic::coff_import_library:
Symtab->addFile(make<ObjFile>(MBRef));
break;
case file_magic::coff_cl_gl_object:
error(Filename + ": is not a native COFF file. Recompile without /GL");
break;
case file_magic::pecoff_executable:
if (Filename.endswith_lower(".dll")) {
error(Filename + ": bad file type. Did you specify a DLL instead of an "
"import library?");
break;
}
LLVM_FALLTHROUGH;
default:
error(MBRef.getBufferIdentifier() + ": unknown file type");
break;
}
}
@ -228,7 +243,29 @@ static bool isDecorated(StringRef Sym) {
void LinkerDriver::parseDirectives(StringRef S) {
ArgParser Parser;
// .drectve is always tokenized using Windows shell rules.
opt::InputArgList Args = Parser.parseDirectives(S);
// /EXPORT: option can appear too many times, processing in fastpath.
opt::InputArgList Args;
std::vector<StringRef> Exports;
std::tie(Args, Exports) = Parser.parseDirectives(S);
for (StringRef E : Exports) {
// If a common header file contains dllexported function
// declarations, many object files may end up with having the
// same /EXPORT options. In order to save cost of parsing them,
// we dedup them first.
if (!DirectivesExports.insert(E).second)
continue;
Export Exp = parseExport(E);
if (Config->Machine == I386 && Config->MinGW) {
if (!isDecorated(Exp.Name))
Exp.Name = Saver.save("_" + Exp.Name);
if (!Exp.ExtName.empty() && !isDecorated(Exp.ExtName))
Exp.ExtName = Saver.save("_" + Exp.ExtName);
}
Exp.Directives = true;
Config->Exports.push_back(Exp);
}
for (auto *Arg : Args) {
switch (Arg->getOption().getUnaliasedOption().getID()) {
@ -245,25 +282,6 @@ void LinkerDriver::parseDirectives(StringRef S) {
case OPT_entry:
Config->Entry = addUndefined(mangle(Arg->getValue()));
break;
case OPT_export: {
// If a common header file contains dllexported function
// declarations, many object files may end up with having the
// same /EXPORT options. In order to save cost of parsing them,
// we dedup them first.
if (!DirectivesExports.insert(Arg->getValue()).second)
break;
Export E = parseExport(Arg->getValue());
if (Config->Machine == I386 && Config->MinGW) {
if (!isDecorated(E.Name))
E.Name = Saver.save("_" + E.Name);
if (!E.ExtName.empty() && !isDecorated(E.ExtName))
E.ExtName = Saver.save("_" + E.ExtName);
}
E.Directives = true;
Config->Exports.push_back(E);
break;
}
case OPT_failifmismatch:
checkFailIfMismatch(Arg->getValue());
break;
@ -316,13 +334,24 @@ StringRef LinkerDriver::doFindFile(StringRef Filename) {
return Filename;
}
static Optional<sys::fs::UniqueID> getUniqueID(StringRef Path) {
sys::fs::UniqueID Ret;
if (sys::fs::getUniqueID(Path, Ret))
return None;
return Ret;
}
// Resolves a file path. This never returns the same path
// (in that case, it returns None).
Optional<StringRef> LinkerDriver::findFile(StringRef Filename) {
StringRef Path = doFindFile(Filename);
bool Seen = !VisitedFiles.insert(Path.lower()).second;
if (Seen)
return None;
if (Optional<sys::fs::UniqueID> ID = getUniqueID(Path)) {
bool Seen = !VisitedFiles.insert(*ID).second;
if (Seen)
return None;
}
if (Path.endswith_lower(".lib"))
VisitedLibs.insert(sys::path::filename(Path));
return Path;
@ -345,11 +374,14 @@ Optional<StringRef> LinkerDriver::findLib(StringRef Filename) {
return None;
if (!VisitedLibs.insert(Filename.lower()).second)
return None;
StringRef Path = doFindLib(Filename);
if (Config->NoDefaultLibs.count(Path))
return None;
if (!VisitedFiles.insert(Path.lower()).second)
return None;
if (Optional<sys::fs::UniqueID> ID = getUniqueID(Path))
if (!VisitedFiles.insert(*ID).second)
return None;
return Path;
}
@ -384,7 +416,24 @@ StringRef LinkerDriver::mangle(StringRef Sym) {
}
// Windows specific -- find default entry point name.
//
// There are four different entry point functions for Windows executables,
// each of which corresponds to a user-defined "main" function. This function
// infers an entry point from a user-defined "main" function.
StringRef LinkerDriver::findDefaultEntry() {
// As a special case, if /nodefaultlib is given, we directly look for an
// entry point. This is because, if no default library is linked, users
// need to define an entry point instead of a "main".
if (Config->NoDefaultLibAll) {
for (StringRef S : {"mainCRTStartup", "wmainCRTStartup",
"WinMainCRTStartup", "wWinMainCRTStartup"}) {
StringRef Entry = Symtab->findMangle(S);
if (!Entry.empty() && !isa<Undefined>(Symtab->find(Entry)))
return mangle(S);
}
return "";
}
// User-defined main functions and their corresponding entry points.
static const char *Entries[][2] = {
{"main", "mainCRTStartup"},
@ -534,10 +583,49 @@ static void createImportLibrary(bool AsLib) {
Exports.push_back(E2);
}
auto E = writeImportLibrary(getImportName(AsLib), getImplibPath(), Exports,
Config->Machine, false);
handleAllErrors(std::move(E),
[&](ErrorInfoBase &EIB) { error(EIB.message()); });
auto HandleError = [](Error &&E) {
handleAllErrors(std::move(E),
[](ErrorInfoBase &EIB) { error(EIB.message()); });
};
std::string LibName = getImportName(AsLib);
std::string Path = getImplibPath();
if (!Config->Incremental) {
HandleError(writeImportLibrary(LibName, Path, Exports, Config->Machine,
Config->MinGW));
return;
}
// If the import library already exists, replace it only if the contents
// have changed.
ErrorOr<std::unique_ptr<MemoryBuffer>> OldBuf = MemoryBuffer::getFile(
Path, /*FileSize*/ -1, /*RequiresNullTerminator*/ false);
if (!OldBuf) {
HandleError(writeImportLibrary(LibName, Path, Exports, Config->Machine,
Config->MinGW));
return;
}
SmallString<128> TmpName;
if (std::error_code EC =
sys::fs::createUniqueFile(Path + ".tmp-%%%%%%%%.lib", TmpName))
fatal("cannot create temporary file for import library " + Path + ": " +
EC.message());
if (Error E = writeImportLibrary(LibName, TmpName, Exports, Config->Machine,
Config->MinGW)) {
HandleError(std::move(E));
return;
}
std::unique_ptr<MemoryBuffer> NewBuf = check(MemoryBuffer::getFile(
TmpName, /*FileSize*/ -1, /*RequiresNullTerminator*/ false));
if ((*OldBuf)->getBuffer() != NewBuf->getBuffer()) {
OldBuf->reset();
HandleError(errorCodeToError(sys::fs::rename(TmpName, Path)));
} else {
sys::fs::remove(TmpName);
}
}
static void parseModuleDefs(StringRef Path) {
@ -570,9 +658,18 @@ static void parseModuleDefs(StringRef Path) {
for (COFFShortExport E1 : M.Exports) {
Export E2;
// In simple cases, only Name is set. Renamed exports are parsed
// and set as "ExtName = Name". If Name has the form "OtherDll.Func",
// it shouldn't be a normal exported function but a forward to another
// DLL instead. This is supported by both MS and GNU linkers.
if (E1.ExtName != E1.Name && StringRef(E1.Name).contains('.')) {
E2.Name = Saver.save(E1.ExtName);
E2.ForwardTo = Saver.save(E1.Name);
Config->Exports.push_back(E2);
continue;
}
E2.Name = Saver.save(E1.Name);
if (E1.isWeak())
E2.ExtName = Saver.save(E1.ExtName);
E2.ExtName = Saver.save(E1.ExtName);
E2.Ordinal = E1.Ordinal;
E2.Noname = E1.Noname;
E2.Data = E1.Data;
@ -635,8 +732,8 @@ filterBitcodeFiles(StringRef Path, std::vector<std::string> &TemporaryFiles) {
log("Creating a temporary archive for " + Path + " to remove bitcode files");
SmallString<128> S;
if (auto EC = sys::fs::createTemporaryFile("lld-" + sys::path::stem(Path),
".lib", S))
if (std::error_code EC = sys::fs::createTemporaryFile(
"lld-" + sys::path::stem(Path), ".lib", S))
fatal("cannot create a temporary file: " + EC.message());
std::string Temp = S.str();
TemporaryFiles.push_back(Temp);
@ -712,6 +809,8 @@ void LinkerDriver::enqueueTask(std::function<void()> Task) {
}
bool LinkerDriver::run() {
ScopedTimer T(InputFileTimer);
bool DidWork = !TaskQueue.empty();
while (!TaskQueue.empty()) {
TaskQueue.front()();
@ -720,6 +819,46 @@ bool LinkerDriver::run() {
return DidWork;
}
// Parse an /order file. If an option is given, the linker places
// COMDAT sections in the same order as their names appear in the
// given file.
static void parseOrderFile(StringRef Arg) {
// For some reason, the MSVC linker requires a filename to be
// preceded by "@".
if (!Arg.startswith("@")) {
error("malformed /order option: '@' missing");
return;
}
// Get a list of all comdat sections for error checking.
DenseSet<StringRef> Set;
for (Chunk *C : Symtab->getChunks())
if (auto *Sec = dyn_cast<SectionChunk>(C))
if (Sec->Sym)
Set.insert(Sec->Sym->getName());
// Open a file.
StringRef Path = Arg.substr(1);
std::unique_ptr<MemoryBuffer> MB = CHECK(
MemoryBuffer::getFile(Path, -1, false, true), "could not open " + Path);
// Parse a file. An order file contains one symbol per line.
// All symbols that were not present in a given order file are
// considered to have the lowest priority 0 and are placed at
// end of an output section.
for (std::string S : args::getLines(MB->getMemBufferRef())) {
if (Config->Machine == I386 && !isDecorated(S))
S = "_" + S;
if (Set.count(S) == 0) {
if (Config->WarnMissingOrderSymbol)
warn("/order:" + Arg + ": missing symbol: " + S + " [LNK4037]");
}
else
Config->Order[S] = INT_MIN + Config->Order.size();
}
}
void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
// If the first command line argument is "/lib", link.exe acts like lib.exe.
// We call our own implementation of lib.exe that understands bitcode files.
@ -735,11 +874,10 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
InitializeAllTargetMCs();
InitializeAllAsmParsers();
InitializeAllAsmPrinters();
InitializeAllDisassemblers();
// Parse command line options.
ArgParser Parser;
opt::InputArgList Args = Parser.parseLINK(ArgsArr.slice(1));
opt::InputArgList Args = Parser.parseLINK(ArgsArr);
// Parse and evaluate -mllvm options.
std::vector<const char *> V;
@ -763,6 +901,10 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
return;
}
if (Args.hasArg(OPT_show_timing))
Config->ShowTiming = true;
ScopedTimer T(Timer::root());
// Handle --version, which is an lld extension. This option is a bit odd
// because it doesn't start with "/", but we deliberately chose "--" to
// avoid conflict with /version and for compatibility with clang-cl.
@ -805,7 +947,9 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
// Handle /ignore
for (auto *Arg : Args.filtered(OPT_ignore)) {
if (StringRef(Arg->getValue()) == "4217")
if (StringRef(Arg->getValue()) == "4037")
Config->WarnMissingOrderSymbol = false;
else if (StringRef(Arg->getValue()) == "4217")
Config->WarnLocallyDefinedImported = false;
// Other warning numbers are ignored.
}
@ -826,6 +970,7 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
// Handle /debug
if (Args.hasArg(OPT_debug, OPT_debug_dwarf, OPT_debug_ghash)) {
Config->Debug = true;
Config->Incremental = true;
if (auto *Arg = Args.getLastArg(OPT_debugtype))
Config->DebugTypes = parseDebugType(Arg->getValue());
else
@ -834,9 +979,17 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
// Handle /pdb
bool ShouldCreatePDB = Args.hasArg(OPT_debug, OPT_debug_ghash);
if (ShouldCreatePDB)
if (ShouldCreatePDB) {
if (auto *Arg = Args.getLastArg(OPT_pdb))
Config->PDBPath = Arg->getValue();
if (auto *Arg = Args.getLastArg(OPT_pdbaltpath))
Config->PDBAltPath = Arg->getValue();
if (Args.hasArg(OPT_natvis))
Config->NatvisFiles = Args.getAllArgValues(OPT_natvis);
if (auto *Arg = Args.getLastArg(OPT_pdb_source_path))
Config->PDBSourcePath = Arg->getValue();
}
// Handle /noentry
if (Args.hasArg(OPT_noentry)) {
@ -860,6 +1013,9 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
DynamicBaseArg->getOption().getID() == OPT_dynamicbase_no)
Config->DynamicBase = false;
// MSDN claims "/FIXED:NO is the default setting for a DLL, and /FIXED is the
// default setting for any other project type.", but link.exe defaults to
// /FIXED:NO for exe outputs as well. Match behavior, not docs.
bool Fixed = Args.hasFlag(OPT_fixed, OPT_fixed_no, false);
if (Fixed) {
if (DynamicBaseArg &&
@ -895,6 +1051,10 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
if (auto *Arg = Args.getLastArg(OPT_stack))
parseNumbers(Arg->getValue(), &Config->StackReserve, &Config->StackCommit);
// Handle /guard:cf
if (auto *Arg = Args.getLastArg(OPT_guard))
parseGuard(Arg->getValue());
// Handle /heap
if (auto *Arg = Args.getLastArg(OPT_heap))
parseNumbers(Arg->getValue(), &Config->HeapReserve, &Config->HeapCommit);
@ -909,6 +1069,23 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
parseSubsystem(Arg->getValue(), &Config->Subsystem, &Config->MajorOSVersion,
&Config->MinorOSVersion);
// Handle /timestamp
if (llvm::opt::Arg *Arg = Args.getLastArg(OPT_timestamp, OPT_repro)) {
if (Arg->getOption().getID() == OPT_repro) {
Config->Timestamp = 0;
Config->Repro = true;
} else {
Config->Repro = false;
StringRef Value(Arg->getValue());
if (Value.getAsInteger(0, Config->Timestamp))
fatal(Twine("invalid timestamp: ") + Value +
". Expected 32-bit integer");
}
} else {
Config->Repro = false;
Config->Timestamp = time(nullptr);
}
// Handle /alternatename
for (auto *Arg : Args.filtered(OPT_alternatename))
parseAlternateName(Arg->getValue());
@ -922,8 +1099,10 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
Config->Implib = Arg->getValue();
// Handle /opt.
bool DoGC = !Args.hasArg(OPT_debug);
unsigned ICFLevel = 1; // 0: off, 1: limited, 2: on
bool DoGC = !Args.hasArg(OPT_debug) || Args.hasArg(OPT_profile);
unsigned ICFLevel =
Args.hasArg(OPT_profile) ? 0 : 1; // 0: off, 1: limited, 2: on
unsigned TailMerge = 1;
for (auto *Arg : Args.filtered(OPT_opt)) {
std::string Str = StringRef(Arg->getValue()).lower();
SmallVector<StringRef, 1> Vec;
@ -937,14 +1116,18 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
ICFLevel = 2;
} else if (S == "noicf") {
ICFLevel = 0;
} else if (S == "lldtailmerge") {
TailMerge = 2;
} else if (S == "nolldtailmerge") {
TailMerge = 0;
} else if (S.startswith("lldlto=")) {
StringRef OptLevel = S.substr(7);
if (OptLevel.getAsInteger(10, Config->LTOOptLevel) ||
Config->LTOOptLevel > 3)
if (OptLevel.getAsInteger(10, Config->LTOO) || Config->LTOO > 3)
error("/opt:lldlto: invalid optimization level: " + OptLevel);
} else if (S.startswith("lldltojobs=")) {
StringRef Jobs = S.substr(11);
if (Jobs.getAsInteger(10, Config->LTOJobs) || Config->LTOJobs == 0)
if (Jobs.getAsInteger(10, Config->ThinLTOJobs) ||
Config->ThinLTOJobs == 0)
error("/opt:lldltojobs: invalid job count: " + Jobs);
} else if (S.startswith("lldltopartitions=")) {
StringRef N = S.substr(17);
@ -965,6 +1148,7 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
ICFLevel = 0;
Config->DoGC = DoGC;
Config->DoICF = ICFLevel > 0;
Config->TailMerge = (TailMerge == 1 && Config->DoICF) || TailMerge == 2;
// Handle /lldsavetemps
if (Args.hasArg(OPT_lldsavetemps))
@ -992,6 +1176,14 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
for (auto *Arg : Args.filtered(OPT_merge))
parseMerge(Arg->getValue());
// Add default section merging rules after user rules. User rules take
// precedence, but we will emit a warning if there is a conflict.
parseMerge(".idata=.rdata");
parseMerge(".didat=.rdata");
parseMerge(".edata=.rdata");
parseMerge(".xdata=.rdata");
parseMerge(".bss=.data");
// Handle /section
for (auto *Arg : Args.filtered(OPT_section))
parseSection(Arg->getValue());
@ -1029,39 +1221,77 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
if (!Config->ManifestInput.empty() &&
Config->Manifest != Configuration::Embed) {
fatal("/MANIFESTINPUT: requires /MANIFEST:EMBED");
fatal("/manifestinput: requires /manifest:embed");
}
// Handle miscellaneous boolean flags.
Config->AllowBind = Args.hasFlag(OPT_allowbind, OPT_allowbind_no, true);
Config->AllowIsolation =
Args.hasFlag(OPT_allowisolation, OPT_allowisolation_no, true);
Config->Incremental =
Args.hasFlag(OPT_incremental, OPT_incremental_no,
!Config->DoGC && !Config->DoICF && !Args.hasArg(OPT_order) &&
!Args.hasArg(OPT_profile));
Config->IntegrityCheck =
Args.hasFlag(OPT_integritycheck, OPT_integritycheck_no, false);
Config->NxCompat = Args.hasFlag(OPT_nxcompat, OPT_nxcompat_no, true);
Config->TerminalServerAware = Args.hasFlag(OPT_tsaware, OPT_tsaware_no, true);
Config->TerminalServerAware =
!Config->DLL && Args.hasFlag(OPT_tsaware, OPT_tsaware_no, true);
Config->DebugDwarf = Args.hasArg(OPT_debug_dwarf);
Config->DebugGHashes = Args.hasArg(OPT_debug_ghash);
Config->DebugSymtab = Args.hasArg(OPT_debug_symtab);
Config->MapFile = getMapFile(Args);
if (Config->Incremental && Args.hasArg(OPT_profile)) {
warn("ignoring '/incremental' due to '/profile' specification");
Config->Incremental = false;
}
if (Config->Incremental && Args.hasArg(OPT_order)) {
warn("ignoring '/incremental' due to '/order' specification");
Config->Incremental = false;
}
if (Config->Incremental && Config->DoGC) {
warn("ignoring '/incremental' because REF is enabled; use '/opt:noref' to "
"disable");
Config->Incremental = false;
}
if (Config->Incremental && Config->DoICF) {
warn("ignoring '/incremental' because ICF is enabled; use '/opt:noicf' to "
"disable");
Config->Incremental = false;
}
if (errorCount())
return;
bool WholeArchiveFlag = Args.hasArg(OPT_wholearchive_flag);
std::set<sys::fs::UniqueID> WholeArchives;
for (auto *Arg : Args.filtered(OPT_wholearchive_file))
if (Optional<StringRef> Path = doFindFile(Arg->getValue()))
if (Optional<sys::fs::UniqueID> ID = getUniqueID(*Path))
WholeArchives.insert(*ID);
// A predicate returning true if a given path is an argument for
// /wholearchive:, or /wholearchive is enabled globally.
// This function is a bit tricky because "foo.obj /wholearchive:././foo.obj"
// needs to be handled as "/wholearchive:foo.obj foo.obj".
auto IsWholeArchive = [&](StringRef Path) -> bool {
if (Args.hasArg(OPT_wholearchive_flag))
return true;
if (Optional<sys::fs::UniqueID> ID = getUniqueID(Path))
return WholeArchives.count(*ID);
return false;
};
// Create a list of input files. Files can be given as arguments
// for /defaultlib option.
std::vector<MemoryBufferRef> MBs;
for (auto *Arg : Args.filtered(OPT_INPUT, OPT_wholearchive_file)) {
switch (Arg->getOption().getID()) {
case OPT_INPUT:
if (Optional<StringRef> Path = findFile(Arg->getValue()))
enqueuePath(*Path, WholeArchiveFlag);
break;
case OPT_wholearchive_file:
if (Optional<StringRef> Path = findFile(Arg->getValue()))
enqueuePath(*Path, true);
break;
}
}
for (auto *Arg : Args.filtered(OPT_INPUT, OPT_wholearchive_file))
if (Optional<StringRef> Path = findFile(Arg->getValue()))
enqueuePath(*Path, IsWholeArchive(*Path));
for (auto *Arg : Args.filtered(OPT_defaultlib))
if (Optional<StringRef> Path = findLib(Arg->getValue()))
enqueuePath(*Path, false);
@ -1165,10 +1395,24 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
getOutputPath((*Args.filtered(OPT_INPUT).begin())->getValue());
}
// Put the PDB next to the image if no /pdb flag was passed.
if (ShouldCreatePDB && Config->PDBPath.empty()) {
Config->PDBPath = Config->OutputFile;
sys::path::replace_extension(Config->PDBPath, ".pdb");
if (ShouldCreatePDB) {
// Put the PDB next to the image if no /pdb flag was passed.
if (Config->PDBPath.empty()) {
Config->PDBPath = Config->OutputFile;
sys::path::replace_extension(Config->PDBPath, ".pdb");
}
// The embedded PDB path should be the absolute path to the PDB if no
// /pdbaltpath flag was passed.
if (Config->PDBAltPath.empty()) {
Config->PDBAltPath = Config->PDBPath;
// It's important to make the path absolute and remove dots. This path
// will eventually be written into the PE header, and certain Microsoft
// tools won't work correctly if these assumptions are not held.
sys::fs::make_absolute(Config->PDBAltPath);
sys::path::remove_dots(Config->PDBAltPath);
}
}
// Set default image base if /base is not given.
@ -1181,11 +1425,9 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
Symtab->addAbsolute("___safe_se_handler_count", 0);
}
// We do not support /guard:cf (control flow protection) yet.
// Define CFG symbols anyway so that we can link MSVC 2015 CRT.
Symtab->addAbsolute(mangle("__guard_fids_count"), 0);
Symtab->addAbsolute(mangle("__guard_fids_table"), 0);
Symtab->addAbsolute(mangle("__guard_flags"), 0x100);
Symtab->addAbsolute(mangle("__guard_flags"), 0);
Symtab->addAbsolute(mangle("__guard_iat_count"), 0);
Symtab->addAbsolute(mangle("__guard_iat_table"), 0);
Symtab->addAbsolute(mangle("__guard_longjmp_count"), 0);
@ -1260,7 +1502,7 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
// Handle /safeseh.
if (Args.hasFlag(OPT_safeseh, OPT_safeseh_no, false)) {
for (ObjFile *File : ObjFile::Instances)
if (!File->SEHCompat)
if (!File->hasSafeSEH())
error("/safeseh: " + File->getName() + " is not compatible with SEH");
if (errorCount())
return;
@ -1280,7 +1522,7 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
E.Name = Def->getName();
E.Sym = Def;
if (Def->getChunk() &&
!(Def->getChunk()->getPermissions() & IMAGE_SCN_MEM_EXECUTE))
!(Def->getChunk()->getOutputCharacteristics() & IMAGE_SCN_MEM_EXECUTE))
E.Data = true;
Config->Exports.push_back(E);
});
@ -1323,6 +1565,12 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
if (Config->Manifest == Configuration::SideBySide)
createSideBySideManifest();
// Handle /order. We want to do this at this moment because we
// need a complete list of comdat sections to warn on nonexistent
// functions.
if (auto *Arg = Args.getLastArg(OPT_order))
parseOrderFile(Arg->getValue());
// Identify unreferenced COMDAT sections.
if (Config->DoGC)
markLive(Symtab->getChunks());
@ -1333,6 +1581,11 @@ void LinkerDriver::link(ArrayRef<const char *> ArgsArr) {
// Write the result.
writeResult();
// Stop early so we can print the results.
Timer::root().stop();
if (Config->ShowTiming)
Timer::root().print();
}
} // namespace coff

21
contrib/llvm/tools/lld/COFF/Driver.h
View File

@ -21,6 +21,7 @@
#include "llvm/Object/COFF.h"
#include "llvm/Option/Arg.h"
#include "llvm/Option/ArgList.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/TarWriter.h"
#include <memory>
#include <set>
@ -36,12 +37,6 @@ using llvm::COFF::MachineTypes;
using llvm::COFF::WindowsSubsystem;
using llvm::Optional;
// Implemented in MarkLive.cpp.
void markLive(ArrayRef<Chunk *> Chunks);
// Implemented in ICF.cpp.
void doICF(ArrayRef<Chunk *> Chunks);
class COFFOptTable : public llvm::opt::OptTable {
public:
COFFOptTable();
@ -56,8 +51,10 @@ class ArgParser {
llvm::opt::InputArgList parse(StringRef S) { return parse(tokenize(S)); }
// Tokenizes a given string and then parses as command line options in
// .drectve section.
llvm::opt::InputArgList parseDirectives(StringRef S);
// .drectve section. /EXPORT options are returned in second element
// to be processed in fastpath.
std::pair<llvm::opt::InputArgList, std::vector<StringRef>>
parseDirectives(StringRef S);
private:
// Parses command line options.
@ -98,7 +95,11 @@ class LinkerDriver {
// Library search path. The first element is always "" (current directory).
std::vector<StringRef> SearchPaths;
std::set<std::string> VisitedFiles;
// We don't want to add the same file more than once.
// Files are uniquified by their filesystem and file number.
std::set<llvm::sys::fs::UniqueID> VisitedFiles;
std::set<std::string> VisitedLibs;
Symbol *addUndefined(StringRef Sym);
@ -143,6 +144,8 @@ StringRef machineToStr(MachineTypes MT);
// Parses a string in the form of "<integer>[,<integer>]".
void parseNumbers(StringRef Arg, uint64_t *Addr, uint64_t *Size = nullptr);
void parseGuard(StringRef Arg);
// Parses a string in the form of "<integer>[.<integer>]".
// Minor's default value is 0.
void parseVersion(StringRef Arg, uint32_t *Major, uint32_t *Minor);

115
contrib/llvm/tools/lld/COFF/DriverUtils.cpp
View File

@ -61,12 +61,7 @@ class Executor {
StringRef Exe = Saver.save(*ExeOrErr);
Args.insert(Args.begin(), Exe);
std::vector<const char *> Vec;
for (StringRef S : Args)
Vec.push_back(S.data());
Vec.push_back(nullptr);
if (sys::ExecuteAndWait(Args[0], Vec.data()) != 0)
if (sys::ExecuteAndWait(Args[0], Args) != 0)
fatal("ExecuteAndWait failed: " +
llvm::join(Args.begin(), Args.end(), " "));
}
@ -128,6 +123,21 @@ void parseVersion(StringRef Arg, uint32_t *Major, uint32_t *Minor) {
fatal("invalid number: " + S2);
}
void parseGuard(StringRef FullArg) {
SmallVector<StringRef, 1> SplitArgs;
FullArg.split(SplitArgs, ",");
for (StringRef Arg : SplitArgs) {
if (Arg.equals_lower("no"))
Config->GuardCF = GuardCFLevel::Off;
else if (Arg.equals_lower("nolongjmp"))
Config->GuardCF = GuardCFLevel::NoLongJmp;
else if (Arg.equals_lower("cf") || Arg.equals_lower("longjmp"))
Config->GuardCF = GuardCFLevel::Full;
else
fatal("invalid argument to /guard: " + Arg);
}
}
// Parses a string in the form of "<subsystem>[,<integer>[.<integer>]]".
void parseSubsystem(StringRef Arg, WindowsSubsystem *Sys, uint32_t *Major,
uint32_t *Minor) {
@ -170,6 +180,10 @@ void parseMerge(StringRef S) {
std::tie(From, To) = S.split('=');
if (From.empty() || To.empty())
fatal("/merge: invalid argument: " + S);
if (From == ".rsrc" || To == ".rsrc")
fatal("/merge: cannot merge '.rsrc' with any section");
if (From == ".reloc" || To == ".reloc")
fatal("/merge: cannot merge '.reloc' with any section");
auto Pair = Config->Merge.insert(std::make_pair(From, To));
bool Inserted = Pair.second;
if (!Inserted) {
@ -418,15 +432,15 @@ static std::string createManifestXml() {
return createManifestXmlWithExternalMt(DefaultXml);
}
static std::unique_ptr<MemoryBuffer>
static std::unique_ptr<WritableMemoryBuffer>
createMemoryBufferForManifestRes(size_t ManifestSize) {
size_t ResSize = alignTo(
object::WIN_RES_MAGIC_SIZE + object::WIN_RES_NULL_ENTRY_SIZE +
sizeof(object::WinResHeaderPrefix) + sizeof(object::WinResIDs) +
sizeof(object::WinResHeaderSuffix) + ManifestSize,
object::WIN_RES_DATA_ALIGNMENT);
return MemoryBuffer::getNewMemBuffer(ResSize,
Config->OutputFile + ".manifest.res");
return WritableMemoryBuffer::getNewMemBuffer(ResSize, Config->OutputFile +
".manifest.res");
}
static void writeResFileHeader(char *&Buf) {
@ -465,16 +479,16 @@ static void writeResEntryHeader(char *&Buf, size_t ManifestSize) {
std::unique_ptr<MemoryBuffer> createManifestRes() {
std::string Manifest = createManifestXml();
std::unique_ptr<MemoryBuffer> Res =
std::unique_ptr<WritableMemoryBuffer> Res =
createMemoryBufferForManifestRes(Manifest.size());
char *Buf = const_cast<char *>(Res->getBufferStart());
char *Buf = Res->getBufferStart();
writeResFileHeader(Buf);
writeResEntryHeader(Buf, Manifest.size());
// Copy the manifest data into the .res file.
std::copy(Manifest.begin(), Manifest.end(), Buf);
return Res;
return std::move(Res);
}
void createSideBySideManifest() {
@ -558,6 +572,12 @@ Export parseExport(StringRef Arg) {
static StringRef undecorate(StringRef Sym) {
if (Config->Machine != I386)
return Sym;
// In MSVC mode, a fully decorated stdcall function is exported
// as-is with the leading underscore (with type IMPORT_NAME).
// In MinGW mode, a decorated stdcall function gets the underscore
// removed, just like normal cdecl functions.
if (Sym.startswith("_") && Sym.contains('@') && !Config->MinGW)
return Sym;
return Sym.startswith("_") ? Sym.substr(1) : Sym;
}
@ -731,6 +751,28 @@ static const llvm::opt::OptTable::Info InfoTable[] = {
COFFOptTable::COFFOptTable() : OptTable(InfoTable, true) {}
// Set color diagnostics according to --color-diagnostics={auto,always,never}
// or --no-color-diagnostics flags.
static void handleColorDiagnostics(opt::InputArgList &Args) {
auto *Arg = Args.getLastArg(OPT_color_diagnostics, OPT_color_diagnostics_eq,
OPT_no_color_diagnostics);
if (!Arg)
return;
if (Arg->getOption().getID() == OPT_color_diagnostics) {
errorHandler().ColorDiagnostics = true;
} else if (Arg->getOption().getID() == OPT_no_color_diagnostics) {
errorHandler().ColorDiagnostics = false;
} else {
StringRef S = Arg->getValue();
if (S == "always")
errorHandler().ColorDiagnostics = true;
else if (S == "never")
errorHandler().ColorDiagnostics = false;
else if (S != "auto")
error("unknown option: --color-diagnostics=" + S);
}
}
static cl::TokenizerCallback getQuotingStyle(opt::InputArgList &Args) {
if (auto *Arg = Args.getLastArg(OPT_rsp_quoting)) {
StringRef S = Arg->getValue();
@ -749,50 +791,73 @@ opt::InputArgList ArgParser::parse(ArrayRef<const char *> Argv) {
// Make InputArgList from string vectors.
unsigned MissingIndex;
unsigned MissingCount;
SmallVector<const char *, 256> Vec(Argv.data(), Argv.data() + Argv.size());
// We need to get the quoting style for response files before parsing all
// options so we parse here before and ignore all the options but
// --rsp-quoting.
opt::InputArgList Args = Table.ParseArgs(Vec, MissingIndex, MissingCount);
opt::InputArgList Args = Table.ParseArgs(Argv, MissingIndex, MissingCount);
// Expand response files (arguments in the form of @<filename>)
// and then parse the argument again.
cl::ExpandResponseFiles(Saver, getQuotingStyle(Args), Vec);
Args = Table.ParseArgs(Vec, MissingIndex, MissingCount);
SmallVector<const char *, 256> ExpandedArgv(Argv.data(), Argv.data() + Argv.size());
cl::ExpandResponseFiles(Saver, getQuotingStyle(Args), ExpandedArgv);
Args = Table.ParseArgs(makeArrayRef(ExpandedArgv).drop_front(), MissingIndex,
MissingCount);
// Print the real command line if response files are expanded.
if (Args.hasArg(OPT_verbose) && Argv.size() != Vec.size()) {
if (Args.hasArg(OPT_verbose) && Argv.size() != ExpandedArgv.size()) {
std::string Msg = "Command line:";
for (const char *S : Vec)
for (const char *S : ExpandedArgv)
Msg += " " + std::string(S);
message(Msg);
}
// Save the command line after response file expansion so we can write it to
// the PDB if necessary.
Config->Argv = {ExpandedArgv.begin(), ExpandedArgv.end()};
// Handle /WX early since it converts missing argument warnings to errors.
errorHandler().FatalWarnings = Args.hasFlag(OPT_WX, OPT_WX_no, false);
if (MissingCount)
fatal(Twine(Args.getArgString(MissingIndex)) + ": missing argument");
handleColorDiagnostics(Args);
for (auto *Arg : Args.filtered(OPT_UNKNOWN))
warn("ignoring unknown argument: " + Arg->getSpelling());
if (Args.hasArg(OPT_lib))
warn("ignoring /lib since it's not the first argument");
return Args;
}
// Tokenizes and parses a given string as command line in .drective section.
opt::InputArgList ArgParser::parseDirectives(StringRef S) {
// Make InputArgList from string vectors.
// /EXPORT options are processed in fastpath.
std::pair<opt::InputArgList, std::vector<StringRef>>
ArgParser::parseDirectives(StringRef S) {
std::vector<StringRef> Exports;
SmallVector<const char *, 16> Rest;
for (StringRef Tok : tokenize(S)) {
if (Tok.startswith_lower("/export:") || Tok.startswith_lower("-export:"))
Exports.push_back(Tok.substr(strlen("/export:")));
else
Rest.push_back(Tok.data());
}
// Make InputArgList from unparsed string vectors.
unsigned MissingIndex;
unsigned MissingCount;
opt::InputArgList Args =
Table.ParseArgs(tokenize(S), MissingIndex, MissingCount);
opt::InputArgList Args = Table.ParseArgs(Rest, MissingIndex, MissingCount);
if (MissingCount)
fatal(Twine(Args.getArgString(MissingIndex)) + ": missing argument");
for (auto *Arg : Args.filtered(OPT_UNKNOWN))
warn("ignoring unknown argument: " + Arg->getSpelling());
return Args;
return {std::move(Args), std::move(Exports)};
}
// link.exe has an interesting feature. If LINK or _LINK_ environment
@ -802,11 +867,11 @@ opt::InputArgList ArgParser::parseLINK(std::vector<const char *> Argv) {
// Concatenate LINK env and command line arguments, and then parse them.
if (Optional<std::string> S = Process::GetEnv("LINK")) {
std::vector<const char *> V = tokenize(*S);
Argv.insert(Argv.begin(), V.begin(), V.end());
Argv.insert(std::next(Argv.begin()), V.begin(), V.end());
}
if (Optional<std::string> S = Process::GetEnv("_LINK_")) {
std::vector<const char *> V = tokenize(*S);
Argv.insert(Argv.begin(), V.begin(), V.end());
Argv.insert(std::next(Argv.begin()), V.begin(), V.end());
}
return parse(Argv);
}

90
contrib/llvm/tools/lld/COFF/ICF.cpp
View File

@ -18,9 +18,11 @@
//
//===----------------------------------------------------------------------===//
#include "ICF.h"
#include "Chunks.h"
#include "Symbols.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Timer.h"
#include "llvm/ADT/Hashing.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/Parallel.h"
@ -34,6 +36,8 @@ using namespace llvm;
namespace lld {
namespace coff {
static Timer ICFTimer("ICF", Timer::root());
class ICF {
public:
void run(ArrayRef<Chunk *> V);
@ -41,6 +45,8 @@ class ICF {
private:
void segregate(size_t Begin, size_t End, bool Constant);
bool assocEquals(const SectionChunk *A, const SectionChunk *B);
bool equalsConstant(const SectionChunk *A, const SectionChunk *B);
bool equalsVariable(const SectionChunk *A, const SectionChunk *B);
@ -61,8 +67,8 @@ class ICF {
// Returns a hash value for S.
uint32_t ICF::getHash(SectionChunk *C) {
return hash_combine(C->getPermissions(), C->SectionName, C->NumRelocs,
C->Alignment, uint32_t(C->Header->SizeOfRawData),
return hash_combine(C->getOutputCharacteristics(), C->SectionName,
C->Relocs.size(), uint32_t(C->Header->SizeOfRawData),
C->Checksum, C->getContents());
}
@ -73,21 +79,27 @@ uint32_t ICF::getHash(SectionChunk *C) {
// 2017) says that /opt:icf folds both functions and read-only data.
// Despite that, the MSVC linker folds only functions. We found
// a few instances of programs that are not safe for data merging.
// Therefore, we merge only functions just like the MSVC tool. However, we merge
// identical .xdata sections, because the address of unwind information is
// insignificant to the user program and the Visual C++ linker does this.
// Therefore, we merge only functions just like the MSVC tool. However, we also
// merge read-only sections in a couple of cases where the address of the
// section is insignificant to the user program and the behaviour matches that
// of the Visual C++ linker.
bool ICF::isEligible(SectionChunk *C) {
// Non-comdat chunks, dead chunks, and writable chunks are not elegible.
bool Writable = C->getPermissions() & llvm::COFF::IMAGE_SCN_MEM_WRITE;
bool Writable = C->getOutputCharacteristics() & llvm::COFF::IMAGE_SCN_MEM_WRITE;
if (!C->isCOMDAT() || !C->isLive() || Writable)
return false;
// Code sections are eligible.
if (C->getPermissions() & llvm::COFF::IMAGE_SCN_MEM_EXECUTE)
if (C->getOutputCharacteristics() & llvm::COFF::IMAGE_SCN_MEM_EXECUTE)
return true;
// .xdata unwind info sections are eligble.
return C->getSectionName().split('$').first == ".xdata";
// .pdata and .xdata unwind info sections are eligible.
StringRef OutSecName = C->getSectionName().split('$').first;
if (OutSecName == ".pdata" || OutSecName == ".xdata")
return true;
// So are vtables.
return C->Sym && C->Sym->getName().startswith("??_7");
}
// Split an equivalence class into smaller classes.
@ -116,10 +128,23 @@ void ICF::segregate(size_t Begin, size_t End, bool Constant) {
}
}
// Returns true if two sections' associative children are equal.
bool ICF::assocEquals(const SectionChunk *A, const SectionChunk *B) {
auto ChildClasses = [&](const SectionChunk *SC) {
std::vector<uint32_t> Classes;
for (const SectionChunk *C : SC->children())
if (!C->SectionName.startswith(".debug") &&
C->SectionName != ".gfids$y" && C->SectionName != ".gljmp$y")
Classes.push_back(C->Class[Cnt % 2]);
return Classes;
};
return ChildClasses(A) == ChildClasses(B);
}
// Compare "non-moving" part of two sections, namely everything
// except relocation targets.
bool ICF::equalsConstant(const SectionChunk *A, const SectionChunk *B) {
if (A->NumRelocs != B->NumRelocs)
if (A->Relocs.size() != B->Relocs.size())
return false;
// Compare relocations.
@ -142,10 +167,11 @@ bool ICF::equalsConstant(const SectionChunk *A, const SectionChunk *B) {
return false;
// Compare section attributes and contents.
return A->getPermissions() == B->getPermissions() &&
A->SectionName == B->SectionName && A->Alignment == B->Alignment &&
return A->getOutputCharacteristics() == B->getOutputCharacteristics() &&
A->SectionName == B->SectionName &&
A->Header->SizeOfRawData == B->Header->SizeOfRawData &&
A->Checksum == B->Checksum && A->getContents() == B->getContents();
A->Checksum == B->Checksum && A->getContents() == B->getContents() &&
assocEquals(A, B);
}
// Compare "moving" part of two sections, namely relocation targets.
@ -161,9 +187,12 @@ bool ICF::equalsVariable(const SectionChunk *A, const SectionChunk *B) {
return D1->getChunk()->Class[Cnt % 2] == D2->getChunk()->Class[Cnt % 2];
return false;
};
return std::equal(A->Relocs.begin(), A->Relocs.end(), B->Relocs.begin(), Eq);
return std::equal(A->Relocs.begin(), A->Relocs.end(), B->Relocs.begin(),
Eq) &&
assocEquals(A, B);
}
// Find the first Chunk after Begin that has a different class from Begin.
size_t ICF::findBoundary(size_t Begin, size_t End) {
for (size_t I = Begin + 1; I < End; ++I)
if (Chunks[Begin]->Class[Cnt % 2] != Chunks[I]->Class[Cnt % 2])
@ -173,11 +202,8 @@ size_t ICF::findBoundary(size_t Begin, size_t End) {
void ICF::forEachClassRange(size_t Begin, size_t End,
std::function<void(size_t, size_t)> Fn) {
if (Begin > 0)
Begin = findBoundary(Begin - 1, End);
while (Begin < End) {
size_t Mid = findBoundary(Begin, Chunks.size());
size_t Mid = findBoundary(Begin, End);
Fn(Begin, Mid);
Begin = Mid;
}
@ -193,12 +219,22 @@ void ICF::forEachClass(std::function<void(size_t, size_t)> Fn) {
return;
}
// Split sections into 256 shards and call Fn in parallel.
size_t NumShards = 256;
// Shard into non-overlapping intervals, and call Fn in parallel.
// The sharding must be completed before any calls to Fn are made
// so that Fn can modify the Chunks in its shard without causing data
// races.
const size_t NumShards = 256;
size_t Step = Chunks.size() / NumShards;
for_each_n(parallel::par, size_t(0), NumShards, [&](size_t I) {
size_t End = (I == NumShards - 1) ? Chunks.size() : (I + 1) * Step;
forEachClassRange(I * Step, End, Fn);
size_t Boundaries[NumShards + 1];
Boundaries[0] = 0;
Boundaries[NumShards] = Chunks.size();
for_each_n(parallel::par, size_t(1), NumShards, [&](size_t I) {
Boundaries[I] = findBoundary((I - 1) * Step, Chunks.size());
});
for_each_n(parallel::par, size_t(1), NumShards + 1, [&](size_t I) {
if (Boundaries[I - 1] < Boundaries[I]) {
forEachClassRange(Boundaries[I - 1], Boundaries[I], Fn);
}
});
++Cnt;
}
@ -207,6 +243,8 @@ void ICF::forEachClass(std::function<void(size_t, size_t)> Fn) {
// Two sections are considered the same if their section headers,
// contents and relocations are all the same.
void ICF::run(ArrayRef<Chunk *> Vec) {
ScopedTimer T(ICFTimer);
// Collect only mergeable sections and group by hash value.
uint32_t NextId = 1;
for (Chunk *C : Vec) {
@ -218,6 +256,12 @@ void ICF::run(ArrayRef<Chunk *> Vec) {
}
}
// Make sure that ICF doesn't merge sections that are being handled by string
// tail merging.
for (auto &P : MergeChunk::Instances)
for (SectionChunk *SC : P.second->Sections)
SC->Class[0] = NextId++;
// Initially, we use hash values to partition sections.
for_each(parallel::par, Chunks.begin(), Chunks.end(), [&](SectionChunk *SC) {
// Set MSB to 1 to avoid collisions with non-hash classs.

21
contrib/llvm/tools/lld/COFF/Strings.h → contrib/llvm/tools/lld/COFF/ICF.h
View File

@ -1,4 +1,4 @@
//===- Strings.h ------------------------------------------------*- C++ -*-===//
//===- ICF.h --------------------------------------------------------------===//
//
// The LLVM Linker
//
@ -7,17 +7,20 @@
//
//===----------------------------------------------------------------------===//
#ifndef LLD_COFF_STRINGS_H
#define LLD_COFF_STRINGS_H
#ifndef LLD_COFF_ICF_H
#define LLD_COFF_ICF_H
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/StringRef.h"
#include <string>
#include "lld/Common/LLVM.h"
#include "llvm/ADT/ArrayRef.h"
namespace lld {
namespace coff {
llvm::Optional<std::string> demangleMSVC(llvm::StringRef S);
}
}
class Chunk;
void doICF(ArrayRef<Chunk *> Chunks);
} // namespace coff
} // namespace lld
#endif

92
contrib/llvm/tools/lld/COFF/InputFiles.cpp
View File

@ -27,6 +27,7 @@
#include "llvm/Support/Error.h"
#include "llvm/Support/ErrorOr.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Path.h"
#include "llvm/Target/TargetOptions.h"
#include <cstring>
#include <system_error>
@ -138,12 +139,13 @@ void ObjFile::initializeChunks() {
if (Sec->Characteristics & IMAGE_SCN_LNK_COMDAT)
SparseChunks[I] = PendingComdat;
else
SparseChunks[I] = readSection(I, nullptr);
SparseChunks[I] = readSection(I, nullptr, "");
}
}
SectionChunk *ObjFile::readSection(uint32_t SectionNumber,
const coff_aux_section_definition *Def) {
const coff_aux_section_definition *Def,
StringRef LeaderName) {
const coff_section *Sec;
StringRef Name;
if (auto EC = COFFObj->getSection(SectionNumber, Sec))
@ -151,15 +153,7 @@ SectionChunk *ObjFile::readSection(uint32_t SectionNumber,
if (auto EC = COFFObj->getSectionName(Sec, Name))
fatal("getSectionName failed: #" + Twine(SectionNumber) + ": " +
EC.message());
if (Name == ".sxdata") {
ArrayRef<uint8_t> Data;
COFFObj->getSectionContents(Sec, Data);
if (Data.size() % 4 != 0)
fatal(".sxdata must be an array of symbol table indices");
SXData = {reinterpret_cast<const ulittle32_t *>(Data.data()),
Data.size() / 4};
return nullptr;
}
if (Name == ".drectve") {
ArrayRef<uint8_t> Data;
COFFObj->getSectionContents(Sec, Data);
@ -177,8 +171,8 @@ SectionChunk *ObjFile::readSection(uint32_t SectionNumber,
// CodeView needs a linker support. We need to interpret and debug
// info, and then write it to a separate .pdb file.
// Ignore debug info unless /debug is given.
if (!Config->Debug && Name.startswith(".debug"))
// Ignore DWARF debug info unless /debug is given.
if (!Config->Debug && Name.startswith(".debug_"))
return nullptr;
if (Sec->Characteristics & llvm::COFF::IMAGE_SCN_LNK_REMOVE)
@ -191,6 +185,18 @@ SectionChunk *ObjFile::readSection(uint32_t SectionNumber,
// linked in the regular manner.
if (C->isCodeView())
DebugChunks.push_back(C);
else if (Config->GuardCF != GuardCFLevel::Off && Name == ".gfids$y")
GuardFidChunks.push_back(C);
else if (Config->GuardCF != GuardCFLevel::Off && Name == ".gljmp$y")
GuardLJmpChunks.push_back(C);
else if (Name == ".sxdata")
SXDataChunks.push_back(C);
else if (Config->TailMerge && Sec->NumberOfRelocations == 0 &&
Name == ".rdata" && LeaderName.startswith("??_C@"))
// COFF sections that look like string literal sections (i.e. no
// relocations, in .rdata, leader symbol name matches the MSVC name mangling
// for string literals) are subject to string tail merging.
MergeChunk::addSection(C);
else
Chunks.push_back(C);
@ -211,7 +217,7 @@ void ObjFile::readAssociativeDefinition(
// the section; otherwise mark it as discarded.
int32_t SectionNumber = Sym.getSectionNumber();
if (Parent) {
SparseChunks[SectionNumber] = readSection(SectionNumber, Def);
SparseChunks[SectionNumber] = readSection(SectionNumber, Def, "");
if (SparseChunks[SectionNumber])
Parent->addAssociative(SparseChunks[SectionNumber]);
} else {
@ -275,6 +281,13 @@ void ObjFile::initializeSymbols() {
if (auto *Def = Sym.getSectionDefinition())
if (Def->Selection == IMAGE_COMDAT_SELECT_ASSOCIATIVE)
readAssociativeDefinition(Sym, Def);
if (SparseChunks[Sym.getSectionNumber()] == PendingComdat) {
StringRef Name;
COFFObj->getSymbolName(Sym, Name);
log("comdat section " + Name +
" without leader and unassociated, discarding");
continue;
}
Symbols[I] = createRegular(Sym);
}
@ -294,43 +307,46 @@ Symbol *ObjFile::createUndefined(COFFSymbolRef Sym) {
Optional<Symbol *> ObjFile::createDefined(
COFFSymbolRef Sym,
std::vector<const coff_aux_section_definition *> &ComdatDefs) {
StringRef Name;
auto GetName = [&]() {
StringRef S;
COFFObj->getSymbolName(Sym, S);
return S;
};
if (Sym.isCommon()) {
auto *C = make<CommonChunk>(Sym);
Chunks.push_back(C);
COFFObj->getSymbolName(Sym, Name);
Symbol *S =
Symtab->addCommon(this, Name, Sym.getValue(), Sym.getGeneric(), C);
return S;
return Symtab->addCommon(this, GetName(), Sym.getValue(), Sym.getGeneric(),
C);
}
if (Sym.isAbsolute()) {
COFFObj->getSymbolName(Sym, Name);
StringRef Name = GetName();
// Skip special symbols.
if (Name == "@comp.id")
return nullptr;
// COFF spec 5.10.1. The .sxdata section.
if (Name == "@feat.00") {
if (Sym.getValue() & 1)
SEHCompat = true;
Feat00Flags = Sym.getValue();
return nullptr;
}
if (Sym.isExternal())
return Symtab->addAbsolute(Name, Sym);
else
return make<DefinedAbsolute>(Name, Sym);
return make<DefinedAbsolute>(Name, Sym);
}
int32_t SectionNumber = Sym.getSectionNumber();
if (SectionNumber == llvm::COFF::IMAGE_SYM_DEBUG)
return nullptr;
// Reserved sections numbers don't have contents.
if (llvm::COFF::isReservedSectionNumber(SectionNumber))
fatal("broken object file: " + toString(this));
fatal(toString(this) + ": " + GetName() +
" should not refer to special section " + Twine(SectionNumber));
// This symbol references a section which is not present in the section
// header.
if ((uint32_t)SectionNumber >= SparseChunks.size())
fatal("broken object file: " + toString(this));
fatal(toString(this) + ": " + GetName() +
" should not refer to non-existent section " + Twine(SectionNumber));
// Handle comdat leader symbols.
if (const coff_aux_section_definition *Def = ComdatDefs[SectionNumber]) {
@ -338,16 +354,16 @@ Optional<Symbol *> ObjFile::createDefined(
Symbol *Leader;
bool Prevailing;
if (Sym.isExternal()) {
COFFObj->getSymbolName(Sym, Name);
std::tie(Leader, Prevailing) =
Symtab->addComdat(this, Name, Sym.getGeneric());
Symtab->addComdat(this, GetName(), Sym.getGeneric());
} else {
Leader = make<DefinedRegular>(this, /*Name*/ "", false,
/*IsExternal*/ false, Sym.getGeneric());
Prevailing = true;
}
if (Prevailing) {
SectionChunk *C = readSection(SectionNumber, Def);
SectionChunk *C = readSection(SectionNumber, Def, GetName());
SparseChunks[SectionNumber] = C;
C->Sym = cast<DefinedRegular>(Leader);
cast<DefinedRegular>(Leader)->Data = &C->Repl;
@ -429,7 +445,8 @@ void ImportFile::parse() {
// address pointed by the __imp_ symbol. (This allows you to call
// DLL functions just like regular non-DLL functions.)
if (Hdr->getType() == llvm::COFF::IMPORT_CODE)
ThunkSym = Symtab->addImportThunk(Name, ImpSym, Hdr->Machine);
ThunkSym = Symtab->addImportThunk(
Name, cast_or_null<DefinedImportData>(ImpSym), Hdr->Machine);
}
void BitcodeFile::parse() {
@ -462,7 +479,7 @@ void BitcodeFile::parse() {
} else {
Sym = Symtab->addRegular(this, SymName);
}
SymbolBodies.push_back(Sym);
Symbols.push_back(Sym);
}
Directives = Obj->getCOFFLinkerOpts();
}
@ -486,10 +503,7 @@ MachineTypes BitcodeFile::getMachineType() {
// Returns the last element of a path, which is supposed to be a filename.
static StringRef getBasename(StringRef Path) {
size_t Pos = Path.find_last_of("\\/");
if (Pos == StringRef::npos)
return Path;
return Path.substr(Pos + 1);
return sys::path::filename(Path, sys::path::Style::windows);
}
// Returns a string in the format of "foo.obj" or "foo.obj(bar.lib)".

51
contrib/llvm/tools/lld/COFF/InputFiles.h
View File

@ -110,6 +110,9 @@ class ObjFile : public InputFile {
MachineTypes getMachineType() override;
ArrayRef<Chunk *> getChunks() { return Chunks; }
ArrayRef<SectionChunk *> getDebugChunks() { return DebugChunks; }
ArrayRef<SectionChunk *> getSXDataChunks() { return SXDataChunks; }
ArrayRef<SectionChunk *> getGuardFidChunks() { return GuardFidChunks; }
ArrayRef<SectionChunk *> getGuardLJmpChunks() { return GuardLJmpChunks; }
ArrayRef<Symbol *> getSymbols() { return Symbols; }
// Returns a Symbol object for the SymbolIndex'th symbol in the
@ -123,13 +126,17 @@ class ObjFile : public InputFile {
static std::vector<ObjFile *> Instances;
// True if this object file is compatible with SEH.
// COFF-specific and x86-only.
bool SEHCompat = false;
// Flags in the absolute @feat.00 symbol if it is present. These usually
// indicate if an object was compiled with certain security features enabled
// like stack guard, safeseh, /guard:cf, or other things.
uint32_t Feat00Flags = 0;
// The symbol table indexes of the safe exception handlers.
// COFF-specific and x86-only.
ArrayRef<llvm::support::ulittle32_t> SXData;
// True if this object file is compatible with SEH. COFF-specific and
// x86-only. COFF spec 5.10.1. The .sxdata section.
bool hasSafeSEH() { return Feat00Flags & 0x1; }
// True if this file was compiled with /guard:cf.
bool hasGuardCF() { return Feat00Flags & 0x800; }
// Pointer to the PDB module descriptor builder. Various debug info records
// will reference object files by "module index", which is here. Things like
@ -143,7 +150,8 @@ class ObjFile : public InputFile {
SectionChunk *
readSection(uint32_t SectionNumber,
const llvm::object::coff_aux_section_definition *Def);
const llvm::object::coff_aux_section_definition *Def,
StringRef LeaderName);
void readAssociativeDefinition(
COFFSymbolRef COFFSym,
@ -165,6 +173,15 @@ class ObjFile : public InputFile {
// CodeView debug info sections.
std::vector<SectionChunk *> DebugChunks;
// Chunks containing symbol table indices of exception handlers. Only used for
// 32-bit x86.
std::vector<SectionChunk *> SXDataChunks;
// Chunks containing symbol table indices of address taken symbols and longjmp
// targets. These are not linked into the final binary when /guard:cf is set.
std::vector<SectionChunk *> GuardFidChunks;
std::vector<SectionChunk *> GuardLJmpChunks;
// This vector contains the same chunks as Chunks, but they are
// indexed such that you can get a SectionChunk by section index.
// Nonexistent section indices are filled with null pointers.
@ -184,15 +201,14 @@ class ObjFile : public InputFile {
// for details about the format.
class ImportFile : public InputFile {
public:
explicit ImportFile(MemoryBufferRef M)
: InputFile(ImportKind, M), Live(!Config->DoGC) {}
explicit ImportFile(MemoryBufferRef M) : InputFile(ImportKind, M) {}
static bool classof(const InputFile *F) { return F->kind() == ImportKind; }
static std::vector<ImportFile *> Instances;
DefinedImportData *ImpSym = nullptr;
DefinedImportThunk *ThunkSym = nullptr;
Symbol *ImpSym = nullptr;
Symbol *ThunkSym = nullptr;
std::string DLLName;
private:
@ -204,12 +220,15 @@ class ImportFile : public InputFile {
Chunk *Location = nullptr;
// We want to eliminate dllimported symbols if no one actually refers them.
// This "Live" bit is used to keep track of which import library members
// These "Live" bits are used to keep track of which import library members
// are actually in use.
//
// If the Live bit is turned off by MarkLive, Writer will ignore dllimported
// symbols provided by this import library member.
bool Live;
// symbols provided by this import library member. We also track whether the
// imported symbol is used separately from whether the thunk is used in order
// to avoid creating unnecessary thunks.
bool Live = !Config->DoGC;
bool ThunkLive = !Config->DoGC;
};
// Used for LTO.
@ -217,7 +236,7 @@ class BitcodeFile : public InputFile {
public:
explicit BitcodeFile(MemoryBufferRef M) : InputFile(BitcodeKind, M) {}
static bool classof(const InputFile *F) { return F->kind() == BitcodeKind; }
ArrayRef<Symbol *> getSymbols() { return SymbolBodies; }
ArrayRef<Symbol *> getSymbols() { return Symbols; }
MachineTypes getMachineType() override;
static std::vector<BitcodeFile *> Instances;
std::unique_ptr<llvm::lto::InputFile> Obj;
@ -225,7 +244,7 @@ class BitcodeFile : public InputFile {
private:
void parse() override;
std::vector<Symbol *> SymbolBodies;
std::vector<Symbol *> Symbols;
};
} // namespace coff

72
contrib/llvm/tools/lld/COFF/LTO.cpp
View File

@ -12,6 +12,7 @@
#include "InputFiles.h"
#include "Symbols.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Strings.h"
#include "lld/Common/TargetOptionsCommandFlags.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/SmallString.h"
@ -40,47 +41,32 @@ using namespace llvm::object;
using namespace lld;
using namespace lld::coff;
static void diagnosticHandler(const DiagnosticInfo &DI) {
SmallString<128> ErrStorage;
raw_svector_ostream OS(ErrStorage);
DiagnosticPrinterRawOStream DP(OS);
DI.print(DP);
warn(ErrStorage);
}
static void checkError(Error E) {
handleAllErrors(std::move(E),
[&](ErrorInfoBase &EIB) { error(EIB.message()); });
}
static void saveBuffer(StringRef Buffer, const Twine &Path) {
std::error_code EC;
raw_fd_ostream OS(Path.str(), EC, sys::fs::OpenFlags::F_None);
if (EC)
error("cannot create " + Path + ": " + EC.message());
OS << Buffer;
}
static std::unique_ptr<lto::LTO> createLTO() {
lto::Config Conf;
Conf.Options = InitTargetOptionsFromCodeGenFlags();
lto::Config C;
C.Options = InitTargetOptionsFromCodeGenFlags();
// Always emit a section per function/datum with LTO. LLVM LTO should get most
// of the benefit of linker GC, but there are still opportunities for ICF.
C.Options.FunctionSections = true;
C.Options.DataSections = true;
// Use static reloc model on 32-bit x86 because it usually results in more
// compact code, and because there are also known code generation bugs when
// using the PIC model (see PR34306).
if (Config->Machine == COFF::IMAGE_FILE_MACHINE_I386)
Conf.RelocModel = Reloc::Static;
C.RelocModel = Reloc::Static;
else
Conf.RelocModel = Reloc::PIC_;
Conf.DisableVerify = true;
Conf.DiagHandler = diagnosticHandler;
Conf.OptLevel = Config->LTOOptLevel;
C.RelocModel = Reloc::PIC_;
C.DisableVerify = true;
C.DiagHandler = diagnosticHandler;
C.OptLevel = Config->LTOO;
if (Config->SaveTemps)
checkError(Conf.addSaveTemps(std::string(Config->OutputFile) + ".",
/*UseInputModulePath*/ true));
checkError(C.addSaveTemps(std::string(Config->OutputFile) + ".",
/*UseInputModulePath*/ true));
lto::ThinBackend Backend;
if (Config->LTOJobs != 0)
Backend = lto::createInProcessThinBackend(Config->LTOJobs);
return llvm::make_unique<lto::LTO>(std::move(Conf), Backend,
if (Config->ThinLTOJobs != 0)
Backend = lto::createInProcessThinBackend(Config->ThinLTOJobs);
return llvm::make_unique<lto::LTO>(std::move(C), Backend,
Config->LTOPartitions);
}
@ -119,7 +105,7 @@ void BitcodeCompiler::add(BitcodeFile &F) {
// and return the resulting objects.
std::vector<StringRef> BitcodeCompiler::compile() {
unsigned MaxTasks = LTOObj->getMaxTasks();
Buff.resize(MaxTasks);
Buf.resize(MaxTasks);
Files.resize(MaxTasks);
// The /lldltocache option specifies the path to a directory in which to cache
@ -127,15 +113,15 @@ std::vector<StringRef> BitcodeCompiler::compile() {
// specified, configure LTO to use it as the cache directory.
lto::NativeObjectCache Cache;
if (!Config->LTOCache.empty())
Cache = check(
lto::localCache(Config->LTOCache,
[&](size_t Task, std::unique_ptr<MemoryBuffer> MB,
StringRef Path) { Files[Task] = std::move(MB); }));
Cache = check(lto::localCache(
Config->LTOCache, [&](size_t Task, std::unique_ptr<MemoryBuffer> MB) {
Files[Task] = std::move(MB);
}));
checkError(LTOObj->run(
[&](size_t Task) {
return llvm::make_unique<lto::NativeObjectStream>(
llvm::make_unique<raw_svector_ostream>(Buff[Task]));
llvm::make_unique<raw_svector_ostream>(Buf[Task]));
},
Cache));
@ -144,15 +130,15 @@ std::vector<StringRef> BitcodeCompiler::compile() {
std::vector<StringRef> Ret;
for (unsigned I = 0; I != MaxTasks; ++I) {
if (Buff[I].empty())
if (Buf[I].empty())
continue;
if (Config->SaveTemps) {
if (I == 0)
saveBuffer(Buff[I], Config->OutputFile + ".lto.obj");
saveBuffer(Buf[I], Config->OutputFile + ".lto.obj");
else
saveBuffer(Buff[I], Config->OutputFile + Twine(I) + ".lto.obj");
saveBuffer(Buf[I], Config->OutputFile + Twine(I) + ".lto.obj");
}
Ret.emplace_back(Buff[I].data(), Buff[I].size());
Ret.emplace_back(Buf[I].data(), Buf[I].size());
}
for (std::unique_ptr<MemoryBuffer> &File : Files)

2
contrib/llvm/tools/lld/COFF/LTO.h
View File

@ -48,7 +48,7 @@ class BitcodeCompiler {
private:
std::unique_ptr<llvm::lto::LTO> LTOObj;
std::vector<SmallString<0>> Buff;
std::vector<SmallString<0>> Buf;
std::vector<std::unique_ptr<MemoryBuffer>> Files;
};
}

12
contrib/llvm/tools/lld/COFF/MapFile.cpp
View File

@ -23,7 +23,6 @@
#include "SymbolTable.h"
#include "Symbols.h"
#include "Writer.h"
#include "lld/Common/ErrorHandler.h"
#include "llvm/Support/Parallel.h"
#include "llvm/Support/raw_ostream.h"
@ -37,14 +36,15 @@ using namespace lld::coff;
typedef DenseMap<const SectionChunk *, SmallVector<DefinedRegular *, 4>>
SymbolMapTy;
static const std::string Indent8 = " "; // 8 spaces
static const std::string Indent16 = " "; // 16 spaces
// Print out the first three columns of a line.
static void writeHeader(raw_ostream &OS, uint64_t Addr, uint64_t Size,
uint64_t Align) {
OS << format("%08llx %08llx %5lld ", Addr, Size, Align);
}
static std::string indent(int Depth) { return std::string(Depth * 8, ' '); }
// Returns a list of all symbols that we want to print out.
static std::vector<DefinedRegular *> getSymbols() {
std::vector<DefinedRegular *> V;
@ -79,7 +79,7 @@ getSymbolStrings(ArrayRef<DefinedRegular *> Syms) {
for_each_n(parallel::par, (size_t)0, Syms.size(), [&](size_t I) {
raw_string_ostream OS(Str[I]);
writeHeader(OS, Syms[I]->getRVA(), 0, 0);
OS << indent(2) << toString(*Syms[I]);
OS << Indent16 << toString(*Syms[I]);
});
DenseMap<DefinedRegular *, std::string> Ret;
@ -108,7 +108,7 @@ void coff::writeMapFile(ArrayRef<OutputSection *> OutputSections) {
// Print out file contents.
for (OutputSection *Sec : OutputSections) {
writeHeader(OS, Sec->getRVA(), Sec->getVirtualSize(), /*Align=*/PageSize);
OS << Sec->getName() << '\n';
OS << Sec->Name << '\n';
for (Chunk *C : Sec->getChunks()) {
auto *SC = dyn_cast<SectionChunk>(C);
@ -116,7 +116,7 @@ void coff::writeMapFile(ArrayRef<OutputSection *> OutputSections) {
continue;
writeHeader(OS, SC->getRVA(), SC->getSize(), SC->Alignment);
OS << indent(1) << SC->File->getName() << ":(" << SC->getSectionName()
OS << Indent8 << SC->File->getName() << ":(" << SC->getSectionName()
<< ")\n";
for (DefinedRegular *Sym : SectionSyms[SC])
OS << SymStr[Sym] << '\n';

7
contrib/llvm/tools/lld/COFF/MarkLive.cpp
View File

@ -9,16 +9,21 @@
#include "Chunks.h"
#include "Symbols.h"
#include "lld/Common/Timer.h"
#include "llvm/ADT/STLExtras.h"
#include <vector>
namespace lld {
namespace coff {
static Timer GCTimer("GC", Timer::root());
// Set live bit on for each reachable chunk. Unmarked (unreachable)
// COMDAT chunks will be ignored by Writer, so they will be excluded
// from the final output.
void markLive(ArrayRef<Chunk *> Chunks) {
ScopedTimer T(GCTimer);
// We build up a worklist of sections which have been marked as live. We only
// push into the worklist when we discover an unmarked section, and we mark
// as we push, so sections never appear twice in the list.
@ -43,7 +48,7 @@ void markLive(ArrayRef<Chunk *> Chunks) {
else if (auto *Sym = dyn_cast<DefinedImportData>(B))
Sym->File->Live = true;
else if (auto *Sym = dyn_cast<DefinedImportThunk>(B))
Sym->WrappedSym->File->Live = true;
Sym->WrappedSym->File->Live = Sym->WrappedSym->File->ThunkLive = true;
};
// Add GC root chunks.

24
contrib/llvm/tools/lld/COFF/MarkLive.h Normal file
View File

@ -0,0 +1,24 @@
//===- MarkLive.h -----------------------------------------------*- C++ -*-===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#ifndef LLD_COFF_MARKLIVE_H
#define LLD_COFF_MARKLIVE_H
#include "lld/Common/LLVM.h"
#include "llvm/ADT/ArrayRef.h"
namespace lld {
namespace coff {
void markLive(ArrayRef<Chunk *> Chunks);
} // namespace coff
} // namespace lld
#endif // LLD_COFF_MARKLIVE_H

2
contrib/llvm/tools/lld/COFF/MinGW.cpp
View File

@ -138,7 +138,7 @@ void coff::writeDefFile(StringRef Name) {
<< "@" << E.Ordinal;
if (auto *Def = dyn_cast_or_null<Defined>(E.Sym)) {
if (Def && Def->getChunk() &&
!(Def->getChunk()->getPermissions() & IMAGE_SCN_MEM_EXECUTE))
!(Def->getChunk()->getOutputCharacteristics() & IMAGE_SCN_MEM_EXECUTE))
OS << " DATA";
}
OS << "\n";

29
contrib/llvm/tools/lld/COFF/Options.td
View File

@ -20,6 +20,10 @@ def align : P<"align", "Section alignment">;
def aligncomm : P<"aligncomm", "Set common symbol alignment">;
def alternatename : P<"alternatename", "Define weak alias">;
def base : P<"base", "Base address of the program">;
def color_diagnostics: Flag<["--"], "color-diagnostics">,
HelpText<"Use colors in diagnostics">;
def color_diagnostics_eq: Joined<["--"], "color-diagnostics=">,
HelpText<"Use colors in diagnostics; one of 'always', 'never', 'auto'">;
def defaultlib : P<"defaultlib", "Add the library to the list of input files">;
def delayload : P<"delayload", "Delay loaded DLL name">;
def entry : P<"entry", "Name of entry point symbol">;
@ -28,9 +32,12 @@ def errorlimit : P<"errorlimit",
def export : P<"export", "Export a function">;
// No help text because /failifmismatch is not intended to be used by the user.
def failifmismatch : P<"failifmismatch", "">;
def guard : P<"guard", "Control flow guard">;
def heap : P<"heap", "Size of the heap">;
def ignore : P<"ignore", "Specify warning codes to ignore">;
def implib : P<"implib", "Import library name">;
def lib : F<"lib">,
HelpText<"Act like lib.exe; must be first argument if present">;
def libpath : P<"libpath", "Additional library search path">;
def linkrepro : P<"linkrepro", "Dump linker invocation and input files for debugging">;
def lldltocache : P<"lldltocache", "Path to ThinLTO cached object file directory">;
@ -42,12 +49,18 @@ def merge : P<"merge", "Combine sections">;
def mllvm : P<"mllvm", "Options to pass to LLVM">;
def nodefaultlib : P<"nodefaultlib", "Remove a default library">;
def opt : P<"opt", "Control optimizations">;
def order : P<"order", "Put functions in order">;
def out : P<"out", "Path to file to write output">;
def natvis : P<"natvis", "Path to natvis file to embed in the PDB">;
def no_color_diagnostics: F<"no-color-diagnostics">,
HelpText<"Do not use colors in diagnostics">;
def pdb : P<"pdb", "PDB file path">;
def pdbaltpath : P<"pdbaltpath", "PDB file path to embed in the image">;
def section : P<"section", "Specify section attributes">;
def stack : P<"stack", "Size of the stack">;
def stub : P<"stub", "Specify DOS stub file">;
def subsystem : P<"subsystem", "Specify subsystem">;
def timestamp : P<"timestamp", "Specify the PE header timestamp">;
def version : P<"version", "Specify a version number in the PE header">;
def wholearchive_file : P<"wholearchive", "Include all object files from this archive">;
@ -72,12 +85,14 @@ def deffile : Joined<["/", "-"], "def:">,
HelpText<"Use module-definition file">;
def debug : F<"debug">, HelpText<"Embed a symbol table in the image">;
def debug_full : F<"debug:full">, Alias<debug>;
def debugtype : P<"debugtype", "Debug Info Options">;
def dll : F<"dll">, HelpText<"Create a DLL">;
def driver : P<"driver", "Generate a Windows NT Kernel Mode Driver">;
def nodefaultlib_all : F<"nodefaultlib">;
def noentry : F<"noentry">;
def profile : F<"profile">;
def repro : F<"Brepro">, HelpText<"Use a hash of the executable as the PE header timestamp">;
def swaprun_cd : F<"swaprun:cd">;
def swaprun_net : F<"swaprun:net">;
def verbose : F<"verbose">;
@ -102,6 +117,12 @@ defm fixed : B<"fixed", "Disable base relocations",
defm highentropyva : B<"highentropyva",
"Enable 64-bit ASLR (default on 64-bit)",
"Disable 64-bit ASLR">;
defm incremental : B<"incremental",
"Keep original import library if contents are unchanged",
"Overwrite import library even if contents are unchanged">;
defm integritycheck : B<"integritycheck",
"Set FORCE_INTEGRITY bit in PE header",
"No effect (default)">;
defm largeaddressaware : B<"largeaddressaware",
"Enable large addresses (default on 64-bit)",
"Disable large addresses (default on 32-bit)">;
@ -120,11 +141,14 @@ def help_q : Flag<["/?", "-?"], "">, Alias<help>;
// LLD extensions
def debug_ghash : F<"debug:ghash">;
def debug_dwarf : F<"debug:dwarf">;
def debug_symtab : F<"debug:symtab">;
def export_all_symbols : F<"export-all-symbols">;
def kill_at : F<"kill-at">;
def lldmingw : F<"lldmingw">;
def msvclto : F<"msvclto">;
def output_def : Joined<["/", "-"], "output-def:">;
def pdb_source_path : P<"pdbsourcepath",
"Base path used to make relative source file path absolute in PDB">;
def rsp_quoting : Joined<["--"], "rsp-quoting=">,
HelpText<"Quoting style for response files, 'windows' (default) or 'posix'">;
def dash_dash_version : Flag<["--"], "version">,
@ -133,6 +157,7 @@ def dash_dash_version : Flag<["--"], "version">,
// Flags for debugging
def lldmap : F<"lldmap">;
def lldmap_file : Joined<["/", "-"], "lldmap:">;
def show_timing : F<"time">;
//==============================================================================
// The flags below do nothing. They are defined only for link.exe compatibility.
@ -147,8 +172,6 @@ multiclass QB<string name> {
def functionpadmin : F<"functionpadmin">;
def ignoreidl : F<"ignoreidl">;
def incremental : F<"incremental">;
def no_incremental : F<"incremental:no">;
def nologo : F<"nologo">;
def throwingnew : F<"throwingnew">;
def editandcontinue : F<"editandcontinue">;
@ -158,8 +181,6 @@ def delay : QF<"delay">;
def errorreport : QF<"errorreport">;
def idlout : QF<"idlout">;
def maxilksize : QF<"maxilksize">;
def natvis : QF<"natvis">;
def pdbaltpath : QF<"pdbaltpath">;
def tlbid : QF<"tlbid">;
def tlbout : QF<"tlbout">;
def verbose_all : QF<"verbose">;

516
contrib/llvm/tools/lld/COFF/PDB.cpp
View File

@ -15,7 +15,7 @@
#include "Symbols.h"
#include "Writer.h"
#include "lld/Common/ErrorHandler.h"
#include "llvm/DebugInfo/CodeView/CVDebugRecord.h"
#include "lld/Common/Timer.h"
#include "llvm/DebugInfo/CodeView/DebugSubsectionRecord.h"
#include "llvm/DebugInfo/CodeView/GlobalTypeTableBuilder.h"
#include "llvm/DebugInfo/CodeView/LazyRandomTypeCollection.h"
@ -45,8 +45,10 @@
#include "llvm/DebugInfo/PDB/Native/TpiStreamBuilder.h"
#include "llvm/DebugInfo/PDB/PDB.h"
#include "llvm/Object/COFF.h"
#include "llvm/Object/CVDebugRecord.h"
#include "llvm/Support/BinaryByteStream.h"
#include "llvm/Support/Endian.h"
#include "llvm/Support/FormatVariadic.h"
#include "llvm/Support/JamCRC.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/ScopedPrinter.h"
@ -61,6 +63,15 @@ using llvm::object::coff_section;
static ExitOnError ExitOnErr;
static Timer TotalPdbLinkTimer("PDB Emission (Cumulative)", Timer::root());
static Timer AddObjectsTimer("Add Objects", TotalPdbLinkTimer);
static Timer TypeMergingTimer("Type Merging", AddObjectsTimer);
static Timer SymbolMergingTimer("Symbol Merging", AddObjectsTimer);
static Timer GlobalsLayoutTimer("Globals Stream Layout", TotalPdbLinkTimer);
static Timer TpiStreamLayoutTimer("TPI Stream Layout", TotalPdbLinkTimer);
static Timer DiskCommitTimer("Commit to Disk", TotalPdbLinkTimer);
namespace {
/// Map from type index and item index in a type server PDB to the
/// corresponding index in the destination PDB.
@ -74,11 +85,19 @@ class PDBLinker {
public:
PDBLinker(SymbolTable *Symtab)
: Alloc(), Symtab(Symtab), Builder(Alloc), TypeTable(Alloc),
IDTable(Alloc), GlobalTypeTable(Alloc), GlobalIDTable(Alloc) {}
IDTable(Alloc), GlobalTypeTable(Alloc), GlobalIDTable(Alloc) {
// This isn't strictly necessary, but link.exe usually puts an empty string
// as the first "valid" string in the string table, so we do the same in
// order to maintain as much byte-for-byte compatibility as possible.
PDBStrTab.insert("");
}
/// Emit the basic PDB structure: initial streams, headers, etc.
void initialize(const llvm::codeview::DebugInfo &BuildId);
/// Add natvis files specified on the command line.
void addNatvisFiles();
/// Link CodeView from each object file in the symbol table into the PDB.
void addObjectsToPDB();
@ -106,9 +125,6 @@ class PDBLinker {
void addSections(ArrayRef<OutputSection *> OutputSections,
ArrayRef<uint8_t> SectionTable);
void addSectionContrib(pdb::DbiModuleDescriptorBuilder &LinkerModule,
OutputSection *OS, Chunk *C);
/// Write the PDB to disk.
void commit();
@ -137,6 +153,11 @@ class PDBLinker {
llvm::SmallString<128> NativePath;
/// A list of other PDBs which are loaded during the linking process and which
/// we need to keep around since the linking operation may reference pointers
/// inside of these PDBs.
llvm::SmallVector<std::unique_ptr<pdb::NativeSession>, 2> LoadedPDBs;
std::vector<pdb::SecMapEntry> SectionMap;
/// Type index mappings of type server PDBs that we've loaded so far.
@ -184,8 +205,8 @@ static bool canUseDebugH(ArrayRef<uint8_t> DebugH) {
DebugH = DebugH.drop_front(sizeof(object::debug_h_header));
return Header->Magic == COFF::DEBUG_HASHES_SECTION_MAGIC &&
Header->Version == 0 &&
Header->HashAlgorithm == uint16_t(GlobalTypeHashAlg::SHA1) &&
(DebugH.size() % 20 == 0);
Header->HashAlgorithm == uint16_t(GlobalTypeHashAlg::SHA1_8) &&
(DebugH.size() % 8 == 0);
}
static Optional<ArrayRef<uint8_t>> getDebugH(ObjFile *File) {
@ -237,6 +258,8 @@ maybeReadTypeServerRecord(CVTypeArray &Types) {
Expected<const CVIndexMap&> PDBLinker::mergeDebugT(ObjFile *File,
CVIndexMap &ObjectIndexMap) {
ScopedTimer T(TypeMergingTimer);
ArrayRef<uint8_t> Data = getDebugSection(File, ".debug$T");
if (Data.empty())
return ObjectIndexMap;
@ -348,10 +371,16 @@ Expected<const CVIndexMap&> PDBLinker::maybeMergeTypeServerPDB(ObjFile *File,
return std::move(E);
}
auto ExpectedTpi = (*ExpectedSession)->getPDBFile().getPDBTpiStream();
pdb::NativeSession *Session = ExpectedSession->get();
// Keep a strong reference to this PDB, so that it's safe to hold pointers
// into the file.
LoadedPDBs.push_back(std::move(*ExpectedSession));
auto ExpectedTpi = Session->getPDBFile().getPDBTpiStream();
if (auto E = ExpectedTpi.takeError())
fatal("Type server does not have TPI stream: " + toString(std::move(E)));
auto ExpectedIpi = (*ExpectedSession)->getPDBFile().getPDBIpiStream();
auto ExpectedIpi = Session->getPDBFile().getPDBIpiStream();
if (auto E = ExpectedIpi.takeError())
fatal("Type server does not have TPI stream: " + toString(std::move(E)));
@ -428,6 +457,38 @@ static void remapTypesInSymbolRecord(ObjFile *File, SymbolKind SymKind,
}
}
static void
recordStringTableReferenceAtOffset(MutableArrayRef<uint8_t> Contents,
uint32_t Offset,
std::vector<ulittle32_t *> &StrTableRefs) {
Contents =
Contents.drop_front(Offset).take_front(sizeof(support::ulittle32_t));
ulittle32_t *Index = reinterpret_cast<ulittle32_t *>(Contents.data());
StrTableRefs.push_back(Index);
}
static void
recordStringTableReferences(SymbolKind Kind, MutableArrayRef<uint8_t> Contents,
std::vector<ulittle32_t *> &StrTableRefs) {
// For now we only handle S_FILESTATIC, but we may need the same logic for
// S_DEFRANGE and S_DEFRANGE_SUBFIELD. However, I cannot seem to generate any
// PDBs that contain these types of records, so because of the uncertainty
// they are omitted here until we can prove that it's necessary.
switch (Kind) {
case SymbolKind::S_FILESTATIC:
// FileStaticSym::ModFileOffset
recordStringTableReferenceAtOffset(Contents, 4, StrTableRefs);
break;
case SymbolKind::S_DEFRANGE:
case SymbolKind::S_DEFRANGE_SUBFIELD:
log("Not fixing up string table reference in S_DEFRANGE / "
"S_DEFRANGE_SUBFIELD record");
break;
default:
break;
}
}
static SymbolKind symbolKind(ArrayRef<uint8_t> RecordData) {
const RecordPrefix *Prefix =
reinterpret_cast<const RecordPrefix *>(RecordData.data());
@ -644,53 +705,65 @@ static void mergeSymbolRecords(BumpPtrAllocator &Alloc, ObjFile *File,
pdb::GSIStreamBuilder &GsiBuilder,
const CVIndexMap &IndexMap,
TypeCollection &IDTable,
std::vector<ulittle32_t *> &StringTableRefs,
BinaryStreamRef SymData) {
// FIXME: Improve error recovery by warning and skipping records when
// possible.
CVSymbolArray Syms;
BinaryStreamReader Reader(SymData);
ExitOnErr(Reader.readArray(Syms, Reader.getLength()));
ArrayRef<uint8_t> SymsBuffer;
cantFail(SymData.readBytes(0, SymData.getLength(), SymsBuffer));
SmallVector<SymbolScope, 4> Scopes;
for (CVSymbol Sym : Syms) {
// Discover type index references in the record. Skip it if we don't know
// where they are.
SmallVector<TiReference, 32> TypeRefs;
if (!discoverTypeIndicesInSymbol(Sym, TypeRefs)) {
log("ignoring unknown symbol record with kind 0x" + utohexstr(Sym.kind()));
continue;
}
// Copy the symbol record so we can mutate it.
MutableArrayRef<uint8_t> NewData = copySymbolForPdb(Sym, Alloc);
auto EC = forEachCodeViewRecord<CVSymbol>(
SymsBuffer, [&](const CVSymbol &Sym) -> llvm::Error {
// Discover type index references in the record. Skip it if we don't
// know where they are.
SmallVector<TiReference, 32> TypeRefs;
if (!discoverTypeIndicesInSymbol(Sym, TypeRefs)) {
log("ignoring unknown symbol record with kind 0x" +
utohexstr(Sym.kind()));
return Error::success();
}
// Re-map all the type index references.
MutableArrayRef<uint8_t> Contents =
NewData.drop_front(sizeof(RecordPrefix));
remapTypesInSymbolRecord(File, Sym.kind(), Contents, IndexMap, TypeRefs);
// Copy the symbol record so we can mutate it.
MutableArrayRef<uint8_t> NewData = copySymbolForPdb(Sym, Alloc);
// An object file may have S_xxx_ID symbols, but these get converted to
// "real" symbols in a PDB.
translateIdSymbols(NewData, IDTable);
// Re-map all the type index references.
MutableArrayRef<uint8_t> Contents =
NewData.drop_front(sizeof(RecordPrefix));
remapTypesInSymbolRecord(File, Sym.kind(), Contents, IndexMap,
TypeRefs);
SymbolKind NewKind = symbolKind(NewData);
// An object file may have S_xxx_ID symbols, but these get converted to
// "real" symbols in a PDB.
translateIdSymbols(NewData, IDTable);
// Fill in "Parent" and "End" fields by maintaining a stack of scopes.
CVSymbol NewSym(NewKind, NewData);
if (symbolOpensScope(NewKind))
scopeStackOpen(Scopes, File->ModuleDBI->getNextSymbolOffset(), NewSym);
else if (symbolEndsScope(NewKind))
scopeStackClose(Scopes, File->ModuleDBI->getNextSymbolOffset(), File);
// If this record refers to an offset in the object file's string table,
// add that item to the global PDB string table and re-write the index.
recordStringTableReferences(Sym.kind(), Contents, StringTableRefs);
// Add the symbol to the globals stream if necessary. Do this before adding
// the symbol to the module since we may need to get the next symbol offset,
// and writing to the module's symbol stream will update that offset.
if (symbolGoesInGlobalsStream(NewSym))
addGlobalSymbol(GsiBuilder, *File, NewSym);
SymbolKind NewKind = symbolKind(NewData);
// Add the symbol to the module.
if (symbolGoesInModuleStream(NewSym))
File->ModuleDBI->addSymbol(NewSym);
}
// Fill in "Parent" and "End" fields by maintaining a stack of scopes.
CVSymbol NewSym(NewKind, NewData);
if (symbolOpensScope(NewKind))
scopeStackOpen(Scopes, File->ModuleDBI->getNextSymbolOffset(),
NewSym);
else if (symbolEndsScope(NewKind))
scopeStackClose(Scopes, File->ModuleDBI->getNextSymbolOffset(), File);
// Add the symbol to the globals stream if necessary. Do this before
// adding the symbol to the module since we may need to get the next
// symbol offset, and writing to the module's symbol stream will update
// that offset.
if (symbolGoesInGlobalsStream(NewSym))
addGlobalSymbol(GsiBuilder, *File, NewSym);
// Add the symbol to the module.
if (symbolGoesInModuleStream(NewSym))
File->ModuleDBI->addSymbol(NewSym);
return Error::success();
});
cantFail(std::move(EC));
}
// Allocate memory for a .debug$S section and relocate it.
@ -704,6 +777,32 @@ static ArrayRef<uint8_t> relocateDebugChunk(BumpPtrAllocator &Alloc,
".debug$S");
}
static pdb::SectionContrib createSectionContrib(const Chunk *C, uint32_t Modi) {
OutputSection *OS = C->getOutputSection();
pdb::SectionContrib SC;
memset(&SC, 0, sizeof(SC));
SC.ISect = OS->SectionIndex;
SC.Off = C->getRVA() - OS->getRVA();
SC.Size = C->getSize();
if (auto *SecChunk = dyn_cast<SectionChunk>(C)) {
SC.Characteristics = SecChunk->Header->Characteristics;
SC.Imod = SecChunk->File->ModuleDBI->getModuleIndex();
ArrayRef<uint8_t> Contents = SecChunk->getContents();
JamCRC CRC(0);
ArrayRef<char> CharContents = makeArrayRef(
reinterpret_cast<const char *>(Contents.data()), Contents.size());
CRC.update(CharContents);
SC.DataCrc = CRC.getCRC();
} else {
SC.Characteristics = OS->Header.Characteristics;
// FIXME: When we start creating DBI for import libraries, use those here.
SC.Imod = Modi;
}
SC.RelocCrc = 0; // FIXME
return SC;
}
void PDBLinker::addObjFile(ObjFile *File) {
// Add a module descriptor for every object file. We need to put an absolute
// path to the object into the PDB. If this is a plain object, we make its
@ -718,6 +817,17 @@ void PDBLinker::addObjFile(ObjFile *File) {
File->ModuleDBI = &ExitOnErr(Builder.getDbiBuilder().addModuleInfo(Name));
File->ModuleDBI->setObjFileName(Path);
auto Chunks = File->getChunks();
uint32_t Modi = File->ModuleDBI->getModuleIndex();
for (Chunk *C : Chunks) {
auto *SecChunk = dyn_cast<SectionChunk>(C);
if (!SecChunk || !SecChunk->isLive())
continue;
pdb::SectionContrib SC = createSectionContrib(SecChunk, Modi);
File->ModuleDBI->setFirstSectionContrib(SC);
break;
}
// Before we can process symbol substreams from .debug$S, we need to process
// type information, file checksums, and the string table. Add type info to
// the PDB first, so that we can get the map from object file type and item
@ -734,7 +844,12 @@ void PDBLinker::addObjFile(ObjFile *File) {
const CVIndexMap &IndexMap = *IndexMapResult;
ScopedTimer T(SymbolMergingTimer);
// Now do all live .debug$S sections.
DebugStringTableSubsectionRef CVStrTab;
DebugChecksumsSubsectionRef Checksums;
std::vector<ulittle32_t *> StringTableReferences;
for (SectionChunk *DebugChunk : File->getDebugChunks()) {
if (!DebugChunk->isLive() || DebugChunk->getSectionName() != ".debug$S")
continue;
@ -748,14 +863,17 @@ void PDBLinker::addObjFile(ObjFile *File) {
BinaryStreamReader Reader(RelocatedDebugContents, support::little);
ExitOnErr(Reader.readArray(Subsections, RelocatedDebugContents.size()));
DebugStringTableSubsectionRef CVStrTab;
DebugChecksumsSubsectionRef Checksums;
for (const DebugSubsectionRecord &SS : Subsections) {
switch (SS.kind()) {
case DebugSubsectionKind::StringTable:
case DebugSubsectionKind::StringTable: {
assert(!CVStrTab.valid() &&
"Encountered multiple string table subsections!");
ExitOnErr(CVStrTab.initialize(SS.getRecordData()));
break;
}
case DebugSubsectionKind::FileChecksums:
assert(!Checksums.valid() &&
"Encountered multiple checksum subsections!");
ExitOnErr(Checksums.initialize(SS.getRecordData()));
break;
case DebugSubsectionKind::Lines:
@ -766,10 +884,12 @@ void PDBLinker::addObjFile(ObjFile *File) {
case DebugSubsectionKind::Symbols:
if (Config->DebugGHashes) {
mergeSymbolRecords(Alloc, File, Builder.getGsiBuilder(), IndexMap,
GlobalIDTable, SS.getRecordData());
GlobalIDTable, StringTableReferences,
SS.getRecordData());
} else {
mergeSymbolRecords(Alloc, File, Builder.getGsiBuilder(), IndexMap,
IDTable, SS.getRecordData());
IDTable, StringTableReferences,
SS.getRecordData());
}
break;
default:
@ -777,25 +897,55 @@ void PDBLinker::addObjFile(ObjFile *File) {
break;
}
}
if (Checksums.valid()) {
// Make a new file checksum table that refers to offsets in the PDB-wide
// string table. Generally the string table subsection appears after the
// checksum table, so we have to do this after looping over all the
// subsections.
if (!CVStrTab.valid())
fatal(".debug$S sections must have both a string table subsection "
"and a checksum subsection table or neither");
auto NewChecksums = make_unique<DebugChecksumsSubsection>(PDBStrTab);
for (FileChecksumEntry &FC : Checksums) {
StringRef FileName = ExitOnErr(CVStrTab.getString(FC.FileNameOffset));
ExitOnErr(Builder.getDbiBuilder().addModuleSourceFile(*File->ModuleDBI,
FileName));
NewChecksums->addChecksum(FileName, FC.Kind, FC.Checksum);
}
File->ModuleDBI->addDebugSubsection(std::move(NewChecksums));
}
}
// We should have seen all debug subsections across the entire object file now
// which means that if a StringTable subsection and Checksums subsection were
// present, now is the time to handle them.
if (!CVStrTab.valid()) {
if (Checksums.valid())
fatal(".debug$S sections with a checksums subsection must also contain a "
"string table subsection");
if (!StringTableReferences.empty())
warn("No StringTable subsection was encountered, but there are string "
"table references");
return;
}
// Rewrite each string table reference based on the value that the string
// assumes in the final PDB.
for (ulittle32_t *Ref : StringTableReferences) {
auto ExpectedString = CVStrTab.getString(*Ref);
if (!ExpectedString) {
warn("Invalid string table reference");
consumeError(ExpectedString.takeError());
continue;
}
*Ref = PDBStrTab.insert(*ExpectedString);
}
// Make a new file checksum table that refers to offsets in the PDB-wide
// string table. Generally the string table subsection appears after the
// checksum table, so we have to do this after looping over all the
// subsections.
auto NewChecksums = make_unique<DebugChecksumsSubsection>(PDBStrTab);
for (FileChecksumEntry &FC : Checksums) {
SmallString<128> FileName = ExitOnErr(CVStrTab.getString(FC.FileNameOffset));
if (!sys::path::is_absolute(FileName) &&
!Config->PDBSourcePath.empty()) {
SmallString<128> AbsoluteFileName = Config->PDBSourcePath;
sys::path::append(AbsoluteFileName, FileName);
sys::path::native(AbsoluteFileName);
sys::path::remove_dots(AbsoluteFileName, /*remove_dot_dots=*/true);
FileName = std::move(AbsoluteFileName);
}
ExitOnErr(Builder.getDbiBuilder().addModuleSourceFile(*File->ModuleDBI,
FileName));
NewChecksums->addChecksum(FileName, FC.Kind, FC.Checksum);
}
File->ModuleDBI->addDebugSubsection(std::move(NewChecksums));
}
static PublicSym32 createPublic(Defined *Def) {
@ -818,12 +968,15 @@ static PublicSym32 createPublic(Defined *Def) {
// Add all object files to the PDB. Merge .debug$T sections into IpiData and
// TpiData.
void PDBLinker::addObjectsToPDB() {
ScopedTimer T1(AddObjectsTimer);
for (ObjFile *File : ObjFile::Instances)
addObjFile(File);
Builder.getStringTableBuilder().setStrings(PDBStrTab);
T1.stop();
// Construct TPI and IPI stream contents.
ScopedTimer T2(TpiStreamLayoutTimer);
if (Config->DebugGHashes) {
addTypeInfo(Builder.getTpiBuilder(), GlobalTypeTable);
addTypeInfo(Builder.getIpiBuilder(), GlobalIDTable);
@ -831,7 +984,9 @@ void PDBLinker::addObjectsToPDB() {
addTypeInfo(Builder.getTpiBuilder(), TypeTable);
addTypeInfo(Builder.getIpiBuilder(), IDTable);
}
T2.stop();
ScopedTimer T3(GlobalsLayoutTimer);
// Compute the public and global symbols.
auto &GsiBuilder = Builder.getGsiBuilder();
std::vector<PublicSym32> Publics;
@ -853,6 +1008,35 @@ void PDBLinker::addObjectsToPDB() {
}
}
void PDBLinker::addNatvisFiles() {
for (StringRef File : Config->NatvisFiles) {
ErrorOr<std::unique_ptr<MemoryBuffer>> DataOrErr =
MemoryBuffer::getFile(File);
if (!DataOrErr) {
warn("Cannot open input file: " + File);
continue;
}
Builder.addInjectedSource(File, std::move(*DataOrErr));
}
}
static codeview::CPUType toCodeViewMachine(COFF::MachineTypes Machine) {
switch (Machine) {
case COFF::IMAGE_FILE_MACHINE_AMD64:
return codeview::CPUType::X64;
case COFF::IMAGE_FILE_MACHINE_ARM:
return codeview::CPUType::ARM7;
case COFF::IMAGE_FILE_MACHINE_ARM64:
return codeview::CPUType::ARM64;
case COFF::IMAGE_FILE_MACHINE_ARMNT:
return codeview::CPUType::ARMNT;
case COFF::IMAGE_FILE_MACHINE_I386:
return codeview::CPUType::Intel80386;
default:
llvm_unreachable("Unsupported CPU Type");
}
}
static void addCommonLinkerModuleSymbols(StringRef Path,
pdb::DbiModuleDescriptorBuilder &Mod,
BumpPtrAllocator &Allocator) {
@ -863,7 +1047,7 @@ static void addCommonLinkerModuleSymbols(StringRef Path,
ONS.Name = "* Linker *";
ONS.Signature = 0;
CS.Machine = Config->is64() ? CPUType::X64 : CPUType::Intel80386;
CS.Machine = toCodeViewMachine(Config->Machine);
// Interestingly, if we set the string to 0.0.0.0, then when trying to view
// local variables WinDbg emits an error that private symbols are not present.
// By setting this to a valid MSVC linker version string, local variables are
@ -914,9 +1098,9 @@ static void addLinkerModuleSectionSymbol(pdb::DbiModuleDescriptorBuilder &Mod,
BumpPtrAllocator &Allocator) {
SectionSym Sym(SymbolRecordKind::SectionSym);
Sym.Alignment = 12; // 2^12 = 4KB
Sym.Characteristics = OS.getCharacteristics();
Sym.Characteristics = OS.Header.Characteristics;
Sym.Length = OS.getVirtualSize();
Sym.Name = OS.getName();
Sym.Name = OS.Name;
Sym.Rva = OS.getRVA();
Sym.SectionNumber = OS.SectionIndex;
Mod.addSymbol(codeview::SymbolSerializer::writeOneSymbol(
@ -928,10 +1112,15 @@ void coff::createPDB(SymbolTable *Symtab,
ArrayRef<OutputSection *> OutputSections,
ArrayRef<uint8_t> SectionTable,
const llvm::codeview::DebugInfo &BuildId) {
ScopedTimer T1(TotalPdbLinkTimer);
PDBLinker PDB(Symtab);
PDB.initialize(BuildId);
PDB.addObjectsToPDB();
PDB.addSections(OutputSections, SectionTable);
PDB.addNatvisFiles();
ScopedTimer T2(DiskCommitTimer);
PDB.commit();
}
@ -945,44 +1134,22 @@ void PDBLinker::initialize(const llvm::codeview::DebugInfo &BuildId) {
// Add an Info stream.
auto &InfoBuilder = Builder.getInfoBuilder();
InfoBuilder.setAge(BuildId.PDB70.Age);
GUID uuid;
memcpy(&uuid, &BuildId.PDB70.Signature, sizeof(uuid));
InfoBuilder.setAge(BuildId.PDB70.Age);
InfoBuilder.setGuid(uuid);
InfoBuilder.setSignature(time(nullptr));
InfoBuilder.setVersion(pdb::PdbRaw_ImplVer::PdbImplVC70);
// Add an empty DBI stream.
pdb::DbiStreamBuilder &DbiBuilder = Builder.getDbiBuilder();
DbiBuilder.setAge(BuildId.PDB70.Age);
DbiBuilder.setVersionHeader(pdb::PdbDbiV70);
ExitOnErr(DbiBuilder.addDbgStream(pdb::DbgHeaderType::NewFPO, {}));
}
void PDBLinker::addSectionContrib(pdb::DbiModuleDescriptorBuilder &LinkerModule,
OutputSection *OS, Chunk *C) {
pdb::SectionContrib SC;
memset(&SC, 0, sizeof(SC));
SC.ISect = OS->SectionIndex;
SC.Off = C->getRVA() - OS->getRVA();
SC.Size = C->getSize();
if (auto *SecChunk = dyn_cast<SectionChunk>(C)) {
SC.Characteristics = SecChunk->Header->Characteristics;
SC.Imod = SecChunk->File->ModuleDBI->getModuleIndex();
ArrayRef<uint8_t> Contents = SecChunk->getContents();
JamCRC CRC(0);
ArrayRef<char> CharContents = makeArrayRef(
reinterpret_cast<const char *>(Contents.data()), Contents.size());
CRC.update(CharContents);
SC.DataCrc = CRC.getCRC();
} else {
SC.Characteristics = OS->getCharacteristics();
// FIXME: When we start creating DBI for import libraries, use those here.
SC.Imod = LinkerModule.getModuleIndex();
}
SC.RelocCrc = 0; // FIXME
Builder.getDbiBuilder().addSectionContrib(SC);
DbiBuilder.setMachineType(Config->Machine);
// Technically we are not link.exe 14.11, but there are known cases where
// debugging tools on Windows expect Microsoft-specific version numbers or
// they fail to work at all. Since we know we produce PDBs that are
// compatible with LINK 14.11, we set that version number here.
DbiBuilder.setBuildNumber(14, 11);
}
void PDBLinker::addSections(ArrayRef<OutputSection *> OutputSections,
@ -1000,8 +1167,11 @@ void PDBLinker::addSections(ArrayRef<OutputSection *> OutputSections,
// Add section contributions. They must be ordered by ascending RVA.
for (OutputSection *OS : OutputSections) {
addLinkerModuleSectionSymbol(LinkerModule, *OS, Alloc);
for (Chunk *C : OS->getChunks())
addSectionContrib(LinkerModule, OS, C);
for (Chunk *C : OS->getChunks()) {
pdb::SectionContrib SC =
createSectionContrib(C, LinkerModule.getModuleIndex());
Builder.getDbiBuilder().addSectionContrib(SC);
}
}
// Add Section Map stream.
@ -1020,3 +1190,145 @@ void PDBLinker::commit() {
// Write to a file.
ExitOnErr(Builder.commit(Config->PDBPath));
}
static Expected<StringRef>
getFileName(const DebugStringTableSubsectionRef &Strings,
const DebugChecksumsSubsectionRef &Checksums, uint32_t FileID) {
auto Iter = Checksums.getArray().at(FileID);
if (Iter == Checksums.getArray().end())
return make_error<CodeViewError>(cv_error_code::no_records);
uint32_t Offset = Iter->FileNameOffset;
return Strings.getString(Offset);
}
static uint32_t getSecrelReloc() {
switch (Config->Machine) {
case AMD64:
return COFF::IMAGE_REL_AMD64_SECREL;
case I386:
return COFF::IMAGE_REL_I386_SECREL;
case ARMNT:
return COFF::IMAGE_REL_ARM_SECREL;
case ARM64:
return COFF::IMAGE_REL_ARM64_SECREL;
default:
llvm_unreachable("unknown machine type");
}
}
// Try to find a line table for the given offset Addr into the given chunk C.
// If a line table was found, the line table, the string and checksum tables
// that are used to interpret the line table, and the offset of Addr in the line
// table are stored in the output arguments. Returns whether a line table was
// found.
static bool findLineTable(const SectionChunk *C, uint32_t Addr,
DebugStringTableSubsectionRef &CVStrTab,
DebugChecksumsSubsectionRef &Checksums,
DebugLinesSubsectionRef &Lines,
uint32_t &OffsetInLinetable) {
ExitOnError ExitOnErr;
uint32_t SecrelReloc = getSecrelReloc();
for (SectionChunk *DbgC : C->File->getDebugChunks()) {
if (DbgC->getSectionName() != ".debug$S")
continue;
// Build a mapping of SECREL relocations in DbgC that refer to C.
DenseMap<uint32_t, uint32_t> Secrels;
for (const coff_relocation &R : DbgC->Relocs) {
if (R.Type != SecrelReloc)
continue;
if (auto *S = dyn_cast_or_null<DefinedRegular>(
C->File->getSymbols()[R.SymbolTableIndex]))
if (S->getChunk() == C)
Secrels[R.VirtualAddress] = S->getValue();
}
ArrayRef<uint8_t> Contents =
consumeDebugMagic(DbgC->getContents(), ".debug$S");
DebugSubsectionArray Subsections;
BinaryStreamReader Reader(Contents, support::little);
ExitOnErr(Reader.readArray(Subsections, Contents.size()));
for (const DebugSubsectionRecord &SS : Subsections) {
switch (SS.kind()) {
case DebugSubsectionKind::StringTable: {
assert(!CVStrTab.valid() &&
"Encountered multiple string table subsections!");
ExitOnErr(CVStrTab.initialize(SS.getRecordData()));
break;
}
case DebugSubsectionKind::FileChecksums:
assert(!Checksums.valid() &&
"Encountered multiple checksum subsections!");
ExitOnErr(Checksums.initialize(SS.getRecordData()));
break;
case DebugSubsectionKind::Lines: {
ArrayRef<uint8_t> Bytes;
auto Ref = SS.getRecordData();
ExitOnErr(Ref.readLongestContiguousChunk(0, Bytes));
size_t OffsetInDbgC = Bytes.data() - DbgC->getContents().data();
// Check whether this line table refers to C.
auto I = Secrels.find(OffsetInDbgC);
if (I == Secrels.end())
break;
// Check whether this line table covers Addr in C.
DebugLinesSubsectionRef LinesTmp;
ExitOnErr(LinesTmp.initialize(BinaryStreamReader(Ref)));
uint32_t OffsetInC = I->second + LinesTmp.header()->RelocOffset;
if (Addr < OffsetInC || Addr >= OffsetInC + LinesTmp.header()->CodeSize)
break;
assert(!Lines.header() &&
"Encountered multiple line tables for function!");
ExitOnErr(Lines.initialize(BinaryStreamReader(Ref)));
OffsetInLinetable = Addr - OffsetInC;
break;
}
default:
break;
}
if (CVStrTab.valid() && Checksums.valid() && Lines.header())
return true;
}
}
return false;
}
// Use CodeView line tables to resolve a file and line number for the given
// offset into the given chunk and return them, or {"", 0} if a line table was
// not found.
std::pair<StringRef, uint32_t> coff::getFileLine(const SectionChunk *C,
uint32_t Addr) {
ExitOnError ExitOnErr;
DebugStringTableSubsectionRef CVStrTab;
DebugChecksumsSubsectionRef Checksums;
DebugLinesSubsectionRef Lines;
uint32_t OffsetInLinetable;
if (!findLineTable(C, Addr, CVStrTab, Checksums, Lines, OffsetInLinetable))
return {"", 0};
uint32_t NameIndex;
uint32_t LineNumber;
for (LineColumnEntry &Entry : Lines) {
for (const LineNumberEntry &LN : Entry.LineNumbers) {
if (LN.Offset > OffsetInLinetable) {
StringRef Filename =
ExitOnErr(getFileName(CVStrTab, Checksums, NameIndex));
return {Filename, LineNumber};
}
LineInfo LI(LN.Flags);
NameIndex = Entry.NameIndex;
LineNumber = LI.getStartLine();
}
}
StringRef Filename = ExitOnErr(getFileName(CVStrTab, Checksums, NameIndex));
return {Filename, LineNumber};
}

4
contrib/llvm/tools/lld/COFF/PDB.h
View File

@ -22,12 +22,16 @@ union DebugInfo;
namespace lld {
namespace coff {
class OutputSection;
class SectionChunk;
class SymbolTable;
void createPDB(SymbolTable *Symtab,
llvm::ArrayRef<OutputSection *> OutputSections,
llvm::ArrayRef<uint8_t> SectionTable,
const llvm::codeview::DebugInfo &BuildId);
std::pair<llvm::StringRef, uint32_t> getFileLine(const SectionChunk *C,
uint32_t Addr);
}
}

35
contrib/llvm/tools/lld/COFF/Strings.cpp
View File

@ -1,35 +0,0 @@
//===- Strings.cpp -------------------------------------------------------===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "Strings.h"
#include <mutex>
#if defined(_MSC_VER)
#include <Windows.h>
#include <DbgHelp.h>
#pragma comment(lib, "dbghelp.lib")
#endif
using namespace lld;
using namespace lld::coff;
using namespace llvm;
Optional<std::string> coff::demangleMSVC(StringRef S) {
#if defined(_MSC_VER)
// UnDecorateSymbolName is not thread-safe, so we need a mutex.
static std::mutex Mu;
std::lock_guard<std::mutex> Lock(Mu);
char Buf[4096];
if (S.startswith("?"))
if (size_t Len = UnDecorateSymbolName(S.str().c_str(), Buf, sizeof(Buf), 0))
return std::string(Buf, Len);
#endif
return None;
}

96
contrib/llvm/tools/lld/COFF/SymbolTable.cpp
View File

@ -11,9 +11,11 @@
#include "Config.h"
#include "Driver.h"
#include "LTO.h"
#include "PDB.h"
#include "Symbols.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Timer.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"
@ -24,6 +26,8 @@ using namespace llvm;
namespace lld {
namespace coff {
static Timer LTOTimer("LTO", Timer::root());
SymbolTable *Symtab;
void SymbolTable::addFile(InputFile *File) {
@ -34,8 +38,9 @@ void SymbolTable::addFile(InputFile *File) {
if (Config->Machine == IMAGE_FILE_MACHINE_UNKNOWN) {
Config->Machine = MT;
} else if (MT != IMAGE_FILE_MACHINE_UNKNOWN && Config->Machine != MT) {
fatal(toString(File) + ": machine type " + machineToStr(MT) +
error(toString(File) + ": machine type " + machineToStr(MT) +
" conflicts with " + machineToStr(Config->Machine));
return;
}
if (auto *F = dyn_cast<ObjFile>(File)) {
@ -61,6 +66,66 @@ static void errorOrWarn(const Twine &S) {
error(S);
}
// Returns the name of the symbol in SC whose value is <= Addr that is closest
// to Addr. This is generally the name of the global variable or function whose
// definition contains Addr.
static StringRef getSymbolName(SectionChunk *SC, uint32_t Addr) {
DefinedRegular *Candidate = nullptr;
for (Symbol *S : SC->File->getSymbols()) {
auto *D = dyn_cast_or_null<DefinedRegular>(S);
if (!D || D->getChunk() != SC || D->getValue() > Addr ||
(Candidate && D->getValue() < Candidate->getValue()))
continue;
Candidate = D;
}
if (!Candidate)
return "";
return Candidate->getName();
}
static std::string getSymbolLocations(ObjFile *File, uint32_t SymIndex) {
struct Location {
StringRef SymName;
std::pair<StringRef, uint32_t> FileLine;
};
std::vector<Location> Locations;
for (Chunk *C : File->getChunks()) {
auto *SC = dyn_cast<SectionChunk>(C);
if (!SC)
continue;
for (const coff_relocation &R : SC->Relocs) {
if (R.SymbolTableIndex != SymIndex)
continue;
std::pair<StringRef, uint32_t> FileLine =
getFileLine(SC, R.VirtualAddress);
StringRef SymName = getSymbolName(SC, R.VirtualAddress);
if (!FileLine.first.empty() || !SymName.empty())
Locations.push_back({SymName, FileLine});
}
}
if (Locations.empty())
return "\n>>> referenced by " + toString(File) + "\n";
std::string Out;
llvm::raw_string_ostream OS(Out);
for (Location Loc : Locations) {
OS << "\n>>> referenced by ";
if (!Loc.FileLine.first.empty())
OS << Loc.FileLine.first << ":" << Loc.FileLine.second
<< "\n>>> ";
OS << toString(File);
if (!Loc.SymName.empty())
OS << ":(" << Loc.SymName << ')';
}
OS << '\n';
return OS.str();
}
void SymbolTable::reportRemainingUndefines() {
SmallPtrSet<Symbol *, 8> Undefs;
DenseMap<Symbol *, Symbol *> LocalImports;
@ -120,20 +185,23 @@ void SymbolTable::reportRemainingUndefines() {
if (Config->WarnLocallyDefinedImported)
if (Symbol *Imp = LocalImports.lookup(B))
warn("<root>: locally defined symbol imported: " + Imp->getName() +
" (defined in " + toString(Imp->getFile()) + ")");
" (defined in " + toString(Imp->getFile()) + ") [LNK4217]");
}
for (ObjFile *File : ObjFile::Instances) {
size_t SymIndex = (size_t)-1;
for (Symbol *Sym : File->getSymbols()) {
++SymIndex;
if (!Sym)
continue;
if (Undefs.count(Sym))
errorOrWarn(toString(File) + ": undefined symbol: " + Sym->getName());
errorOrWarn("undefined symbol: " + Sym->getName() +
getSymbolLocations(File, SymIndex));
if (Config->WarnLocallyDefinedImported)
if (Symbol *Imp = LocalImports.lookup(Sym))
warn(toString(File) + ": locally defined symbol imported: " +
Imp->getName() + " (defined in " + toString(Imp->getFile()) +
")");
") [LNK4217]");
}
}
}
@ -142,7 +210,7 @@ std::pair<Symbol *, bool> SymbolTable::insert(StringRef Name) {
Symbol *&Sym = SymMap[CachedHashStringRef(Name)];
if (Sym)
return {Sym, false};
Sym = (Symbol *)make<SymbolUnion>();
Sym = reinterpret_cast<Symbol *>(make<SymbolUnion>());
Sym->IsUsedInRegularObj = false;
Sym->PendingArchiveLoad = false;
return {Sym, true};
@ -274,30 +342,29 @@ Symbol *SymbolTable::addCommon(InputFile *F, StringRef N, uint64_t Size,
return S;
}
DefinedImportData *SymbolTable::addImportData(StringRef N, ImportFile *F) {
Symbol *SymbolTable::addImportData(StringRef N, ImportFile *F) {
Symbol *S;
bool WasInserted;
std::tie(S, WasInserted) = insert(N);
S->IsUsedInRegularObj = true;
if (WasInserted || isa<Undefined>(S) || isa<Lazy>(S)) {
replaceSymbol<DefinedImportData>(S, N, F);
return cast<DefinedImportData>(S);
return S;
}
reportDuplicate(S, F);
return nullptr;
}
DefinedImportThunk *SymbolTable::addImportThunk(StringRef Name,
DefinedImportData *ID,
uint16_t Machine) {
Symbol *SymbolTable::addImportThunk(StringRef Name, DefinedImportData *ID,
uint16_t Machine) {
Symbol *S;
bool WasInserted;
std::tie(S, WasInserted) = insert(Name);
S->IsUsedInRegularObj = true;
if (WasInserted || isa<Undefined>(S) || isa<Lazy>(S)) {
replaceSymbol<DefinedImportThunk>(S, Name, ID, Machine);
return cast<DefinedImportThunk>(S);
return S;
}
reportDuplicate(S, ID->File);
@ -314,10 +381,7 @@ std::vector<Chunk *> SymbolTable::getChunks() {
}
Symbol *SymbolTable::find(StringRef Name) {
auto It = SymMap.find(CachedHashStringRef(Name));
if (It == SymMap.end())
return nullptr;
return It->second;
return SymMap.lookup(CachedHashStringRef(Name));
}
Symbol *SymbolTable::findUnderscore(StringRef Name) {
@ -384,6 +448,8 @@ std::vector<StringRef> SymbolTable::compileBitcodeFiles() {
void SymbolTable::addCombinedLTOObjects() {
if (BitcodeFile::Instances.empty())
return;
ScopedTimer T(LTOTimer);
for (StringRef Object : compileBitcodeFiles()) {
auto *Obj = make<ObjFile>(MemoryBufferRef(Object, "lto.tmp"));
Obj->parse();

6
contrib/llvm/tools/lld/COFF/SymbolTable.h
View File

@ -92,9 +92,9 @@ class SymbolTable {
Symbol *addCommon(InputFile *F, StringRef N, uint64_t Size,
const llvm::object::coff_symbol_generic *S = nullptr,
CommonChunk *C = nullptr);
DefinedImportData *addImportData(StringRef N, ImportFile *F);
DefinedImportThunk *addImportThunk(StringRef Name, DefinedImportData *S,
uint16_t Machine);
Symbol *addImportData(StringRef N, ImportFile *F);
Symbol *addImportThunk(StringRef Name, DefinedImportData *S,
uint16_t Machine);
void reportDuplicate(Symbol *Existing, InputFile *NewFile);

8
contrib/llvm/tools/lld/COFF/Symbols.cpp
View File

@ -9,9 +9,9 @@
#include "Symbols.h"
#include "InputFiles.h"
#include "Strings.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Strings.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"
@ -21,7 +21,7 @@ using namespace llvm::object;
// Returns a symbol name for an error message.
std::string lld::toString(coff::Symbol &B) {
if (Optional<std::string> S = coff::demangleMSVC(B.getName()))
if (Optional<std::string> S = lld::demangleMSVC(B.getName()))
return ("\"" + *S + "\" (" + B.getName() + ")").str();
return B.getName();
}
@ -58,7 +58,7 @@ bool Symbol::isLive() const {
if (auto *Imp = dyn_cast<DefinedImportData>(this))
return Imp->File->Live;
if (auto *Imp = dyn_cast<DefinedImportThunk>(this))
return Imp->WrappedSym->File->Live;
return Imp->WrappedSym->File->ThunkLive;
// Assume any other kind of symbol is live.
return true;
}
@ -71,7 +71,7 @@ COFFSymbolRef DefinedCOFF::getCOFFSymbol() {
return COFFSymbolRef(reinterpret_cast<const coff_symbol32 *>(Sym));
}
uint16_t DefinedAbsolute::OutputSectionIndex = 0;
uint16_t DefinedAbsolute::NumOutputSections;
static Chunk *makeImportThunk(DefinedImportData *S, uint16_t Machine) {
if (Machine == AMD64)

11
contrib/llvm/tools/lld/COFF/Symbols.h
View File

@ -213,11 +213,10 @@ class DefinedAbsolute : public Defined {
uint64_t getRVA() { return VA - Config->ImageBase; }
void setVA(uint64_t V) { VA = V; }
// The sentinel absolute symbol section index. Section index relocations
// against absolute symbols resolve to this 16 bit number, and it is the
// largest valid section index plus one. This is written by the Writer.
static uint16_t OutputSectionIndex;
uint16_t getSecIdx() { return OutputSectionIndex; }
// Section index relocations against absolute symbols resolve to
// this 16 bit number, and it is the largest valid section index
// plus one. This variable keeps it.
static uint16_t NumOutputSections;
private:
uint64_t VA;
@ -416,6 +415,8 @@ union SymbolUnion {
template <typename T, typename... ArgT>
void replaceSymbol(Symbol *S, ArgT &&... Arg) {
static_assert(std::is_trivially_destructible<T>(),
"Symbol types must be trivially destructible");
static_assert(sizeof(T) <= sizeof(SymbolUnion), "Symbol too small");
static_assert(alignof(T) <= alignof(SymbolUnion),
"SymbolUnion not aligned enough");

829
contrib/llvm/tools/lld/COFF/Writer.cpp
View File

File diff suppressed because it is too large Load Diff

16
contrib/llvm/tools/lld/COFF/Writer.h
View File

@ -13,6 +13,7 @@
#include "Chunks.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/Object/COFF.h"
#include <chrono>
#include <cstdint>
#include <vector>
@ -29,16 +30,14 @@ void writeResult();
// non-overlapping file offsets and RVAs.
class OutputSection {
public:
OutputSection(llvm::StringRef N) : Name(N), Header({}) {}
void setRVA(uint64_t);
void setFileOffset(uint64_t);
OutputSection(llvm::StringRef N, uint32_t Chars) : Name(N) {
Header.Characteristics = Chars;
}
void addChunk(Chunk *C);
llvm::StringRef getName() { return Name; }
void merge(OutputSection *Other);
ArrayRef<Chunk *> getChunks() { return Chunks; }
void addPermissions(uint32_t C);
void setPermissions(uint32_t C);
uint32_t getPermissions() { return Header.Characteristics & PermMask; }
uint32_t getCharacteristics() { return Header.Characteristics; }
uint64_t getRVA() { return Header.VirtualAddress; }
uint64_t getFileOff() { return Header.PointerToRawData; }
void writeHeaderTo(uint8_t *Buf);
@ -60,9 +59,10 @@ class OutputSection {
// N.B. The section index is one based.
uint32_t SectionIndex = 0;
private:
llvm::StringRef Name;
llvm::object::coff_section Header;
llvm::object::coff_section Header = {};
private:
uint32_t StringTableOff = 0;
std::vector<Chunk *> Chunks;
};

18
contrib/llvm/tools/lld/Common/Args.cpp
View File

@ -18,13 +18,17 @@ using namespace llvm;
using namespace lld;
int lld::args::getInteger(opt::InputArgList &Args, unsigned Key, int Default) {
int V = Default;
if (auto *Arg = Args.getLastArg(Key)) {
StringRef S = Arg->getValue();
if (!to_integer(S, V, 10))
error(Arg->getSpelling() + ": number expected, but got '" + S + "'");
}
return V;
auto *A = Args.getLastArg(Key);
if (!A)
return Default;
int V;
if (to_integer(A->getValue(), V, 10))
return V;
StringRef Spelling = Args.getArgString(A->getIndex());
error(Spelling + ": number expected, but got '" + A->getValue() + "'");
return 0;
}
std::vector<StringRef> lld::args::getStrings(opt::InputArgList &Args, int Id) {

1
contrib/llvm/tools/lld/Common/CMakeLists.txt
View File

@ -10,6 +10,7 @@ add_lld_library(lldCommon
Strings.cpp
TargetOptionsCommandFlags.cpp
Threads.cpp
Timer.cpp
Version.cpp
ADDITIONAL_HEADER_DIRS

27
contrib/llvm/tools/lld/Common/ErrorHandler.cpp
View File

@ -12,7 +12,8 @@
#include "lld/Common/Threads.h"
#include "llvm/ADT/Twine.h"
#include "llvm/Support/Error.h"
#include "llvm/IR/DiagnosticInfo.h"
#include "llvm/IR/DiagnosticPrinter.h"
#include "llvm/Support/ManagedStatic.h"
#include "llvm/Support/raw_ostream.h"
#include <mutex>
@ -59,6 +60,30 @@ void lld::exitLld(int Val) {
_exit(Val);
}
void lld::diagnosticHandler(const DiagnosticInfo &DI) {
SmallString<128> S;
raw_svector_ostream OS(S);
DiagnosticPrinterRawOStream DP(OS);
DI.print(DP);
switch (DI.getSeverity()) {
case DS_Error:
error(S);
break;
case DS_Warning:
warn(S);
break;
case DS_Remark:
case DS_Note:
message(S);
break;
}
}
void lld::checkError(Error E) {
handleAllErrors(std::move(E),
[&](ErrorInfoBase &EIB) { error(EIB.message()); });
}
void ErrorHandler::print(StringRef S, raw_ostream::Colors C) {
*ErrorOS << LogName << ": ";
if (ColorDiagnostics) {

77
contrib/llvm/tools/lld/Common/Strings.cpp
View File

@ -8,7 +8,21 @@
//===----------------------------------------------------------------------===//
#include "lld/Common/Strings.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/LLVM.h"
#include "llvm/Demangle/Demangle.h"
#include "llvm/Support/GlobPattern.h"
#include <algorithm>
#include <mutex>
#include <vector>
#if defined(_MSC_VER)
#include <Windows.h>
// DbgHelp.h must be included after Windows.h.
#include <DbgHelp.h>
#pragma comment(lib, "dbghelp.lib")
#endif
using namespace llvm;
using namespace lld;
@ -30,3 +44,66 @@ Optional<std::string> lld::demangleItanium(StringRef Name) {
free(Buf);
return S;
}
Optional<std::string> lld::demangleMSVC(StringRef S) {
#if defined(_MSC_VER)
// UnDecorateSymbolName is not thread-safe, so we need a mutex.
static std::mutex Mu;
std::lock_guard<std::mutex> Lock(Mu);
char Buf[4096];
if (S.startswith("?"))
if (size_t Len = UnDecorateSymbolName(S.str().c_str(), Buf, sizeof(Buf), 0))
return std::string(Buf, Len);
#endif
return None;
}
StringMatcher::StringMatcher(ArrayRef<StringRef> Pat) {
for (StringRef S : Pat) {
Expected<GlobPattern> Pat = GlobPattern::create(S);
if (!Pat)
error(toString(Pat.takeError()));
else
Patterns.push_back(*Pat);
}
}
bool StringMatcher::match(StringRef S) const {
for (const GlobPattern &Pat : Patterns)
if (Pat.match(S))
return true;
return false;
}
// Converts a hex string (e.g. "deadbeef") to a vector.
std::vector<uint8_t> lld::parseHex(StringRef S) {
std::vector<uint8_t> Hex;
while (!S.empty()) {
StringRef B = S.substr(0, 2);
S = S.substr(2);
uint8_t H;
if (!to_integer(B, H, 16)) {
error("not a hexadecimal value: " + B);
return {};
}
Hex.push_back(H);
}
return Hex;
}
// Returns true if S is valid as a C language identifier.
bool lld::isValidCIdentifier(StringRef S) {
return !S.empty() && (isAlpha(S[0]) || S[0] == '_') &&
std::all_of(S.begin() + 1, S.end(),
[](char C) { return C == '_' || isAlnum(C); });
}
// Write the contents of the a buffer to a file
void lld::saveBuffer(StringRef Buffer, const Twine &Path) {
std::error_code EC;
raw_fd_ostream OS(Path.str(), EC, sys::fs::OpenFlags::F_None);
if (EC)
error("cannot create " + Path + ": " + EC.message());
OS << Buffer;
}

8
contrib/llvm/tools/lld/Common/TargetOptionsCommandFlags.cpp
View File

@ -8,7 +8,7 @@
//===----------------------------------------------------------------------===//
//
// This file exists as a place for global variables defined in LLVM's
// CodeGen/CommandFlags.def. By putting the resulting object file in
// CodeGen/CommandFlags.inc. By putting the resulting object file in
// an archive and linking with it, the definitions will automatically be
// included when needed and skipped when already present.
//
@ -16,12 +16,12 @@
#include "lld/Common/TargetOptionsCommandFlags.h"
#include "llvm/CodeGen/CommandFlags.def"
#include "llvm/CodeGen/CommandFlags.inc"
#include "llvm/Target/TargetOptions.h"
// Define an externally visible version of
// InitTargetOptionsFromCodeGenFlags, so that its functionality can be
// used without having to include llvm/CodeGen/CommandFlags.def, which
// used without having to include llvm/CodeGen/CommandFlags.inc, which
// would lead to multiple definitions of the command line flags.
llvm::TargetOptions lld::InitTargetOptionsFromCodeGenFlags() {
return ::InitTargetOptionsFromCodeGenFlags();
@ -30,3 +30,5 @@ llvm::TargetOptions lld::InitTargetOptionsFromCodeGenFlags() {
llvm::Optional<llvm::CodeModel::Model> lld::GetCodeModelFromCMModel() {
return getCodeModel();
}
std::string lld::GetCPUStr() { return ::getCPUStr(); }

80
contrib/llvm/tools/lld/Common/Timer.cpp Normal file
View File

@ -0,0 +1,80 @@
//===- Timer.cpp ----------------------------------------------------------===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "lld/Common/Timer.h"
#include "lld/Common/ErrorHandler.h"
#include "llvm/Support/Format.h"
using namespace lld;
using namespace llvm;
ScopedTimer::ScopedTimer(Timer &T) : T(&T) { T.start(); }
void ScopedTimer::stop() {
if (!T)
return;
T->stop();
T = nullptr;
}
ScopedTimer::~ScopedTimer() { stop(); }
Timer::Timer(llvm::StringRef Name) : Name(Name), Parent(nullptr) {}
Timer::Timer(llvm::StringRef Name, Timer &Parent)
: Name(Name), Parent(&Parent) {}
void Timer::start() {
if (Parent && Total.count() == 0)
Parent->Children.push_back(this);
StartTime = std::chrono::high_resolution_clock::now();
}
void Timer::stop() {
Total += (std::chrono::high_resolution_clock::now() - StartTime);
}
Timer &Timer::root() {
static Timer RootTimer("Total Link Time");
return RootTimer;
}
void Timer::print() {
double TotalDuration = static_cast<double>(root().millis());
// We want to print the grand total under all the intermediate phases, so we
// print all children first, then print the total under that.
for (const auto &Child : Children)
Child->print(1, TotalDuration);
message(std::string(49, '-'));
root().print(0, root().millis(), false);
}
double Timer::millis() const {
return std::chrono::duration_cast<std::chrono::duration<double, std::milli>>(
Total)
.count();
}
void Timer::print(int Depth, double TotalDuration, bool Recurse) const {
double P = 100.0 * millis() / TotalDuration;
SmallString<32> Str;
llvm::raw_svector_ostream Stream(Str);
std::string S = std::string(Depth * 2, ' ') + Name + std::string(":");
Stream << format("%-30s%5d ms (%5.1f%%)", S.c_str(), (int)millis(), P);
message(Str);
if (Recurse) {
for (const auto &Child : Children)
Child->print(Depth + 1, TotalDuration);
}
}

14
contrib/llvm/tools/lld/ELF/AArch64ErrataFix.cpp
View File

@ -34,12 +34,11 @@
#include "LinkerScript.h"
#include "OutputSections.h"
#include "Relocations.h"
#include "Strings.h"
#include "Symbols.h"
#include "SyntheticSections.h"
#include "Target.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Strings.h"
#include "llvm/Support/Endian.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
@ -342,7 +341,7 @@ static bool is843419ErratumSequence(uint32_t Instr1, uint32_t Instr2,
// patch or 0 if no patch required.
static uint64_t scanCortexA53Errata843419(InputSection *IS, uint64_t &Off,
uint64_t Limit) {
uint64_t ISAddr = IS->getParent()->Addr + IS->OutSecOff;
uint64_t ISAddr = IS->getVA(0);
// Advance Off so that (ISAddr + Off) modulo 0x1000 is at least 0xff8.
uint64_t InitialPageOff = (ISAddr + Off) & 0xfff;
@ -406,7 +405,7 @@ lld::elf::Patch843419Section::Patch843419Section(InputSection *P, uint64_t Off)
}
uint64_t lld::elf::Patch843419Section::getLDSTAddr() const {
return Patchee->getParent()->Addr + Patchee->OutSecOff + PatcheeOffset;
return Patchee->getVA(PatcheeOffset);
}
void lld::elf::Patch843419Section::writeTo(uint8_t *Buf) {
@ -555,9 +554,8 @@ static void implementPatch(uint64_t AdrpAddr, uint64_t PatcheeOffset,
if (RelIt != IS->Relocations.end() && RelIt->Type == R_AARCH64_JUMP26)
return;
if (Config->Verbose)
message("detected cortex-a53-843419 erratum sequence starting at " +
utohexstr(AdrpAddr) + " in unpatched output.");
log("detected cortex-a53-843419 erratum sequence starting at " +
utohexstr(AdrpAddr) + " in unpatched output.");
auto *PS = make<Patch843419Section>(IS, PatcheeOffset);
Patches.push_back(PS);
@ -603,7 +601,7 @@ AArch64Err843419Patcher::patchInputSectionDescription(
(DataSym == MapSyms.end()) ? IS->Data.size() : (*DataSym)->Value;
while (Off < Limit) {
uint64_t StartAddr = IS->getParent()->Addr + IS->OutSecOff + Off;
uint64_t StartAddr = IS->getVA(Off);
if (uint64_t PatcheeOffset = scanCortexA53Errata843419(IS, Off, Limit))
implementPatch(StartAddr, PatcheeOffset, IS, Patches);
}

1
contrib/llvm/tools/lld/ELF/AArch64ErrataFix.h
View File

@ -11,7 +11,6 @@
#define LLD_ELF_AARCH64ERRATAFIX_H
#include "lld/Common/LLVM.h"
#include <map>
#include <vector>

57
contrib/llvm/tools/lld/ELF/Arch/AArch64.cpp
View File

@ -34,7 +34,7 @@ class AArch64 final : public TargetInfo {
AArch64();
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
bool isPicRel(RelType Type) const override;
RelType getDynRel(RelType Type) const override;
void writeGotPlt(uint8_t *Buf, const Symbol &S) const override;
void writePltHeader(uint8_t *Buf) const override;
void writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr, uint64_t PltEntryAddr,
@ -89,6 +89,11 @@ RelExpr AArch64::getRelExpr(RelType Type, const Symbol &S,
return R_TLSDESC_CALL;
case R_AARCH64_TLSLE_ADD_TPREL_HI12:
case R_AARCH64_TLSLE_ADD_TPREL_LO12_NC:
case R_AARCH64_TLSLE_LDST8_TPREL_LO12_NC:
case R_AARCH64_TLSLE_LDST16_TPREL_LO12_NC:
case R_AARCH64_TLSLE_LDST32_TPREL_LO12_NC:
case R_AARCH64_TLSLE_LDST64_TPREL_LO12_NC:
case R_AARCH64_TLSLE_LDST128_TPREL_LO12_NC:
return R_TLS;
case R_AARCH64_CALL26:
case R_AARCH64_CONDBR19:
@ -144,8 +149,10 @@ bool AArch64::usesOnlyLowPageBits(RelType Type) const {
}
}
bool AArch64::isPicRel(RelType Type) const {
return Type == R_AARCH64_ABS32 || Type == R_AARCH64_ABS64;
RelType AArch64::getDynRel(RelType Type) const {
if (Type == R_AARCH64_ABS32 || Type == R_AARCH64_ABS64)
return Type;
return R_AARCH64_NONE;
}
void AArch64::writeGotPlt(uint8_t *Buf, const Symbol &) const {
@ -240,12 +247,12 @@ void AArch64::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
switch (Type) {
case R_AARCH64_ABS16:
case R_AARCH64_PREL16:
checkIntUInt<16>(Loc, Val, Type);
checkIntUInt(Loc, Val, 16, Type);
write16le(Loc, Val);
break;
case R_AARCH64_ABS32:
case R_AARCH64_PREL32:
checkIntUInt<32>(Loc, Val, Type);
checkIntUInt(Loc, Val, 32, Type);
write32le(Loc, Val);
break;
case R_AARCH64_ABS64:
@ -260,11 +267,11 @@ void AArch64::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_AARCH64_ADR_PREL_PG_HI21:
case R_AARCH64_TLSIE_ADR_GOTTPREL_PAGE21:
case R_AARCH64_TLSDESC_ADR_PAGE21:
checkInt<33>(Loc, Val, Type);
checkInt(Loc, Val, 33, Type);
write32AArch64Addr(Loc, Val >> 12);
break;
case R_AARCH64_ADR_PREL_LO21:
checkInt<21>(Loc, Val, Type);
checkInt(Loc, Val, 21, Type);
write32AArch64Addr(Loc, Val);
break;
case R_AARCH64_JUMP26:
@ -278,38 +285,40 @@ void AArch64::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
write32le(Loc, 0x14000000);
LLVM_FALLTHROUGH;
case R_AARCH64_CALL26:
checkInt<28>(Loc, Val, Type);
checkInt(Loc, Val, 28, Type);
or32le(Loc, (Val & 0x0FFFFFFC) >> 2);
break;
case R_AARCH64_CONDBR19:
case R_AARCH64_LD_PREL_LO19:
checkAlignment<4>(Loc, Val, Type);
checkInt<21>(Loc, Val, Type);
checkAlignment(Loc, Val, 4, Type);
checkInt(Loc, Val, 21, Type);
or32le(Loc, (Val & 0x1FFFFC) << 3);
break;
case R_AARCH64_LD64_GOT_LO12_NC:
case R_AARCH64_TLSIE_LD64_GOTTPREL_LO12_NC:
case R_AARCH64_TLSDESC_LD64_LO12:
checkAlignment<8>(Loc, Val, Type);
or32le(Loc, (Val & 0xFF8) << 7);
break;
case R_AARCH64_LDST8_ABS_LO12_NC:
case R_AARCH64_TLSLE_LDST8_TPREL_LO12_NC:
or32AArch64Imm(Loc, getBits(Val, 0, 11));
break;
case R_AARCH64_LDST16_ABS_LO12_NC:
checkAlignment<2>(Loc, Val, Type);
case R_AARCH64_TLSLE_LDST16_TPREL_LO12_NC:
checkAlignment(Loc, Val, 2, Type);
or32AArch64Imm(Loc, getBits(Val, 1, 11));
break;
case R_AARCH64_LDST32_ABS_LO12_NC:
checkAlignment<4>(Loc, Val, Type);
case R_AARCH64_TLSLE_LDST32_TPREL_LO12_NC:
checkAlignment(Loc, Val, 4, Type);
or32AArch64Imm(Loc, getBits(Val, 2, 11));
break;
case R_AARCH64_LDST64_ABS_LO12_NC:
checkAlignment<8>(Loc, Val, Type);
case R_AARCH64_LD64_GOT_LO12_NC:
case R_AARCH64_TLSIE_LD64_GOTTPREL_LO12_NC:
case R_AARCH64_TLSLE_LDST64_TPREL_LO12_NC:
case R_AARCH64_TLSDESC_LD64_LO12:
checkAlignment(Loc, Val, 8, Type);
or32AArch64Imm(Loc, getBits(Val, 3, 11));
break;
case R_AARCH64_LDST128_ABS_LO12_NC:
checkAlignment<16>(Loc, Val, Type);
case R_AARCH64_TLSLE_LDST128_TPREL_LO12_NC:
checkAlignment(Loc, Val, 16, Type);
or32AArch64Imm(Loc, getBits(Val, 4, 11));
break;
case R_AARCH64_MOVW_UABS_G0_NC:
@ -325,11 +334,11 @@ void AArch64::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
or32le(Loc, (Val & 0xFFFF000000000000) >> 43);
break;
case R_AARCH64_TSTBR14:
checkInt<16>(Loc, Val, Type);
checkInt(Loc, Val, 16, Type);
or32le(Loc, (Val & 0xFFFC) << 3);
break;
case R_AARCH64_TLSLE_ADD_TPREL_HI12:
checkInt<24>(Loc, Val, Type);
checkInt(Loc, Val, 24, Type);
or32AArch64Imm(Loc, Val >> 12);
break;
case R_AARCH64_TLSLE_ADD_TPREL_LO12_NC:
@ -353,7 +362,7 @@ void AArch64::relaxTlsGdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const {
// movk x0, #0x10
// nop
// nop
checkUInt<32>(Loc, Val, Type);
checkUInt(Loc, Val, 32, Type);
switch (Type) {
case R_AARCH64_TLSDESC_ADD_LO12:
@ -403,7 +412,7 @@ void AArch64::relaxTlsGdToIe(uint8_t *Loc, RelType Type, uint64_t Val) const {
}
void AArch64::relaxTlsIeToLe(uint8_t *Loc, RelType Type, uint64_t Val) const {
checkUInt<32>(Loc, Val, Type);
checkUInt(Loc, Val, 32, Type);
if (Type == R_AARCH64_TLSIE_ADR_GOTTPREL_PAGE21) {
// Generate MOVZ.

2
contrib/llvm/tools/lld/ELF/Arch/AMDGPU.cpp
View File

@ -66,6 +66,7 @@ void AMDGPU::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
write32le(Loc, Val);
break;
case R_AMDGPU_ABS64:
case R_AMDGPU_REL64:
write64le(Loc, Val);
break;
case R_AMDGPU_GOTPCREL32_HI:
@ -86,6 +87,7 @@ RelExpr AMDGPU::getRelExpr(RelType Type, const Symbol &S,
case R_AMDGPU_REL32:
case R_AMDGPU_REL32_LO:
case R_AMDGPU_REL32_HI:
case R_AMDGPU_REL64:
return R_PC;
case R_AMDGPU_GOTPCREL:
case R_AMDGPU_GOTPCREL32_LO:

30
contrib/llvm/tools/lld/ELF/Arch/ARM.cpp
View File

@ -29,7 +29,6 @@ class ARM final : public TargetInfo {
uint32_t calcEFlags() const override;
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
bool isPicRel(RelType Type) const override;
RelType getDynRel(RelType Type) const override;
int64_t getImplicitAddend(const uint8_t *Buf, RelType Type) const override;
void writeGotPlt(uint8_t *Buf, const Symbol &S) const override;
@ -55,6 +54,7 @@ ARM::ARM() {
TlsGotRel = R_ARM_TLS_TPOFF32;
TlsModuleIndexRel = R_ARM_TLS_DTPMOD32;
TlsOffsetRel = R_ARM_TLS_DTPOFF32;
GotBaseSymInGotPlt = false;
GotEntrySize = 4;
GotPltEntrySize = 4;
PltEntrySize = 16;
@ -161,18 +161,10 @@ RelExpr ARM::getRelExpr(RelType Type, const Symbol &S,
}
}
bool ARM::isPicRel(RelType Type) const {
return (Type == R_ARM_TARGET1 && !Config->Target1Rel) ||
(Type == R_ARM_ABS32);
}
RelType ARM::getDynRel(RelType Type) const {
if (Type == R_ARM_TARGET1 && !Config->Target1Rel)
if ((Type == R_ARM_ABS32) || (Type == R_ARM_TARGET1 && !Config->Target1Rel))
return R_ARM_ABS32;
if (Type == R_ARM_ABS32)
return Type;
// Keep it going with a dummy value so that we can find more reloc errors.
return R_ARM_ABS32;
return R_ARM_NONE;
}
void ARM::writeGotPlt(uint8_t *Buf, const Symbol &) const {
@ -392,7 +384,7 @@ void ARM::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
write32le(Loc, 1);
break;
case R_ARM_PREL31:
checkInt<31>(Loc, Val, Type);
checkInt(Loc, Val, 31, Type);
write32le(Loc, (read32le(Loc) & 0x80000000) | (Val & ~0x80000000));
break;
case R_ARM_CALL:
@ -401,7 +393,7 @@ void ARM::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
if (Val & 1) {
// If bit 0 of Val is 1 the target is Thumb, we must select a BLX.
// The BLX encoding is 0xfa:H:imm24 where Val = imm24:H:'1'
checkInt<26>(Loc, Val, Type);
checkInt(Loc, Val, 26, Type);
write32le(Loc, 0xfa000000 | // opcode
((Val & 2) << 23) | // H
((Val >> 2) & 0x00ffffff)); // imm24
@ -416,16 +408,16 @@ void ARM::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_ARM_JUMP24:
case R_ARM_PC24:
case R_ARM_PLT32:
checkInt<26>(Loc, Val, Type);
checkInt(Loc, Val, 26, Type);
write32le(Loc, (read32le(Loc) & ~0x00ffffff) | ((Val >> 2) & 0x00ffffff));
break;
case R_ARM_THM_JUMP11:
checkInt<12>(Loc, Val, Type);
checkInt(Loc, Val, 12, Type);
write16le(Loc, (read32le(Loc) & 0xf800) | ((Val >> 1) & 0x07ff));
break;
case R_ARM_THM_JUMP19:
// Encoding T3: Val = S:J2:J1:imm6:imm11:0
checkInt<21>(Loc, Val, Type);
checkInt(Loc, Val, 21, Type);
write16le(Loc,
(read16le(Loc) & 0xfbc0) | // opcode cond
((Val >> 10) & 0x0400) | // S
@ -451,7 +443,7 @@ void ARM::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_ARM_THM_JUMP24:
// Encoding B T4, BL T1, BLX T2: Val = S:I1:I2:imm10:imm11:0
// FIXME: Use of I1 and I2 require v6T2ops
checkInt<25>(Loc, Val, Type);
checkInt(Loc, Val, 25, Type);
write16le(Loc,
0xf000 | // opcode
((Val >> 14) & 0x0400) | // S
@ -469,14 +461,14 @@ void ARM::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
break;
case R_ARM_MOVT_ABS:
case R_ARM_MOVT_PREL:
checkInt<32>(Loc, Val, Type);
checkInt(Loc, Val, 32, Type);
write32le(Loc, (read32le(Loc) & ~0x000f0fff) |
(((Val >> 16) & 0xf000) << 4) | ((Val >> 16) & 0xfff));
break;
case R_ARM_THM_MOVT_ABS:
case R_ARM_THM_MOVT_PREL:
// Encoding T1: A = imm4:i:imm3:imm8
checkInt<32>(Loc, Val, Type);
checkInt(Loc, Val, 32, Type);
write16le(Loc,
0xf2c0 | // opcode
((Val >> 17) & 0x0400) | // i

97
contrib/llvm/tools/lld/ELF/Arch/Hexagon.cpp Normal file
View File

@ -0,0 +1,97 @@
//===-- Hexagon.cpp -------------------------------------------------------===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "InputFiles.h"
#include "Symbols.h"
#include "Target.h"
#include "lld/Common/ErrorHandler.h"
#include "llvm/BinaryFormat/ELF.h"
#include "llvm/Object/ELF.h"
#include "llvm/Support/Endian.h"
using namespace llvm;
using namespace llvm::object;
using namespace llvm::support::endian;
using namespace llvm::ELF;
using namespace lld;
using namespace lld::elf;
namespace {
class Hexagon final : public TargetInfo {
public:
uint32_t calcEFlags() const override;
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
void relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const override;
};
} // namespace
// Support V60 only at the moment.
uint32_t Hexagon::calcEFlags() const { return 0x60; }
static uint32_t applyMask(uint32_t Mask, uint32_t Data) {
uint32_t Result = 0;
size_t Off = 0;
for (size_t Bit = 0; Bit != 32; ++Bit) {
uint32_t ValBit = (Data >> Off) & 1;
uint32_t MaskBit = (Mask >> Bit) & 1;
if (MaskBit) {
Result |= (ValBit << Bit);
++Off;
}
}
return Result;
}
RelExpr Hexagon::getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const {
switch (Type) {
case R_HEX_B15_PCREL:
case R_HEX_B15_PCREL_X:
case R_HEX_B22_PCREL:
case R_HEX_B22_PCREL_X:
case R_HEX_B32_PCREL_X:
return R_PC;
default:
return R_ABS;
}
}
static void or32le(uint8_t *P, int32_t V) { write32le(P, read32le(P) | V); }
void Hexagon::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
switch (Type) {
case R_HEX_NONE:
break;
case R_HEX_B15_PCREL:
or32le(Loc, applyMask(0x00df20fe, Val >> 2));
break;
case R_HEX_B15_PCREL_X:
or32le(Loc, applyMask(0x00df20fe, Val & 0x3f));
break;
case R_HEX_B22_PCREL:
or32le(Loc, applyMask(0x1ff3ffe, Val >> 2));
break;
case R_HEX_B22_PCREL_X:
or32le(Loc, applyMask(0x1ff3ffe, Val & 0x3f));
break;
case R_HEX_B32_PCREL_X:
or32le(Loc, applyMask(0x0fff3fff, Val >> 6));
break;
default:
error(getErrorLocation(Loc) + "unrecognized reloc " + toString(Type));
break;
}
}
TargetInfo *elf::getHexagonTargetInfo() {
static Hexagon Target;
return &Target;
}

167
contrib/llvm/tools/lld/ELF/Arch/Mips.cpp
View File

@ -32,7 +32,6 @@ template <class ELFT> class MIPS final : public TargetInfo {
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
int64_t getImplicitAddend(const uint8_t *Buf, RelType Type) const override;
bool isPicRel(RelType Type) const override;
RelType getDynRel(RelType Type) const override;
void writeGotPlt(uint8_t *Buf, const Symbol &S) const override;
void writePltHeader(uint8_t *Buf) const override;
@ -50,6 +49,7 @@ template <class ELFT> MIPS<ELFT>::MIPS() {
DefaultMaxPageSize = 65536;
GotEntrySize = sizeof(typename ELFT::uint);
GotPltEntrySize = sizeof(typename ELFT::uint);
GotBaseSymInGotPlt = false;
PltEntrySize = 16;
PltHeaderSize = 32;
CopyRel = R_MIPS_COPY;
@ -101,8 +101,6 @@ RelExpr MIPS<ELFT>::getRelExpr(RelType Type, const Symbol &S,
case R_MIPS_HIGHEST:
case R_MICROMIPS_HI16:
case R_MICROMIPS_LO16:
case R_MICROMIPS_HIGHER:
case R_MICROMIPS_HIGHEST:
// R_MIPS_HI16/R_MIPS_LO16 relocations against _gp_disp calculate
// offset between start of function and 'gp' value which by default
// equal to the start of .got section. In that case we consider these
@ -124,8 +122,6 @@ RelExpr MIPS<ELFT>::getRelExpr(RelType Type, const Symbol &S,
case R_MIPS_TLS_TPREL_LO16:
case R_MIPS_TLS_TPREL32:
case R_MIPS_TLS_TPREL64:
case R_MICROMIPS_GOT_OFST:
case R_MICROMIPS_SUB:
case R_MICROMIPS_TLS_DTPREL_HI16:
case R_MICROMIPS_TLS_DTPREL_LO16:
case R_MICROMIPS_TLS_TPREL_HI16:
@ -155,7 +151,6 @@ RelExpr MIPS<ELFT>::getRelExpr(RelType Type, const Symbol &S,
case R_MIPS_GOT_DISP:
case R_MIPS_TLS_GOTTPREL:
case R_MICROMIPS_CALL16:
case R_MICROMIPS_GOT_DISP:
case R_MICROMIPS_TLS_GOTTPREL:
return R_MIPS_GOT_OFF;
case R_MIPS_CALL_HI16:
@ -168,7 +163,6 @@ RelExpr MIPS<ELFT>::getRelExpr(RelType Type, const Symbol &S,
case R_MICROMIPS_GOT_LO16:
return R_MIPS_GOT_OFF32;
case R_MIPS_GOT_PAGE:
case R_MICROMIPS_GOT_PAGE:
return R_MIPS_GOT_LOCAL_PAGE;
case R_MIPS_TLS_GD:
case R_MICROMIPS_TLS_GD:
@ -183,12 +177,10 @@ RelExpr MIPS<ELFT>::getRelExpr(RelType Type, const Symbol &S,
}
}
template <class ELFT> bool MIPS<ELFT>::isPicRel(RelType Type) const {
return Type == R_MIPS_32 || Type == R_MIPS_64;
}
template <class ELFT> RelType MIPS<ELFT>::getDynRel(RelType Type) const {
return RelativeRel;
if (Type == R_MIPS_32 || Type == R_MIPS_64)
return RelativeRel;
return R_MIPS_NONE;
}
template <class ELFT>
@ -213,8 +205,8 @@ template <endianness E> static uint32_t readShuffle(const uint8_t *Loc) {
}
template <endianness E>
static void writeRelocation(uint8_t *Loc, uint64_t V, uint8_t BitsSize,
uint8_t Shift) {
static void writeValue(uint8_t *Loc, uint64_t V, uint8_t BitsSize,
uint8_t Shift) {
uint32_t Instr = read32<E>(Loc);
uint32_t Mask = 0xffffffff >> (32 - BitsSize);
uint32_t Data = (Instr & ~Mask) | ((V >> Shift) & Mask);
@ -222,14 +214,14 @@ static void writeRelocation(uint8_t *Loc, uint64_t V, uint8_t BitsSize,
}
template <endianness E>
static void writeMicroRelocation32(uint8_t *Loc, uint64_t V, uint8_t BitsSize,
uint8_t Shift) {
static void writeShuffleValue(uint8_t *Loc, uint64_t V, uint8_t BitsSize,
uint8_t Shift) {
// See comments in readShuffle for purpose of this code.
uint16_t *Words = (uint16_t *)Loc;
if (E == support::little)
std::swap(Words[0], Words[1]);
writeRelocation<E>(Loc, V, BitsSize, Shift);
writeValue<E>(Loc, V, BitsSize, Shift);
if (E == support::little)
std::swap(Words[0], Words[1]);
@ -301,9 +293,9 @@ template <class ELFT> void MIPS<ELFT>::writePltHeader(uint8_t *Buf) const {
write32<E>(Buf + 28, 0x2718fffe); // subu $24, $24, 2
uint64_t GotPlt = InX::GotPlt->getVA();
writeRelocation<E>(Buf, GotPlt + 0x8000, 16, 16);
writeRelocation<E>(Buf + 4, GotPlt, 16, 0);
writeRelocation<E>(Buf + 8, GotPlt, 16, 0);
writeValue<E>(Buf, GotPlt + 0x8000, 16, 16);
writeValue<E>(Buf + 4, GotPlt, 16, 0);
writeValue<E>(Buf + 8, GotPlt, 16, 0);
}
template <class ELFT>
@ -338,9 +330,9 @@ void MIPS<ELFT>::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
write32<E>(Buf + 4, 0x8df90000); // l[wd] $25, %lo(.got.plt entry)($15)
write32<E>(Buf + 8, JrInst); // jr $25 / jr.hb $25
write32<E>(Buf + 12, 0x25f80000); // addiu $24, $15, %lo(.got.plt entry)
writeRelocation<E>(Buf, GotPltEntryAddr + 0x8000, 16, 16);
writeRelocation<E>(Buf + 4, GotPltEntryAddr, 16, 0);
writeRelocation<E>(Buf + 12, GotPltEntryAddr, 16, 0);
writeValue<E>(Buf, GotPltEntryAddr + 0x8000, 16, 16);
writeValue<E>(Buf + 4, GotPltEntryAddr, 16, 0);
writeValue<E>(Buf + 12, GotPltEntryAddr, 16, 0);
}
template <class ELFT>
@ -459,9 +451,6 @@ calculateMipsRelChain(uint8_t *Loc, RelType Type, uint64_t Val) {
return std::make_pair(Type2, Val);
if (Type2 == R_MIPS_SUB && (Type3 == R_MIPS_HI16 || Type3 == R_MIPS_LO16))
return std::make_pair(Type3, -Val);
if (Type2 == R_MICROMIPS_SUB &&
(Type3 == R_MICROMIPS_HI16 || Type3 == R_MICROMIPS_LO16))
return std::make_pair(Type3, -Val);
error(getErrorLocation(Loc) + "unsupported relocations combination " +
Twine(Type));
return std::make_pair(Type & 0xff, Val);
@ -471,6 +460,9 @@ template <class ELFT>
void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
const endianness E = ELFT::TargetEndianness;
if (ELFT::Is64Bits || Config->MipsN32Abi)
std::tie(Type, Val) = calculateMipsRelChain(Loc, Type, Val);
// Thread pointer and DRP offsets from the start of TLS data area.
// https://www.linux-mips.org/wiki/NPTL
if (Type == R_MIPS_TLS_DTPREL_HI16 || Type == R_MIPS_TLS_DTPREL_LO16 ||
@ -485,9 +477,6 @@ void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
Val -= 0x7000;
}
if (ELFT::Is64Bits || Config->MipsN32Abi)
std::tie(Type, Val) = calculateMipsRelChain(Loc, Type, Val);
switch (Type) {
case R_MIPS_32:
case R_MIPS_GPREL32:
@ -501,25 +490,25 @@ void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
write64<E>(Loc, Val);
break;
case R_MIPS_26:
writeRelocation<E>(Loc, Val, 26, 2);
writeValue<E>(Loc, Val, 26, 2);
break;
case R_MIPS_GOT16:
// The R_MIPS_GOT16 relocation's value in "relocatable" linking mode
// is updated addend (not a GOT index). In that case write high 16 bits
// to store a correct addend value.
if (Config->Relocatable) {
writeRelocation<E>(Loc, Val + 0x8000, 16, 16);
writeValue<E>(Loc, Val + 0x8000, 16, 16);
} else {
checkInt<16>(Loc, Val, Type);
writeRelocation<E>(Loc, Val, 16, 0);
checkInt(Loc, Val, 16, Type);
writeValue<E>(Loc, Val, 16, 0);
}
break;
case R_MICROMIPS_GOT16:
if (Config->Relocatable) {
writeMicroRelocation32<E>(Loc, Val + 0x8000, 16, 16);
writeShuffleValue<E>(Loc, Val + 0x8000, 16, 16);
} else {
checkInt<16>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 16, 0);
checkInt(Loc, Val, 16, Type);
writeShuffleValue<E>(Loc, Val, 16, 0);
}
break;
case R_MIPS_CALL16:
@ -529,7 +518,7 @@ void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_MIPS_TLS_GD:
case R_MIPS_TLS_GOTTPREL:
case R_MIPS_TLS_LDM:
checkInt<16>(Loc, Val, Type);
checkInt(Loc, Val, 16, Type);
LLVM_FALLTHROUGH;
case R_MIPS_CALL_LO16:
case R_MIPS_GOT_LO16:
@ -538,28 +527,25 @@ void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_MIPS_PCLO16:
case R_MIPS_TLS_DTPREL_LO16:
case R_MIPS_TLS_TPREL_LO16:
writeRelocation<E>(Loc, Val, 16, 0);
writeValue<E>(Loc, Val, 16, 0);
break;
case R_MICROMIPS_GOT_DISP:
case R_MICROMIPS_GOT_PAGE:
case R_MICROMIPS_GPREL16:
case R_MICROMIPS_TLS_GD:
case R_MICROMIPS_TLS_LDM:
checkInt<16>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 16, 0);
checkInt(Loc, Val, 16, Type);
writeShuffleValue<E>(Loc, Val, 16, 0);
break;
case R_MICROMIPS_CALL16:
case R_MICROMIPS_CALL_LO16:
case R_MICROMIPS_GOT_OFST:
case R_MICROMIPS_LO16:
case R_MICROMIPS_TLS_DTPREL_LO16:
case R_MICROMIPS_TLS_GOTTPREL:
case R_MICROMIPS_TLS_TPREL_LO16:
writeMicroRelocation32<E>(Loc, Val, 16, 0);
writeShuffleValue<E>(Loc, Val, 16, 0);
break;
case R_MICROMIPS_GPREL7_S2:
checkInt<7>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 7, 2);
checkInt(Loc, Val, 7, Type);
writeShuffleValue<E>(Loc, Val, 7, 2);
break;
case R_MIPS_CALL_HI16:
case R_MIPS_GOT_HI16:
@ -567,86 +553,80 @@ void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_MIPS_PCHI16:
case R_MIPS_TLS_DTPREL_HI16:
case R_MIPS_TLS_TPREL_HI16:
writeRelocation<E>(Loc, Val + 0x8000, 16, 16);
writeValue<E>(Loc, Val + 0x8000, 16, 16);
break;
case R_MICROMIPS_CALL_HI16:
case R_MICROMIPS_GOT_HI16:
case R_MICROMIPS_HI16:
case R_MICROMIPS_TLS_DTPREL_HI16:
case R_MICROMIPS_TLS_TPREL_HI16:
writeMicroRelocation32<E>(Loc, Val + 0x8000, 16, 16);
writeShuffleValue<E>(Loc, Val + 0x8000, 16, 16);
break;
case R_MIPS_HIGHER:
writeRelocation<E>(Loc, Val + 0x80008000, 16, 32);
writeValue<E>(Loc, Val + 0x80008000, 16, 32);
break;
case R_MIPS_HIGHEST:
writeRelocation<E>(Loc, Val + 0x800080008000, 16, 48);
break;
case R_MICROMIPS_HIGHER:
writeMicroRelocation32<E>(Loc, Val + 0x80008000, 16, 32);
break;
case R_MICROMIPS_HIGHEST:
writeMicroRelocation32<E>(Loc, Val + 0x800080008000, 16, 48);
writeValue<E>(Loc, Val + 0x800080008000, 16, 48);
break;
case R_MIPS_JALR:
case R_MICROMIPS_JALR:
// Ignore this optimization relocation for now
break;
case R_MIPS_PC16:
checkAlignment<4>(Loc, Val, Type);
checkInt<18>(Loc, Val, Type);
writeRelocation<E>(Loc, Val, 16, 2);
checkAlignment(Loc, Val, 4, Type);
checkInt(Loc, Val, 18, Type);
writeValue<E>(Loc, Val, 16, 2);
break;
case R_MIPS_PC19_S2:
checkAlignment<4>(Loc, Val, Type);
checkInt<21>(Loc, Val, Type);
writeRelocation<E>(Loc, Val, 19, 2);
checkAlignment(Loc, Val, 4, Type);
checkInt(Loc, Val, 21, Type);
writeValue<E>(Loc, Val, 19, 2);
break;
case R_MIPS_PC21_S2:
checkAlignment<4>(Loc, Val, Type);
checkInt<23>(Loc, Val, Type);
writeRelocation<E>(Loc, Val, 21, 2);
checkAlignment(Loc, Val, 4, Type);
checkInt(Loc, Val, 23, Type);
writeValue<E>(Loc, Val, 21, 2);
break;
case R_MIPS_PC26_S2:
checkAlignment<4>(Loc, Val, Type);
checkInt<28>(Loc, Val, Type);
writeRelocation<E>(Loc, Val, 26, 2);
checkAlignment(Loc, Val, 4, Type);
checkInt(Loc, Val, 28, Type);
writeValue<E>(Loc, Val, 26, 2);
break;
case R_MIPS_PC32:
writeRelocation<E>(Loc, Val, 32, 0);
writeValue<E>(Loc, Val, 32, 0);
break;
case R_MICROMIPS_26_S1:
case R_MICROMIPS_PC26_S1:
checkInt<27>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 26, 1);
checkInt(Loc, Val, 27, Type);
writeShuffleValue<E>(Loc, Val, 26, 1);
break;
case R_MICROMIPS_PC7_S1:
checkInt<8>(Loc, Val, Type);
checkInt(Loc, Val, 8, Type);
writeMicroRelocation16<E>(Loc, Val, 7, 1);
break;
case R_MICROMIPS_PC10_S1:
checkInt<11>(Loc, Val, Type);
checkInt(Loc, Val, 11, Type);
writeMicroRelocation16<E>(Loc, Val, 10, 1);
break;
case R_MICROMIPS_PC16_S1:
checkInt<17>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 16, 1);
checkInt(Loc, Val, 17, Type);
writeShuffleValue<E>(Loc, Val, 16, 1);
break;
case R_MICROMIPS_PC18_S3:
checkInt<21>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 18, 3);
checkInt(Loc, Val, 21, Type);
writeShuffleValue<E>(Loc, Val, 18, 3);
break;
case R_MICROMIPS_PC19_S2:
checkInt<21>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 19, 2);
checkInt(Loc, Val, 21, Type);
writeShuffleValue<E>(Loc, Val, 19, 2);
break;
case R_MICROMIPS_PC21_S1:
checkInt<22>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 21, 1);
checkInt(Loc, Val, 22, Type);
writeShuffleValue<E>(Loc, Val, 21, 1);
break;
case R_MICROMIPS_PC23_S2:
checkInt<25>(Loc, Val, Type);
writeMicroRelocation32<E>(Loc, Val, 23, 2);
checkInt(Loc, Val, 25, Type);
writeShuffleValue<E>(Loc, Val, 23, 2);
break;
default:
error(getErrorLocation(Loc) + "unrecognized reloc " + Twine(Type));
@ -655,19 +635,26 @@ void MIPS<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
template <class ELFT> bool MIPS<ELFT>::usesOnlyLowPageBits(RelType Type) const {
return Type == R_MIPS_LO16 || Type == R_MIPS_GOT_OFST ||
Type == R_MICROMIPS_LO16 || Type == R_MICROMIPS_GOT_OFST;
Type == R_MICROMIPS_LO16;
}
// Return true if the symbol is a PIC function.
template <class ELFT> bool elf::isMipsPIC(const Defined *Sym) {
typedef typename ELFT::Ehdr Elf_Ehdr;
if (!Sym->Section || !Sym->isFunc())
if (!Sym->isFunc())
return false;
auto *Sec = cast<InputSectionBase>(Sym->Section);
const Elf_Ehdr *Hdr = Sec->template getFile<ELFT>()->getObj().getHeader();
return (Sym->StOther & STO_MIPS_MIPS16) == STO_MIPS_PIC ||
(Hdr->e_flags & EF_MIPS_PIC);
if (Sym->StOther & STO_MIPS_PIC)
return true;
if (!Sym->Section)
return false;
ObjFile<ELFT> *File =
cast<InputSectionBase>(Sym->Section)->template getFile<ELFT>();
if (!File)
return false;
return File->getObj().getHeader()->e_flags & EF_MIPS_PIC;
}
template <class ELFT> TargetInfo *elf::getMipsTargetInfo() {

37
contrib/llvm/tools/lld/ELF/Arch/MipsArchTree.cpp
View File

@ -65,25 +65,30 @@ static StringRef getNanName(bool IsNan2008) {
static StringRef getFpName(bool IsFp64) { return IsFp64 ? "64" : "32"; }
static void checkFlags(ArrayRef<FileFlags> Files) {
assert(!Files.empty() && "expected non-empty file list");
uint32_t ABI = Files[0].Flags & (EF_MIPS_ABI | EF_MIPS_ABI2);
bool Nan = Files[0].Flags & EF_MIPS_NAN2008;
bool Fp = Files[0].Flags & EF_MIPS_FP64;
for (const FileFlags &F : Files.slice(1)) {
for (const FileFlags &F : Files) {
if (Config->Is64 && F.Flags & EF_MIPS_MICROMIPS)
error(toString(F.File) + ": microMIPS 64-bit is not supported");
uint32_t ABI2 = F.Flags & (EF_MIPS_ABI | EF_MIPS_ABI2);
if (ABI != ABI2)
error("target ABI '" + getAbiName(ABI) + "' is incompatible with '" +
getAbiName(ABI2) + "': " + toString(F.File));
error(toString(F.File) + ": ABI '" + getAbiName(ABI2) +
"' is incompatible with target ABI '" + getAbiName(ABI) + "'");
bool Nan2 = F.Flags & EF_MIPS_NAN2008;
if (Nan != Nan2)
error("target -mnan=" + getNanName(Nan) + " is incompatible with -mnan=" +
getNanName(Nan2) + ": " + toString(F.File));
error(toString(F.File) + ": -mnan=" + getNanName(Nan2) +
" is incompatible with target -mnan=" + getNanName(Nan));
bool Fp2 = F.Flags & EF_MIPS_FP64;
if (Fp != Fp2)
error("target -mfp" + getFpName(Fp) + " is incompatible with -mfp" +
getFpName(Fp2) + ": " + toString(F.File));
error(toString(F.File) + ": -mfp" + getFpName(Fp2) +
" is incompatible with target -mfp" + getFpName(Fp));
}
}
@ -102,11 +107,13 @@ static uint32_t getPicFlags(ArrayRef<FileFlags> Files) {
for (const FileFlags &F : Files.slice(1)) {
bool IsPic2 = F.Flags & (EF_MIPS_PIC | EF_MIPS_CPIC);
if (IsPic && !IsPic2)
warn("linking abicalls code " + toString(Files[0].File) +
" with non-abicalls file: " + toString(F.File));
warn(toString(F.File) +
": linking non-abicalls code with abicalls code " +
toString(Files[0].File));
if (!IsPic && IsPic2)
warn("linking non-abicalls code " + toString(Files[0].File) +
" with abicalls file: " + toString(F.File));
warn(toString(F.File) +
": linking abicalls code with non-abicalls code " +
toString(Files[0].File));
}
// Compute the result PIC/non-PIC flag.
@ -326,7 +333,7 @@ static StringRef getMipsFpAbiName(uint8_t FpAbi) {
case Mips::Val_GNU_MIPS_ABI_FP_SOFT:
return "-msoft-float";
case Mips::Val_GNU_MIPS_ABI_FP_OLD_64:
return "-mips32r2 -mfp64 (old)";
return "-mgp32 -mfp64 (old)";
case Mips::Val_GNU_MIPS_ABI_FP_XX:
return "-mfpxx";
case Mips::Val_GNU_MIPS_ABI_FP_64:
@ -343,9 +350,9 @@ uint8_t elf::getMipsFpAbiFlag(uint8_t OldFlag, uint8_t NewFlag,
if (compareMipsFpAbi(NewFlag, OldFlag) >= 0)
return NewFlag;
if (compareMipsFpAbi(OldFlag, NewFlag) < 0)
error("target floating point ABI '" + getMipsFpAbiName(OldFlag) +
"' is incompatible with '" + getMipsFpAbiName(NewFlag) +
"': " + FileName);
error(FileName + ": floating point ABI '" + getMipsFpAbiName(NewFlag) +
"' is incompatible with target floating point ABI '" +
getMipsFpAbiName(OldFlag) + "'");
return OldFlag;
}

7
contrib/llvm/tools/lld/ELF/Arch/PPC.cpp
View File

@ -21,13 +21,18 @@ using namespace lld::elf;
namespace {
class PPC final : public TargetInfo {
public:
PPC() { GotBaseSymOff = 0x8000; }
PPC();
void relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const override;
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
};
} // namespace
PPC::PPC() {
GotBaseSymOff = 0x8000;
GotBaseSymInGotPlt = false;
}
RelExpr PPC::getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const {
switch (Type) {

423
contrib/llvm/tools/lld/ELF/Arch/PPC64.cpp
View File

@ -14,12 +14,14 @@
#include "llvm/Support/Endian.h"
using namespace llvm;
using namespace llvm::object;
using namespace llvm::support::endian;
using namespace llvm::ELF;
using namespace lld;
using namespace lld::elf;
static uint64_t PPC64TocOffset = 0x8000;
static uint64_t DynamicThreadPointerOffset = 0x8000;
uint64_t elf::getPPC64TocBase() {
// The TOC consists of sections .got, .toc, .tocbss, .plt in that order. The
@ -39,11 +41,21 @@ namespace {
class PPC64 final : public TargetInfo {
public:
PPC64();
uint32_t calcEFlags() const override;
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
void writePltHeader(uint8_t *Buf) const override;
void writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr, uint64_t PltEntryAddr,
int32_t Index, unsigned RelOff) const override;
void relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const override;
void writeGotHeader(uint8_t *Buf) const override;
bool needsThunk(RelExpr Expr, RelType Type, const InputFile *File,
uint64_t BranchAddr, const Symbol &S) const override;
RelExpr adjustRelaxExpr(RelType Type, const uint8_t *Data,
RelExpr Expr) const override;
void relaxTlsGdToIe(uint8_t *Loc, RelType Type, uint64_t Val) const override;
void relaxTlsGdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const override;
void relaxTlsLdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const override;
};
} // namespace
@ -51,21 +63,35 @@ class PPC64 final : public TargetInfo {
// #higher(value), #highera(value), #highest(value), and #highesta(value)
// macros defined in section 4.5.1. Relocation Types of the PPC-elf64abi
// document.
static uint16_t applyPPCLo(uint64_t V) { return V; }
static uint16_t applyPPCHi(uint64_t V) { return V >> 16; }
static uint16_t applyPPCHa(uint64_t V) { return (V + 0x8000) >> 16; }
static uint16_t applyPPCHigher(uint64_t V) { return V >> 32; }
static uint16_t applyPPCHighera(uint64_t V) { return (V + 0x8000) >> 32; }
static uint16_t applyPPCHighest(uint64_t V) { return V >> 48; }
static uint16_t applyPPCHighesta(uint64_t V) { return (V + 0x8000) >> 48; }
static uint16_t lo(uint64_t V) { return V; }
static uint16_t hi(uint64_t V) { return V >> 16; }
static uint16_t ha(uint64_t V) { return (V + 0x8000) >> 16; }
static uint16_t higher(uint64_t V) { return V >> 32; }
static uint16_t highera(uint64_t V) { return (V + 0x8000) >> 32; }
static uint16_t highest(uint64_t V) { return V >> 48; }
static uint16_t highesta(uint64_t V) { return (V + 0x8000) >> 48; }
PPC64::PPC64() {
PltRel = GotRel = R_PPC64_GLOB_DAT;
GotRel = R_PPC64_GLOB_DAT;
PltRel = R_PPC64_JMP_SLOT;
RelativeRel = R_PPC64_RELATIVE;
IRelativeRel = R_PPC64_IRELATIVE;
GotEntrySize = 8;
PltEntrySize = 4;
GotPltEntrySize = 8;
PltEntrySize = 32;
PltHeaderSize = 0;
GotBaseSymInGotPlt = false;
GotBaseSymOff = 0x8000;
GotHeaderEntriesNum = 1;
GotPltHeaderEntriesNum = 2;
PltHeaderSize = 60;
NeedsThunks = true;
TcbSize = 8;
TlsTpOffset = 0x7000;
TlsModuleIndexRel = R_PPC64_DTPMOD64;
TlsOffsetRel = R_PPC64_DTPREL64;
TlsGotRel = R_PPC64_TPREL64;
// We need 64K pages (at least under glibc/Linux, the loader won't
// set different permissions on a finer granularity than that).
@ -80,6 +106,110 @@ PPC64::PPC64() {
// And because the lowest non-zero 256M boundary is 0x10000000, PPC64 linkers
// use 0x10000000 as the starting address.
DefaultImageBase = 0x10000000;
TrapInstr =
(Config->IsLE == sys::IsLittleEndianHost) ? 0x7fe00008 : 0x0800e07f;
}
static uint32_t getEFlags(InputFile *File) {
if (Config->EKind == ELF64BEKind)
return cast<ObjFile<ELF64BE>>(File)->getObj().getHeader()->e_flags;
return cast<ObjFile<ELF64LE>>(File)->getObj().getHeader()->e_flags;
}
// This file implements v2 ABI. This function makes sure that all
// object files have v2 or an unspecified version as an ABI version.
uint32_t PPC64::calcEFlags() const {
for (InputFile *F : ObjectFiles) {
uint32_t Flag = getEFlags(F);
if (Flag == 1)
error(toString(F) + ": ABI version 1 is not supported");
else if (Flag > 2)
error(toString(F) + ": unrecognized e_flags: " + Twine(Flag));
}
return 2;
}
void PPC64::relaxTlsGdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const {
// Reference: 3.7.4.2 of the 64-bit ELF V2 abi supplement.
// The general dynamic code sequence for a global `x` will look like:
// Instruction Relocation Symbol
// addis r3, r2, x@got@tlsgd@ha R_PPC64_GOT_TLSGD16_HA x
// addi r3, r3, x@got@tlsgd@l R_PPC64_GOT_TLSGD16_LO x
// bl __tls_get_addr(x@tlsgd) R_PPC64_TLSGD x
// R_PPC64_REL24 __tls_get_addr
// nop None None
// Relaxing to local exec entails converting:
// addis r3, r2, x@got@tlsgd@ha into nop
// addi r3, r3, x@got@tlsgd@l into addis r3, r13, x@tprel@ha
// bl __tls_get_addr(x@tlsgd) into nop
// nop into addi r3, r3, x@tprel@l
uint32_t EndianOffset = Config->EKind == ELF64BEKind ? 2U : 0U;
switch (Type) {
case R_PPC64_GOT_TLSGD16_HA:
write32(Loc - EndianOffset, 0x60000000); // nop
break;
case R_PPC64_GOT_TLSGD16_LO:
write32(Loc - EndianOffset, 0x3c6d0000); // addis r3, r13
relocateOne(Loc, R_PPC64_TPREL16_HA, Val);
break;
case R_PPC64_TLSGD:
write32(Loc, 0x60000000); // nop
write32(Loc + 4, 0x38630000); // addi r3, r3
relocateOne(Loc + 4 + EndianOffset, R_PPC64_TPREL16_LO, Val);
break;
default:
llvm_unreachable("unsupported relocation for TLS GD to LE relaxation");
}
}
void PPC64::relaxTlsLdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const {
// Reference: 3.7.4.3 of the 64-bit ELF V2 abi supplement.
// The local dynamic code sequence for a global `x` will look like:
// Instruction Relocation Symbol
// addis r3, r2, x@got@tlsld@ha R_PPC64_GOT_TLSLD16_HA x
// addi r3, r3, x@got@tlsld@l R_PPC64_GOT_TLSLD16_LO x
// bl __tls_get_addr(x@tlsgd) R_PPC64_TLSLD x
// R_PPC64_REL24 __tls_get_addr
// nop None None
// Relaxing to local exec entails converting:
// addis r3, r2, x@got@tlsld@ha into nop
// addi r3, r3, x@got@tlsld@l into addis r3, r13, 0
// bl __tls_get_addr(x@tlsgd) into nop
// nop into addi r3, r3, 4096
uint32_t EndianOffset = Config->EKind == ELF64BEKind ? 2U : 0U;
switch (Type) {
case R_PPC64_GOT_TLSLD16_HA:
write32(Loc - EndianOffset, 0x60000000); // nop
break;
case R_PPC64_GOT_TLSLD16_LO:
write32(Loc - EndianOffset, 0x3c6d0000); // addis r3, r13, 0
break;
case R_PPC64_TLSLD:
write32(Loc, 0x60000000); // nop
write32(Loc + 4, 0x38631000); // addi r3, r3, 4096
break;
case R_PPC64_DTPREL16:
case R_PPC64_DTPREL16_HA:
case R_PPC64_DTPREL16_HI:
case R_PPC64_DTPREL16_DS:
case R_PPC64_DTPREL16_LO:
case R_PPC64_DTPREL16_LO_DS:
case R_PPC64_GOT_DTPREL16_HA:
case R_PPC64_GOT_DTPREL16_LO_DS:
case R_PPC64_GOT_DTPREL16_DS:
case R_PPC64_GOT_DTPREL16_HI:
relocateOne(Loc, Type, Val);
break;
default:
llvm_unreachable("unsupported relocation for TLS LD to LE relaxation");
}
}
RelExpr PPC64::getRelExpr(RelType Type, const Symbol &S,
@ -95,48 +225,162 @@ RelExpr PPC64::getRelExpr(RelType Type, const Symbol &S,
case R_PPC64_TOC:
return R_PPC_TOC;
case R_PPC64_REL24:
return R_PPC_PLT_OPD;
return R_PPC_CALL_PLT;
case R_PPC64_REL16_LO:
case R_PPC64_REL16_HA:
case R_PPC64_REL32:
case R_PPC64_REL64:
return R_PC;
case R_PPC64_GOT_TLSGD16:
case R_PPC64_GOT_TLSGD16_HA:
case R_PPC64_GOT_TLSGD16_HI:
case R_PPC64_GOT_TLSGD16_LO:
return R_TLSGD_GOT;
case R_PPC64_GOT_TLSLD16:
case R_PPC64_GOT_TLSLD16_HA:
case R_PPC64_GOT_TLSLD16_HI:
case R_PPC64_GOT_TLSLD16_LO:
return R_TLSLD_GOT;
case R_PPC64_GOT_TPREL16_HA:
case R_PPC64_GOT_TPREL16_LO_DS:
case R_PPC64_GOT_TPREL16_DS:
case R_PPC64_GOT_TPREL16_HI:
return R_GOT_OFF;
case R_PPC64_GOT_DTPREL16_HA:
case R_PPC64_GOT_DTPREL16_LO_DS:
case R_PPC64_GOT_DTPREL16_DS:
case R_PPC64_GOT_DTPREL16_HI:
return R_TLSLD_GOT_OFF;
case R_PPC64_TPREL16:
case R_PPC64_TPREL16_HA:
case R_PPC64_TPREL16_LO:
case R_PPC64_TPREL16_HI:
case R_PPC64_TPREL16_DS:
case R_PPC64_TPREL16_LO_DS:
case R_PPC64_TPREL16_HIGHER:
case R_PPC64_TPREL16_HIGHERA:
case R_PPC64_TPREL16_HIGHEST:
case R_PPC64_TPREL16_HIGHESTA:
return R_TLS;
case R_PPC64_DTPREL16:
case R_PPC64_DTPREL16_DS:
case R_PPC64_DTPREL16_HA:
case R_PPC64_DTPREL16_HI:
case R_PPC64_DTPREL16_HIGHER:
case R_PPC64_DTPREL16_HIGHERA:
case R_PPC64_DTPREL16_HIGHEST:
case R_PPC64_DTPREL16_HIGHESTA:
case R_PPC64_DTPREL16_LO:
case R_PPC64_DTPREL16_LO_DS:
case R_PPC64_DTPREL64:
return R_ABS;
case R_PPC64_TLSGD:
return R_TLSDESC_CALL;
case R_PPC64_TLSLD:
return R_TLSLD_HINT;
case R_PPC64_TLS:
return R_HINT;
default:
return R_ABS;
}
}
void PPC64::writeGotHeader(uint8_t *Buf) const {
write64(Buf, getPPC64TocBase());
}
void PPC64::writePltHeader(uint8_t *Buf) const {
// The generic resolver stub goes first.
write32(Buf + 0, 0x7c0802a6); // mflr r0
write32(Buf + 4, 0x429f0005); // bcl 20,4*cr7+so,8 <_glink+0x8>
write32(Buf + 8, 0x7d6802a6); // mflr r11
write32(Buf + 12, 0x7c0803a6); // mtlr r0
write32(Buf + 16, 0x7d8b6050); // subf r12, r11, r12
write32(Buf + 20, 0x380cffcc); // subi r0,r12,52
write32(Buf + 24, 0x7800f082); // srdi r0,r0,62,2
write32(Buf + 28, 0xe98b002c); // ld r12,44(r11)
write32(Buf + 32, 0x7d6c5a14); // add r11,r12,r11
write32(Buf + 36, 0xe98b0000); // ld r12,0(r11)
write32(Buf + 40, 0xe96b0008); // ld r11,8(r11)
write32(Buf + 44, 0x7d8903a6); // mtctr r12
write32(Buf + 48, 0x4e800420); // bctr
// The 'bcl' instruction will set the link register to the address of the
// following instruction ('mflr r11'). Here we store the offset from that
// instruction to the first entry in the GotPlt section.
int64_t GotPltOffset = InX::GotPlt->getVA() - (InX::Plt->getVA() + 8);
write64(Buf + 52, GotPltOffset);
}
void PPC64::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
uint64_t PltEntryAddr, int32_t Index,
unsigned RelOff) const {
uint64_t Off = GotPltEntryAddr - getPPC64TocBase();
// FIXME: What we should do, in theory, is get the offset of the function
// descriptor in the .opd section, and use that as the offset from %r2 (the
// TOC-base pointer). Instead, we have the GOT-entry offset, and that will
// be a pointer to the function descriptor in the .opd section. Using
// this scheme is simpler, but requires an extra indirection per PLT dispatch.
write32be(Buf, 0xf8410028); // std %r2, 40(%r1)
write32be(Buf + 4, 0x3d620000 | applyPPCHa(Off)); // addis %r11, %r2, X@ha
write32be(Buf + 8, 0xe98b0000 | applyPPCLo(Off)); // ld %r12, X@l(%r11)
write32be(Buf + 12, 0xe96c0000); // ld %r11,0(%r12)
write32be(Buf + 16, 0x7d6903a6); // mtctr %r11
write32be(Buf + 20, 0xe84c0008); // ld %r2,8(%r12)
write32be(Buf + 24, 0xe96c0010); // ld %r11,16(%r12)
write32be(Buf + 28, 0x4e800420); // bctr
int32_t Offset = PltHeaderSize + Index * PltEntrySize;
// bl __glink_PLTresolve
write32(Buf, 0x48000000 | ((-Offset) & 0x03FFFFFc));
}
static std::pair<RelType, uint64_t> toAddr16Rel(RelType Type, uint64_t Val) {
uint64_t V = Val - PPC64TocOffset;
// Relocations relative to the toc-base need to be adjusted by the Toc offset.
uint64_t TocBiasedVal = Val - PPC64TocOffset;
// Relocations relative to dtv[dtpmod] need to be adjusted by the DTP offset.
uint64_t DTPBiasedVal = Val - DynamicThreadPointerOffset;
switch (Type) {
// TOC biased relocation.
case R_PPC64_GOT_TLSGD16:
case R_PPC64_GOT_TLSLD16:
case R_PPC64_TOC16:
return {R_PPC64_ADDR16, V};
return {R_PPC64_ADDR16, TocBiasedVal};
case R_PPC64_TOC16_DS:
return {R_PPC64_ADDR16_DS, V};
case R_PPC64_GOT_TPREL16_DS:
case R_PPC64_GOT_DTPREL16_DS:
return {R_PPC64_ADDR16_DS, TocBiasedVal};
case R_PPC64_GOT_TLSGD16_HA:
case R_PPC64_GOT_TLSLD16_HA:
case R_PPC64_GOT_TPREL16_HA:
case R_PPC64_GOT_DTPREL16_HA:
case R_PPC64_TOC16_HA:
return {R_PPC64_ADDR16_HA, V};
return {R_PPC64_ADDR16_HA, TocBiasedVal};
case R_PPC64_GOT_TLSGD16_HI:
case R_PPC64_GOT_TLSLD16_HI:
case R_PPC64_GOT_TPREL16_HI:
case R_PPC64_GOT_DTPREL16_HI:
case R_PPC64_TOC16_HI:
return {R_PPC64_ADDR16_HI, V};
return {R_PPC64_ADDR16_HI, TocBiasedVal};
case R_PPC64_GOT_TLSGD16_LO:
case R_PPC64_GOT_TLSLD16_LO:
case R_PPC64_TOC16_LO:
return {R_PPC64_ADDR16_LO, V};
return {R_PPC64_ADDR16_LO, TocBiasedVal};
case R_PPC64_TOC16_LO_DS:
return {R_PPC64_ADDR16_LO_DS, V};
case R_PPC64_GOT_TPREL16_LO_DS:
case R_PPC64_GOT_DTPREL16_LO_DS:
return {R_PPC64_ADDR16_LO_DS, TocBiasedVal};
// Dynamic Thread pointer biased relocation types.
case R_PPC64_DTPREL16:
return {R_PPC64_ADDR16, DTPBiasedVal};
case R_PPC64_DTPREL16_DS:
return {R_PPC64_ADDR16_DS, DTPBiasedVal};
case R_PPC64_DTPREL16_HA:
return {R_PPC64_ADDR16_HA, DTPBiasedVal};
case R_PPC64_DTPREL16_HI:
return {R_PPC64_ADDR16_HI, DTPBiasedVal};
case R_PPC64_DTPREL16_HIGHER:
return {R_PPC64_ADDR16_HIGHER, DTPBiasedVal};
case R_PPC64_DTPREL16_HIGHERA:
return {R_PPC64_ADDR16_HIGHERA, DTPBiasedVal};
case R_PPC64_DTPREL16_HIGHEST:
return {R_PPC64_ADDR16_HIGHEST, DTPBiasedVal};
case R_PPC64_DTPREL16_HIGHESTA:
return {R_PPC64_ADDR16_HIGHESTA, DTPBiasedVal};
case R_PPC64_DTPREL16_LO:
return {R_PPC64_ADDR16_LO, DTPBiasedVal};
case R_PPC64_DTPREL16_LO_DS:
return {R_PPC64_ADDR16_LO_DS, DTPBiasedVal};
case R_PPC64_DTPREL64:
return {R_PPC64_ADDR64, DTPBiasedVal};
default:
return {Type, Val};
}
@ -149,68 +393,139 @@ void PPC64::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
switch (Type) {
case R_PPC64_ADDR14: {
checkAlignment<4>(Loc, Val, Type);
checkAlignment(Loc, Val, 4, Type);
// Preserve the AA/LK bits in the branch instruction
uint8_t AALK = Loc[3];
write16be(Loc + 2, (AALK & 3) | (Val & 0xfffc));
write16(Loc + 2, (AALK & 3) | (Val & 0xfffc));
break;
}
case R_PPC64_ADDR16:
checkInt<16>(Loc, Val, Type);
write16be(Loc, Val);
case R_PPC64_TPREL16:
checkInt(Loc, Val, 16, Type);
write16(Loc, Val);
break;
case R_PPC64_ADDR16_DS:
checkInt<16>(Loc, Val, Type);
write16be(Loc, (read16be(Loc) & 3) | (Val & ~3));
case R_PPC64_TPREL16_DS:
checkInt(Loc, Val, 16, Type);
write16(Loc, (read16(Loc) & 3) | (Val & ~3));
break;
case R_PPC64_ADDR16_HA:
case R_PPC64_REL16_HA:
write16be(Loc, applyPPCHa(Val));
case R_PPC64_TPREL16_HA:
write16(Loc, ha(Val));
break;
case R_PPC64_ADDR16_HI:
case R_PPC64_REL16_HI:
write16be(Loc, applyPPCHi(Val));
case R_PPC64_TPREL16_HI:
write16(Loc, hi(Val));
break;
case R_PPC64_ADDR16_HIGHER:
write16be(Loc, applyPPCHigher(Val));
case R_PPC64_TPREL16_HIGHER:
write16(Loc, higher(Val));
break;
case R_PPC64_ADDR16_HIGHERA:
write16be(Loc, applyPPCHighera(Val));
case R_PPC64_TPREL16_HIGHERA:
write16(Loc, highera(Val));
break;
case R_PPC64_ADDR16_HIGHEST:
write16be(Loc, applyPPCHighest(Val));
case R_PPC64_TPREL16_HIGHEST:
write16(Loc, highest(Val));
break;
case R_PPC64_ADDR16_HIGHESTA:
write16be(Loc, applyPPCHighesta(Val));
case R_PPC64_TPREL16_HIGHESTA:
write16(Loc, highesta(Val));
break;
case R_PPC64_ADDR16_LO:
write16be(Loc, applyPPCLo(Val));
case R_PPC64_REL16_LO:
case R_PPC64_TPREL16_LO:
write16(Loc, lo(Val));
break;
case R_PPC64_ADDR16_LO_DS:
case R_PPC64_REL16_LO:
write16be(Loc, (read16be(Loc) & 3) | (applyPPCLo(Val) & ~3));
case R_PPC64_TPREL16_LO_DS:
write16(Loc, (read16(Loc) & 3) | (lo(Val) & ~3));
break;
case R_PPC64_ADDR32:
case R_PPC64_REL32:
checkInt<32>(Loc, Val, Type);
write32be(Loc, Val);
checkInt(Loc, Val, 32, Type);
write32(Loc, Val);
break;
case R_PPC64_ADDR64:
case R_PPC64_REL64:
case R_PPC64_TOC:
write64be(Loc, Val);
write64(Loc, Val);
break;
case R_PPC64_REL24: {
uint32_t Mask = 0x03FFFFFC;
checkInt<24>(Loc, Val, Type);
write32be(Loc, (read32be(Loc) & ~Mask) | (Val & Mask));
checkInt(Loc, Val, 24, Type);
write32(Loc, (read32(Loc) & ~Mask) | (Val & Mask));
break;
}
case R_PPC64_DTPREL64:
write64(Loc, Val - DynamicThreadPointerOffset);
break;
default:
error(getErrorLocation(Loc) + "unrecognized reloc " + Twine(Type));
}
}
bool PPC64::needsThunk(RelExpr Expr, RelType Type, const InputFile *File,
uint64_t BranchAddr, const Symbol &S) const {
// If a function is in the plt it needs to be called through
// a call stub.
return Type == R_PPC64_REL24 && S.isInPlt();
}
RelExpr PPC64::adjustRelaxExpr(RelType Type, const uint8_t *Data,
RelExpr Expr) const {
if (Expr == R_RELAX_TLS_GD_TO_IE)
return R_RELAX_TLS_GD_TO_IE_GOT_OFF;
if (Expr == R_RELAX_TLS_LD_TO_LE)
return R_RELAX_TLS_LD_TO_LE_ABS;
return Expr;
}
// Reference: 3.7.4.1 of the 64-bit ELF V2 abi supplement.
// The general dynamic code sequence for a global `x` uses 4 instructions.
// Instruction Relocation Symbol
// addis r3, r2, x@got@tlsgd@ha R_PPC64_GOT_TLSGD16_HA x
// addi r3, r3, x@got@tlsgd@l R_PPC64_GOT_TLSGD16_LO x
// bl __tls_get_addr(x@tlsgd) R_PPC64_TLSGD x
// R_PPC64_REL24 __tls_get_addr
// nop None None
//
// Relaxing to initial-exec entails:
// 1) Convert the addis/addi pair that builds the address of the tls_index
// struct for 'x' to an addis/ld pair that loads an offset from a got-entry.
// 2) Convert the call to __tls_get_addr to a nop.
// 3) Convert the nop following the call to an add of the loaded offset to the
// thread pointer.
// Since the nop must directly follow the call, the R_PPC64_TLSGD relocation is
// used as the relaxation hint for both steps 2 and 3.
void PPC64::relaxTlsGdToIe(uint8_t *Loc, RelType Type, uint64_t Val) const {
switch (Type) {
case R_PPC64_GOT_TLSGD16_HA:
// This is relaxed from addis rT, r2, sym@got@tlsgd@ha to
// addis rT, r2, sym@got@tprel@ha.
relocateOne(Loc, R_PPC64_GOT_TPREL16_HA, Val);
return;
case R_PPC64_GOT_TLSGD16_LO: {
// Relax from addi r3, rA, sym@got@tlsgd@l to
// ld r3, sym@got@tprel@l(rA)
uint32_t EndianOffset = Config->EKind == ELF64BEKind ? 2U : 0U;
uint32_t InputRegister = (read32(Loc - EndianOffset) & (0x1f << 16));
write32(Loc - EndianOffset, 0xE8600000 | InputRegister);
relocateOne(Loc, R_PPC64_GOT_TPREL16_LO_DS, Val);
return;
}
case R_PPC64_TLSGD:
write32(Loc, 0x60000000); // bl __tls_get_addr(sym@tlsgd) --> nop
write32(Loc + 4, 0x7c636A14); // nop --> add r3, r3, r13
return;
default:
llvm_unreachable("unsupported relocation for TLS GD to IE relaxation");
}
}
TargetInfo *elf::getPPC64TargetInfo() {
static PPC64 Target;
return &Target;

12
contrib/llvm/tools/lld/ELF/Arch/SPARCV9.cpp
View File

@ -77,23 +77,23 @@ void SPARCV9::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_SPARC_32:
case R_SPARC_UA32:
// V-word32
checkUInt<32>(Loc, Val, Type);
checkUInt(Loc, Val, 32, Type);
write32be(Loc, Val);
break;
case R_SPARC_DISP32:
// V-disp32
checkInt<32>(Loc, Val, Type);
checkInt(Loc, Val, 32, Type);
write32be(Loc, Val);
break;
case R_SPARC_WDISP30:
case R_SPARC_WPLT30:
// V-disp30
checkInt<32>(Loc, Val, Type);
checkInt(Loc, Val, 32, Type);
write32be(Loc, (read32be(Loc) & ~0x3fffffff) | ((Val >> 2) & 0x3fffffff));
break;
case R_SPARC_22:
// V-imm22
checkUInt<22>(Loc, Val, Type);
checkUInt(Loc, Val, 22, Type);
write32be(Loc, (read32be(Loc) & ~0x003fffff) | (Val & 0x003fffff));
break;
case R_SPARC_GOT22:
@ -103,7 +103,7 @@ void SPARCV9::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
break;
case R_SPARC_WDISP19:
// V-disp19
checkInt<21>(Loc, Val, Type);
checkInt(Loc, Val, 21, Type);
write32be(Loc, (read32be(Loc) & ~0x0007ffff) | ((Val >> 2) & 0x0007ffff));
break;
case R_SPARC_GOT10:
@ -137,7 +137,7 @@ void SPARCV9::writePlt(uint8_t *Buf, uint64_t GotEntryAddr,
};
memcpy(Buf, PltData, sizeof(PltData));
uint64_t Off = PltHeaderSize + Index * PltEntrySize;
uint64_t Off = getPltEntryOffset(Index);
relocateOne(Buf, R_SPARC_22, Off);
relocateOne(Buf + 4, R_SPARC_WDISP19, -(Off + 4 - PltEntrySize));
}

64
contrib/llvm/tools/lld/ELF/Arch/X86.cpp
View File

@ -46,7 +46,6 @@ class X86 : public TargetInfo {
} // namespace
X86::X86() {
GotBaseSymOff = -1;
CopyRel = R_386_COPY;
GotRel = R_386_GLOB_DAT;
PltRel = R_386_JUMP_SLOT;
@ -74,9 +73,9 @@ RelExpr X86::getRelExpr(RelType Type, const Symbol &S,
case R_386_TLS_LDO_32:
return R_ABS;
case R_386_TLS_GD:
return R_TLSGD;
return R_TLSGD_GOT_FROM_END;
case R_386_TLS_LDM:
return R_TLSLD;
return R_TLSLD_GOT_FROM_END;
case R_386_PLT32:
return R_PLT_PC;
case R_386_PC8:
@ -224,7 +223,7 @@ void X86::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
}
write32le(Buf + 7, RelOff);
write32le(Buf + 12, -Index * PltEntrySize - PltHeaderSize - 16);
write32le(Buf + 12, -getPltEntryOffset(Index) - 16);
}
int64_t X86::getImplicitAddend(const uint8_t *Buf, RelType Type) const {
@ -256,15 +255,15 @@ void X86::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
// R_386_{PC,}{8,16} are not part of the i386 psABI, but they are
// being used for some 16-bit programs such as boot loaders, so
// we want to support them.
checkUInt<8>(Loc, Val, Type);
checkIntUInt(Loc, Val, 8, Type);
*Loc = Val;
break;
case R_386_PC8:
checkInt<8>(Loc, Val, Type);
checkInt(Loc, Val, 8, Type);
*Loc = Val;
break;
case R_386_16:
checkUInt<16>(Loc, Val, Type);
checkIntUInt(Loc, Val, 16, Type);
write16le(Loc, Val);
break;
case R_386_PC16:
@ -278,7 +277,7 @@ void X86::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
// current location subtracted from it.
// We just check that Val fits in 17 bits. This misses some cases, but
// should have no false positives.
checkInt<17>(Loc, Val, Type);
checkInt(Loc, Val, 17, Type);
write16le(Loc, Val);
break;
case R_386_32:
@ -301,7 +300,7 @@ void X86::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_386_TLS_LE_32:
case R_386_TLS_TPOFF:
case R_386_TLS_TPOFF32:
checkInt<32>(Loc, Val, Type);
checkInt(Loc, Val, 32, Type);
write32le(Loc, Val);
break;
default:
@ -444,6 +443,7 @@ void RetpolinePic::writePltHeader(uint8_t *Buf) const {
0x89, 0xc8, // 2b: mov %ecx, %eax
0x59, // 2d: pop %ecx
0xc3, // 2e: ret
0xcc, // 2f: int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
@ -457,21 +457,23 @@ void RetpolinePic::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
uint64_t PltEntryAddr, int32_t Index,
unsigned RelOff) const {
const uint8_t Insn[] = {
0x50, // pushl %eax
0x8b, 0x83, 0, 0, 0, 0, // mov foo@GOT(%ebx), %eax
0xe8, 0, 0, 0, 0, // call plt+0x20
0xe9, 0, 0, 0, 0, // jmp plt+0x12
0x68, 0, 0, 0, 0, // pushl $reloc_offset
0xe9, 0, 0, 0, 0, // jmp plt+0
0x50, // pushl %eax
0x8b, 0x83, 0, 0, 0, 0, // mov foo@GOT(%ebx), %eax
0xe8, 0, 0, 0, 0, // call plt+0x20
0xe9, 0, 0, 0, 0, // jmp plt+0x12
0x68, 0, 0, 0, 0, // pushl $reloc_offset
0xe9, 0, 0, 0, 0, // jmp plt+0
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
uint32_t Ebx = InX::Got->getVA() + InX::Got->getSize();
unsigned Off = getPltEntryOffset(Index);
write32le(Buf + 3, GotPltEntryAddr - Ebx);
write32le(Buf + 8, -Index * PltEntrySize - PltHeaderSize - 12 + 32);
write32le(Buf + 13, -Index * PltEntrySize - PltHeaderSize - 17 + 18);
write32le(Buf + 8, -Off - 12 + 32);
write32le(Buf + 13, -Off - 17 + 18);
write32le(Buf + 18, RelOff);
write32le(Buf + 23, -Index * PltEntrySize - PltHeaderSize - 27);
write32le(Buf + 23, -Off - 27);
}
RetpolineNoPic::RetpolineNoPic() {
@ -484,7 +486,7 @@ void RetpolineNoPic::writeGotPlt(uint8_t *Buf, const Symbol &S) const {
}
void RetpolineNoPic::writePltHeader(uint8_t *Buf) const {
const uint8_t PltData[] = {
const uint8_t Insn[] = {
0xff, 0x35, 0, 0, 0, 0, // 0: pushl GOTPLT+4
0x50, // 6: pushl %eax
0xa1, 0, 0, 0, 0, // 7: mov GOTPLT+8, %eax
@ -500,8 +502,9 @@ void RetpolineNoPic::writePltHeader(uint8_t *Buf) const {
0x89, 0xc8, // 2b: mov %ecx, %eax
0x59, // 2d: pop %ecx
0xc3, // 2e: ret
0xcc, // 2f: int3; padding
};
memcpy(Buf, PltData, sizeof(PltData));
memcpy(Buf, Insn, sizeof(Insn));
uint32_t GotPlt = InX::GotPlt->getVA();
write32le(Buf + 2, GotPlt + 4);
@ -512,20 +515,23 @@ void RetpolineNoPic::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
uint64_t PltEntryAddr, int32_t Index,
unsigned RelOff) const {
const uint8_t Insn[] = {
0x50, // 0: pushl %eax
0xa1, 0, 0, 0, 0, // 1: mov foo_in_GOT, %eax
0xe8, 0, 0, 0, 0, // 6: call plt+0x20
0xe9, 0, 0, 0, 0, // b: jmp plt+0x11
0x68, 0, 0, 0, 0, // 10: pushl $reloc_offset
0xe9, 0, 0, 0, 0, // 15: jmp plt+0
0x50, // 0: pushl %eax
0xa1, 0, 0, 0, 0, // 1: mov foo_in_GOT, %eax
0xe8, 0, 0, 0, 0, // 6: call plt+0x20
0xe9, 0, 0, 0, 0, // b: jmp plt+0x11
0x68, 0, 0, 0, 0, // 10: pushl $reloc_offset
0xe9, 0, 0, 0, 0, // 15: jmp plt+0
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // 1a: int3; padding
0xcc, // 1f: int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
unsigned Off = getPltEntryOffset(Index);
write32le(Buf + 2, GotPltEntryAddr);
write32le(Buf + 7, -Index * PltEntrySize - PltHeaderSize - 11 + 32);
write32le(Buf + 12, -Index * PltEntrySize - PltHeaderSize - 16 + 17);
write32le(Buf + 7, -Off - 11 + 32);
write32le(Buf + 12, -Off - 16 + 17);
write32le(Buf + 17, RelOff);
write32le(Buf + 22, -Index * PltEntrySize - PltHeaderSize - 26);
write32le(Buf + 22, -Off - 26);
}
TargetInfo *elf::getX86TargetInfo() {

110
contrib/llvm/tools/lld/ELF/Arch/X86_64.cpp
View File

@ -28,7 +28,7 @@ template <class ELFT> class X86_64 : public TargetInfo {
X86_64();
RelExpr getRelExpr(RelType Type, const Symbol &S,
const uint8_t *Loc) const override;
bool isPicRel(RelType Type) const override;
RelType getDynRel(RelType Type) const override;
void writeGotPltHeader(uint8_t *Buf) const override;
void writeGotPlt(uint8_t *Buf, const Symbol &S) const override;
void writePltHeader(uint8_t *Buf) const override;
@ -43,6 +43,8 @@ template <class ELFT> class X86_64 : public TargetInfo {
void relaxTlsGdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const override;
void relaxTlsIeToLe(uint8_t *Loc, RelType Type, uint64_t Val) const override;
void relaxTlsLdToLe(uint8_t *Loc, RelType Type, uint64_t Val) const override;
bool adjustPrologueForCrossSplitStack(uint8_t *Loc,
uint8_t *End) const override;
private:
void relaxGotNoPic(uint8_t *Loc, uint64_t Val, uint8_t Op,
@ -51,7 +53,6 @@ template <class ELFT> class X86_64 : public TargetInfo {
} // namespace
template <class ELFT> X86_64<ELFT>::X86_64() {
GotBaseSymOff = -1;
CopyRel = R_X86_64_COPY;
GotRel = R_X86_64_GLOB_DAT;
PltRel = R_X86_64_JUMP_SLOT;
@ -106,6 +107,11 @@ RelExpr X86_64<ELFT>::getRelExpr(RelType Type, const Symbol &S,
case R_X86_64_REX_GOTPCRELX:
case R_X86_64_GOTTPOFF:
return R_GOT_PC;
case R_X86_64_GOTOFF64:
return R_GOTREL_FROM_END;
case R_X86_64_GOTPC32:
case R_X86_64_GOTPC64:
return R_GOTONLY_PC_FROM_END;
case R_X86_64_NONE:
return R_NONE;
default:
@ -124,7 +130,7 @@ template <class ELFT> void X86_64<ELFT>::writeGotPltHeader(uint8_t *Buf) const {
template <class ELFT>
void X86_64<ELFT>::writeGotPlt(uint8_t *Buf, const Symbol &S) const {
// See comments in X86::writeGotPlt.
write32le(Buf, S.getPltVA() + 6);
write64le(Buf, S.getPltVA() + 6);
}
template <class ELFT> void X86_64<ELFT>::writePltHeader(uint8_t *Buf) const {
@ -153,12 +159,14 @@ void X86_64<ELFT>::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
write32le(Buf + 2, GotPltEntryAddr - PltEntryAddr - 6);
write32le(Buf + 7, Index);
write32le(Buf + 12, -Index * PltEntrySize - PltHeaderSize - 16);
write32le(Buf + 12, -getPltEntryOffset(Index) - 16);
}
template <class ELFT> bool X86_64<ELFT>::isPicRel(RelType Type) const {
return Type != R_X86_64_PC32 && Type != R_X86_64_32 &&
Type != R_X86_64_TPOFF32;
template <class ELFT> RelType X86_64<ELFT>::getDynRel(RelType Type) const {
if (Type == R_X86_64_64 || Type == R_X86_64_PC64 || Type == R_X86_64_SIZE32 ||
Type == R_X86_64_SIZE64)
return Type;
return R_X86_64_NONE;
}
template <class ELFT>
@ -285,20 +293,21 @@ template <class ELFT>
void X86_64<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
switch (Type) {
case R_X86_64_8:
checkUInt<8>(Loc, Val, Type);
checkUInt(Loc, Val, 8, Type);
*Loc = Val;
break;
case R_X86_64_16:
checkUInt<16>(Loc, Val, Type);
checkUInt(Loc, Val, 16, Type);
write16le(Loc, Val);
break;
case R_X86_64_32:
checkUInt<32>(Loc, Val, Type);
checkUInt(Loc, Val, 32, Type);
write32le(Loc, Val);
break;
case R_X86_64_32S:
case R_X86_64_TPOFF32:
case R_X86_64_GOT32:
case R_X86_64_GOTPC32:
case R_X86_64_GOTPCREL:
case R_X86_64_GOTPCRELX:
case R_X86_64_REX_GOTPCRELX:
@ -309,7 +318,7 @@ void X86_64<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_X86_64_TLSLD:
case R_X86_64_DTPOFF32:
case R_X86_64_SIZE32:
checkInt<32>(Loc, Val, Type);
checkInt(Loc, Val, 32, Type);
write32le(Loc, Val);
break;
case R_X86_64_64:
@ -318,6 +327,8 @@ void X86_64<ELFT>::relocateOne(uint8_t *Loc, RelType Type, uint64_t Val) const {
case R_X86_64_PC64:
case R_X86_64_SIZE64:
case R_X86_64_GOT64:
case R_X86_64_GOTOFF64:
case R_X86_64_GOTPC64:
write64le(Loc, Val);
break;
default:
@ -460,6 +471,55 @@ void X86_64<ELFT>::relaxGot(uint8_t *Loc, uint64_t Val) const {
write32le(Loc - 1, Val + 1);
}
// This anonymous namespace works around a warning bug in
// old versions of gcc. See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56480
namespace {
// A split-stack prologue starts by checking the amount of stack remaining
// in one of two ways:
// A) Comparing of the stack pointer to a field in the tcb.
// B) Or a load of a stack pointer offset with an lea to r10 or r11.
template <>
bool X86_64<ELF64LE>::adjustPrologueForCrossSplitStack(uint8_t *Loc,
uint8_t *End) const {
// Replace "cmp %fs:0x70,%rsp" and subsequent branch
// with "stc, nopl 0x0(%rax,%rax,1)"
if (Loc + 8 < End && memcmp(Loc, "\x64\x48\x3b\x24\x25", 4) == 0) {
memcpy(Loc, "\xf9\x0f\x1f\x84\x00\x00\x00\x00", 8);
return true;
}
// Adjust "lea -0x200(%rsp),%r10" to lea "-0x4200(%rsp),%r10"
if (Loc + 7 < End && memcmp(Loc, "\x4c\x8d\x94\x24\x00\xfe\xff", 7) == 0) {
memcpy(Loc, "\x4c\x8d\x94\x24\x00\xbe\xff", 7);
return true;
}
// Adjust "lea -0x200(%rsp),%r11" to lea "-0x4200(%rsp),%r11"
if (Loc + 7 < End && memcmp(Loc, "\x4c\x8d\x9c\x24\x00\xfe\xff", 7) == 0) {
memcpy(Loc, "\x4c\x8d\x9c\x24\x00\xbe\xff", 7);
return true;
}
return false;
}
template <>
bool X86_64<ELF32LE>::adjustPrologueForCrossSplitStack(uint8_t *Loc,
uint8_t *End) const {
llvm_unreachable("Target doesn't support split stacks.");
}
} // namespace
// These nonstandard PLT entries are to migtigate Spectre v2 security
// vulnerability. In order to mitigate Spectre v2, we want to avoid indirect
// branch instructions such as `jmp *GOTPLT(%rip)`. So, in the following PLT
// entries, we use a CALL followed by MOV and RET to do the same thing as an
// indirect jump. That instruction sequence is so-called "retpoline".
//
// We have two types of retpoline PLTs as a size optimization. If `-z now`
// is specified, all dynamic symbols are resolved at load-time. Thus, when
// that option is given, we can omit code for symbol lazy resolution.
namespace {
template <class ELFT> class Retpoline : public X86_64<ELFT> {
public:
@ -487,7 +547,7 @@ template <class ELFT> Retpoline<ELFT>::Retpoline() {
template <class ELFT>
void Retpoline<ELFT>::writeGotPlt(uint8_t *Buf, const Symbol &S) const {
write32le(Buf, S.getPltVA() + 17);
write64le(Buf, S.getPltVA() + 17);
}
template <class ELFT> void Retpoline<ELFT>::writePltHeader(uint8_t *Buf) const {
@ -501,6 +561,8 @@ template <class ELFT> void Retpoline<ELFT>::writePltHeader(uint8_t *Buf) const {
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // 19: int3; .align 16
0x4c, 0x89, 0x1c, 0x24, // 20: next: mov %r11, (%rsp)
0xc3, // 24: ret
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // 25: int3; padding
0xcc, 0xcc, 0xcc, 0xcc, // 2c: int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
@ -516,14 +578,15 @@ void Retpoline<ELFT>::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
unsigned RelOff) const {
const uint8_t Insn[] = {
0x4c, 0x8b, 0x1d, 0, 0, 0, 0, // 0: mov foo@GOTPLT(%rip), %r11
0xe8, 0, 0, 0, 0, // 7: callq plt+0x20
0xe9, 0, 0, 0, 0, // c: jmp plt+0x12
0x68, 0, 0, 0, 0, // 11: pushq <relocation index>
0xe9, 0, 0, 0, 0, // 16: jmp plt+0
0xe8, 0, 0, 0, 0, // 7: callq plt+0x20
0xe9, 0, 0, 0, 0, // c: jmp plt+0x12
0x68, 0, 0, 0, 0, // 11: pushq <relocation index>
0xe9, 0, 0, 0, 0, // 16: jmp plt+0
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // 1b: int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
uint64_t Off = TargetInfo::PltHeaderSize + TargetInfo::PltEntrySize * Index;
uint64_t Off = TargetInfo::getPltEntryOffset(Index);
write32le(Buf + 3, GotPltEntryAddr - PltEntryAddr - 7);
write32le(Buf + 8, -Off - 12 + 32);
@ -547,6 +610,9 @@ void RetpolineZNow<ELFT>::writePltHeader(uint8_t *Buf) const {
0xcc, 0xcc, 0xcc, 0xcc, // c: int3; .align 16
0x4c, 0x89, 0x1c, 0x24, // 10: next: mov %r11, (%rsp)
0xc3, // 14: ret
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // 15: int3; padding
0xcc, 0xcc, 0xcc, 0xcc, 0xcc, // 1a: int3; padding
0xcc, // 1f: int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
}
@ -556,17 +622,17 @@ void RetpolineZNow<ELFT>::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr,
uint64_t PltEntryAddr, int32_t Index,
unsigned RelOff) const {
const uint8_t Insn[] = {
0x4c, 0x8b, 0x1d, 0, 0, 0, 0, // mov foo@GOTPLT(%rip), %r11
0xe9, 0, 0, 0, 0, // jmp plt+0
0x4c, 0x8b, 0x1d, 0, 0, 0, 0, // mov foo@GOTPLT(%rip), %r11
0xe9, 0, 0, 0, 0, // jmp plt+0
0xcc, 0xcc, 0xcc, 0xcc, // int3; padding
};
memcpy(Buf, Insn, sizeof(Insn));
write32le(Buf + 3, GotPltEntryAddr - PltEntryAddr - 7);
write32le(Buf + 8,
-Index * TargetInfo::PltEntrySize - TargetInfo::PltHeaderSize - 12);
write32le(Buf + 8, -TargetInfo::getPltEntryOffset(Index) - 12);
}
template <class ELFT> TargetInfo *getTargetInfo() {
template <class ELFT> static TargetInfo *getTargetInfo() {
if (Config->ZRetpolineplt) {
if (Config->ZNow) {
static RetpolineZNow<ELFT> T;

4
contrib/llvm/tools/lld/ELF/CMakeLists.txt
View File

@ -12,6 +12,7 @@ add_lld_library(lldELF
Arch/AMDGPU.cpp
Arch/ARM.cpp
Arch/AVR.cpp
Arch/Hexagon.cpp
Arch/Mips.cpp
Arch/MipsArchTree.cpp
Arch/PPC.cpp
@ -19,6 +20,7 @@ add_lld_library(lldELF
Arch/SPARCV9.cpp
Arch/X86.cpp
Arch/X86_64.cpp
CallGraphSort.cpp
Driver.cpp
DriverUtils.cpp
EhFrame.cpp
@ -35,7 +37,6 @@ add_lld_library(lldELF
Relocations.cpp
ScriptLexer.cpp
ScriptParser.cpp
Strings.cpp
SymbolTable.cpp
Symbols.cpp
SyntheticSections.cpp
@ -46,6 +47,7 @@ add_lld_library(lldELF
LINK_COMPONENTS
${LLVM_TARGETS_TO_BUILD}
BinaryFormat
BitWriter
Core
DebugInfoDWARF
LTO

249
contrib/llvm/tools/lld/ELF/CallGraphSort.cpp Normal file
View File

@ -0,0 +1,249 @@
//===- CallGraphSort.cpp --------------------------------------------------===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
///
/// Implementation of Call-Chain Clustering from: Optimizing Function Placement
/// for Large-Scale Data-Center Applications
/// https://research.fb.com/wp-content/uploads/2017/01/cgo2017-hfsort-final1.pdf
///
/// The goal of this algorithm is to improve runtime performance of the final
/// executable by arranging code sections such that page table and i-cache
/// misses are minimized.
///
/// Definitions:
/// * Cluster
/// * An ordered list of input sections which are layed out as a unit. At the
/// beginning of the algorithm each input section has its own cluster and
/// the weight of the cluster is the sum of the weight of all incomming
/// edges.
/// * Call-Chain Clustering (C³) Heuristic
/// * Defines when and how clusters are combined. Pick the highest weighted
/// input section then add it to its most likely predecessor if it wouldn't
/// penalize it too much.
/// * Density
/// * The weight of the cluster divided by the size of the cluster. This is a
/// proxy for the ammount of execution time spent per byte of the cluster.
///
/// It does so given a call graph profile by the following:
/// * Build a weighted call graph from the call graph profile
/// * Sort input sections by weight
/// * For each input section starting with the highest weight
/// * Find its most likely predecessor cluster
/// * Check if the combined cluster would be too large, or would have too low
/// a density.
/// * If not, then combine the clusters.
/// * Sort non-empty clusters by density
///
//===----------------------------------------------------------------------===//
#include "CallGraphSort.h"
#include "OutputSections.h"
#include "SymbolTable.h"
#include "Symbols.h"
using namespace llvm;
using namespace lld;
using namespace lld::elf;
namespace {
struct Edge {
int From;
uint64_t Weight;
};
struct Cluster {
Cluster(int Sec, size_t S) {
Sections.push_back(Sec);
Size = S;
}
double getDensity() const {
if (Size == 0)
return 0;
return double(Weight) / double(Size);
}
std::vector<int> Sections;
size_t Size = 0;
uint64_t Weight = 0;
uint64_t InitialWeight = 0;
std::vector<Edge> Preds;
};
class CallGraphSort {
public:
CallGraphSort();
DenseMap<const InputSectionBase *, int> run();
private:
std::vector<Cluster> Clusters;
std::vector<const InputSectionBase *> Sections;
void groupClusters();
};
// Maximum ammount the combined cluster density can be worse than the original
// cluster to consider merging.
constexpr int MAX_DENSITY_DEGRADATION = 8;
// Maximum cluster size in bytes.
constexpr uint64_t MAX_CLUSTER_SIZE = 1024 * 1024;
} // end anonymous namespace
// Take the edge list in Config->CallGraphProfile, resolve symbol names to
// Symbols, and generate a graph between InputSections with the provided
// weights.
CallGraphSort::CallGraphSort() {
llvm::MapVector<std::pair<const InputSectionBase *, const InputSectionBase *>,
uint64_t> &Profile = Config->CallGraphProfile;
DenseMap<const InputSectionBase *, int> SecToCluster;
auto GetOrCreateNode = [&](const InputSectionBase *IS) -> int {
auto Res = SecToCluster.insert(std::make_pair(IS, Clusters.size()));
if (Res.second) {
Sections.push_back(IS);
Clusters.emplace_back(Clusters.size(), IS->getSize());
}
return Res.first->second;
};
// Create the graph.
for (const auto &C : Profile) {
const auto *FromSB = cast<InputSectionBase>(C.first.first->Repl);
const auto *ToSB = cast<InputSectionBase>(C.first.second->Repl);
uint64_t Weight = C.second;
// Ignore edges between input sections belonging to different output
// sections. This is done because otherwise we would end up with clusters
// containing input sections that can't actually be placed adjacently in the
// output. This messes with the cluster size and density calculations. We
// would also end up moving input sections in other output sections without
// moving them closer to what calls them.
if (FromSB->getOutputSection() != ToSB->getOutputSection())
continue;
int From = GetOrCreateNode(FromSB);
int To = GetOrCreateNode(ToSB);
Clusters[To].Weight += Weight;
if (From == To)
continue;
// Add an edge
Clusters[To].Preds.push_back({From, Weight});
}
for (Cluster &C : Clusters)
C.InitialWeight = C.Weight;
}
// It's bad to merge clusters which would degrade the density too much.
static bool isNewDensityBad(Cluster &A, Cluster &B) {
double NewDensity = double(A.Weight + B.Weight) / double(A.Size + B.Size);
if (NewDensity < A.getDensity() / MAX_DENSITY_DEGRADATION)
return true;
return false;
}
static void mergeClusters(Cluster &Into, Cluster &From) {
Into.Sections.insert(Into.Sections.end(), From.Sections.begin(),
From.Sections.end());
Into.Size += From.Size;
Into.Weight += From.Weight;
From.Sections.clear();
From.Size = 0;
From.Weight = 0;
}
// Group InputSections into clusters using the Call-Chain Clustering heuristic
// then sort the clusters by density.
void CallGraphSort::groupClusters() {
std::vector<int> SortedSecs(Clusters.size());
std::vector<Cluster *> SecToCluster(Clusters.size());
for (int SI = 0, SE = Clusters.size(); SI != SE; ++SI) {
SortedSecs[SI] = SI;
SecToCluster[SI] = &Clusters[SI];
}
std::stable_sort(SortedSecs.begin(), SortedSecs.end(), [&](int A, int B) {
return Clusters[B].getDensity() < Clusters[A].getDensity();
});
for (int SI : SortedSecs) {
// Clusters[SI] is the same as SecToClusters[SI] here because it has not
// been merged into another cluster yet.
Cluster &C = Clusters[SI];
int BestPred = -1;
uint64_t BestWeight = 0;
for (Edge &E : C.Preds) {
if (BestPred == -1 || E.Weight > BestWeight) {
BestPred = E.From;
BestWeight = E.Weight;
}
}
// don't consider merging if the edge is unlikely.
if (BestWeight * 10 <= C.InitialWeight)
continue;
Cluster *PredC = SecToCluster[BestPred];
if (PredC == &C)
continue;
if (C.Size + PredC->Size > MAX_CLUSTER_SIZE)
continue;
if (isNewDensityBad(*PredC, C))
continue;
// NOTE: Consider using a disjoint-set to track section -> cluster mapping
// if this is ever slow.
for (int SI : C.Sections)
SecToCluster[SI] = PredC;
mergeClusters(*PredC, C);
}
// Remove empty or dead nodes. Invalidates all cluster indices.
llvm::erase_if(Clusters, [](const Cluster &C) {
return C.Size == 0 || C.Sections.empty();
});
// Sort by density.
std::stable_sort(Clusters.begin(), Clusters.end(),
[](const Cluster &A, const Cluster &B) {
return A.getDensity() > B.getDensity();
});
}
DenseMap<const InputSectionBase *, int> CallGraphSort::run() {
groupClusters();
// Generate order.
llvm::DenseMap<const InputSectionBase *, int> OrderMap;
ssize_t CurOrder = 1;
for (const Cluster &C : Clusters)
for (int SecIndex : C.Sections)
OrderMap[Sections[SecIndex]] = CurOrder++;
return OrderMap;
}
// Sort sections by the profile data provided by -callgraph-profile-file
//
// This first builds a call graph based on the profile data then merges sections
// according to the C³ huristic. All clusters are then sorted by a density
// metric to further improve locality.
DenseMap<const InputSectionBase *, int> elf::computeCallGraphProfileOrder() {
return CallGraphSort().run();
}

23
contrib/llvm/tools/lld/ELF/CallGraphSort.h Normal file
View File

@ -0,0 +1,23 @@
//===- CallGraphSort.h ------------------------------------------*- C++ -*-===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#ifndef LLD_ELF_CALL_GRAPH_SORT_H
#define LLD_ELF_CALL_GRAPH_SORT_H
#include "llvm/ADT/DenseMap.h"
namespace lld {
namespace elf {
class InputSectionBase;
llvm::DenseMap<const InputSectionBase *, int> computeCallGraphProfileOrder();
} // namespace elf
} // namespace lld
#endif

53
contrib/llvm/tools/lld/ELF/Config.h
View File

@ -10,6 +10,7 @@
#ifndef LLD_ELF_CONFIG_H
#define LLD_ELF_CONFIG_H
#include "lld/Common/ErrorHandler.h"
#include "llvm/ADT/MapVector.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/StringSet.h"
@ -17,13 +18,13 @@
#include "llvm/Support/CachePruning.h"
#include "llvm/Support/CodeGen.h"
#include "llvm/Support/Endian.h"
#include <vector>
namespace lld {
namespace elf {
class InputFile;
class InputSectionBase;
enum ELFKind {
ELFNoneKind,
@ -39,6 +40,9 @@ enum class BuildIdKind { None, Fast, Md5, Sha1, Hexstring, Uuid };
// For --discard-{all,locals,none}.
enum class DiscardPolicy { Default, All, Locals, None };
// For --icf={none,safe,all}.
enum class ICFLevel { None, Safe, All };
// For --strip-{all,debug}.
enum class StripPolicy { None, All, Debug };
@ -79,21 +83,27 @@ struct Configuration {
llvm::StringMap<uint64_t> SectionStartMap;
llvm::StringRef Chroot;
llvm::StringRef DynamicLinker;
llvm::StringRef DwoDir;
llvm::StringRef Entry;
llvm::StringRef Emulation;
llvm::StringRef Fini;
llvm::StringRef Init;
llvm::StringRef LTOAAPipeline;
llvm::StringRef LTONewPmPasses;
llvm::StringRef LTOObjPath;
llvm::StringRef LTOSampleProfile;
llvm::StringRef MapFile;
llvm::StringRef OutputFile;
llvm::StringRef OptRemarksFilename;
llvm::StringRef ProgName;
llvm::StringRef SoName;
llvm::StringRef Sysroot;
llvm::StringRef ThinLTOCacheDir;
llvm::StringRef ThinLTOIndexOnlyArg;
std::pair<llvm::StringRef, llvm::StringRef> ThinLTOObjectSuffixReplace;
std::pair<llvm::StringRef, llvm::StringRef> ThinLTOPrefixReplace;
std::string Rpath;
std::vector<VersionDefinition> VersionDefinitions;
std::vector<llvm::StringRef> Argv;
std::vector<llvm::StringRef> AuxiliaryList;
std::vector<llvm::StringRef> FilterList;
std::vector<llvm::StringRef> SearchPaths;
@ -103,15 +113,20 @@ struct Configuration {
std::vector<SymbolVersion> VersionScriptGlobals;
std::vector<SymbolVersion> VersionScriptLocals;
std::vector<uint8_t> BuildIdVector;
llvm::MapVector<std::pair<const InputSectionBase *, const InputSectionBase *>,
uint64_t>
CallGraphProfile;
bool AllowMultipleDefinition;
bool AndroidPackDynRelocs = false;
bool AndroidPackDynRelocs;
bool ARMHasBlx = false;
bool ARMHasMovtMovw = false;
bool ARMJ1J2BranchEncoding = false;
bool AsNeeded = false;
bool Bsymbolic;
bool BsymbolicFunctions;
bool CheckSections;
bool CompressDebugSections;
bool Cref;
bool DefineCommon;
bool Demangle = true;
bool DisableVerify;
@ -123,14 +138,15 @@ struct Configuration {
bool GcSections;
bool GdbIndex;
bool GnuHash = false;
bool GnuUnique;
bool HasDynamicList = false;
bool HasDynSymTab;
bool ICF;
bool ICFData;
bool IgnoreDataAddressEquality;
bool IgnoreFunctionAddressEquality;
bool LTODebugPassManager;
bool LTONewPassManager;
bool MergeArmExidx;
bool MipsN32Abi = false;
bool NoGnuUnique;
bool NoUndefinedVersion;
bool NoinhibitExec;
bool Nostdlib;
bool OFormatBinary;
@ -138,7 +154,9 @@ struct Configuration {
bool OptRemarksWithHotness;
bool Pie;
bool PrintGcSections;
bool PrintIcfSections;
bool Relocatable;
bool RelrPackDynRelocs;
bool SaveTemps;
bool SingleRoRx;
bool Shared;
@ -146,13 +164,21 @@ struct Configuration {
bool SysvHash = false;
bool Target1Rel;
bool Trace;
bool Verbose;
bool ThinLTOEmitImportsFiles;
bool ThinLTOIndexOnly;
bool UndefinedVersion;
bool UseAndroidRelrTags = false;
bool WarnBackrefs;
bool WarnCommon;
bool WarnMissingEntry;
bool WarnSymbolOrdering;
bool WriteAddends;
bool ZCombreloc;
bool ZCopyreloc;
bool ZExecstack;
bool ZHazardplt;
bool ZNocopyreloc;
bool ZInitfirst;
bool ZKeepTextSectionPrefix;
bool ZNodelete;
bool ZNodlopen;
bool ZNow;
@ -161,9 +187,9 @@ struct Configuration {
bool ZRodynamic;
bool ZText;
bool ZRetpolineplt;
bool ExitEarly;
bool ZWxneeded;
DiscardPolicy Discard;
ICFLevel ICF;
OrphanHandlingPolicy OrphanHandling;
SortSectionPolicy SortSection;
StripPolicy Strip;
@ -175,6 +201,7 @@ struct Configuration {
uint16_t EMachine = llvm::ELF::EM_NONE;
llvm::Optional<uint64_t> ImageBase;
uint64_t MaxPageSize;
uint64_t MipsGotSize;
uint64_t ZStackSize;
unsigned LTOPartitions;
unsigned LTOO;
@ -240,6 +267,12 @@ struct Configuration {
// The only instance of Configuration struct.
extern Configuration *Config;
static inline void errorOrWarn(const Twine &Msg) {
if (!Config->NoinhibitExec)
error(Msg);
else
warn(Msg);
}
} // namespace elf
} // namespace lld

582
contrib/llvm/tools/lld/ELF/Driver.cpp
View File

@ -30,9 +30,9 @@
#include "InputFiles.h"
#include "InputSection.h"
#include "LinkerScript.h"
#include "MarkLive.h"
#include "OutputSections.h"
#include "ScriptParser.h"
#include "Strings.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "SyntheticSections.h"
@ -42,12 +42,16 @@
#include "lld/Common/Driver.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Strings.h"
#include "lld/Common/TargetOptionsCommandFlags.h"
#include "lld/Common/Threads.h"
#include "lld/Common/Version.h"
#include "llvm/ADT/SetVector.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringSwitch.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Compression.h"
#include "llvm/Support/LEB128.h"
#include "llvm/Support/Path.h"
#include "llvm/Support/TarWriter.h"
#include "llvm/Support/TargetSelect.h"
@ -66,16 +70,18 @@ using namespace lld::elf;
Configuration *elf::Config;
LinkerDriver *elf::Driver;
static void setConfigs();
static void setConfigs(opt::InputArgList &Args);
bool elf::link(ArrayRef<const char *> Args, bool CanExitEarly,
raw_ostream &Error) {
errorHandler().LogName = Args[0];
errorHandler().LogName = sys::path::filename(Args[0]);
errorHandler().ErrorLimitExceededMsg =
"too many errors emitted, stopping now (use "
"-error-limit=0 to see all errors)";
errorHandler().ErrorOS = &Error;
errorHandler().ExitEarly = CanExitEarly;
errorHandler().ColorDiagnostics = Error.has_colors();
InputSections.clear();
OutputSections.clear();
Tar = nullptr;
@ -88,14 +94,14 @@ bool elf::link(ArrayRef<const char *> Args, bool CanExitEarly,
Driver = make<LinkerDriver>();
Script = make<LinkerScript>();
Symtab = make<SymbolTable>();
Config->Argv = {Args.begin(), Args.end()};
Config->ProgName = Args[0];
Driver->main(Args, CanExitEarly);
Driver->main(Args);
// Exit immediately if we don't need to return to the caller.
// This saves time because the overhead of calling destructors
// for all globally-allocated objects is not negligible.
if (Config->ExitEarly)
if (CanExitEarly)
exitLld(errorCount() ? 1 : 0);
freeArena();
@ -113,7 +119,8 @@ static std::tuple<ELFKind, uint16_t, uint8_t> parseEmulation(StringRef Emul) {
std::pair<ELFKind, uint16_t> Ret =
StringSwitch<std::pair<ELFKind, uint16_t>>(S)
.Cases("aarch64elf", "aarch64linux", {ELF64LEKind, EM_AARCH64})
.Cases("aarch64elf", "aarch64linux", "aarch64_elf64_le_vec",
{ELF64LEKind, EM_AARCH64})
.Cases("armelf", "armelf_linux_eabi", {ELF32LEKind, EM_ARM})
.Case("elf32_x86_64", {ELF32LEKind, EM_X86_64})
.Cases("elf32btsmip", "elf32btsmipn32", {ELF32BEKind, EM_MIPS})
@ -122,6 +129,7 @@ static std::tuple<ELFKind, uint16_t, uint8_t> parseEmulation(StringRef Emul) {
.Case("elf64btsmip", {ELF64BEKind, EM_MIPS})
.Case("elf64ltsmip", {ELF64LEKind, EM_MIPS})
.Case("elf64ppc", {ELF64BEKind, EM_PPC64})
.Case("elf64lppc", {ELF64LEKind, EM_PPC64})
.Cases("elf_amd64", "elf_x86_64", {ELF64LEKind, EM_X86_64})
.Case("elf_i386", {ELF32LEKind, EM_386})
.Case("elf_iamcu", {ELF32LEKind, EM_IAMCU})
@ -228,11 +236,15 @@ void LinkerDriver::addFile(StringRef Path, bool WithLOption) {
Files.push_back(
createSharedFile(MBRef, WithLOption ? path::filename(Path) : Path));
return;
default:
case file_magic::bitcode:
case file_magic::elf_relocatable:
if (InLib)
Files.push_back(make<LazyObjFile>(MBRef, "", 0));
else
Files.push_back(createObjectFile(MBRef));
break;
default:
error(Path + ": unknown file type");
}
}
@ -248,18 +260,11 @@ void LinkerDriver::addLibrary(StringRef Name) {
// LTO calls LLVM functions to compile bitcode files to native code.
// Technically this can be delayed until we read bitcode files, but
// we don't bother to do lazily because the initialization is fast.
static void initLLVM(opt::InputArgList &Args) {
static void initLLVM() {
InitializeAllTargets();
InitializeAllTargetMCs();
InitializeAllAsmPrinters();
InitializeAllAsmParsers();
// Parse and evaluate -mllvm options.
std::vector<const char *> V;
V.push_back("lld (LLVM option parsing)");
for (auto *Arg : Args.filtered(OPT_mllvm))
V.push_back(Arg->getValue());
cl::ParseCommandLineOptions(V.size(), V.data());
}
// Some command line options or some combinations of them are not allowed.
@ -290,7 +295,9 @@ static void checkOptions(opt::InputArgList &Args) {
error("-r and -shared may not be used together");
if (Config->GcSections)
error("-r and --gc-sections may not be used together");
if (Config->ICF)
if (Config->GdbIndex)
error("-r and --gdb-index may not be used together");
if (Config->ICF != ICFLevel::None)
error("-r and --icf may not be used together");
if (Config->Pie)
error("-r and -pie may not be used together");
@ -310,7 +317,37 @@ static bool hasZOption(opt::InputArgList &Args, StringRef Key) {
return false;
}
void LinkerDriver::main(ArrayRef<const char *> ArgsArr, bool CanExitEarly) {
static bool getZFlag(opt::InputArgList &Args, StringRef K1, StringRef K2,
bool Default) {
for (auto *Arg : Args.filtered_reverse(OPT_z)) {
if (K1 == Arg->getValue())
return true;
if (K2 == Arg->getValue())
return false;
}
return Default;
}
static bool isKnown(StringRef S) {
return S == "combreloc" || S == "copyreloc" || S == "defs" ||
S == "execstack" || S == "hazardplt" || S == "initfirst" ||
S == "keep-text-section-prefix" || S == "lazy" || S == "muldefs" ||
S == "nocombreloc" || S == "nocopyreloc" || S == "nodelete" ||
S == "nodlopen" || S == "noexecstack" ||
S == "nokeep-text-section-prefix" || S == "norelro" || S == "notext" ||
S == "now" || S == "origin" || S == "relro" || S == "retpolineplt" ||
S == "rodynamic" || S == "text" || S == "wxneeded" ||
S.startswith("max-page-size=") || S.startswith("stack-size=");
}
// Report an error for an unknown -z option.
static void checkZOptions(opt::InputArgList &Args) {
for (auto *Arg : Args.filtered(OPT_z))
if (!isKnown(Arg->getValue()))
error("unknown -z value: " + StringRef(Arg->getValue()));
}
void LinkerDriver::main(ArrayRef<const char *> ArgsArr) {
ELFOptTable Parser;
opt::InputArgList Args = Parser.parse(ArgsArr.slice(1));
@ -319,7 +356,7 @@ void LinkerDriver::main(ArrayRef<const char *> ArgsArr, bool CanExitEarly) {
// Handle -help
if (Args.hasArg(OPT_help)) {
printHelp(ArgsArr[0]);
printHelp();
return;
}
@ -348,9 +385,6 @@ void LinkerDriver::main(ArrayRef<const char *> ArgsArr, bool CanExitEarly) {
if (Args.hasArg(OPT_version))
return;
Config->ExitEarly = CanExitEarly && !Args.hasArg(OPT_full_shutdown);
errorHandler().ExitEarly = Config->ExitEarly;
if (const char *Path = getReproduceOption(Args)) {
// Note that --reproduce is a debug option so you can ignore it
// if you are trying to understand the whole picture of the code.
@ -368,10 +402,14 @@ void LinkerDriver::main(ArrayRef<const char *> ArgsArr, bool CanExitEarly) {
}
readConfigs(Args);
initLLVM(Args);
checkZOptions(Args);
initLLVM();
createFiles(Args);
if (errorCount())
return;
inferMachineType();
setConfigs();
setConfigs(Args);
checkOptions(Args);
if (errorCount())
return;
@ -482,6 +520,15 @@ static StringRef getDynamicLinker(opt::InputArgList &Args) {
return Arg->getValue();
}
static ICFLevel getICF(opt::InputArgList &Args) {
auto *Arg = Args.getLastArg(OPT_icf_none, OPT_icf_safe, OPT_icf_all);
if (!Arg || Arg->getOption().getID() == OPT_icf_none)
return ICFLevel::None;
if (Arg->getOption().getID() == OPT_icf_safe)
return ICFLevel::Safe;
return ICFLevel::All;
}
static StripPolicy getStrip(opt::InputArgList &Args) {
if (Args.hasArg(OPT_relocatable))
return StripPolicy::None;
@ -556,6 +603,8 @@ getBuildId(opt::InputArgList &Args) {
return {BuildIdKind::Fast, {}};
StringRef S = Arg->getValue();
if (S == "fast")
return {BuildIdKind::Fast, {}};
if (S == "md5")
return {BuildIdKind::Md5, {}};
if (S == "sha1" || S == "tree")
@ -570,6 +619,57 @@ getBuildId(opt::InputArgList &Args) {
return {BuildIdKind::None, {}};
}
static std::pair<bool, bool> getPackDynRelocs(opt::InputArgList &Args) {
StringRef S = Args.getLastArgValue(OPT_pack_dyn_relocs, "none");
if (S == "android")
return {true, false};
if (S == "relr")
return {false, true};
if (S == "android+relr")
return {true, true};
if (S != "none")
error("unknown -pack-dyn-relocs format: " + S);
return {false, false};
}
static void readCallGraph(MemoryBufferRef MB) {
// Build a map from symbol name to section
DenseMap<StringRef, const Symbol *> SymbolNameToSymbol;
for (InputFile *File : ObjectFiles)
for (Symbol *Sym : File->getSymbols())
SymbolNameToSymbol[Sym->getName()] = Sym;
for (StringRef L : args::getLines(MB)) {
SmallVector<StringRef, 3> Fields;
L.split(Fields, ' ');
uint64_t Count;
if (Fields.size() != 3 || !to_integer(Fields[2], Count))
fatal(MB.getBufferIdentifier() + ": parse error");
const Symbol *FromSym = SymbolNameToSymbol.lookup(Fields[0]);
const Symbol *ToSym = SymbolNameToSymbol.lookup(Fields[1]);
if (Config->WarnSymbolOrdering) {
if (!FromSym)
warn(MB.getBufferIdentifier() + ": no such symbol: " + Fields[0]);
if (!ToSym)
warn(MB.getBufferIdentifier() + ": no such symbol: " + Fields[1]);
}
if (!FromSym || !ToSym || Count == 0)
continue;
warnUnorderableSymbol(FromSym);
warnUnorderableSymbol(ToSym);
const Defined *FromSymD = dyn_cast<Defined>(FromSym);
const Defined *ToSymD = dyn_cast<Defined>(ToSym);
if (!FromSymD || !ToSymD)
continue;
const auto *FromSB = dyn_cast_or_null<InputSectionBase>(FromSymD->Section);
const auto *ToSB = dyn_cast_or_null<InputSectionBase>(ToSymD->Section);
if (!FromSB || !ToSB)
continue;
Config->CallGraphProfile[std::make_pair(FromSB, ToSB)] += Count;
}
}
static bool getCompressDebugSections(opt::InputArgList &Args) {
StringRef S = Args.getLastArgValue(OPT_compress_debug_sections, "none");
if (S == "none")
@ -581,54 +681,98 @@ static bool getCompressDebugSections(opt::InputArgList &Args) {
return true;
}
static int parseInt(StringRef S, opt::Arg *Arg) {
int V = 0;
if (!to_integer(S, V, 10))
error(Arg->getSpelling() + ": number expected, but got '" + S + "'");
return V;
static std::pair<StringRef, StringRef> getOldNewOptions(opt::InputArgList &Args,
unsigned Id) {
auto *Arg = Args.getLastArg(Id);
if (!Arg)
return {"", ""};
StringRef S = Arg->getValue();
std::pair<StringRef, StringRef> Ret = S.split(';');
if (Ret.second.empty())
error(Arg->getSpelling() + " expects 'old;new' format, but got " + S);
return Ret;
}
// Parse the symbol ordering file and warn for any duplicate entries.
static std::vector<StringRef> getSymbolOrderingFile(MemoryBufferRef MB) {
SetVector<StringRef> Names;
for (StringRef S : args::getLines(MB))
if (!Names.insert(S) && Config->WarnSymbolOrdering)
warn(MB.getBufferIdentifier() + ": duplicate ordered symbol: " + S);
return Names.takeVector();
}
static void parseClangOption(StringRef Opt, const Twine &Msg) {
std::string Err;
raw_string_ostream OS(Err);
const char *Argv[] = {Config->ProgName.data(), Opt.data()};
if (cl::ParseCommandLineOptions(2, Argv, "", &OS))
return;
OS.flush();
error(Msg + ": " + StringRef(Err).trim());
}
// Initializes Config members by the command line options.
void LinkerDriver::readConfigs(opt::InputArgList &Args) {
errorHandler().Verbose = Args.hasArg(OPT_verbose);
errorHandler().FatalWarnings =
Args.hasFlag(OPT_fatal_warnings, OPT_no_fatal_warnings, false);
ThreadsEnabled = Args.hasFlag(OPT_threads, OPT_no_threads, true);
Config->AllowMultipleDefinition =
Args.hasArg(OPT_allow_multiple_definition) || hasZOption(Args, "muldefs");
Args.hasFlag(OPT_allow_multiple_definition,
OPT_no_allow_multiple_definition, false) ||
hasZOption(Args, "muldefs");
Config->AuxiliaryList = args::getStrings(Args, OPT_auxiliary);
Config->Bsymbolic = Args.hasArg(OPT_Bsymbolic);
Config->BsymbolicFunctions = Args.hasArg(OPT_Bsymbolic_functions);
Config->CheckSections =
Args.hasFlag(OPT_check_sections, OPT_no_check_sections, true);
Config->Chroot = Args.getLastArgValue(OPT_chroot);
Config->CompressDebugSections = getCompressDebugSections(Args);
Config->Cref = Args.hasFlag(OPT_cref, OPT_no_cref, false);
Config->DefineCommon = Args.hasFlag(OPT_define_common, OPT_no_define_common,
!Args.hasArg(OPT_relocatable));
Config->Demangle = Args.hasFlag(OPT_demangle, OPT_no_demangle, true);
Config->DisableVerify = Args.hasArg(OPT_disable_verify);
Config->Discard = getDiscard(Args);
Config->DwoDir = Args.getLastArgValue(OPT_plugin_opt_dwo_dir_eq);
Config->DynamicLinker = getDynamicLinker(Args);
Config->EhFrameHdr =
Args.hasFlag(OPT_eh_frame_hdr, OPT_no_eh_frame_hdr, false);
Config->EmitRelocs = Args.hasArg(OPT_emit_relocs);
Config->EnableNewDtags = !Args.hasArg(OPT_disable_new_dtags);
Config->EnableNewDtags =
Args.hasFlag(OPT_enable_new_dtags, OPT_disable_new_dtags, true);
Config->Entry = Args.getLastArgValue(OPT_entry);
Config->ExportDynamic =
Args.hasFlag(OPT_export_dynamic, OPT_no_export_dynamic, false);
errorHandler().FatalWarnings =
Args.hasFlag(OPT_fatal_warnings, OPT_no_fatal_warnings, false);
Config->FilterList = args::getStrings(Args, OPT_filter);
Config->Fini = Args.getLastArgValue(OPT_fini, "_fini");
Config->FixCortexA53Errata843419 = Args.hasArg(OPT_fix_cortex_a53_843419);
Config->GcSections = Args.hasFlag(OPT_gc_sections, OPT_no_gc_sections, false);
Config->GnuUnique = Args.hasFlag(OPT_gnu_unique, OPT_no_gnu_unique, true);
Config->GdbIndex = Args.hasFlag(OPT_gdb_index, OPT_no_gdb_index, false);
Config->ICF = Args.hasFlag(OPT_icf_all, OPT_icf_none, false);
Config->ICFData = Args.hasArg(OPT_icf_data);
Config->ICF = getICF(Args);
Config->IgnoreDataAddressEquality =
Args.hasArg(OPT_ignore_data_address_equality);
Config->IgnoreFunctionAddressEquality =
Args.hasArg(OPT_ignore_function_address_equality);
Config->Init = Args.getLastArgValue(OPT_init, "_init");
Config->LTOAAPipeline = Args.getLastArgValue(OPT_lto_aa_pipeline);
Config->LTODebugPassManager = Args.hasArg(OPT_lto_debug_pass_manager);
Config->LTONewPassManager = Args.hasArg(OPT_lto_new_pass_manager);
Config->LTONewPmPasses = Args.getLastArgValue(OPT_lto_newpm_passes);
Config->LTOO = args::getInteger(Args, OPT_lto_O, 2);
Config->LTOObjPath = Args.getLastArgValue(OPT_plugin_opt_obj_path_eq);
Config->LTOPartitions = args::getInteger(Args, OPT_lto_partitions, 1);
Config->LTOSampleProfile = Args.getLastArgValue(OPT_lto_sample_profile);
Config->MapFile = Args.getLastArgValue(OPT_Map);
Config->NoGnuUnique = Args.hasArg(OPT_no_gnu_unique);
Config->MipsGotSize = args::getInteger(Args, OPT_mips_got_size, 0xfff0);
Config->MergeArmExidx =
Args.hasFlag(OPT_merge_exidx_entries, OPT_no_merge_exidx_entries, true);
Config->NoUndefinedVersion = Args.hasArg(OPT_no_undefined_version);
Config->NoinhibitExec = Args.hasArg(OPT_noinhibit_exec);
Config->Nostdlib = Args.hasArg(OPT_nostdlib);
Config->OFormatBinary = isOutputFormatBinary(Args);
@ -638,7 +782,9 @@ void LinkerDriver::readConfigs(opt::InputArgList &Args) {
Config->Optimize = args::getInteger(Args, OPT_O, 1);
Config->OrphanHandling = getOrphanHandling(Args);
Config->OutputFile = Args.getLastArgValue(OPT_o);
Config->Pie = Args.hasFlag(OPT_pie, OPT_nopie, false);
Config->Pie = Args.hasFlag(OPT_pie, OPT_no_pie, false);
Config->PrintIcfSections =
Args.hasFlag(OPT_print_icf_sections, OPT_no_print_icf_sections, false);
Config->PrintGcSections =
Args.hasFlag(OPT_print_gc_sections, OPT_no_print_gc_sections, false);
Config->Rpath = getRpath(Args);
@ -658,48 +804,58 @@ void LinkerDriver::readConfigs(opt::InputArgList &Args) {
Config->ThinLTOCachePolicy = CHECK(
parseCachePruningPolicy(Args.getLastArgValue(OPT_thinlto_cache_policy)),
"--thinlto-cache-policy: invalid cache policy");
Config->ThinLTOEmitImportsFiles =
Args.hasArg(OPT_plugin_opt_thinlto_emit_imports_files);
Config->ThinLTOIndexOnly = Args.hasArg(OPT_plugin_opt_thinlto_index_only) ||
Args.hasArg(OPT_plugin_opt_thinlto_index_only_eq);
Config->ThinLTOIndexOnlyArg =
Args.getLastArgValue(OPT_plugin_opt_thinlto_index_only_eq);
Config->ThinLTOJobs = args::getInteger(Args, OPT_thinlto_jobs, -1u);
ThreadsEnabled = Args.hasFlag(OPT_threads, OPT_no_threads, true);
Config->ThinLTOObjectSuffixReplace =
getOldNewOptions(Args, OPT_plugin_opt_thinlto_object_suffix_replace_eq);
Config->ThinLTOPrefixReplace =
getOldNewOptions(Args, OPT_plugin_opt_thinlto_prefix_replace_eq);
Config->Trace = Args.hasArg(OPT_trace);
Config->Undefined = args::getStrings(Args, OPT_undefined);
Config->UndefinedVersion =
Args.hasFlag(OPT_undefined_version, OPT_no_undefined_version, true);
Config->UseAndroidRelrTags = Args.hasFlag(
OPT_use_android_relr_tags, OPT_no_use_android_relr_tags, false);
Config->UnresolvedSymbols = getUnresolvedSymbolPolicy(Args);
Config->Verbose = Args.hasArg(OPT_verbose);
errorHandler().Verbose = Config->Verbose;
Config->WarnCommon = Args.hasArg(OPT_warn_common);
Config->ZCombreloc = !hasZOption(Args, "nocombreloc");
Config->ZExecstack = hasZOption(Args, "execstack");
Config->WarnBackrefs =
Args.hasFlag(OPT_warn_backrefs, OPT_no_warn_backrefs, false);
Config->WarnCommon = Args.hasFlag(OPT_warn_common, OPT_no_warn_common, false);
Config->WarnSymbolOrdering =
Args.hasFlag(OPT_warn_symbol_ordering, OPT_no_warn_symbol_ordering, true);
Config->ZCombreloc = getZFlag(Args, "combreloc", "nocombreloc", true);
Config->ZCopyreloc = getZFlag(Args, "copyreloc", "nocopyreloc", true);
Config->ZExecstack = getZFlag(Args, "execstack", "noexecstack", false);
Config->ZHazardplt = hasZOption(Args, "hazardplt");
Config->ZNocopyreloc = hasZOption(Args, "nocopyreloc");
Config->ZInitfirst = hasZOption(Args, "initfirst");
Config->ZKeepTextSectionPrefix = getZFlag(
Args, "keep-text-section-prefix", "nokeep-text-section-prefix", false);
Config->ZNodelete = hasZOption(Args, "nodelete");
Config->ZNodlopen = hasZOption(Args, "nodlopen");
Config->ZNow = hasZOption(Args, "now");
Config->ZNow = getZFlag(Args, "now", "lazy", false);
Config->ZOrigin = hasZOption(Args, "origin");
Config->ZRelro = !hasZOption(Args, "norelro");
Config->ZRelro = getZFlag(Args, "relro", "norelro", true);
Config->ZRetpolineplt = hasZOption(Args, "retpolineplt");
Config->ZRodynamic = hasZOption(Args, "rodynamic");
Config->ZStackSize = args::getZOptionValue(Args, OPT_z, "stack-size", 0);
Config->ZText = !hasZOption(Args, "notext");
Config->ZText = getZFlag(Args, "text", "notext", true);
Config->ZWxneeded = hasZOption(Args, "wxneeded");
// Parse LTO plugin-related options for compatibility with gold.
for (auto *Arg : Args.filtered(OPT_plugin_opt, OPT_plugin_opt_eq)) {
StringRef S = Arg->getValue();
if (S == "disable-verify")
Config->DisableVerify = true;
else if (S == "save-temps")
Config->SaveTemps = true;
else if (S.startswith("O"))
Config->LTOO = parseInt(S.substr(1), Arg);
else if (S.startswith("lto-partitions="))
Config->LTOPartitions = parseInt(S.substr(15), Arg);
else if (S.startswith("jobs="))
Config->ThinLTOJobs = parseInt(S.substr(5), Arg);
else if (!S.startswith("/") && !S.startswith("-fresolution=") &&
!S.startswith("-pass-through=") && !S.startswith("mcpu=") &&
!S.startswith("thinlto") && S != "-function-sections" &&
S != "-data-sections")
error(Arg->getSpelling() + ": unknown option: " + S);
}
// Parse LTO options.
if (auto *Arg = Args.getLastArg(OPT_plugin_opt_mcpu_eq))
parseClangOption(Saver.save("-mcpu=" + StringRef(Arg->getValue())),
Arg->getSpelling());
for (auto *Arg : Args.filtered(OPT_plugin_opt))
parseClangOption(Arg->getValue(), Arg->getSpelling());
// Parse -mllvm options.
for (auto *Arg : Args.filtered(OPT_mllvm))
parseClangOption(Arg->getValue(), Arg->getSpelling());
if (Config->LTOO > 3)
error("invalid optimization level for LTO: " + Twine(Config->LTOO));
@ -742,17 +898,12 @@ void LinkerDriver::readConfigs(opt::InputArgList &Args) {
std::tie(Config->BuildId, Config->BuildIdVector) = getBuildId(Args);
if (auto *Arg = Args.getLastArg(OPT_pack_dyn_relocs_eq)) {
StringRef S = Arg->getValue();
if (S == "android")
Config->AndroidPackDynRelocs = true;
else if (S != "none")
error("unknown -pack-dyn-relocs format: " + S);
}
std::tie(Config->AndroidPackDynRelocs, Config->RelrPackDynRelocs) =
getPackDynRelocs(Args);
if (auto *Arg = Args.getLastArg(OPT_symbol_ordering_file))
if (Optional<MemoryBufferRef> Buffer = readFile(Arg->getValue()))
Config->SymbolOrderingFile = args::getLines(*Buffer);
Config->SymbolOrderingFile = getSymbolOrderingFile(*Buffer);
// If --retain-symbol-file is used, we'll keep only the symbols listed in
// the file and discard all others.
@ -780,32 +931,67 @@ void LinkerDriver::readConfigs(opt::InputArgList &Args) {
{Arg->getValue(), /*IsExternCpp*/ false, /*HasWildcard*/ false});
}
// If --export-dynamic-symbol=foo is given and symbol foo is defined in
// an object file in an archive file, that object file should be pulled
// out and linked. (It doesn't have to behave like that from technical
// point of view, but this is needed for compatibility with GNU.)
for (auto *Arg : Args.filtered(OPT_export_dynamic_symbol))
Config->Undefined.push_back(Arg->getValue());
for (auto *Arg : Args.filtered(OPT_version_script))
if (Optional<MemoryBufferRef> Buffer = readFile(Arg->getValue()))
readVersionScript(*Buffer);
if (Optional<std::string> Path = searchScript(Arg->getValue())) {
if (Optional<MemoryBufferRef> Buffer = readFile(*Path))
readVersionScript(*Buffer);
} else {
error(Twine("cannot find version script ") + Arg->getValue());
}
}
// Some Config members do not directly correspond to any particular
// command line options, but computed based on other Config values.
// This function initialize such members. See Config.h for the details
// of these values.
static void setConfigs() {
static void setConfigs(opt::InputArgList &Args) {
ELFKind Kind = Config->EKind;
uint16_t Machine = Config->EMachine;
// There is an ILP32 ABI for x86-64, although it's not very popular.
// It is called the x32 ABI.
bool IsX32 = (Kind == ELF32LEKind && Machine == EM_X86_64);
Config->CopyRelocs = (Config->Relocatable || Config->EmitRelocs);
Config->Is64 = (Kind == ELF64LEKind || Kind == ELF64BEKind);
Config->IsLE = (Kind == ELF32LEKind || Kind == ELF64LEKind);
Config->Endianness =
Config->IsLE ? support::endianness::little : support::endianness::big;
Config->IsMips64EL = (Kind == ELF64LEKind && Machine == EM_MIPS);
Config->IsRela = Config->Is64 || IsX32 || Config->MipsN32Abi;
Config->Pic = Config->Pie || Config->Shared;
Config->Wordsize = Config->Is64 ? 8 : 4;
// There is an ILP32 ABI for x86-64, although it's not very popular.
// It is called the x32 ABI.
bool IsX32 = (Kind == ELF32LEKind && Machine == EM_X86_64);
// ELF defines two different ways to store relocation addends as shown below:
//
// Rel: Addends are stored to the location where relocations are applied.
// Rela: Addends are stored as part of relocation entry.
//
// In other words, Rela makes it easy to read addends at the price of extra
// 4 or 8 byte for each relocation entry. We don't know why ELF defined two
// different mechanisms in the first place, but this is how the spec is
// defined.
//
// You cannot choose which one, Rel or Rela, you want to use. Instead each
// ABI defines which one you need to use. The following expression expresses
// that.
Config->IsRela =
(Config->Is64 || IsX32 || Machine == EM_PPC) && Machine != EM_MIPS;
// If the output uses REL relocations we must store the dynamic relocation
// addends to the output sections. We also store addends for RELA relocations
// if --apply-dynamic-relocs is used.
// We default to not writing the addends when using RELA relocations since
// any standard conforming tool can find it in r_addend.
Config->WriteAddends = Args.hasFlag(OPT_apply_dynamic_relocs,
OPT_no_apply_dynamic_relocs, false) ||
!Config->IsRela;
}
// Returns a value of "-format" option.
@ -820,6 +1006,10 @@ static bool getBinaryOption(StringRef S) {
}
void LinkerDriver::createFiles(opt::InputArgList &Args) {
// For --{push,pop}-state.
std::vector<std::tuple<bool, bool, bool>> Stack;
// Iterate over argv to process input files and positional arguments.
for (auto *Arg : Args) {
switch (Arg->getOption().getUnaliasedOption().getID()) {
case OPT_library:
@ -828,8 +1018,15 @@ void LinkerDriver::createFiles(opt::InputArgList &Args) {
case OPT_INPUT:
addFile(Arg->getValue(), /*WithLOption=*/false);
break;
case OPT_defsym: {
StringRef From;
StringRef To;
std::tie(From, To) = StringRef(Arg->getValue()).split('=');
readDefsym(From, MemoryBufferRef(To, "-defsym"));
break;
}
case OPT_script:
if (Optional<std::string> Path = searchLinkerScript(Arg->getValue())) {
if (Optional<std::string> Path = searchScript(Arg->getValue())) {
if (Optional<MemoryBufferRef> MB = readFile(*Path))
readLinkerScript(*MB);
break;
@ -857,11 +1054,48 @@ void LinkerDriver::createFiles(opt::InputArgList &Args) {
case OPT_no_whole_archive:
InWholeArchive = false;
break;
case OPT_just_symbols:
if (Optional<MemoryBufferRef> MB = readFile(Arg->getValue())) {
Files.push_back(createObjectFile(*MB));
Files.back()->JustSymbols = true;
}
break;
case OPT_start_group:
if (InputFile::IsInGroup)
error("nested --start-group");
InputFile::IsInGroup = true;
break;
case OPT_end_group:
if (!InputFile::IsInGroup)
error("stray --end-group");
InputFile::IsInGroup = false;
++InputFile::NextGroupId;
break;
case OPT_start_lib:
if (InLib)
error("nested --start-lib");
if (InputFile::IsInGroup)
error("may not nest --start-lib in --start-group");
InLib = true;
InputFile::IsInGroup = true;
break;
case OPT_end_lib:
if (!InLib)
error("stray --end-lib");
InLib = false;
InputFile::IsInGroup = false;
++InputFile::NextGroupId;
break;
case OPT_push_state:
Stack.emplace_back(Config->AsNeeded, Config->Static, InWholeArchive);
break;
case OPT_pop_state:
if (Stack.empty()) {
error("unbalanced --push-state/--pop-state");
break;
}
std::tie(Config->AsNeeded, Config->Static, InWholeArchive) = Stack.back();
Stack.pop_back();
break;
}
}
@ -934,14 +1168,6 @@ static DenseSet<StringRef> getExcludeLibs(opt::InputArgList &Args) {
return Ret;
}
static Optional<StringRef> getArchiveName(InputFile *File) {
if (isa<ArchiveFile>(File))
return File->getName();
if (!File->ArchiveName.empty())
return File->ArchiveName;
return None;
}
// Handles the -exclude-libs option. If a static library file is specified
// by the -exclude-libs option, all public symbols from the archive become
// private unless otherwise specified by version scripts or something.
@ -949,16 +1175,132 @@ static Optional<StringRef> getArchiveName(InputFile *File) {
//
// This is not a popular option, but some programs such as bionic libc use it.
template <class ELFT>
static void excludeLibs(opt::InputArgList &Args, ArrayRef<InputFile *> Files) {
static void excludeLibs(opt::InputArgList &Args) {
DenseSet<StringRef> Libs = getExcludeLibs(Args);
bool All = Libs.count("ALL");
for (InputFile *File : Files)
if (Optional<StringRef> Archive = getArchiveName(File))
if (All || Libs.count(path::filename(*Archive)))
auto Visit = [&](InputFile *File) {
if (!File->ArchiveName.empty())
if (All || Libs.count(path::filename(File->ArchiveName)))
for (Symbol *Sym : File->getSymbols())
if (!Sym->isLocal())
if (!Sym->isLocal() && Sym->File == File)
Sym->VersionId = VER_NDX_LOCAL;
};
for (InputFile *File : ObjectFiles)
Visit(File);
for (BitcodeFile *File : BitcodeFiles)
Visit(File);
}
// Force Sym to be entered in the output. Used for -u or equivalent.
template <class ELFT> static void handleUndefined(StringRef Name) {
Symbol *Sym = Symtab->find(Name);
if (!Sym)
return;
// Since symbol S may not be used inside the program, LTO may
// eliminate it. Mark the symbol as "used" to prevent it.
Sym->IsUsedInRegularObj = true;
if (Sym->isLazy())
Symtab->fetchLazy<ELFT>(Sym);
}
template <class ELFT> static bool shouldDemote(Symbol &Sym) {
// If all references to a DSO happen to be weak, the DSO is not added to
// DT_NEEDED. If that happens, we need to eliminate shared symbols created
// from the DSO. Otherwise, they become dangling references that point to a
// non-existent DSO.
if (auto *S = dyn_cast<SharedSymbol>(&Sym))
return !S->getFile<ELFT>().IsNeeded;
// We are done processing archives, so lazy symbols that were used but not
// found can be converted to undefined. We could also just delete the other
// lazy symbols, but that seems to be more work than it is worth.
return Sym.isLazy() && Sym.IsUsedInRegularObj;
}
// Some files, such as .so or files between -{start,end}-lib may be removed
// after their symbols are added to the symbol table. If that happens, we
// need to remove symbols that refer files that no longer exist, so that
// they won't appear in the symbol table of the output file.
//
// We remove symbols by demoting them to undefined symbol.
template <class ELFT> static void demoteSymbols() {
for (Symbol *Sym : Symtab->getSymbols()) {
if (shouldDemote<ELFT>(*Sym)) {
bool Used = Sym->Used;
replaceSymbol<Undefined>(Sym, nullptr, Sym->getName(), Sym->Binding,
Sym->StOther, Sym->Type);
Sym->Used = Used;
}
}
}
// The section referred to by S is considered address-significant. Set the
// KeepUnique flag on the section if appropriate.
static void markAddrsig(Symbol *S) {
if (auto *D = dyn_cast_or_null<Defined>(S))
if (D->Section)
// We don't need to keep text sections unique under --icf=all even if they
// are address-significant.
if (Config->ICF == ICFLevel::Safe || !(D->Section->Flags & SHF_EXECINSTR))
D->Section->KeepUnique = true;
}
// Record sections that define symbols mentioned in --keep-unique <symbol>
// and symbols referred to by address-significance tables. These sections are
// ineligible for ICF.
template <class ELFT>
static void findKeepUniqueSections(opt::InputArgList &Args) {
for (auto *Arg : Args.filtered(OPT_keep_unique)) {
StringRef Name = Arg->getValue();
auto *D = dyn_cast_or_null<Defined>(Symtab->find(Name));
if (!D || !D->Section) {
warn("could not find symbol " + Name + " to keep unique");
continue;
}
D->Section->KeepUnique = true;
}
// --icf=all --ignore-data-address-equality means that we can ignore
// the dynsym and address-significance tables entirely.
if (Config->ICF == ICFLevel::All && Config->IgnoreDataAddressEquality)
return;
// Symbols in the dynsym could be address-significant in other executables
// or DSOs, so we conservatively mark them as address-significant.
for (Symbol *S : Symtab->getSymbols())
if (S->includeInDynsym())
markAddrsig(S);
// Visit the address-significance table in each object file and mark each
// referenced symbol as address-significant.
for (InputFile *F : ObjectFiles) {
auto *Obj = cast<ObjFile<ELFT>>(F);
ArrayRef<Symbol *> Syms = Obj->getSymbols();
if (Obj->AddrsigSec) {
ArrayRef<uint8_t> Contents =
check(Obj->getObj().getSectionContents(Obj->AddrsigSec));
const uint8_t *Cur = Contents.begin();
while (Cur != Contents.end()) {
unsigned Size;
const char *Err;
uint64_t SymIndex = decodeULEB128(Cur, &Size, Contents.end(), &Err);
if (Err)
fatal(toString(F) + ": could not decode addrsig section: " + Err);
markAddrsig(Syms[SymIndex]);
Cur += Size;
}
} else {
// If an object file does not have an address-significance table,
// conservatively mark all of its symbols as address-significant.
for (Symbol *S : Syms)
markAddrsig(S);
}
}
}
// Do actual linking. Note that when this function is called,
@ -1009,14 +1351,6 @@ template <class ELFT> void LinkerDriver::link(opt::InputArgList &Args) {
for (InputFile *F : Files)
Symtab->addFile<ELFT>(F);
// Process -defsym option.
for (auto *Arg : Args.filtered(OPT_defsym)) {
StringRef From;
StringRef To;
std::tie(From, To) = StringRef(Arg->getValue()).split('=');
readDefsym(From, MemoryBufferRef(To, "-defsym"));
}
// Now that we have every file, we can decide if we will need a
// dynamic symbol table.
// We need one if we were asked to export dynamic symbols or if we are
@ -1033,24 +1367,29 @@ template <class ELFT> void LinkerDriver::link(opt::InputArgList &Args) {
// Handle the `--undefined <sym>` options.
for (StringRef S : Config->Undefined)
Symtab->fetchIfLazy<ELFT>(S);
handleUndefined<ELFT>(S);
// If an entry symbol is in a static archive, pull out that file now
// to complete the symbol table. After this, no new names except a
// few linker-synthesized ones will be added to the symbol table.
Symtab->fetchIfLazy<ELFT>(Config->Entry);
handleUndefined<ELFT>(Config->Entry);
// Return if there were name resolution errors.
if (errorCount())
return;
// Handle undefined symbols in DSOs.
if (!Config->Shared)
Symtab->scanShlibUndefined<ELFT>();
// Now when we read all script files, we want to finalize order of linker
// script commands, which can be not yet final because of INSERT commands.
Script->processInsertCommands();
// We want to declare linker script's symbols early,
// so that we can version them.
// They also might be exported if referenced by DSOs.
Script->declareSymbols();
// Handle the -exclude-libs option.
if (Args.hasArg(OPT_exclude_libs))
excludeLibs<ELFT>(Args, Files);
excludeLibs<ELFT>(Args);
// Create ElfHeader early. We need a dummy section in
// addReservedSymbols to mark the created symbols as not absolute.
@ -1073,10 +1412,18 @@ template <class ELFT> void LinkerDriver::link(opt::InputArgList &Args) {
for (auto *Arg : Args.filtered(OPT_wrap))
Symtab->addSymbolWrap<ELFT>(Arg->getValue());
// Do link-time optimization if given files are LLVM bitcode files.
// This compiles bitcode files into real object files.
Symtab->addCombinedLTOObject<ELFT>();
if (errorCount())
return;
// If -thinlto-index-only is given, we should create only "index
// files" and not object files. Index file creation is already done
// in addCombinedLTOObject, so we are done if that's the case.
if (Config->ThinLTOIndexOnly)
return;
// Apply symbol renames for -wrap.
Symtab->applySymbolWrap();
@ -1123,11 +1470,20 @@ template <class ELFT> void LinkerDriver::link(opt::InputArgList &Args) {
// Do size optimizations: garbage collection, merging of SHF_MERGE sections
// and identical code folding.
markLive<ELFT>();
decompressSections();
splitSections<ELFT>();
markLive<ELFT>();
demoteSymbols<ELFT>();
mergeSections();
if (Config->ICF)
if (Config->ICF != ICFLevel::None) {
findKeepUniqueSections<ELFT>(Args);
doIcf<ELFT>();
}
// Read the callgraph now that we know what was gced or icfed
if (auto *Arg = Args.getLastArg(OPT_call_graph_ordering_file))
if (Optional<MemoryBufferRef> Buffer = readFile(Arg->getValue()))
readCallGraph(*Buffer);
// Write the result to the file.
writeResult<ELFT>();

6
contrib/llvm/tools/lld/ELF/Driver.h
View File

@ -26,7 +26,7 @@ extern class LinkerDriver *Driver;
class LinkerDriver {
public:
void main(ArrayRef<const char *> Args, bool CanExitEarly);
void main(ArrayRef<const char *> Args);
void addFile(StringRef Path, bool WithLOption);
void addLibrary(StringRef Name);
@ -63,11 +63,11 @@ enum {
#undef OPTION
};
void printHelp(const char *Argv0);
void printHelp();
std::string createResponseFile(const llvm::opt::InputArgList &Args);
llvm::Optional<std::string> findFromSearchPaths(StringRef Path);
llvm::Optional<std::string> searchLinkerScript(StringRef Path);
llvm::Optional<std::string> searchScript(StringRef Path);
llvm::Optional<std::string> searchLibrary(StringRef Path);
} // namespace elf

55
contrib/llvm/tools/lld/ELF/DriverUtils.cpp
View File

@ -29,6 +29,7 @@
using namespace llvm;
using namespace llvm::sys;
using namespace llvm::opt;
using namespace lld;
using namespace lld::elf;
@ -58,18 +59,18 @@ static void handleColorDiagnostics(opt::InputArgList &Args) {
OPT_no_color_diagnostics);
if (!Arg)
return;
else if (Arg->getOption().getID() == OPT_color_diagnostics)
if (Arg->getOption().getID() == OPT_color_diagnostics) {
errorHandler().ColorDiagnostics = true;
else if (Arg->getOption().getID() == OPT_no_color_diagnostics)
} else if (Arg->getOption().getID() == OPT_no_color_diagnostics) {
errorHandler().ColorDiagnostics = false;
else {
} else {
StringRef S = Arg->getValue();
if (S == "always")
errorHandler().ColorDiagnostics = true;
else if (S == "never")
errorHandler().ColorDiagnostics = false;
else if (S != "auto")
error("unknown option: -color-diagnostics=" + S);
error("unknown option: --color-diagnostics=" + S);
}
}
@ -87,6 +88,29 @@ static cl::TokenizerCallback getQuotingStyle(opt::InputArgList &Args) {
return cl::TokenizeGNUCommandLine;
}
// Gold LTO plugin takes a `--plugin-opt foo=bar` option as an alias for
// `--plugin-opt=foo=bar`. We want to handle `--plugin-opt=foo=` as an
// option name and `bar` as a value. Unfortunately, OptParser cannot
// handle an option with a space in it.
//
// In this function, we concatenate command line arguments so that
// `--plugin-opt <foo>` is converted to `--plugin-opt=<foo>`. This is a
// bit hacky, but looks like it is still better than handling --plugin-opt
// options by hand.
static void concatLTOPluginOptions(SmallVectorImpl<const char *> &Args) {
SmallVector<const char *, 256> V;
for (size_t I = 0, E = Args.size(); I != E; ++I) {
StringRef S = Args[I];
if ((S == "-plugin-opt" || S == "--plugin-opt") && I + 1 != E) {
V.push_back(Saver.save(S + "=" + Args[I + 1]).data());
++I;
} else {
V.push_back(Args[I]);
}
}
Args = std::move(V);
}
// Parses a given list of options.
opt::InputArgList ELFOptTable::parse(ArrayRef<const char *> Argv) {
// Make InputArgList from string vectors.
@ -102,6 +126,7 @@ opt::InputArgList ELFOptTable::parse(ArrayRef<const char *> Argv) {
// Expand response files (arguments in the form of @<filename>)
// and then parse the argument again.
cl::ExpandResponseFiles(Saver, getQuotingStyle(Args), Vec);
concatLTOPluginOptions(Vec);
Args = this->ParseArgs(Vec, MissingIndex, MissingCount);
handleColorDiagnostics(Args);
@ -113,9 +138,9 @@ opt::InputArgList ELFOptTable::parse(ArrayRef<const char *> Argv) {
return Args;
}
void elf::printHelp(const char *Argv0) {
ELFOptTable().PrintHelp(outs(), Argv0, "lld", false /*ShowHidden*/,
true /*ShowAllAliases*/);
void elf::printHelp() {
ELFOptTable().PrintHelp(outs(), Config->ProgName.data(), "lld",
false /*ShowHidden*/, true /*ShowAllAliases*/);
outs() << "\n";
// Scripts generated by Libtool versions up to at least 2.4.6 (the most
@ -123,13 +148,7 @@ void elf::printHelp(const char *Argv0) {
// in a message for the -help option. If it doesn't match, the scripts
// assume that the linker doesn't support very basic features such as
// shared libraries. Therefore, we need to print out at least "elf".
// Here, we print out all the targets that we support.
outs() << Argv0 << ": supported targets: "
<< "elf32-i386 elf32-iamcu elf32-littlearm elf32-ntradbigmips "
<< "elf32-ntradlittlemips elf32-powerpc elf32-tradbigmips "
<< "elf32-tradlittlemips elf32-x86-64 "
<< "elf64-amdgpu elf64-littleaarch64 elf64-powerpc elf64-tradbigmips "
<< "elf64-tradlittlemips elf64-x86-64\n";
outs() << Config->ProgName << ": supported targets: elf\n";
}
// Reconstructs command line arguments so that so that you can re-run
@ -208,10 +227,10 @@ Optional<std::string> elf::searchLibrary(StringRef Name) {
return None;
}
// If a linker script doesn't exist in the current directory, we also look for
// the script in the '-L' search paths. This matches the behaviour of both '-T'
// and linker script INPUT() directives in ld.bfd.
Optional<std::string> elf::searchLinkerScript(StringRef Name) {
// If a linker/version script doesn't exist in the current directory, we also
// look for the script in the '-L' search paths. This matches the behaviour of
// '-T', --version-script=, and linker script INPUT() command in ld.bfd.
Optional<std::string> elf::searchScript(StringRef Name) {
if (fs::exists(Name))
return Name.str();
return findFromSearchPaths(Name);

8
contrib/llvm/tools/lld/ELF/EhFrame.cpp
View File

@ -20,18 +20,16 @@
#include "Config.h"
#include "InputSection.h"
#include "Relocations.h"
#include "Strings.h"
#include "Target.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Strings.h"
#include "llvm/BinaryFormat/Dwarf.h"
#include "llvm/Object/ELF.h"
#include "llvm/Support/Endian.h"
using namespace llvm;
using namespace llvm::ELF;
using namespace llvm::dwarf;
using namespace llvm::object;
using namespace llvm::support::endian;
using namespace lld;
using namespace lld::elf;
@ -73,7 +71,7 @@ size_t EhReader::readEhRecordSize() {
// First 4 bytes of CIE/FDE is the size of the record.
// If it is 0xFFFFFFFF, the next 8 bytes contain the size instead,
// but we do not support that format yet.
uint64_t V = read32(D.data(), Config->Endianness);
uint64_t V = read32(D.data());
if (V == UINT32_MAX)
failOn(D.data(), "CIE/FDE too large");
uint64_t Size = V + 4;

2
contrib/llvm/tools/lld/ELF/Filesystem.cpp
View File

@ -44,7 +44,7 @@ using namespace lld::elf;
// The calling thread returns almost immediately.
void elf::unlinkAsync(StringRef Path) {
// Removing a file is async on windows.
#if defined(LLVM_ON_WIN32)
#if defined(_WIN32)
sys::fs::remove(Path);
#else
if (!ThreadsEnabled || !sys::fs::exists(Path) ||

2
contrib/llvm/tools/lld/ELF/GdbIndex.cpp
View File

@ -34,7 +34,7 @@ template <class ELFT> LLDDwarfObj<ELFT>::LLDDwarfObj(ObjFile<ELFT> *Obj) {
.Case(".debug_ranges", &RangeSection)
.Case(".debug_line", &LineSection)
.Default(nullptr)) {
Sec->maybeUncompress();
Sec->maybeDecompress();
M->Data = toStringRef(Sec->Data);
M->Sec = Sec;
continue;

1
contrib/llvm/tools/lld/ELF/GdbIndex.h
View File

@ -49,7 +49,6 @@ template <class ELFT> class LLDDwarfObj final : public llvm::DWARFObject {
return LineSection;
}
StringRef getFileName() const override { return ""; }
StringRef getCUIndexSection() const override { return ""; }
StringRef getAbbrevSection() const override { return AbbrevSection; }
StringRef getStringSection() const override { return StrSection; }
StringRef getGnuPubNamesSection() const override {

116
contrib/llvm/tools/lld/ELF/ICF.cpp
View File

@ -77,6 +77,8 @@
#include "Config.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "SyntheticSections.h"
#include "Writer.h"
#include "lld/Common/Threads.h"
#include "llvm/ADT/Hashing.h"
#include "llvm/BinaryFormat/ELF.h"
@ -112,9 +114,9 @@ template <class ELFT> class ICF {
size_t findBoundary(size_t Begin, size_t End);
void forEachClassRange(size_t Begin, size_t End,
std::function<void(size_t, size_t)> Fn);
llvm::function_ref<void(size_t, size_t)> Fn);
void forEachClass(std::function<void(size_t, size_t)> Fn);
void forEachClass(llvm::function_ref<void(size_t, size_t)> Fn);
std::vector<InputSection *> Sections;
@ -161,15 +163,38 @@ template <class ELFT> static uint32_t getHash(InputSection *S) {
// Returns true if section S is subject of ICF.
static bool isEligible(InputSection *S) {
// Don't merge read only data sections unless --icf-data was passed.
if (!(S->Flags & SHF_EXECINSTR) && !Config->ICFData)
if (!S->Live || S->KeepUnique || !(S->Flags & SHF_ALLOC))
return false;
// .init and .fini contains instructions that must be executed to
// initialize and finalize the process. They cannot and should not
// be merged.
return S->Live && (S->Flags & SHF_ALLOC) && !(S->Flags & SHF_WRITE) &&
S->Name != ".init" && S->Name != ".fini";
// Don't merge writable sections. .data.rel.ro sections are marked as writable
// but are semantically read-only.
if ((S->Flags & SHF_WRITE) && S->Name != ".data.rel.ro" &&
!S->Name.startswith(".data.rel.ro."))
return false;
// SHF_LINK_ORDER sections are ICF'd as a unit with their dependent sections,
// so we don't consider them for ICF individually.
if (S->Flags & SHF_LINK_ORDER)
return false;
// Don't merge synthetic sections as their Data member is not valid and empty.
// The Data member needs to be valid for ICF as it is used by ICF to determine
// the equality of section contents.
if (isa<SyntheticSection>(S))
return false;
// .init and .fini contains instructions that must be executed to initialize
// and finalize the process. They cannot and should not be merged.
if (S->Name == ".init" || S->Name == ".fini")
return false;
// A user program may enumerate sections named with a C identifier using
// __start_* and __stop_* symbols. We cannot ICF any such sections because
// that could change program semantics.
if (isValidCIdentifier(S->Name))
return false;
return true;
}
// Split an equivalence class into smaller classes.
@ -214,9 +239,6 @@ template <class ELFT>
template <class RelTy>
bool ICF<ELFT>::constantEq(const InputSection *SecA, ArrayRef<RelTy> RA,
const InputSection *SecB, ArrayRef<RelTy> RB) {
if (RA.size() != RB.size())
return false;
for (size_t I = 0; I < RA.size(); ++I) {
if (RA[I].r_offset != RB[I].r_offset ||
RA[I].getType(Config->IsMips64EL) != RB[I].getType(Config->IsMips64EL))
@ -284,6 +306,13 @@ bool ICF<ELFT>::equalsConstant(const InputSection *A, const InputSection *B) {
A->getSize() != B->getSize() || A->Data != B->Data)
return false;
// If two sections have different output sections, we cannot merge them.
// FIXME: This doesn't do the right thing in the case where there is a linker
// script. We probably need to move output section assignment before ICF to
// get the correct behaviour here.
if (getOutputSectionName(A) != getOutputSectionName(B))
return false;
if (A->AreRelocsRela)
return constantEq(A, A->template relas<ELFT>(), B,
B->template relas<ELFT>());
@ -350,17 +379,12 @@ template <class ELFT> size_t ICF<ELFT>::findBoundary(size_t Begin, size_t End) {
// vector. Therefore, Sections vector can be considered as contiguous
// groups of sections, grouped by the class.
//
// This function calls Fn on every group that starts within [Begin, End).
// Note that a group must start in that range but doesn't necessarily
// have to end before End.
// This function calls Fn on every group within [Begin, End).
template <class ELFT>
void ICF<ELFT>::forEachClassRange(size_t Begin, size_t End,
std::function<void(size_t, size_t)> Fn) {
if (Begin > 0)
Begin = findBoundary(Begin - 1, End);
llvm::function_ref<void(size_t, size_t)> Fn) {
while (Begin < End) {
size_t Mid = findBoundary(Begin, Sections.size());
size_t Mid = findBoundary(Begin, End);
Fn(Begin, Mid);
Begin = Mid;
}
@ -368,7 +392,7 @@ void ICF<ELFT>::forEachClassRange(size_t Begin, size_t End,
// Call Fn on each equivalence class.
template <class ELFT>
void ICF<ELFT>::forEachClass(std::function<void(size_t, size_t)> Fn) {
void ICF<ELFT>::forEachClass(llvm::function_ref<void(size_t, size_t)> Fn) {
// If threading is disabled or the number of sections are
// too small to use threading, call Fn sequentially.
if (!ThreadsEnabled || Sections.size() < 1024) {
@ -380,16 +404,32 @@ void ICF<ELFT>::forEachClass(std::function<void(size_t, size_t)> Fn) {
Current = Cnt % 2;
Next = (Cnt + 1) % 2;
// Split sections into 256 shards and call Fn in parallel.
size_t NumShards = 256;
// Shard into non-overlapping intervals, and call Fn in parallel.
// The sharding must be completed before any calls to Fn are made
// so that Fn can modify the Chunks in its shard without causing data
// races.
const size_t NumShards = 256;
size_t Step = Sections.size() / NumShards;
parallelForEachN(0, NumShards, [&](size_t I) {
size_t End = (I == NumShards - 1) ? Sections.size() : (I + 1) * Step;
forEachClassRange(I * Step, End, Fn);
size_t Boundaries[NumShards + 1];
Boundaries[0] = 0;
Boundaries[NumShards] = Sections.size();
parallelForEachN(1, NumShards, [&](size_t I) {
Boundaries[I] = findBoundary((I - 1) * Step, Sections.size());
});
parallelForEachN(1, NumShards + 1, [&](size_t I) {
if (Boundaries[I - 1] < Boundaries[I])
forEachClassRange(Boundaries[I - 1], Boundaries[I], Fn);
});
++Cnt;
}
static void print(const Twine &S) {
if (Config->PrintIcfSections)
message(S);
}
// The main function of ICF.
template <class ELFT> void ICF<ELFT>::run() {
// Collect sections to merge.
@ -401,7 +441,7 @@ template <class ELFT> void ICF<ELFT>::run() {
// Initially, we use hash values to partition sections.
parallelForEach(Sections, [&](InputSection *S) {
// Set MSB to 1 to avoid collisions with non-hash IDs.
S->Class[0] = getHash<ELFT>(S) | (1 << 31);
S->Class[0] = getHash<ELFT>(S) | (1U << 31);
});
// From now on, sections in Sections vector are ordered so that sections
@ -424,25 +464,21 @@ template <class ELFT> void ICF<ELFT>::run() {
log("ICF needed " + Twine(Cnt) + " iterations");
// Merge sections by the equivalence class.
forEachClass([&](size_t Begin, size_t End) {
forEachClassRange(0, Sections.size(), [&](size_t Begin, size_t End) {
if (End - Begin == 1)
return;
log("selected " + Sections[Begin]->Name);
print("selected section " + toString(Sections[Begin]));
for (size_t I = Begin + 1; I < End; ++I) {
log(" removed " + Sections[I]->Name);
print(" removing identical section " + toString(Sections[I]));
Sections[Begin]->replace(Sections[I]);
// At this point we know sections merged are fully identical and hence
// we want to remove duplicate implicit dependencies such as link order
// and relocation sections.
for (InputSection *IS : Sections[I]->DependentSections)
IS->Live = false;
}
});
// Mark ARM Exception Index table sections that refer to folded code
// sections as not live. These sections have an implict dependency
// via the link order dependency.
if (Config->EMachine == EM_ARM)
for (InputSectionBase *Sec : InputSections)
if (auto *S = dyn_cast<InputSection>(Sec))
if (S->Flags & SHF_LINK_ORDER)
S->Live = S->getLinkOrderDep()->Live;
}
// ICF entry point function.

4
contrib/llvm/tools/lld/ELF/ICF.h
View File

@ -12,8 +12,10 @@
namespace lld {
namespace elf {
template <class ELFT> void doIcf();
}
} // namespace elf
} // namespace lld
#endif

626
contrib/llvm/tools/lld/ELF/InputFiles.cpp
View File

@ -38,14 +38,23 @@ using namespace llvm::sys::fs;
using namespace lld;
using namespace lld::elf;
bool InputFile::IsInGroup;
uint32_t InputFile::NextGroupId;
std::vector<BinaryFile *> elf::BinaryFiles;
std::vector<BitcodeFile *> elf::BitcodeFiles;
std::vector<LazyObjFile *> elf::LazyObjFiles;
std::vector<InputFile *> elf::ObjectFiles;
std::vector<InputFile *> elf::SharedFiles;
TarWriter *elf::Tar;
InputFile::InputFile(Kind K, MemoryBufferRef M) : MB(M), FileKind(K) {}
InputFile::InputFile(Kind K, MemoryBufferRef M)
: MB(M), GroupId(NextGroupId), FileKind(K) {
// All files within the same --{start,end}-group get the same group ID.
// Otherwise, a new file will get a new group ID.
if (!IsInGroup)
++NextGroupId;
}
Optional<MemoryBufferRef> elf::readFile(StringRef Path) {
// The --chroot option changes our virtual root directory.
@ -55,7 +64,7 @@ Optional<MemoryBufferRef> elf::readFile(StringRef Path) {
log(Path);
auto MBOrErr = MemoryBuffer::getFile(Path);
auto MBOrErr = MemoryBuffer::getFile(Path, -1, false);
if (auto EC = MBOrErr.getError()) {
error("cannot open " + Path + ": " + EC.message());
return None;
@ -115,51 +124,60 @@ std::string InputFile::getSrcMsg(const Symbol &Sym, InputSectionBase &Sec,
}
template <class ELFT> void ObjFile<ELFT>::initializeDwarf() {
DWARFContext Dwarf(make_unique<LLDDwarfObj<ELFT>>(this));
const DWARFObject &Obj = Dwarf.getDWARFObj();
DwarfLine.reset(new DWARFDebugLine);
Dwarf = llvm::make_unique<DWARFContext>(make_unique<LLDDwarfObj<ELFT>>(this));
const DWARFObject &Obj = Dwarf->getDWARFObj();
DWARFDataExtractor LineData(Obj, Obj.getLineSection(), Config->IsLE,
Config->Wordsize);
// The second parameter is offset in .debug_line section
// for compilation unit (CU) of interest. We have only one
// CU (object file), so offset is always 0.
// FIXME: Provide the associated DWARFUnit if there is one. DWARF v5
// needs it in order to find indirect strings.
const DWARFDebugLine::LineTable *LT =
DwarfLine->getOrParseLineTable(LineData, 0, nullptr);
// Return if there is no debug information about CU available.
if (!Dwarf.getNumCompileUnits())
return;
// Loop over variable records and insert them to VariableLoc.
DWARFCompileUnit *CU = Dwarf.getCompileUnitAtIndex(0);
for (const auto &Entry : CU->dies()) {
DWARFDie Die(CU, &Entry);
// Skip all tags that are not variables.
if (Die.getTag() != dwarf::DW_TAG_variable)
for (std::unique_ptr<DWARFCompileUnit> &CU : Dwarf->compile_units()) {
auto Report = [](Error Err) {
handleAllErrors(std::move(Err),
[](ErrorInfoBase &Info) { warn(Info.message()); });
};
Expected<const DWARFDebugLine::LineTable *> ExpectedLT =
Dwarf->getLineTableForUnit(CU.get(), Report);
const DWARFDebugLine::LineTable *LT = nullptr;
if (ExpectedLT)
LT = *ExpectedLT;
else
Report(ExpectedLT.takeError());
if (!LT)
continue;
LineTables.push_back(LT);
// Skip if a local variable because we don't need them for generating error
// messages. In general, only non-local symbols can fail to be linked.
if (!dwarf::toUnsigned(Die.find(dwarf::DW_AT_external), 0))
continue;
// Loop over variable records and insert them to VariableLoc.
for (const auto &Entry : CU->dies()) {
DWARFDie Die(CU.get(), &Entry);
// Skip all tags that are not variables.
if (Die.getTag() != dwarf::DW_TAG_variable)
continue;
// Get the source filename index for the variable.
unsigned File = dwarf::toUnsigned(Die.find(dwarf::DW_AT_decl_file), 0);
if (!LT->hasFileAtIndex(File))
continue;
// Skip if a local variable because we don't need them for generating
// error messages. In general, only non-local symbols can fail to be
// linked.
if (!dwarf::toUnsigned(Die.find(dwarf::DW_AT_external), 0))
continue;
// Get the line number on which the variable is declared.
unsigned Line = dwarf::toUnsigned(Die.find(dwarf::DW_AT_decl_line), 0);
// Get the source filename index for the variable.
unsigned File = dwarf::toUnsigned(Die.find(dwarf::DW_AT_decl_file), 0);
if (!LT->hasFileAtIndex(File))
continue;
// Get the name of the variable and add the collected information to
// VariableLoc. Usually Name is non-empty, but it can be empty if the input
// object file lacks some debug info.
StringRef Name = dwarf::toString(Die.find(dwarf::DW_AT_name), "");
if (!Name.empty())
VariableLoc.insert({Name, {File, Line}});
// Get the line number on which the variable is declared.
unsigned Line = dwarf::toUnsigned(Die.find(dwarf::DW_AT_decl_line), 0);
// Here we want to take the variable name to add it into VariableLoc.
// Variable can have regular and linkage name associated. At first, we try
// to get linkage name as it can be different, for example when we have
// two variables in different namespaces of the same object. Use common
// name otherwise, but handle the case when it also absent in case if the
// input object file lacks some debug info.
StringRef Name =
dwarf::toString(Die.find(dwarf::DW_AT_linkage_name),
dwarf::toString(Die.find(dwarf::DW_AT_name), ""));
if (!Name.empty())
VariableLoc.insert({Name, {LT, File, Line}});
}
}
}
@ -170,11 +188,6 @@ Optional<std::pair<std::string, unsigned>>
ObjFile<ELFT>::getVariableLoc(StringRef Name) {
llvm::call_once(InitDwarfLine, [this]() { initializeDwarf(); });
// There is always only one CU so it's offset is 0.
const DWARFDebugLine::LineTable *LT = DwarfLine->getLineTable(0);
if (!LT)
return None;
// Return if we have no debug information about data object.
auto It = VariableLoc.find(Name);
if (It == VariableLoc.end())
@ -182,12 +195,12 @@ ObjFile<ELFT>::getVariableLoc(StringRef Name) {
// Take file name string from line table.
std::string FileName;
if (!LT->getFileNameByIndex(
It->second.first /* File */, nullptr,
if (!It->second.LT->getFileNameByIndex(
It->second.File, nullptr,
DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, FileName))
return None;
return std::make_pair(FileName, It->second.second /*Line*/);
return std::make_pair(FileName, It->second.Line);
}
// Returns source line information for a given offset
@ -197,29 +210,15 @@ Optional<DILineInfo> ObjFile<ELFT>::getDILineInfo(InputSectionBase *S,
uint64_t Offset) {
llvm::call_once(InitDwarfLine, [this]() { initializeDwarf(); });
// The offset to CU is 0.
const DWARFDebugLine::LineTable *Tbl = DwarfLine->getLineTable(0);
if (!Tbl)
return None;
// Use fake address calcuated by adding section file offset and offset in
// section. See comments for ObjectInfo class.
DILineInfo Info;
Tbl->getFileLineInfoForAddress(
S->getOffsetInFile() + Offset, nullptr,
DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, Info);
if (Info.Line == 0)
return None;
return Info;
}
// Returns source line information for a given offset
// using DWARF debug info.
template <class ELFT>
std::string ObjFile<ELFT>::getLineInfo(InputSectionBase *S, uint64_t Offset) {
if (Optional<DILineInfo> Info = getDILineInfo(S, Offset))
return Info->FileName + ":" + std::to_string(Info->Line);
return "";
for (const llvm::DWARFDebugLine::LineTable *LT : LineTables)
if (LT->getFileLineInfoForAddress(
S->getOffsetInFile() + Offset, nullptr,
DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, Info))
return Info;
return None;
}
// Returns "<internal>", "foo.a(bar.o)" or "baz.o".
@ -249,7 +248,7 @@ ELFFileBase<ELFT>::ELFFileBase(Kind K, MemoryBufferRef MB) : InputFile(K, MB) {
template <class ELFT>
typename ELFT::SymRange ELFFileBase<ELFT>::getGlobalELFSyms() {
return makeArrayRef(ELFSyms.begin() + FirstNonLocal, ELFSyms.end());
return makeArrayRef(ELFSyms.begin() + FirstGlobal, ELFSyms.end());
}
template <class ELFT>
@ -260,9 +259,9 @@ uint32_t ELFFileBase<ELFT>::getSectionIndex(const Elf_Sym &Sym) const {
template <class ELFT>
void ELFFileBase<ELFT>::initSymtab(ArrayRef<Elf_Shdr> Sections,
const Elf_Shdr *Symtab) {
FirstNonLocal = Symtab->sh_info;
FirstGlobal = Symtab->sh_info;
ELFSyms = CHECK(getObj().symbols(Symtab), this);
if (FirstNonLocal == 0 || FirstNonLocal > ELFSyms.size())
if (FirstGlobal == 0 || FirstGlobal > ELFSyms.size())
fatal(toString(this) + ": invalid sh_info in symbol table");
StringTable =
@ -278,17 +277,22 @@ ObjFile<ELFT>::ObjFile(MemoryBufferRef M, StringRef ArchiveName)
template <class ELFT> ArrayRef<Symbol *> ObjFile<ELFT>::getLocalSymbols() {
if (this->Symbols.empty())
return {};
return makeArrayRef(this->Symbols).slice(1, this->FirstNonLocal - 1);
return makeArrayRef(this->Symbols).slice(1, this->FirstGlobal - 1);
}
template <class ELFT> ArrayRef<Symbol *> ObjFile<ELFT>::getGlobalSymbols() {
return makeArrayRef(this->Symbols).slice(this->FirstNonLocal);
return makeArrayRef(this->Symbols).slice(this->FirstGlobal);
}
template <class ELFT>
void ObjFile<ELFT>::parse(DenseSet<CachedHashStringRef> &ComdatGroups) {
// Read section and symbol tables.
initializeSections(ComdatGroups);
// Read a section table. JustSymbols is usually false.
if (this->JustSymbols)
initializeJustSymbols();
else
initializeSections(ComdatGroups);
// Read a symbol table.
initializeSymbols();
}
@ -312,7 +316,7 @@ StringRef ObjFile<ELFT>::getShtGroupSignature(ArrayRef<Elf_Shdr> Sections,
// we use a section name as a signature.
//
// Such SHT_GROUP sections are invalid from the perspective of the ELF
// standard, but GNU gold 1.14 (the neweset version as of July 2017) or
// standard, but GNU gold 1.14 (the newest version as of July 2017) or
// older produce such sections as outputs for the -r option, so we need
// a bug-compatibility.
if (Signature.empty() && Sym->getType() == STT_SECTION)
@ -332,9 +336,19 @@ ObjFile<ELFT>::getShtGroupEntries(const Elf_Shdr &Sec) {
}
template <class ELFT> bool ObjFile<ELFT>::shouldMerge(const Elf_Shdr &Sec) {
// We don't merge sections if -O0 (default is -O1). This makes sometimes
// the linker significantly faster, although the output will be bigger.
if (Config->Optimize == 0)
// On a regular link we don't merge sections if -O0 (default is -O1). This
// sometimes makes the linker significantly faster, although the output will
// be bigger.
//
// Doing the same for -r would create a problem as it would combine sections
// with different sh_entsize. One option would be to just copy every SHF_MERGE
// section as is to the output. While this would produce a valid ELF file with
// usable SHF_MERGE sections, tools like (llvm-)?dwarfdump get confused when
// they see two .debug_str. We could have separate logic for combining
// SHF_MERGE sections based both on their name and sh_entsize, but that seems
// to be more trouble than it is worth. Instead, we just use the regular (-O1)
// logic for -r.
if (Config->Optimize == 0 && !Config->Relocatable)
return false;
// A mergeable section with size 0 is useless because they don't have
@ -365,12 +379,33 @@ template <class ELFT> bool ObjFile<ELFT>::shouldMerge(const Elf_Shdr &Sec) {
return true;
}
// This is for --just-symbols.
//
// --just-symbols is a very minor feature that allows you to link your
// output against other existing program, so that if you load both your
// program and the other program into memory, your output can refer the
// other program's symbols.
//
// When the option is given, we link "just symbols". The section table is
// initialized with null pointers.
template <class ELFT> void ObjFile<ELFT>::initializeJustSymbols() {
ArrayRef<Elf_Shdr> ObjSections = CHECK(this->getObj().sections(), this);
this->Sections.resize(ObjSections.size());
for (const Elf_Shdr &Sec : ObjSections) {
if (Sec.sh_type != SHT_SYMTAB)
continue;
this->initSymtab(ObjSections, &Sec);
return;
}
}
template <class ELFT>
void ObjFile<ELFT>::initializeSections(
DenseSet<CachedHashStringRef> &ComdatGroups) {
const ELFFile<ELFT> &Obj = this->getObj();
ArrayRef<Elf_Shdr> ObjSections = CHECK(this->getObj().sections(), this);
ArrayRef<Elf_Shdr> ObjSections = CHECK(Obj.sections(), this);
uint64_t Size = ObjSections.size();
this->Sections.resize(Size);
this->SectionStringTable =
@ -385,6 +420,17 @@ void ObjFile<ELFT>::initializeSections(
// if -r is given, we'll let the final link discard such sections.
// This is compatible with GNU.
if ((Sec.sh_flags & SHF_EXCLUDE) && !Config->Relocatable) {
if (Sec.sh_type == SHT_LLVM_ADDRSIG) {
// We ignore the address-significance table if we know that the object
// file was created by objcopy or ld -r. This is because these tools
// will reorder the symbols in the symbol table, invalidating the data
// in the address-significance table, which refers to symbols by index.
if (Sec.sh_link != 0)
this->AddrsigSec = &Sec;
else if (Config->ICF == ICFLevel::Safe)
warn(toString(this) + ": --icf=safe is incompatible with object "
"files created using objcopy or ld -r");
}
this->Sections[I] = &InputSection::Discarded;
continue;
}
@ -435,8 +481,15 @@ void ObjFile<ELFT>::initializeSections(
if (Sec.sh_link >= this->Sections.size())
fatal(toString(this) +
": invalid sh_link index: " + Twine(Sec.sh_link));
this->Sections[Sec.sh_link]->DependentSections.push_back(
cast<InputSection>(this->Sections[I]));
InputSectionBase *LinkSec = this->Sections[Sec.sh_link];
InputSection *IS = cast<InputSection>(this->Sections[I]);
LinkSec->DependentSections.push_back(IS);
if (!isa<InputSection>(LinkSec))
error("a section " + IS->Name +
" with SHF_LINK_ORDER should not refer a non-regular "
"section: " +
toString(LinkSec));
}
}
}
@ -531,10 +584,11 @@ InputSectionBase *ObjFile<ELFT>::createInputSection(const Elf_Shdr &Sec) {
}
case SHT_RELA:
case SHT_REL: {
// Find the relocation target section and associate this
// section with it. Target can be discarded, for example
// if it is a duplicated member of SHT_GROUP section, we
// do not create or proccess relocatable sections then.
// Find a relocation target section and associate this section with that.
// Target may have been discarded if it is in a different section group
// and the group is discarded, even though it's a violation of the
// spec. We handle that situation gracefully by discarding dangling
// relocation sections.
InputSectionBase *Target = getRelocTarget(Sec);
if (!Target)
return nullptr;
@ -549,32 +603,28 @@ InputSectionBase *ObjFile<ELFT>::createInputSection(const Elf_Shdr &Sec) {
fatal(toString(this) +
": multiple relocation sections to one section are not supported");
// Mergeable sections with relocations are tricky because relocations
// need to be taken into account when comparing section contents for
// merging. It's not worth supporting such mergeable sections because
// they are rare and it'd complicates the internal design (we usually
// have to determine if two sections are mergeable early in the link
// process much before applying relocations). We simply handle mergeable
// sections with relocations as non-mergeable.
// ELF spec allows mergeable sections with relocations, but they are
// rare, and it is in practice hard to merge such sections by contents,
// because applying relocations at end of linking changes section
// contents. So, we simply handle such sections as non-mergeable ones.
// Degrading like this is acceptable because section merging is optional.
if (auto *MS = dyn_cast<MergeInputSection>(Target)) {
Target = toRegularSection(MS);
this->Sections[Sec.sh_info] = Target;
}
size_t NumRelocations;
if (Sec.sh_type == SHT_RELA) {
ArrayRef<Elf_Rela> Rels = CHECK(this->getObj().relas(&Sec), this);
Target->FirstRelocation = Rels.begin();
NumRelocations = Rels.size();
Target->NumRelocations = Rels.size();
Target->AreRelocsRela = true;
} else {
ArrayRef<Elf_Rel> Rels = CHECK(this->getObj().rels(&Sec), this);
Target->FirstRelocation = Rels.begin();
NumRelocations = Rels.size();
Target->NumRelocations = Rels.size();
Target->AreRelocsRela = false;
}
assert(isUInt<31>(NumRelocations));
Target->NumRelocations = NumRelocations;
assert(isUInt<31>(Target->NumRelocations));
// Relocation sections processed by the linker are usually removed
// from the output, so returning `nullptr` for the normal case.
@ -606,13 +656,24 @@ InputSectionBase *ObjFile<ELFT>::createInputSection(const Elf_Shdr &Sec) {
if (Name == ".note.GNU-stack")
return &InputSection::Discarded;
// Split stacks is a feature to support a discontiguous stack. At least
// as of 2017, it seems that the feature is not being used widely.
// Only GNU gold supports that. We don't. For the details about that,
// see https://gcc.gnu.org/wiki/SplitStacks
// Split stacks is a feature to support a discontiguous stack,
// commonly used in the programming language Go. For the details,
// see https://gcc.gnu.org/wiki/SplitStacks. An object file compiled
// for split stack will include a .note.GNU-split-stack section.
if (Name == ".note.GNU-split-stack") {
error(toString(this) +
": object file compiled with -fsplit-stack is not supported");
if (Config->Relocatable) {
error("Cannot mix split-stack and non-split-stack in a relocatable link");
return &InputSection::Discarded;
}
this->SplitStack = true;
return &InputSection::Discarded;
}
// An object file cmpiled for split stack, but where some of the
// functions were compiled with the no_split_stack_attribute will
// include a .note.GNU-no-split-stack section.
if (Name == ".note.GNU-no-split-stack") {
this->SomeNoSplitStack = true;
return &InputSection::Discarded;
}
@ -624,6 +685,14 @@ InputSectionBase *ObjFile<ELFT>::createInputSection(const Elf_Shdr &Sec) {
if (Name.startswith(".gnu.linkonce."))
return &InputSection::Discarded;
// If we are creating a new .build-id section, strip existing .build-id
// sections so that the output won't have more than one .build-id.
// This is not usually a problem because input object files normally don't
// have .build-id sections, but you can create such files by
// "ld.{bfd,gold,lld} -r --build-id", and we want to guard against it.
if (Name == ".note.gnu.build-id" && Config->BuildId != BuildIdKind::None)
return &InputSection::Discarded;
// The linker merges EH (exception handling) frames and creates a
// .eh_frame_hdr section for runtime. So we handle them with a special
// class. For relocatable outputs, they are just passed through.
@ -705,33 +774,33 @@ ArchiveFile::ArchiveFile(std::unique_ptr<Archive> &&File)
File(std::move(File)) {}
template <class ELFT> void ArchiveFile::parse() {
Symbols.reserve(File->getNumberOfSymbols());
for (const Archive::Symbol &Sym : File->symbols())
Symbols.push_back(Symtab->addLazyArchive<ELFT>(Sym.getName(), *this, Sym));
Symtab->addLazyArchive<ELFT>(Sym.getName(), *this, Sym);
}
// Returns a buffer pointing to a member file containing a given symbol.
std::pair<MemoryBufferRef, uint64_t>
ArchiveFile::getMember(const Archive::Symbol *Sym) {
InputFile *ArchiveFile::fetch(const Archive::Symbol &Sym) {
Archive::Child C =
CHECK(Sym->getMember(), toString(this) +
": could not get the member for symbol " +
Sym->getName());
CHECK(Sym.getMember(), toString(this) +
": could not get the member for symbol " +
Sym.getName());
if (!Seen.insert(C.getChildOffset()).second)
return {MemoryBufferRef(), 0};
return nullptr;
MemoryBufferRef Ret =
MemoryBufferRef MB =
CHECK(C.getMemoryBufferRef(),
toString(this) +
": could not get the buffer for the member defining symbol " +
Sym->getName());
Sym.getName());
if (C.getParent()->isThin() && Tar)
Tar->append(relativeToRoot(CHECK(C.getFullName(), this)), Ret.getBuffer());
if (C.getParent()->isThin())
return {Ret, 0};
return {Ret, C.getChildOffset()};
if (Tar && C.getParent()->isThin())
Tar->append(relativeToRoot(CHECK(C.getFullName(), this)), MB.getBuffer());
InputFile *File = createObjectFile(
MB, getName(), C.getParent()->isThin() ? 0 : C.getChildOffset());
File->GroupId = GroupId;
return File;
}
template <class ELFT>
@ -788,34 +857,42 @@ template <class ELFT> void SharedFile<ELFT>::parseSoName() {
}
}
// Parses ".gnu.version" section which is a parallel array for the symbol table.
// If a given file doesn't have ".gnu.version" section, returns VER_NDX_GLOBAL.
template <class ELFT> std::vector<uint32_t> SharedFile<ELFT>::parseVersyms() {
size_t Size = this->ELFSyms.size() - this->FirstGlobal;
if (!VersymSec)
return std::vector<uint32_t>(Size, VER_NDX_GLOBAL);
const char *Base = this->MB.getBuffer().data();
const Elf_Versym *Versym =
reinterpret_cast<const Elf_Versym *>(Base + VersymSec->sh_offset) +
this->FirstGlobal;
std::vector<uint32_t> Ret(Size);
for (size_t I = 0; I < Size; ++I)
Ret[I] = Versym[I].vs_index;
return Ret;
}
// Parse the version definitions in the object file if present. Returns a vector
// whose nth element contains a pointer to the Elf_Verdef for version identifier
// n. Version identifiers that are not definitions map to nullptr. The array
// always has at least length 1.
// n. Version identifiers that are not definitions map to nullptr.
template <class ELFT>
std::vector<const typename ELFT::Verdef *>
SharedFile<ELFT>::parseVerdefs(const Elf_Versym *&Versym) {
std::vector<const Elf_Verdef *> Verdefs(1);
// We only need to process symbol versions for this DSO if it has both a
// versym and a verdef section, which indicates that the DSO contains symbol
// version definitions.
if (!VersymSec || !VerdefSec)
return Verdefs;
// The location of the first global versym entry.
const char *Base = this->MB.getBuffer().data();
Versym = reinterpret_cast<const Elf_Versym *>(Base + VersymSec->sh_offset) +
this->FirstNonLocal;
std::vector<const typename ELFT::Verdef *> SharedFile<ELFT>::parseVerdefs() {
if (!VerdefSec)
return {};
// We cannot determine the largest verdef identifier without inspecting
// every Elf_Verdef, but both bfd and gold assign verdef identifiers
// sequentially starting from 1, so we predict that the largest identifier
// will be VerdefCount.
unsigned VerdefCount = VerdefSec->sh_info;
Verdefs.resize(VerdefCount + 1);
std::vector<const Elf_Verdef *> Verdefs(VerdefCount + 1);
// Build the Verdefs array by following the chain of Elf_Verdef objects
// from the start of the .gnu.version_d section.
const char *Base = this->MB.getBuffer().data();
const char *Verdef = Base + VerdefSec->sh_offset;
for (unsigned I = 0; I != VerdefCount; ++I) {
auto *CurVerdef = reinterpret_cast<const Elf_Verdef *>(Verdef);
@ -829,82 +906,99 @@ SharedFile<ELFT>::parseVerdefs(const Elf_Versym *&Versym) {
return Verdefs;
}
// Fully parse the shared object file. This must be called after parseSoName().
template <class ELFT> void SharedFile<ELFT>::parseRest() {
// Create mapping from version identifiers to Elf_Verdef entries.
const Elf_Versym *Versym = nullptr;
Verdefs = parseVerdefs(Versym);
// We do not usually care about alignments of data in shared object
// files because the loader takes care of it. However, if we promote a
// DSO symbol to point to .bss due to copy relocation, we need to keep
// the original alignment requirements. We infer it in this function.
template <class ELFT>
uint32_t SharedFile<ELFT>::getAlignment(ArrayRef<Elf_Shdr> Sections,
const Elf_Sym &Sym) {
uint64_t Ret = UINT64_MAX;
if (Sym.st_value)
Ret = 1ULL << countTrailingZeros((uint64_t)Sym.st_value);
if (0 < Sym.st_shndx && Sym.st_shndx < Sections.size())
Ret = std::min<uint64_t>(Ret, Sections[Sym.st_shndx].sh_addralign);
return (Ret > UINT32_MAX) ? 0 : Ret;
}
// Fully parse the shared object file. This must be called after parseSoName().
//
// This function parses symbol versions. If a DSO has version information,
// the file has a ".gnu.version_d" section which contains symbol version
// definitions. Each symbol is associated to one version through a table in
// ".gnu.version" section. That table is a parallel array for the symbol
// table, and each table entry contains an index in ".gnu.version_d".
//
// The special index 0 is reserved for VERF_NDX_LOCAL and 1 is for
// VER_NDX_GLOBAL. There's no table entry for these special versions in
// ".gnu.version_d".
//
// The file format for symbol versioning is perhaps a bit more complicated
// than necessary, but you can easily understand the code if you wrap your
// head around the data structure described above.
template <class ELFT> void SharedFile<ELFT>::parseRest() {
Verdefs = parseVerdefs(); // parse .gnu.version_d
std::vector<uint32_t> Versyms = parseVersyms(); // parse .gnu.version
ArrayRef<Elf_Shdr> Sections = CHECK(this->getObj().sections(), this);
// System libraries can have a lot of symbols with versions. Using a
// fixed buffer for computing the versions name (foo@ver) can save a
// lot of allocations.
SmallString<0> VersionedNameBuffer;
// Add symbols to the symbol table.
Elf_Sym_Range Syms = this->getGlobalELFSyms();
for (const Elf_Sym &Sym : Syms) {
unsigned VersymIndex = VER_NDX_GLOBAL;
if (Versym) {
VersymIndex = Versym->vs_index;
++Versym;
}
bool Hidden = VersymIndex & VERSYM_HIDDEN;
VersymIndex = VersymIndex & ~VERSYM_HIDDEN;
ArrayRef<Elf_Sym> Syms = this->getGlobalELFSyms();
for (size_t I = 0; I < Syms.size(); ++I) {
const Elf_Sym &Sym = Syms[I];
StringRef Name = CHECK(Sym.getName(this->StringTable), this);
if (Sym.isUndefined()) {
Undefs.push_back(Name);
Symbol *S = Symtab->addUndefined<ELFT>(Name, Sym.getBinding(),
Sym.st_other, Sym.getType(),
/*CanOmitFromDynSym=*/false, this);
S->ExportDynamic = true;
continue;
}
// ELF spec requires that all local symbols precede weak or global
// symbols in each symbol table, and the index of first non-local symbol
// is stored to sh_info. If a local symbol appears after some non-local
// symbol, that's a violation of the spec.
if (Sym.getBinding() == STB_LOCAL) {
warn("found local symbol '" + Name +
"' in global part of symbol table in file " + toString(this));
continue;
}
if (Config->EMachine == EM_MIPS) {
// FIXME: MIPS BFD linker puts _gp_disp symbol into DSO files
// and incorrectly assigns VER_NDX_LOCAL to this section global
// symbol. Here is a workaround for this bug.
if (Versym && VersymIndex == VER_NDX_LOCAL && Name == "_gp_disp")
continue;
}
// MIPS BFD linker puts _gp_disp symbol into DSO files and incorrectly
// assigns VER_NDX_LOCAL to this section global symbol. Here is a
// workaround for this bug.
uint32_t Idx = Versyms[I] & ~VERSYM_HIDDEN;
if (Config->EMachine == EM_MIPS && Idx == VER_NDX_LOCAL &&
Name == "_gp_disp")
continue;
const Elf_Verdef *Ver = nullptr;
if (VersymIndex != VER_NDX_GLOBAL) {
if (VersymIndex >= Verdefs.size() || VersymIndex == VER_NDX_LOCAL) {
error("corrupt input file: version definition index " +
Twine(VersymIndex) + " for symbol " + Name +
" is out of bounds\n>>> defined in " + toString(this));
continue;
}
Ver = Verdefs[VersymIndex];
} else {
VersymIndex = 0;
}
// We do not usually care about alignments of data in shared object
// files because the loader takes care of it. However, if we promote a
// DSO symbol to point to .bss due to copy relocation, we need to keep
// the original alignment requirements. We infer it here.
uint64_t Alignment = 1;
if (Sym.st_value)
Alignment = 1ULL << countTrailingZeros((uint64_t)Sym.st_value);
if (0 < Sym.st_shndx && Sym.st_shndx < Sections.size()) {
uint64_t SecAlign = Sections[Sym.st_shndx].sh_addralign;
Alignment = std::min(Alignment, SecAlign);
}
if (Alignment > UINT32_MAX)
error(toString(this) + ": alignment too large: " + Name);
if (!Hidden)
Symtab->addShared(Name, *this, Sym, Alignment, VersymIndex);
uint64_t Alignment = getAlignment(Sections, Sym);
if (!(Versyms[I] & VERSYM_HIDDEN))
Symtab->addShared(Name, *this, Sym, Alignment, Idx);
// Also add the symbol with the versioned name to handle undefined symbols
// with explicit versions.
if (Ver) {
StringRef VerName = this->StringTable.data() + Ver->getAux()->vda_name;
Name = Saver.save(Name + "@" + VerName);
Symtab->addShared(Name, *this, Sym, Alignment, VersymIndex);
if (Idx == VER_NDX_GLOBAL)
continue;
if (Idx >= Verdefs.size() || Idx == VER_NDX_LOCAL) {
error("corrupt input file: version definition index " + Twine(Idx) +
" for symbol " + Name + " is out of bounds\n>>> defined in " +
toString(this));
continue;
}
StringRef VerName =
this->StringTable.data() + Verdefs[Idx]->getAux()->vda_name;
VersionedNameBuffer.clear();
Name = (Name + "@" + VerName).toStringRef(VersionedNameBuffer);
Symtab->addShared(Saver.save(Name), *this, Sym, Alignment, Idx);
}
}
@ -937,8 +1031,9 @@ static uint8_t getBitcodeMachineKind(StringRef Path, const Triple &T) {
case Triple::x86_64:
return EM_X86_64;
default:
fatal(Path + ": could not infer e_machine from bitcode target triple " +
error(Path + ": could not infer e_machine from bitcode target triple " +
T.str());
return EM_NONE;
}
}
@ -947,17 +1042,21 @@ BitcodeFile::BitcodeFile(MemoryBufferRef MB, StringRef ArchiveName,
: InputFile(BitcodeKind, MB) {
this->ArchiveName = ArchiveName;
// Here we pass a new MemoryBufferRef which is identified by ArchiveName
// (the fully resolved path of the archive) + member name + offset of the
// member in the archive.
// ThinLTO uses the MemoryBufferRef identifier to access its internal
// data structures and if two archives define two members with the same name,
// this causes a collision which result in only one of the objects being
// taken into consideration at LTO time (which very likely causes undefined
// symbols later in the link stage).
MemoryBufferRef MBRef(MB.getBuffer(),
Saver.save(ArchiveName + MB.getBufferIdentifier() +
utostr(OffsetInArchive)));
std::string Path = MB.getBufferIdentifier().str();
if (Config->ThinLTOIndexOnly)
Path = replaceThinLTOSuffix(MB.getBufferIdentifier());
// ThinLTO assumes that all MemoryBufferRefs given to it have a unique
// name. If two archives define two members with the same name, this
// causes a collision which result in only one of the objects being taken
// into consideration at LTO time (which very likely causes undefined
// symbols later in the link stage). So we append file offset to make
// filename unique.
MemoryBufferRef MBRef(
MB.getBuffer(),
Saver.save(ArchiveName + Path +
(ArchiveName.empty() ? "" : utostr(OffsetInArchive))));
Obj = CHECK(lto::InputFile::create(MBRef), this);
Triple T(Obj->getTargetTriple());
@ -981,7 +1080,7 @@ template <class ELFT>
static Symbol *createBitcodeSymbol(const std::vector<bool> &KeptComdats,
const lto::InputFile::Symbol &ObjSym,
BitcodeFile &F) {
StringRef NameRef = Saver.save(ObjSym.getName());
StringRef Name = Saver.save(ObjSym.getName());
uint32_t Binding = ObjSym.isWeak() ? STB_WEAK : STB_GLOBAL;
uint8_t Type = ObjSym.isTLS() ? STT_TLS : STT_NOTYPE;
@ -990,20 +1089,20 @@ static Symbol *createBitcodeSymbol(const std::vector<bool> &KeptComdats,
int C = ObjSym.getComdatIndex();
if (C != -1 && !KeptComdats[C])
return Symtab->addUndefined<ELFT>(NameRef, Binding, Visibility, Type,
return Symtab->addUndefined<ELFT>(Name, Binding, Visibility, Type,
CanOmitFromDynSym, &F);
if (ObjSym.isUndefined())
return Symtab->addUndefined<ELFT>(NameRef, Binding, Visibility, Type,
return Symtab->addUndefined<ELFT>(Name, Binding, Visibility, Type,
CanOmitFromDynSym, &F);
if (ObjSym.isCommon())
return Symtab->addCommon(NameRef, ObjSym.getCommonSize(),
return Symtab->addCommon(Name, ObjSym.getCommonSize(),
ObjSym.getCommonAlignment(), Binding, Visibility,
STT_OBJECT, F);
return Symtab->addBitcode(NameRef, Binding, Visibility, Type,
CanOmitFromDynSym, F);
return Symtab->addBitcode(Name, Binding, Visibility, Type, CanOmitFromDynSym,
F);
}
template <class ELFT>
@ -1038,8 +1137,8 @@ static ELFKind getELFKind(MemoryBufferRef MB) {
void BinaryFile::parse() {
ArrayRef<uint8_t> Data = toArrayRef(MB.getBuffer());
auto *Section = make<InputSection>(nullptr, SHF_ALLOC | SHF_WRITE,
SHT_PROGBITS, 8, Data, ".data");
auto *Section = make<InputSection>(this, SHF_ALLOC | SHF_WRITE, SHT_PROGBITS,
8, Data, ".data");
Sections.push_back(Section);
// For each input file foo that is embedded to a result as a binary
@ -1059,11 +1158,6 @@ void BinaryFile::parse() {
Data.size(), 0, STB_GLOBAL, nullptr, nullptr);
}
static bool isBitcode(MemoryBufferRef MB) {
using namespace sys::fs;
return identify_magic(MB.getBuffer()) == file_magic::bitcode;
}
InputFile *elf::createObjectFile(MemoryBufferRef MB, StringRef ArchiveName,
uint64_t OffsetInArchive) {
if (isBitcode(MB))
@ -1099,9 +1193,9 @@ InputFile *elf::createSharedFile(MemoryBufferRef MB, StringRef DefaultSoName) {
}
MemoryBufferRef LazyObjFile::getBuffer() {
if (Seen)
if (AddedToLink)
return MemoryBufferRef();
Seen = true;
AddedToLink = true;
return MB;
}
@ -1109,68 +1203,74 @@ InputFile *LazyObjFile::fetch() {
MemoryBufferRef MBRef = getBuffer();
if (MBRef.getBuffer().empty())
return nullptr;
return createObjectFile(MBRef, ArchiveName, OffsetInArchive);
InputFile *File = createObjectFile(MBRef, ArchiveName, OffsetInArchive);
File->GroupId = GroupId;
return File;
}
template <class ELFT> void LazyObjFile::parse() {
for (StringRef Sym : getSymbolNames())
Symtab->addLazyObject<ELFT>(Sym, *this);
}
template <class ELFT> std::vector<StringRef> LazyObjFile::getElfSymbols() {
typedef typename ELFT::Shdr Elf_Shdr;
typedef typename ELFT::Sym Elf_Sym;
typedef typename ELFT::SymRange Elf_Sym_Range;
ELFFile<ELFT> Obj = check(ELFFile<ELFT>::create(this->MB.getBuffer()));
ArrayRef<Elf_Shdr> Sections = CHECK(Obj.sections(), this);
for (const Elf_Shdr &Sec : Sections) {
if (Sec.sh_type != SHT_SYMTAB)
continue;
Elf_Sym_Range Syms = CHECK(Obj.symbols(&Sec), this);
uint32_t FirstNonLocal = Sec.sh_info;
StringRef StringTable =
CHECK(Obj.getStringTableForSymtab(Sec, Sections), this);
std::vector<StringRef> V;
for (const Elf_Sym &Sym : Syms.slice(FirstNonLocal))
if (Sym.st_shndx != SHN_UNDEF)
V.push_back(CHECK(Sym.getName(StringTable), this));
return V;
// A lazy object file wraps either a bitcode file or an ELF file.
if (isBitcode(this->MB)) {
std::unique_ptr<lto::InputFile> Obj =
CHECK(lto::InputFile::create(this->MB), this);
for (const lto::InputFile::Symbol &Sym : Obj->symbols())
if (!Sym.isUndefined())
Symtab->addLazyObject<ELFT>(Saver.save(Sym.getName()), *this);
return;
}
return {};
}
std::vector<StringRef> LazyObjFile::getBitcodeSymbols() {
std::unique_ptr<lto::InputFile> Obj =
CHECK(lto::InputFile::create(this->MB), this);
std::vector<StringRef> V;
for (const lto::InputFile::Symbol &Sym : Obj->symbols())
if (!Sym.isUndefined())
V.push_back(Saver.save(Sym.getName()));
return V;
}
// Returns a vector of globally-visible defined symbol names.
std::vector<StringRef> LazyObjFile::getSymbolNames() {
if (isBitcode(this->MB))
return getBitcodeSymbols();
switch (getELFKind(this->MB)) {
case ELF32LEKind:
return getElfSymbols<ELF32LE>();
addElfSymbols<ELF32LE>();
return;
case ELF32BEKind:
return getElfSymbols<ELF32BE>();
addElfSymbols<ELF32BE>();
return;
case ELF64LEKind:
return getElfSymbols<ELF64LE>();
addElfSymbols<ELF64LE>();
return;
case ELF64BEKind:
return getElfSymbols<ELF64BE>();
addElfSymbols<ELF64BE>();
return;
default:
llvm_unreachable("getELFKind");
}
}
template <class ELFT> void LazyObjFile::addElfSymbols() {
ELFFile<ELFT> Obj = check(ELFFile<ELFT>::create(MB.getBuffer()));
ArrayRef<typename ELFT::Shdr> Sections = CHECK(Obj.sections(), this);
for (const typename ELFT::Shdr &Sec : Sections) {
if (Sec.sh_type != SHT_SYMTAB)
continue;
typename ELFT::SymRange Syms = CHECK(Obj.symbols(&Sec), this);
uint32_t FirstGlobal = Sec.sh_info;
StringRef StringTable =
CHECK(Obj.getStringTableForSymtab(Sec, Sections), this);
for (const typename ELFT::Sym &Sym : Syms.slice(FirstGlobal))
if (Sym.st_shndx != SHN_UNDEF)
Symtab->addLazyObject<ELFT>(CHECK(Sym.getName(StringTable), this),
*this);
return;
}
}
std::string elf::replaceThinLTOSuffix(StringRef Path) {
StringRef Suffix = Config->ThinLTOObjectSuffixReplace.first;
StringRef Repl = Config->ThinLTOObjectSuffixReplace.second;
if (!Path.endswith(Suffix)) {
error("-thinlto-object-suffix-replace=" + Suffix + ";" + Repl +
" was given, but " + Path + " does not end with the suffix");
return "";
}
return (Path.drop_back(Suffix.size()) + Repl).str();
}
template void ArchiveFile::parse<ELF32LE>();
template void ArchiveFile::parse<ELF32BE>();
template void ArchiveFile::parse<ELF64LE>();

82
contrib/llvm/tools/lld/ELF/InputFiles.h
View File

@ -12,22 +12,20 @@
#include "Config.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/LLVM.h"
#include "lld/Common/Reproduce.h"
#include "llvm/ADT/CachedHashString.h"
#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
#include "llvm/IR/Comdat.h"
#include "llvm/Object/Archive.h"
#include "llvm/Object/ELF.h"
#include "llvm/Object/IRObjectFile.h"
#include "llvm/Support/Threading.h"
#include <map>
namespace llvm {
class DWARFDebugLine;
class TarWriter;
struct DILineInfo;
namespace lto {
@ -48,7 +46,6 @@ namespace elf {
using llvm::object::Archive;
class Lazy;
class Symbol;
// If -reproduce option is given, all input files are written
@ -90,15 +87,15 @@ class InputFile {
// Returns object file symbols. It is a runtime error to call this
// function on files of other types.
ArrayRef<Symbol *> getSymbols() {
assert(FileKind == ObjKind || FileKind == BitcodeKind ||
FileKind == ArchiveKind);
assert(FileKind == BinaryKind || FileKind == ObjKind ||
FileKind == BitcodeKind);
return Symbols;
}
// Filename of .a which contained this file. If this file was
// not in an archive file, it is the empty string. We use this
// string for creating error messages.
StringRef ArchiveName;
std::string ArchiveName;
// If this is an architecture-specific file, the following members
// have ELF type (i.e. ELF{32,64}{LE,BE}) and target machine type.
@ -112,6 +109,20 @@ class InputFile {
std::string getSrcMsg(const Symbol &Sym, InputSectionBase &Sec,
uint64_t Offset);
// True if this is an argument for --just-symbols. Usually false.
bool JustSymbols = false;
// GroupId is used for --warn-backrefs which is an optional error
// checking feature. All files within the same --{start,end}-group or
// --{start,end}-lib get the same group ID. Otherwise, each file gets a new
// group ID. For more info, see checkDependency() in SymbolTable.cpp.
uint32_t GroupId;
static bool IsInGroup;
static uint32_t NextGroupId;
// Index of MIPS GOT built for this file.
llvm::Optional<size_t> MipsGotIndex;
protected:
InputFile(Kind K, MemoryBufferRef M);
std::vector<InputSectionBase *> Sections;
@ -144,7 +155,7 @@ template <typename ELFT> class ELFFileBase : public InputFile {
protected:
ArrayRef<Elf_Sym> ELFSyms;
uint32_t FirstNonLocal = 0;
uint32_t FirstGlobal = 0;
ArrayRef<Elf_Word> SymtabSHNDX;
StringRef StringTable;
void initSymtab(ArrayRef<Elf_Shdr> Sections, const Elf_Shdr *Symtab);
@ -183,9 +194,6 @@ template <class ELFT> class ObjFile : public ELFFileBase<ELFT> {
return getSymbol(SymIndex);
}
// Returns source line information for a given offset.
// If no information is available, returns "".
std::string getLineInfo(InputSectionBase *S, uint64_t Offset);
llvm::Optional<llvm::DILineInfo> getDILineInfo(InputSectionBase *, uint64_t);
llvm::Optional<std::pair<std::string, unsigned>> getVariableLoc(StringRef Name);
@ -199,10 +207,22 @@ template <class ELFT> class ObjFile : public ELFFileBase<ELFT> {
// symbol table.
StringRef SourceFile;
// True if the file defines functions compiled with
// -fsplit-stack. Usually false.
bool SplitStack = false;
// True if the file defines functions compiled with -fsplit-stack,
// but had one or more functions with the no_split_stack attribute.
bool SomeNoSplitStack = false;
// Pointer to this input file's .llvm_addrsig section, if it has one.
const Elf_Shdr *AddrsigSec = nullptr;
private:
void
initializeSections(llvm::DenseSet<llvm::CachedHashStringRef> &ComdatGroups);
void initializeSymbols();
void initializeJustSymbols();
void initializeDwarf();
InputSectionBase *getRelocTarget(const Elf_Shdr &Sec);
InputSectionBase *createInputSection(const Elf_Shdr &Sec);
@ -218,8 +238,14 @@ template <class ELFT> class ObjFile : public ELFFileBase<ELFT> {
// reporting. Linker may find reasonable number of errors in a
// single object file, so we cache debugging information in order to
// parse it only once for each object file we link.
std::unique_ptr<llvm::DWARFDebugLine> DwarfLine;
llvm::DenseMap<StringRef, std::pair<unsigned, unsigned>> VariableLoc;
std::unique_ptr<llvm::DWARFContext> Dwarf;
std::vector<const llvm::DWARFDebugLine::LineTable *> LineTables;
struct VarLoc {
const llvm::DWARFDebugLine::LineTable *LT;
unsigned File;
unsigned Line;
};
llvm::DenseMap<StringRef, VarLoc> VariableLoc;
llvm::once_flag InitDwarfLine;
};
@ -243,13 +269,11 @@ class LazyObjFile : public InputFile {
template <class ELFT> void parse();
MemoryBufferRef getBuffer();
InputFile *fetch();
bool AddedToLink = false;
private:
std::vector<StringRef> getSymbolNames();
template <class ELFT> std::vector<StringRef> getElfSymbols();
std::vector<StringRef> getBitcodeSymbols();
template <class ELFT> void addElfSymbols();
bool Seen = false;
uint64_t OffsetInArchive;
};
@ -260,11 +284,11 @@ class ArchiveFile : public InputFile {
static bool classof(const InputFile *F) { return F->kind() == ArchiveKind; }
template <class ELFT> void parse();
// Returns a memory buffer for a given symbol and the offset in the archive
// for the member. An empty memory buffer and an offset of zero
// is returned if we have already returned the same memory buffer.
// (So that we don't instantiate same members more than once.)
std::pair<MemoryBufferRef, uint64_t> getMember(const Archive::Symbol *Sym);
// Pulls out an object file that contains a definition for Sym and
// returns it. If the same file was instantiated before, this
// function returns a nullptr (so we don't instantiate the same file
// more than once.)
InputFile *fetch(const Archive::Symbol &Sym);
private:
std::unique_ptr<Archive> File;
@ -291,7 +315,6 @@ template <class ELFT> class SharedFile : public ELFFileBase<ELFT> {
typedef typename ELFT::Verdef Elf_Verdef;
typedef typename ELFT::Versym Elf_Versym;
std::vector<StringRef> Undefs;
const Elf_Shdr *VersymSec = nullptr;
const Elf_Shdr *VerdefSec = nullptr;
@ -299,8 +322,6 @@ template <class ELFT> class SharedFile : public ELFFileBase<ELFT> {
std::vector<const Elf_Verdef *> Verdefs;
std::string SoName;
llvm::ArrayRef<StringRef> getUndefinedSymbols() { return Undefs; }
static bool classof(const InputFile *F) {
return F->kind() == Base::SharedKind;
}
@ -309,7 +330,9 @@ template <class ELFT> class SharedFile : public ELFFileBase<ELFT> {
void parseSoName();
void parseRest();
std::vector<const Elf_Verdef *> parseVerdefs(const Elf_Versym *&Versym);
uint32_t getAlignment(ArrayRef<Elf_Shdr> Sections, const Elf_Sym &Sym);
std::vector<const Elf_Verdef *> parseVerdefs();
std::vector<uint32_t> parseVersyms();
struct NeededVer {
// The string table offset of the version name in the output file.
@ -338,8 +361,15 @@ InputFile *createObjectFile(MemoryBufferRef MB, StringRef ArchiveName = "",
uint64_t OffsetInArchive = 0);
InputFile *createSharedFile(MemoryBufferRef MB, StringRef DefaultSoName);
inline bool isBitcode(MemoryBufferRef MB) {
return identify_magic(MB.getBuffer()) == llvm::file_magic::bitcode;
}
std::string replaceThinLTOSuffix(StringRef Path);
extern std::vector<BinaryFile *> BinaryFiles;
extern std::vector<BitcodeFile *> BitcodeFiles;
extern std::vector<LazyObjFile *> LazyObjFiles;
extern std::vector<InputFile *> ObjectFiles;
extern std::vector<InputFile *> SharedFiles;

460
contrib/llvm/tools/lld/ELF/InputSection.cpp
View File

@ -14,6 +14,7 @@
#include "LinkerScript.h"
#include "OutputSections.h"
#include "Relocations.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "SyntheticSections.h"
#include "Target.h"
@ -26,7 +27,10 @@
#include "llvm/Support/Endian.h"
#include "llvm/Support/Threading.h"
#include "llvm/Support/xxhash.h"
#include <algorithm>
#include <mutex>
#include <set>
#include <vector>
using namespace llvm;
using namespace llvm::ELF;
@ -45,32 +49,6 @@ std::string lld::toString(const InputSectionBase *Sec) {
return (toString(Sec->File) + ":(" + Sec->Name + ")").str();
}
DenseMap<SectionBase *, int> elf::buildSectionOrder() {
DenseMap<SectionBase *, int> SectionOrder;
if (Config->SymbolOrderingFile.empty())
return SectionOrder;
// Build a map from symbols to their priorities. Symbols that didn't
// appear in the symbol ordering file have the lowest priority 0.
// All explicitly mentioned symbols have negative (higher) priorities.
DenseMap<StringRef, int> SymbolOrder;
int Priority = -Config->SymbolOrderingFile.size();
for (StringRef S : Config->SymbolOrderingFile)
SymbolOrder.insert({S, Priority++});
// Build a map from sections to their priorities.
for (InputFile *File : ObjectFiles) {
for (Symbol *Sym : File->getSymbols()) {
auto *D = dyn_cast<Defined>(Sym);
if (!D || !D->Section)
continue;
int &Priority = SectionOrder[D->Section];
Priority = std::min(Priority, SymbolOrder.lookup(D->getName()));
}
}
return SectionOrder;
}
template <class ELFT>
static ArrayRef<uint8_t> getSectionContents(ObjFile<ELFT> &File,
const typename ELFT::Shdr &Hdr) {
@ -168,12 +146,8 @@ uint64_t SectionBase::getOffset(uint64_t Offset) const {
return Offset == uint64_t(-1) ? OS->Size : Offset;
}
case Regular:
return cast<InputSection>(this)->OutSecOff + Offset;
case Synthetic: {
auto *IS = cast<InputSection>(this);
// For synthetic sections we treat offset -1 as the end of the section.
return IS->OutSecOff + (Offset == uint64_t(-1) ? IS->getSize() : Offset);
}
case Synthetic:
return cast<InputSection>(this)->getOffset(Offset);
case EHFrame:
// The file crtbeginT.o has relocations pointing to the start of an empty
// .eh_frame that is known to be the first in the link. It does that to
@ -182,16 +156,21 @@ uint64_t SectionBase::getOffset(uint64_t Offset) const {
case Merge:
const MergeInputSection *MS = cast<MergeInputSection>(this);
if (InputSection *IS = MS->getParent())
return IS->OutSecOff + MS->getOffset(Offset);
return MS->getOffset(Offset);
return IS->getOffset(MS->getParentOffset(Offset));
return MS->getParentOffset(Offset);
}
llvm_unreachable("invalid section kind");
}
uint64_t SectionBase::getVA(uint64_t Offset) const {
const OutputSection *Out = getOutputSection();
return (Out ? Out->Addr : 0) + getOffset(Offset);
}
OutputSection *SectionBase::getOutputSection() {
InputSection *Sec;
if (auto *IS = dyn_cast<InputSection>(this))
return IS->getParent();
Sec = IS;
else if (auto *MS = dyn_cast<MergeInputSection>(this))
Sec = MS->getParent();
else if (auto *EH = dyn_cast<EhInputSection>(this))
@ -201,34 +180,50 @@ OutputSection *SectionBase::getOutputSection() {
return Sec ? Sec->getParent() : nullptr;
}
// Uncompress section contents if required. Note that this function
// Decompress section contents if required. Note that this function
// is called from parallelForEach, so it must be thread-safe.
void InputSectionBase::maybeUncompress() {
if (UncompressBuf || !Decompressor::isCompressedELFSection(Flags, Name))
void InputSectionBase::maybeDecompress() {
if (DecompressBuf)
return;
if (!(Flags & SHF_COMPRESSED) && !Name.startswith(".zdebug"))
return;
// Decompress a section.
Decompressor Dec = check(Decompressor::create(Name, toStringRef(Data),
Config->IsLE, Config->Is64));
size_t Size = Dec.getDecompressedSize();
UncompressBuf.reset(new char[Size]());
if (Error E = Dec.decompress({UncompressBuf.get(), Size}))
DecompressBuf.reset(new char[Size + Name.size()]());
if (Error E = Dec.decompress({DecompressBuf.get(), Size}))
fatal(toString(this) +
": decompress failed: " + llvm::toString(std::move(E)));
Data = makeArrayRef((uint8_t *)UncompressBuf.get(), Size);
Data = makeArrayRef((uint8_t *)DecompressBuf.get(), Size);
Flags &= ~(uint64_t)SHF_COMPRESSED;
// A section name may have been altered if compressed. If that's
// the case, restore the original name. (i.e. ".zdebug_" -> ".debug_")
if (Name.startswith(".zdebug")) {
DecompressBuf[Size] = '.';
memcpy(&DecompressBuf[Size + 1], Name.data() + 2, Name.size() - 2);
Name = StringRef(&DecompressBuf[Size], Name.size() - 1);
}
}
InputSection *InputSectionBase::getLinkOrderDep() const {
if ((Flags & SHF_LINK_ORDER) && Link != 0) {
InputSectionBase *L = File->getSections()[Link];
if (auto *IS = dyn_cast<InputSection>(L))
return IS;
error("a section with SHF_LINK_ORDER should not refer a non-regular "
"section: " +
toString(L));
}
assert(Link);
assert(Flags & SHF_LINK_ORDER);
return cast<InputSection>(File->getSections()[Link]);
}
// Find a function symbol that encloses a given location.
template <class ELFT>
Defined *InputSectionBase::getEnclosingFunction(uint64_t Offset) {
for (Symbol *B : File->getSymbols())
if (Defined *D = dyn_cast<Defined>(B))
if (D->Section == this && D->Type == STT_FUNC &&
D->Value <= Offset && Offset < D->Value + D->Size)
return D;
return nullptr;
}
@ -241,9 +236,8 @@ std::string InputSectionBase::getLocation(uint64_t Offset) {
.str();
// First check if we can get desired values from debugging information.
std::string LineInfo = getFile<ELFT>()->getLineInfo(this, Offset);
if (!LineInfo.empty())
return LineInfo;
if (Optional<DILineInfo> Info = getFile<ELFT>()->getDILineInfo(this, Offset))
return Info->FileName + ":" + std::to_string(Info->Line);
// File->SourceFile contains STT_FILE symbol that contains a
// source file name. If it's missing, we use an object file name.
@ -251,12 +245,8 @@ std::string InputSectionBase::getLocation(uint64_t Offset) {
if (SrcFile.empty())
SrcFile = toString(File);
// Find a function symbol that encloses a given location.
for (Symbol *B : File->getSymbols())
if (auto *D = dyn_cast<Defined>(B))
if (D->Section == this && D->Type == STT_FUNC)
if (D->Value <= Offset && Offset < D->Value + D->Size)
return SrcFile + ":(function " + toString(*D) + ")";
if (Defined *D = getEnclosingFunction<ELFT>(Offset))
return SrcFile + ":(function " + toString(*D) + ")";
// If there's no symbol, print out the offset in the section.
return (SrcFile + ":(" + Name + "+0x" + utohexstr(Offset) + ")").str();
@ -292,7 +282,7 @@ std::string InputSectionBase::getObjMsg(uint64_t Off) {
std::string Archive;
if (!File->ArchiveName.empty())
Archive = (" in archive " + File->ArchiveName).str();
Archive = " in archive " + File->ArchiveName;
// Find a symbol that encloses a given location.
for (Symbol *B : File->getSymbols())
@ -345,8 +335,9 @@ template <class ELFT> void InputSection::copyShtGroup(uint8_t *Buf) {
*To++ = Sections[Idx]->getOutputSection()->SectionIndex;
}
InputSectionBase *InputSection::getRelocatedSection() {
assert(Type == SHT_RELA || Type == SHT_REL);
InputSectionBase *InputSection::getRelocatedSection() const {
if (!File || (Type != SHT_RELA && Type != SHT_REL))
return nullptr;
ArrayRef<InputSectionBase *> Sections = File->getSections();
return Sections[Info];
}
@ -365,12 +356,12 @@ void InputSection::copyRelocations(uint8_t *Buf, ArrayRef<RelTy> Rels) {
auto *P = reinterpret_cast<typename ELFT::Rela *>(Buf);
Buf += sizeof(RelTy);
if (Config->IsRela)
if (RelTy::IsRela)
P->r_addend = getAddend<ELFT>(Rel);
// Output section VA is zero for -r, so r_offset is an offset within the
// section, but for --emit-relocs it is an virtual address.
P->r_offset = Sec->getOutputSection()->Addr + Sec->getOffset(Rel.r_offset);
P->r_offset = Sec->getVA(Rel.r_offset);
P->setSymbolAndType(InX::SymTab->getSymbolIndex(&Sym), Type,
Config->IsMips64EL);
@ -395,17 +386,32 @@ void InputSection::copyRelocations(uint8_t *Buf, ArrayRef<RelTy> Rels) {
continue;
}
if (Config->IsRela) {
P->r_addend =
Sym.getVA(getAddend<ELFT>(Rel)) - Section->getOutputSection()->Addr;
} else if (Config->Relocatable) {
const uint8_t *BufLoc = Sec->Data.begin() + Rel.r_offset;
Sec->Relocations.push_back({R_ABS, Type, Rel.r_offset,
Target->getImplicitAddend(BufLoc, Type),
&Sym});
}
}
int64_t Addend = getAddend<ELFT>(Rel);
const uint8_t *BufLoc = Sec->Data.begin() + Rel.r_offset;
if (!RelTy::IsRela)
Addend = Target->getImplicitAddend(BufLoc, Type);
if (Config->EMachine == EM_MIPS && Config->Relocatable &&
Target->getRelExpr(Type, Sym, BufLoc) == R_MIPS_GOTREL) {
// Some MIPS relocations depend on "gp" value. By default,
// this value has 0x7ff0 offset from a .got section. But
// relocatable files produced by a complier or a linker
// might redefine this default value and we must use it
// for a calculation of the relocation result. When we
// generate EXE or DSO it's trivial. Generating a relocatable
// output is more difficult case because the linker does
// not calculate relocations in this mode and loses
// individual "gp" values used by each input object file.
// As a workaround we add the "gp" value to the relocation
// addend and save it back to the file.
Addend += Sec->getFile<ELFT>()->MipsGp0;
}
if (RelTy::IsRela)
P->r_addend = Sym.getVA(Addend) - Section->getOutputSection()->Addr;
else if (Config->Relocatable)
Sec->Relocations.push_back({R_ABS, Type, Rel.r_offset, Addend, &Sym});
}
}
}
@ -481,14 +487,17 @@ static uint64_t getARMStaticBase(const Symbol &Sym) {
return OS->PtLoad->FirstSec->Addr;
}
static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
const Symbol &Sym, RelExpr Expr) {
static uint64_t getRelocTargetVA(const InputFile *File, RelType Type, int64_t A,
uint64_t P, const Symbol &Sym, RelExpr Expr) {
switch (Expr) {
case R_INVALID:
return 0;
case R_ABS:
case R_RELAX_TLS_LD_TO_LE_ABS:
case R_RELAX_GOT_PC_NOPIC:
return Sym.getVA(A);
case R_ADDEND:
return A;
case R_ARM_SBREL:
return Sym.getVA(A) - getARMStaticBase(Sym);
case R_GOT:
@ -505,7 +514,9 @@ static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
case R_GOT_FROM_END:
case R_RELAX_TLS_GD_TO_IE_END:
return Sym.getGotOffset() + A - InX::Got->getSize();
case R_TLSLD_GOT_OFF:
case R_GOT_OFF:
case R_RELAX_TLS_GD_TO_IE_GOT_OFF:
return Sym.getGotOffset() + A;
case R_GOT_PAGE_PC:
case R_RELAX_TLS_GD_TO_IE_PAGE_PC:
@ -516,11 +527,12 @@ static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
case R_HINT:
case R_NONE:
case R_TLSDESC_CALL:
case R_TLSLD_HINT:
llvm_unreachable("cannot relocate hint relocs");
case R_MIPS_GOTREL:
return Sym.getVA(A) - InX::MipsGot->getGp();
return Sym.getVA(A) - InX::MipsGot->getGp(File);
case R_MIPS_GOT_GP:
return InX::MipsGot->getGp() + A;
return InX::MipsGot->getGp(File) + A;
case R_MIPS_GOT_GP_PC: {
// R_MIPS_LO16 expression has R_MIPS_GOT_GP_PC type iif the target
// is _gp_disp symbol. In that case we should use the following
@ -529,7 +541,7 @@ static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
// microMIPS variants of these relocations use slightly different
// expressions: AHL + GP - P + 3 for %lo() and AHL + GP - P - 1 for %hi()
// to correctly handle less-sugnificant bit of the microMIPS symbol.
uint64_t V = InX::MipsGot->getGp() + A - P;
uint64_t V = InX::MipsGot->getGp(File) + A - P;
if (Type == R_MIPS_LO16 || Type == R_MICROMIPS_LO16)
V += 4;
if (Type == R_MICROMIPS_LO16 || Type == R_MICROMIPS_HI16)
@ -540,21 +552,23 @@ static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
// If relocation against MIPS local symbol requires GOT entry, this entry
// should be initialized by 'page address'. This address is high 16-bits
// of sum the symbol's value and the addend.
return InX::MipsGot->getVA() + InX::MipsGot->getPageEntryOffset(Sym, A) -
InX::MipsGot->getGp();
return InX::MipsGot->getVA() +
InX::MipsGot->getPageEntryOffset(File, Sym, A) -
InX::MipsGot->getGp(File);
case R_MIPS_GOT_OFF:
case R_MIPS_GOT_OFF32:
// In case of MIPS if a GOT relocation has non-zero addend this addend
// should be applied to the GOT entry content not to the GOT entry offset.
// That is why we use separate expression type.
return InX::MipsGot->getVA() + InX::MipsGot->getSymEntryOffset(Sym, A) -
InX::MipsGot->getGp();
return InX::MipsGot->getVA() +
InX::MipsGot->getSymEntryOffset(File, Sym, A) -
InX::MipsGot->getGp(File);
case R_MIPS_TLSGD:
return InX::MipsGot->getVA() + InX::MipsGot->getTlsOffset() +
InX::MipsGot->getGlobalDynOffset(Sym) - InX::MipsGot->getGp();
return InX::MipsGot->getVA() + InX::MipsGot->getGlobalDynOffset(File, Sym) -
InX::MipsGot->getGp(File);
case R_MIPS_TLSLD:
return InX::MipsGot->getVA() + InX::MipsGot->getTlsOffset() +
InX::MipsGot->getTlsIndexOff() - InX::MipsGot->getGp();
return InX::MipsGot->getVA() + InX::MipsGot->getTlsIndexOffset(File) -
InX::MipsGot->getGp(File);
case R_PAGE_PC:
case R_PLT_PAGE_PC: {
uint64_t Dest;
@ -583,25 +597,27 @@ static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
case R_PLT:
return Sym.getPltVA() + A;
case R_PLT_PC:
case R_PPC_PLT_OPD:
case R_PPC_CALL_PLT:
return Sym.getPltVA() + A - P;
case R_PPC_OPD: {
case R_PPC_CALL: {
uint64_t SymVA = Sym.getVA(A);
// If we have an undefined weak symbol, we might get here with a symbol
// address of zero. That could overflow, but the code must be unreachable,
// so don't bother doing anything at all.
if (!SymVA)
return 0;
if (Out::Opd) {
// If this is a local call, and we currently have the address of a
// function-descriptor, get the underlying code address instead.
uint64_t OpdStart = Out::Opd->Addr;
uint64_t OpdEnd = OpdStart + Out::Opd->Size;
bool InOpd = OpdStart <= SymVA && SymVA < OpdEnd;
if (InOpd)
SymVA = read64be(&Out::OpdBuf[SymVA - OpdStart]);
}
return SymVA - P;
// PPC64 V2 ABI describes two entry points to a function. The global entry
// point sets up the TOC base pointer. When calling a local function, the
// call should branch to the local entry point rather than the global entry
// point. Section 3.4.1 describes using the 3 most significant bits of the
// st_other field to find out how many instructions there are between the
// local and global entry point.
uint8_t StOther = (Sym.StOther >> 5) & 7;
if (StOther == 0 || StOther == 1)
return SymVA - P;
return SymVA - P + (1LL << StOther);
}
case R_PPC_TOC:
return getPPC64TocBase() + A;
@ -618,25 +634,44 @@ static uint64_t getRelocTargetVA(RelType Type, int64_t A, uint64_t P,
// statically to zero.
if (Sym.isTls() && Sym.isUndefWeak())
return 0;
if (Target->TcbSize)
// For TLS variant 1 the TCB is a fixed size, whereas for TLS variant 2 the
// TCB is on unspecified size and content. Targets that implement variant 1
// should set TcbSize.
if (Target->TcbSize) {
// PPC64 V2 ABI has the thread pointer offset into the middle of the TLS
// storage area by TlsTpOffset for efficient addressing TCB and up to
// 4KB – 8 B of other thread library information (placed before the TCB).
// Subtracting this offset will get the address of the first TLS block.
if (Target->TlsTpOffset)
return Sym.getVA(A) - Target->TlsTpOffset;
// If thread pointer is not offset into the middle, the first thing in the
// TLS storage area is the TCB. Add the TcbSize to get the address of the
// first TLS block.
return Sym.getVA(A) + alignTo(Target->TcbSize, Out::TlsPhdr->p_align);
}
return Sym.getVA(A) - Out::TlsPhdr->p_memsz;
case R_RELAX_TLS_GD_TO_LE_NEG:
case R_NEG_TLS:
return Out::TlsPhdr->p_memsz - Sym.getVA(A);
case R_SIZE:
return A; // Sym.getSize was already folded into the addend.
return Sym.getSize() + A;
case R_TLSDESC:
return InX::Got->getGlobalDynAddr(Sym) + A;
case R_TLSDESC_PAGE:
return getAArch64Page(InX::Got->getGlobalDynAddr(Sym) + A) -
getAArch64Page(P);
case R_TLSGD:
case R_TLSGD_GOT:
return InX::Got->getGlobalDynOffset(Sym) + A;
case R_TLSGD_GOT_FROM_END:
return InX::Got->getGlobalDynOffset(Sym) + A - InX::Got->getSize();
case R_TLSGD_PC:
return InX::Got->getGlobalDynAddr(Sym) + A - P;
case R_TLSLD:
case R_TLSLD_GOT_FROM_END:
return InX::Got->getTlsIndexOff() + A - InX::Got->getSize();
case R_TLSLD_GOT:
return InX::Got->getTlsIndexOff() + A;
case R_TLSLD_PC:
return InX::Got->getTlsIndexVA() + A - P;
}
@ -656,6 +691,14 @@ void InputSection::relocateNonAlloc(uint8_t *Buf, ArrayRef<RelTy> Rels) {
for (const RelTy &Rel : Rels) {
RelType Type = Rel.getType(Config->IsMips64EL);
// GCC 8.0 or earlier have a bug that they emit R_386_GOTPC relocations
// against _GLOBAL_OFFSET_TABLE_ for .debug_info. The bug has been fixed
// in 2017 (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82630), but we
// need to keep this bug-compatible code for a while.
if (Config->EMachine == EM_386 && Type == R_386_GOTPC)
continue;
uint64_t Offset = getOffset(Rel.r_offset);
uint8_t *BufLoc = Buf + Offset;
int64_t Addend = getAddend<ELFT>(Rel);
@ -666,17 +709,27 @@ void InputSection::relocateNonAlloc(uint8_t *Buf, ArrayRef<RelTy> Rels) {
RelExpr Expr = Target->getRelExpr(Type, Sym, BufLoc);
if (Expr == R_NONE)
continue;
if (Expr != R_ABS) {
// GCC 8.0 or earlier have a bug that it emits R_386_GOTPC relocations
// against _GLOBAL_OFFSET_TABLE for .debug_info. The bug seems to have
// been fixed in 2017: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82630,
// but we need to keep this bug-compatible code for a while.
if (Config->EMachine == EM_386 && Type == R_386_GOTPC)
continue;
error(getLocation<ELFT>(Offset) + ": has non-ABS relocation " +
toString(Type) + " against symbol '" + toString(Sym) + "'");
return;
if (Expr != R_ABS) {
std::string Msg = getLocation<ELFT>(Offset) +
": has non-ABS relocation " + toString(Type) +
" against symbol '" + toString(Sym) + "'";
if (Expr != R_PC) {
error(Msg);
return;
}
// If the control reaches here, we found a PC-relative relocation in a
// non-ALLOC section. Since non-ALLOC section is not loaded into memory
// at runtime, the notion of PC-relative doesn't make sense here. So,
// this is a usage error. However, GNU linkers historically accept such
// relocations without any errors and relocate them as if they were at
// address 0. For bug-compatibilty, we accept them with warnings. We
// know Steel Bank Common Lisp as of 2018 have this bug.
warn(Msg);
Target->relocateOne(BufLoc, Type,
SignExtend64<Bits>(Sym.getVA(Addend - Offset)));
continue;
}
if (Sym.isTls() && !Out::TlsPhdr)
@ -705,6 +758,9 @@ static void relocateNonAllocForRelocatable(InputSection *Sec, uint8_t *Buf) {
template <class ELFT>
void InputSectionBase::relocate(uint8_t *Buf, uint8_t *BufEnd) {
if (Flags & SHF_EXECINSTR)
adjustSplitStackFunctionPrologues<ELFT>(Buf, BufEnd);
if (Flags & SHF_ALLOC) {
relocateAlloc(Buf, BufEnd);
return;
@ -724,14 +780,17 @@ void InputSectionBase::relocateAlloc(uint8_t *Buf, uint8_t *BufEnd) {
const unsigned Bits = Config->Wordsize * 8;
for (const Relocation &Rel : Relocations) {
uint64_t Offset = getOffset(Rel.Offset);
uint64_t Offset = Rel.Offset;
if (auto *Sec = dyn_cast<InputSection>(this))
Offset += Sec->OutSecOff;
uint8_t *BufLoc = Buf + Offset;
RelType Type = Rel.Type;
uint64_t AddrLoc = getOutputSection()->Addr + Offset;
RelExpr Expr = Rel.Expr;
uint64_t TargetVA = SignExtend64(
getRelocTargetVA(Type, Rel.Addend, AddrLoc, *Rel.Sym, Expr), Bits);
getRelocTargetVA(File, Type, Rel.Addend, AddrLoc, *Rel.Sym, Expr),
Bits);
switch (Expr) {
case R_RELAX_GOT_PC:
@ -742,6 +801,7 @@ void InputSectionBase::relocateAlloc(uint8_t *Buf, uint8_t *BufEnd) {
Target->relaxTlsIeToLe(BufLoc, Type, TargetVA);
break;
case R_RELAX_TLS_LD_TO_LE:
case R_RELAX_TLS_LD_TO_LE_ABS:
Target->relaxTlsLdToLe(BufLoc, Type, TargetVA);
break;
case R_RELAX_TLS_GD_TO_LE:
@ -750,15 +810,28 @@ void InputSectionBase::relocateAlloc(uint8_t *Buf, uint8_t *BufEnd) {
break;
case R_RELAX_TLS_GD_TO_IE:
case R_RELAX_TLS_GD_TO_IE_ABS:
case R_RELAX_TLS_GD_TO_IE_GOT_OFF:
case R_RELAX_TLS_GD_TO_IE_PAGE_PC:
case R_RELAX_TLS_GD_TO_IE_END:
Target->relaxTlsGdToIe(BufLoc, Type, TargetVA);
break;
case R_PPC_PLT_OPD:
case R_PPC_CALL:
// If this is a call to __tls_get_addr, it may be part of a TLS
// sequence that has been relaxed and turned into a nop. In this
// case, we don't want to handle it as a call.
if (read32(BufLoc) == 0x60000000) // nop
break;
// Patch a nop (0x60000000) to a ld.
if (BufLoc + 8 <= BufEnd && read32be(BufLoc + 4) == 0x60000000)
write32be(BufLoc + 4, 0xe8410028); // ld %r2, 40(%r1)
LLVM_FALLTHROUGH;
if (Rel.Sym->NeedsTocRestore) {
if (BufLoc + 8 > BufEnd || read32(BufLoc + 4) != 0x60000000) {
error(getErrorLocation(BufLoc) + "call lacks nop, can't restore toc");
break;
}
write32(BufLoc + 4, 0xe8410018); // ld %r2, 24(%r1)
}
Target->relocateOne(BufLoc, Type, TargetVA);
break;
default:
Target->relocateOne(BufLoc, Type, TargetVA);
break;
@ -766,6 +839,103 @@ void InputSectionBase::relocateAlloc(uint8_t *Buf, uint8_t *BufEnd) {
}
}
// For each function-defining prologue, find any calls to __morestack,
// and replace them with calls to __morestack_non_split.
static void switchMorestackCallsToMorestackNonSplit(
llvm::DenseSet<Defined *>& Prologues,
std::vector<Relocation *>& MorestackCalls) {
// If the target adjusted a function's prologue, all calls to
// __morestack inside that function should be switched to
// __morestack_non_split.
Symbol *MoreStackNonSplit = Symtab->find("__morestack_non_split");
// Sort both collections to compare addresses efficiently.
llvm::sort(MorestackCalls.begin(), MorestackCalls.end(),
[](const Relocation *L, const Relocation *R) {
return L->Offset < R->Offset;
});
std::vector<Defined *> Functions(Prologues.begin(), Prologues.end());
llvm::sort(
Functions.begin(), Functions.end(),
[](const Defined *L, const Defined *R) { return L->Value < R->Value; });
auto It = MorestackCalls.begin();
for (Defined *F : Functions) {
// Find the first call to __morestack within the function.
while (It != MorestackCalls.end() && (*It)->Offset < F->Value)
++It;
// Adjust all calls inside the function.
while (It != MorestackCalls.end() && (*It)->Offset < F->Value + F->Size) {
(*It)->Sym = MoreStackNonSplit;
++It;
}
}
}
static bool
enclosingPrologueAdjusted(uint64_t Offset,
const llvm::DenseSet<Defined *> &Prologues) {
for (Defined *F : Prologues)
if (F->Value <= Offset && Offset < F->Value + F->Size)
return true;
return false;
}
// If a function compiled for split stack calls a function not
// compiled for split stack, then the caller needs its prologue
// adjusted to ensure that the called function will have enough stack
// available. Find those functions, and adjust their prologues.
template <class ELFT>
void InputSectionBase::adjustSplitStackFunctionPrologues(uint8_t *Buf,
uint8_t *End) {
if (!getFile<ELFT>()->SplitStack)
return;
llvm::DenseSet<Defined *> AdjustedPrologues;
std::vector<Relocation *> MorestackCalls;
for (Relocation &Rel : Relocations) {
// Local symbols can't possibly be cross-calls, and should have been
// resolved long before this line.
if (Rel.Sym->isLocal())
continue;
Defined *D = dyn_cast<Defined>(Rel.Sym);
// A reference to an undefined symbol was an error, and should not
// have gotten to this point.
if (!D)
continue;
// Ignore calls into the split-stack api.
if (D->getName().startswith("__morestack")) {
if (D->getName().equals("__morestack"))
MorestackCalls.push_back(&Rel);
continue;
}
// A relocation to non-function isn't relevant. Sometimes
// __morestack is not marked as a function, so this check comes
// after the name check.
if (D->Type != STT_FUNC)
continue;
if (enclosingPrologueAdjusted(Rel.Offset, AdjustedPrologues))
continue;
if (Defined *F = getEnclosingFunction<ELFT>(Rel.Offset)) {
if (Target->adjustPrologueForCrossSplitStack(Buf + F->Value, End)) {
AdjustedPrologues.insert(F);
continue;
}
}
if (!getFile<ELFT>()->SomeNoSplitStack)
error("function call at " + getErrorLocation(Buf + Rel.Offset) +
"crosses a split-stack boundary, but unable " +
"to adjust the enclosing function's prologue");
}
switchMorestackCallsToMorestackNonSplit(AdjustedPrologues, MorestackCalls);
}
template <class ELFT> void InputSection::writeTo(uint8_t *Buf) {
if (Type == SHT_NOBITS)
return;
@ -837,10 +1007,6 @@ static unsigned getReloc(IntTy Begin, IntTy Size, const ArrayRef<RelTy> &Rels,
// .eh_frame is a sequence of CIE or FDE records.
// This function splits an input section into records and returns them.
template <class ELFT> void EhInputSection::split() {
// Early exit if already split.
if (!Pieces.empty())
return;
if (AreRelocsRela)
split<ELFT>(relas<ELFT>());
else
@ -935,15 +1101,9 @@ void MergeInputSection::splitIntoPieces() {
else
splitNonStrings(Data, Entsize);
if (Config->GcSections && (Flags & SHF_ALLOC))
for (uint64_t Off : LiveOffsets)
getSectionPiece(Off)->Live = true;
}
// Do binary search to get a section piece at a given input offset.
SectionPiece *MergeInputSection::getSectionPiece(uint64_t Offset) {
auto *This = static_cast<const MergeInputSection *>(this);
return const_cast<SectionPiece *>(This->getSectionPiece(Offset));
OffsetMap.reserve(Pieces.size());
for (size_t I = 0, E = Pieces.size(); I != E; ++I)
OffsetMap[Pieces[I].InputOff] = I;
}
template <class It, class T, class Compare>
@ -959,32 +1119,34 @@ static It fastUpperBound(It First, It Last, const T &Value, Compare Comp) {
return Comp(Value, *First) ? First : First + 1;
}
const SectionPiece *MergeInputSection::getSectionPiece(uint64_t Offset) const {
if (Data.size() <= Offset)
fatal(toString(this) + ": entry is past the end of the section");
// Do binary search to get a section piece at a given input offset.
static SectionPiece *findSectionPiece(MergeInputSection *Sec, uint64_t Offset) {
if (Sec->Data.size() <= Offset)
fatal(toString(Sec) + ": entry is past the end of the section");
// Find the element this offset points to.
auto I = fastUpperBound(
Pieces.begin(), Pieces.end(), Offset,
Sec->Pieces.begin(), Sec->Pieces.end(), Offset,
[](const uint64_t &A, const SectionPiece &B) { return A < B.InputOff; });
--I;
return &*I;
}
SectionPiece *MergeInputSection::getSectionPiece(uint64_t Offset) {
// Find a piece starting at a given offset.
auto It = OffsetMap.find(Offset);
if (It != OffsetMap.end())
return &Pieces[It->second];
// If Offset is not at beginning of a section piece, it is not in the map.
// In that case we need to search from the original section piece vector.
return findSectionPiece(this, Offset);
}
// Returns the offset in an output section for a given input offset.
// Because contents of a mergeable section is not contiguous in output,
// it is not just an addition to a base output offset.
uint64_t MergeInputSection::getOffset(uint64_t Offset) const {
if (!Live)
return 0;
// Initialize OffsetMap lazily.
llvm::call_once(InitOffsetMap, [&] {
OffsetMap.reserve(Pieces.size());
for (size_t I = 0; I < Pieces.size(); ++I)
OffsetMap[Pieces[I].InputOff] = I;
});
uint64_t MergeInputSection::getParentOffset(uint64_t Offset) const {
// Find a string starting at a given offset.
auto It = OffsetMap.find(Offset);
if (It != OffsetMap.end())
@ -992,10 +1154,8 @@ uint64_t MergeInputSection::getOffset(uint64_t Offset) const {
// If Offset is not at beginning of a section piece, it is not in the map.
// In that case we need to search from the original section piece vector.
const SectionPiece &Piece = *getSectionPiece(Offset);
if (!Piece.Live)
return 0;
const SectionPiece &Piece =
*findSectionPiece(const_cast<MergeInputSection *>(this), Offset);
uint64_t Addend = Offset - Piece.InputOff;
return Piece.OutputOff + Addend;
}

65
contrib/llvm/tools/lld/ELF/InputSection.h
View File

@ -18,8 +18,6 @@
#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/TinyPtrVector.h"
#include "llvm/Object/ELF.h"
#include "llvm/Support/Threading.h"
#include <mutex>
namespace lld {
namespace elf {
@ -63,6 +61,9 @@ class SectionBase {
unsigned Bss : 1;
// Set for sections that should not be folded by ICF.
unsigned KeepUnique : 1;
// These corresponds to the fields in Elf_Shdr.
uint32_t Alignment;
uint64_t Flags;
@ -80,13 +81,15 @@ class SectionBase {
// section.
uint64_t getOffset(uint64_t Offset) const;
uint64_t getVA(uint64_t Offset = 0) const;
protected:
SectionBase(Kind SectionKind, StringRef Name, uint64_t Flags,
uint64_t Entsize, uint64_t Alignment, uint32_t Type,
uint32_t Info, uint32_t Link)
: Name(Name), Repl(this), SectionKind(SectionKind), Live(false),
Bss(false), Alignment(Alignment), Flags(Flags), Entsize(Entsize),
Type(Type), Link(Link), Info(Info) {}
Bss(false), KeepUnique(false), Alignment(Alignment), Flags(Flags),
Entsize(Entsize), Type(Type), Link(Link), Info(Info) {}
};
// This corresponds to a section of an input file.
@ -103,7 +106,7 @@ class InputSectionBase : public SectionBase {
static bool classof(const SectionBase *S) { return S->kind() != Output; }
// The file which contains this section. It's dynamic type is always
// The file which contains this section. Its dynamic type is always
// ObjFile<ELFT>, but in order to avoid ELFT, we use InputFile as
// its static type.
InputFile *File;
@ -161,10 +164,15 @@ class InputSectionBase : public SectionBase {
InputSection *getLinkOrderDep() const;
// Get the function symbol that encloses this offset from within the
// section.
template <class ELFT>
Defined *getEnclosingFunction(uint64_t Offset);
// Compilers emit zlib-compressed debug sections if the -gz option
// is given. This function checks if this section is compressed, and
// if so, decompress in memory.
void maybeUncompress();
void maybeDecompress();
// Returns a source location string. Used to construct an error message.
template <class ELFT> std::string getLocation(uint64_t Offset);
@ -182,6 +190,15 @@ class InputSectionBase : public SectionBase {
// This vector contains such "cooked" relocations.
std::vector<Relocation> Relocations;
// A function compiled with -fsplit-stack calling a function
// compiled without -fsplit-stack needs its prologue adjusted. Find
// such functions and adjust their prologues. This is very similar
// to relocation. See https://gcc.gnu.org/wiki/SplitStacks for more
// information.
template <typename ELFT>
void adjustSplitStackFunctionPrologues(uint8_t *Buf, uint8_t *End);
template <typename T> llvm::ArrayRef<T> getDataAs() const {
size_t S = Data.size();
assert(S % sizeof(T) == 0);
@ -189,9 +206,9 @@ class InputSectionBase : public SectionBase {
}
private:
// A pointer that owns uncompressed data if a section is compressed by zlib.
// A pointer that owns decompressed data if a section is compressed by zlib.
// Since the feature is not used often, this is usually a nullptr.
std::unique_ptr<char[]> UncompressBuf;
std::unique_ptr<char[]> DecompressBuf;
};
// SectionPiece represents a piece of splittable section contents.
@ -200,7 +217,7 @@ class InputSectionBase : public SectionBase {
// be found by looking at the next one).
struct SectionPiece {
SectionPiece(size_t Off, uint32_t Hash, bool Live)
: InputOff(Off), Hash(Hash), OutputOff(-1),
: InputOff(Off), Hash(Hash), OutputOff(0),
Live(Live || !Config->GcSections) {}
uint32_t InputOff;
@ -223,19 +240,14 @@ class MergeInputSection : public InputSectionBase {
static bool classof(const SectionBase *S) { return S->kind() == Merge; }
void splitIntoPieces();
// Mark the piece at a given offset live. Used by GC.
void markLiveAt(uint64_t Offset) {
if (this->Flags & llvm::ELF::SHF_ALLOC)
LiveOffsets.insert(Offset);
}
// Translate an offset in the input section to an offset
// in the output section.
uint64_t getOffset(uint64_t Offset) const;
// Translate an offset in the input section to an offset in the parent
// MergeSyntheticSection.
uint64_t getParentOffset(uint64_t Offset) const;
// Splittable sections are handled as a sequence of data
// rather than a single large blob of data.
std::vector<SectionPiece> Pieces;
llvm::DenseMap<uint32_t, uint32_t> OffsetMap;
// Returns I'th piece's data. This function is very hot when
// string merging is enabled, so we want to inline.
@ -249,18 +261,15 @@ class MergeInputSection : public InputSectionBase {
// Returns the SectionPiece at a given input section offset.
SectionPiece *getSectionPiece(uint64_t Offset);
const SectionPiece *getSectionPiece(uint64_t Offset) const;
const SectionPiece *getSectionPiece(uint64_t Offset) const {
return const_cast<MergeInputSection *>(this)->getSectionPiece(Offset);
}
SyntheticSection *getParent() const;
private:
void splitStrings(ArrayRef<uint8_t> A, size_t Size);
void splitNonStrings(ArrayRef<uint8_t> A, size_t Size);
mutable llvm::DenseMap<uint32_t, uint32_t> OffsetMap;
mutable llvm::once_flag InitOffsetMap;
llvm::DenseSet<uint64_t> LiveOffsets;
};
struct EhSectionPiece {
@ -310,6 +319,8 @@ class InputSection : public InputSectionBase {
// beginning of the output section.
template <class ELFT> void writeTo(uint8_t *Buf);
uint64_t getOffset(uint64_t Offset) const { return OutSecOff + Offset; }
OutputSection *getParent() const;
// This variable has two usages. Initially, it represents an index in the
@ -320,7 +331,7 @@ class InputSection : public InputSectionBase {
static bool classof(const SectionBase *S);
InputSectionBase *getRelocatedSection();
InputSectionBase *getRelocatedSection() const;
template <class ELFT, class RelTy>
void relocateNonAlloc(uint8_t *Buf, llvm::ArrayRef<RelTy> Rels);
@ -342,10 +353,6 @@ class InputSection : public InputSectionBase {
// The list of all input sections.
extern std::vector<InputSectionBase *> InputSections;
// Builds section order for handling --symbol-ordering-file.
llvm::DenseMap<SectionBase *, int> buildSectionOrder();
} // namespace elf
std::string toString(const elf::InputSectionBase *);

218
contrib/llvm/tools/lld/ELF/LTO.cpp
View File

@ -20,6 +20,8 @@
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/Twine.h"
#include "llvm/BinaryFormat/ELF.h"
#include "llvm/Bitcode/BitcodeReader.h"
#include "llvm/Bitcode/BitcodeWriter.h"
#include "llvm/IR/DiagnosticPrinter.h"
#include "llvm/LTO/Caching.h"
#include "llvm/LTO/Config.h"
@ -29,7 +31,6 @@
#include "llvm/Support/Error.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
#include <cstddef>
#include <memory>
@ -44,70 +45,92 @@ using namespace llvm::ELF;
using namespace lld;
using namespace lld::elf;
// This is for use when debugging LTO.
static void saveBuffer(StringRef Buffer, const Twine &Path) {
// Creates an empty file to store a list of object files for final
// linking of distributed ThinLTO.
static std::unique_ptr<raw_fd_ostream> openFile(StringRef File) {
std::error_code EC;
raw_fd_ostream OS(Path.str(), EC, sys::fs::OpenFlags::F_None);
if (EC)
error("cannot create " + Path + ": " + EC.message());
OS << Buffer;
auto Ret =
llvm::make_unique<raw_fd_ostream>(File, EC, sys::fs::OpenFlags::F_None);
if (EC) {
error("cannot open " + File + ": " + EC.message());
return nullptr;
}
return Ret;
}
static void diagnosticHandler(const DiagnosticInfo &DI) {
SmallString<128> ErrStorage;
raw_svector_ostream OS(ErrStorage);
DiagnosticPrinterRawOStream DP(OS);
DI.print(DP);
warn(ErrStorage);
static std::string getThinLTOOutputFile(StringRef ModulePath) {
return lto::getThinLTOOutputFile(ModulePath,
Config->ThinLTOPrefixReplace.first,
Config->ThinLTOPrefixReplace.second);
}
static void checkError(Error E) {
handleAllErrors(std::move(E),
[&](ErrorInfoBase &EIB) { error(EIB.message()); });
}
static std::unique_ptr<lto::LTO> createLTO() {
lto::Config Conf;
static lto::Config createConfig() {
lto::Config C;
// LLD supports the new relocations.
Conf.Options = InitTargetOptionsFromCodeGenFlags();
Conf.Options.RelaxELFRelocations = true;
C.Options = InitTargetOptionsFromCodeGenFlags();
C.Options.RelaxELFRelocations = true;
// Always emit a section per function/datum with LTO.
Conf.Options.FunctionSections = true;
Conf.Options.DataSections = true;
C.Options.FunctionSections = true;
C.Options.DataSections = true;
if (Config->Relocatable)
Conf.RelocModel = None;
C.RelocModel = None;
else if (Config->Pic)
Conf.RelocModel = Reloc::PIC_;
C.RelocModel = Reloc::PIC_;
else
Conf.RelocModel = Reloc::Static;
Conf.CodeModel = GetCodeModelFromCMModel();
Conf.DisableVerify = Config->DisableVerify;
Conf.DiagHandler = diagnosticHandler;
Conf.OptLevel = Config->LTOO;
C.RelocModel = Reloc::Static;
C.CodeModel = GetCodeModelFromCMModel();
C.DisableVerify = Config->DisableVerify;
C.DiagHandler = diagnosticHandler;
C.OptLevel = Config->LTOO;
C.CPU = GetCPUStr();
// Set up a custom pipeline if we've been asked to.
Conf.OptPipeline = Config->LTONewPmPasses;
Conf.AAPipeline = Config->LTOAAPipeline;
C.OptPipeline = Config->LTONewPmPasses;
C.AAPipeline = Config->LTOAAPipeline;
// Set up optimization remarks if we've been asked to.
Conf.RemarksFilename = Config->OptRemarksFilename;
Conf.RemarksWithHotness = Config->OptRemarksWithHotness;
C.RemarksFilename = Config->OptRemarksFilename;
C.RemarksWithHotness = Config->OptRemarksWithHotness;
C.SampleProfile = Config->LTOSampleProfile;
C.UseNewPM = Config->LTONewPassManager;
C.DebugPassManager = Config->LTODebugPassManager;
C.DwoDir = Config->DwoDir;
if (Config->SaveTemps)
checkError(Conf.addSaveTemps(std::string(Config->OutputFile) + ".",
/*UseInputModulePath*/ true));
lto::ThinBackend Backend;
if (Config->ThinLTOJobs != -1u)
Backend = lto::createInProcessThinBackend(Config->ThinLTOJobs);
return llvm::make_unique<lto::LTO>(std::move(Conf), Backend,
Config->LTOPartitions);
checkError(C.addSaveTemps(Config->OutputFile.str() + ".",
/*UseInputModulePath*/ true));
return C;
}
BitcodeCompiler::BitcodeCompiler() : LTOObj(createLTO()) {
BitcodeCompiler::BitcodeCompiler() {
// Initialize LTOObj.
lto::ThinBackend Backend;
if (Config->ThinLTOIndexOnly) {
StringRef Path = Config->ThinLTOIndexOnlyArg;
if (!Path.empty())
IndexFile = openFile(Path);
auto OnIndexWrite = [&](const std::string &Identifier) {
ObjectToIndexFileState[Identifier] = true;
};
Backend = lto::createWriteIndexesThinBackend(
Config->ThinLTOPrefixReplace.first, Config->ThinLTOPrefixReplace.second,
Config->ThinLTOEmitImportsFiles, IndexFile.get(), OnIndexWrite);
} else if (Config->ThinLTOJobs != -1U) {
Backend = lto::createInProcessThinBackend(Config->ThinLTOJobs);
}
LTOObj = llvm::make_unique<lto::LTO>(createConfig(), Backend,
Config->LTOPartitions);
// Initialize UsedStartStop.
for (Symbol *Sym : Symtab->getSymbols()) {
StringRef Name = Sym->getName();
for (StringRef Prefix : {"__start_", "__stop_"})
@ -125,20 +148,20 @@ static void undefine(Symbol *S) {
void BitcodeCompiler::add(BitcodeFile &F) {
lto::InputFile &Obj = *F.Obj;
unsigned SymNum = 0;
std::vector<Symbol *> Syms = F.getSymbols();
bool IsExec = !Config->Shared && !Config->Relocatable;
if (Config->ThinLTOIndexOnly)
ObjectToIndexFileState.insert({Obj.getName(), false});
ArrayRef<Symbol *> Syms = F.getSymbols();
ArrayRef<lto::InputFile::Symbol> ObjSyms = Obj.symbols();
std::vector<lto::SymbolResolution> Resols(Syms.size());
DenseSet<StringRef> ScriptSymbols;
for (BaseCommand *Base : Script->SectionCommands)
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base))
ScriptSymbols.insert(Cmd->Name);
// Provide a resolution to the LTO API for each symbol.
for (const lto::InputFile::Symbol &ObjSym : Obj.symbols()) {
Symbol *Sym = Syms[SymNum];
lto::SymbolResolution &R = Resols[SymNum];
++SymNum;
for (size_t I = 0, E = Syms.size(); I != E; ++I) {
Symbol *Sym = Syms[I];
const lto::InputFile::Symbol &ObjSym = ObjSyms[I];
lto::SymbolResolution &R = Resols[I];
// Ideally we shouldn't check for SF_Undefined but currently IRObjectFile
// reports two symbols for module ASM defined. Without this check, lld
@ -156,25 +179,46 @@ void BitcodeCompiler::add(BitcodeFile &F) {
R.VisibleToRegularObj = Config->Relocatable || Sym->IsUsedInRegularObj ||
(R.Prevailing && Sym->includeInDynsym()) ||
UsedStartStop.count(ObjSym.getSectionName());
const auto *DR = dyn_cast<Defined>(Sym);
R.FinalDefinitionInLinkageUnit =
(IsExec || Sym->Visibility != STV_DEFAULT) && DR &&
// Skip absolute symbols from ELF objects, otherwise PC-rel relocations
// will be generated by for them, triggering linker errors.
// Symbol section is always null for bitcode symbols, hence the check
// for isElf(). Skip linker script defined symbols as well: they have
// no File defined.
!(DR->Section == nullptr && (!Sym->File || Sym->File->isElf()));
if (R.Prevailing)
undefine(Sym);
// We tell LTO to not apply interprocedural optimization for following
// symbols because otherwise LTO would inline them while their values are
// still not final:
// 1) Aliased (with --defsym) or wrapped (with --wrap) symbols.
// 2) Symbols redefined in linker script.
R.LinkerRedefined = !Sym->CanInline || ScriptSymbols.count(Sym->getName());
// We tell LTO to not apply interprocedural optimization for wrapped
// (with --wrap) symbols because otherwise LTO would inline them while
// their values are still not final.
R.LinkerRedefined = !Sym->CanInline;
}
checkError(LTOObj->add(std::move(F.Obj), Resols));
}
static void createEmptyIndex(StringRef ModulePath) {
std::string Path = replaceThinLTOSuffix(getThinLTOOutputFile(ModulePath));
std::unique_ptr<raw_fd_ostream> OS = openFile(Path + ".thinlto.bc");
if (!OS)
return;
ModuleSummaryIndex M(/*HaveGVs*/ false);
M.setSkipModuleByDistributedBackend();
WriteIndexToFile(M, *OS);
if (Config->ThinLTOEmitImportsFiles)
openFile(Path + ".imports");
}
// Merge all the bitcode files we have seen, codegen the result
// and return the resulting ObjectFile(s).
std::vector<InputFile *> BitcodeCompiler::compile() {
std::vector<InputFile *> Ret;
unsigned MaxTasks = LTOObj->getMaxTasks();
Buff.resize(MaxTasks);
Buf.resize(MaxTasks);
Files.resize(MaxTasks);
// The --thinlto-cache-dir option specifies the path to a directory in which
@ -184,35 +228,67 @@ std::vector<InputFile *> BitcodeCompiler::compile() {
if (!Config->ThinLTOCacheDir.empty())
Cache = check(
lto::localCache(Config->ThinLTOCacheDir,
[&](size_t Task, std::unique_ptr<MemoryBuffer> MB,
StringRef Path) { Files[Task] = std::move(MB); }));
[&](size_t Task, std::unique_ptr<MemoryBuffer> MB) {
Files[Task] = std::move(MB);
}));
checkError(LTOObj->run(
[&](size_t Task) {
return llvm::make_unique<lto::NativeObjectStream>(
llvm::make_unique<raw_svector_ostream>(Buff[Task]));
llvm::make_unique<raw_svector_ostream>(Buf[Task]));
},
Cache));
// Emit empty index files for non-indexed files
if (Config->ThinLTOIndexOnly) {
for (auto &Identifier : ObjectToIndexFileState)
if (!Identifier.getValue()) {
std::string Path = getThinLTOOutputFile(Identifier.getKey());
openFile(Path + ".thinlto.bc");
if (Config->ThinLTOEmitImportsFiles)
openFile(Path + ".imports");
}
}
// If LazyObjFile has not been added to link, emit empty index files.
// This is needed because this is what GNU gold plugin does and we have a
// distributed build system that depends on that behavior.
if (Config->ThinLTOIndexOnly) {
for (LazyObjFile *F : LazyObjFiles)
if (!F->AddedToLink && isBitcode(F->MB))
createEmptyIndex(F->getName());
if (!Config->LTOObjPath.empty())
saveBuffer(Buf[0], Config->LTOObjPath);
// ThinLTO with index only option is required to generate only the index
// files. After that, we exit from linker and ThinLTO backend runs in a
// distributed environment.
if (IndexFile)
IndexFile->close();
return {};
}
if (!Config->ThinLTOCacheDir.empty())
pruneCache(Config->ThinLTOCacheDir, Config->ThinLTOCachePolicy);
std::vector<InputFile *> Ret;
for (unsigned I = 0; I != MaxTasks; ++I) {
if (Buff[I].empty())
if (Buf[I].empty())
continue;
if (Config->SaveTemps) {
if (I == 0)
saveBuffer(Buff[I], Config->OutputFile + ".lto.o");
saveBuffer(Buf[I], Config->OutputFile + ".lto.o");
else
saveBuffer(Buff[I], Config->OutputFile + Twine(I) + ".lto.o");
saveBuffer(Buf[I], Config->OutputFile + Twine(I) + ".lto.o");
}
InputFile *Obj = createObjectFile(MemoryBufferRef(Buff[I], "lto.tmp"));
InputFile *Obj = createObjectFile(MemoryBufferRef(Buf[I], "lto.tmp"));
Ret.push_back(Obj);
}
for (std::unique_ptr<MemoryBuffer> &File : Files)
if (File)
Ret.push_back(createObjectFile(*File));
return Ret;
}

6
contrib/llvm/tools/lld/ELF/LTO.h
View File

@ -24,6 +24,7 @@
#include "lld/Common/LLVM.h"
#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/Support/raw_ostream.h"
#include <memory>
#include <vector>
@ -38,6 +39,7 @@ namespace elf {
class BitcodeFile;
class InputFile;
class LazyObjFile;
class BitcodeCompiler {
public:
@ -49,9 +51,11 @@ class BitcodeCompiler {
private:
std::unique_ptr<llvm::lto::LTO> LTOObj;
std::vector<SmallString<0>> Buff;
std::vector<SmallString<0>> Buf;
std::vector<std::unique_ptr<MemoryBuffer>> Files;
llvm::DenseSet<StringRef> UsedStartStop;
std::unique_ptr<llvm::raw_fd_ostream> IndexFile;
llvm::StringMap<bool> ObjectToIndexFileState;
};
} // namespace elf
} // namespace lld

373
contrib/llvm/tools/lld/ELF/LinkerScript.cpp
View File

@ -15,13 +15,13 @@
#include "Config.h"
#include "InputSection.h"
#include "OutputSections.h"
#include "Strings.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "SyntheticSections.h"
#include "Target.h"
#include "Writer.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Strings.h"
#include "lld/Common/Threads.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
@ -74,7 +74,7 @@ uint64_t ExprValue::getSectionOffset() const {
// If the alignment is trivial, we don't have to compute the full
// value to know the offset. This allows this function to succeed in
// cases where the output section is not yet known.
if (Alignment == 1)
if (Alignment == 1 && (!Sec || !Sec->getOutputSection()))
return Val;
return getValue() - getSecAddr();
}
@ -102,28 +102,67 @@ OutputSection *LinkerScript::getOrCreateOutputSection(StringRef Name) {
return CmdRef;
}
// Expands the memory region by the specified size.
static void expandMemoryRegion(MemoryRegion *MemRegion, uint64_t Size,
StringRef RegionName, StringRef SecName) {
MemRegion->CurPos += Size;
uint64_t NewSize = MemRegion->CurPos - MemRegion->Origin;
if (NewSize > MemRegion->Length)
error("section '" + SecName + "' will not fit in region '" + RegionName +
"': overflowed by " + Twine(NewSize - MemRegion->Length) + " bytes");
}
void LinkerScript::expandMemoryRegions(uint64_t Size) {
if (Ctx->MemRegion)
expandMemoryRegion(Ctx->MemRegion, Size, Ctx->MemRegion->Name,
Ctx->OutSec->Name);
if (Ctx->LMARegion)
expandMemoryRegion(Ctx->LMARegion, Size, Ctx->LMARegion->Name,
Ctx->OutSec->Name);
}
void LinkerScript::expandOutputSection(uint64_t Size) {
Ctx->OutSec->Size += Size;
expandMemoryRegions(Size);
}
void LinkerScript::setDot(Expr E, const Twine &Loc, bool InSec) {
uint64_t Val = E().getValue();
if (Val < Dot && InSec)
error(Loc + ": unable to move location counter backward for: " +
Ctx->OutSec->Name);
Dot = Val;
// Update to location counter means update to section size.
if (InSec)
Ctx->OutSec->Size = Dot - Ctx->OutSec->Addr;
expandOutputSection(Val - Dot);
else
expandMemoryRegions(Val - Dot);
Dot = Val;
}
// Used for handling linker symbol assignments, for both finalizing
// their values and doing early declarations. Returns true if symbol
// should be defined from linker script.
static bool shouldDefineSym(SymbolAssignment *Cmd) {
if (Cmd->Name == ".")
return false;
if (!Cmd->Provide)
return true;
// If a symbol was in PROVIDE(), we need to define it only
// when it is a referenced undefined symbol.
Symbol *B = Symtab->find(Cmd->Name);
if (B && !B->isDefined())
return true;
return false;
}
// This function is called from processSectionCommands,
// while we are fixing the output section layout.
void LinkerScript::addSymbol(SymbolAssignment *Cmd) {
if (Cmd->Name == ".")
return;
// If a symbol was in PROVIDE(), we need to define it only when
// it is a referenced undefined symbol.
Symbol *B = Symtab->find(Cmd->Name);
if (Cmd->Provide && (!B || B->isDefined()))
if (!shouldDefineSym(Cmd))
return;
// Define a symbol.
@ -153,6 +192,76 @@ void LinkerScript::addSymbol(SymbolAssignment *Cmd) {
Cmd->Sym = cast<Defined>(Sym);
}
// This function is called from LinkerScript::declareSymbols.
// It creates a placeholder symbol if needed.
static void declareSymbol(SymbolAssignment *Cmd) {
if (!shouldDefineSym(Cmd))
return;
// We can't calculate final value right now.
Symbol *Sym;
uint8_t Visibility = Cmd->Hidden ? STV_HIDDEN : STV_DEFAULT;
std::tie(Sym, std::ignore) = Symtab->insert(Cmd->Name, /*Type*/ 0, Visibility,
/*CanOmitFromDynSym*/ false,
/*File*/ nullptr);
replaceSymbol<Defined>(Sym, nullptr, Cmd->Name, STB_GLOBAL, Visibility,
STT_NOTYPE, 0, 0, nullptr);
Cmd->Sym = cast<Defined>(Sym);
Cmd->Provide = false;
}
// This method is used to handle INSERT AFTER statement. Here we rebuild
// the list of script commands to mix sections inserted into.
void LinkerScript::processInsertCommands() {
std::vector<BaseCommand *> V;
auto Insert = [&](std::vector<BaseCommand *> &From) {
V.insert(V.end(), From.begin(), From.end());
From.clear();
};
for (BaseCommand *Base : SectionCommands) {
if (auto *OS = dyn_cast<OutputSection>(Base)) {
Insert(InsertBeforeCommands[OS->Name]);
V.push_back(Base);
Insert(InsertAfterCommands[OS->Name]);
continue;
}
V.push_back(Base);
}
for (auto &Cmds : {InsertBeforeCommands, InsertAfterCommands})
for (const std::pair<StringRef, std::vector<BaseCommand *>> &P : Cmds)
if (!P.second.empty())
error("unable to INSERT AFTER/BEFORE " + P.first +
": section not defined");
SectionCommands = std::move(V);
}
// Symbols defined in script should not be inlined by LTO. At the same time
// we don't know their final values until late stages of link. Here we scan
// over symbol assignment commands and create placeholder symbols if needed.
void LinkerScript::declareSymbols() {
assert(!Ctx);
for (BaseCommand *Base : SectionCommands) {
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base)) {
declareSymbol(Cmd);
continue;
}
// If the output section directive has constraints,
// we can't say for sure if it is going to be included or not.
// Skip such sections for now. Improve the checks if we ever
// need symbols from that sections to be declared early.
auto *Sec = cast<OutputSection>(Base);
if (Sec->Constraint != ConstraintKind::NoConstraint)
continue;
for (BaseCommand *Base2 : Sec->SectionCommands)
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base2))
declareSymbol(Cmd);
}
}
// This function is called from assignAddresses, while we are
// fixing the output section addresses. This function is supposed
// to set the final value for a given symbol assignment.
@ -249,23 +358,11 @@ static void sortSections(MutableArrayRef<InputSection *> Vec,
// --sort-section is handled as an inner SORT command.
// 3. If one SORT command is given, and if it is SORT_NONE, don't sort.
// 4. If no SORT command is given, sort according to --sort-section.
// 5. If no SORT commands are given and --sort-section is not specified,
// apply sorting provided by --symbol-ordering-file if any exist.
static void sortInputSections(
MutableArrayRef<InputSection *> Vec, const SectionPattern &Pat,
const DenseMap<SectionBase *, int> &Order) {
static void sortInputSections(MutableArrayRef<InputSection *> Vec,
const SectionPattern &Pat) {
if (Pat.SortOuter == SortSectionPolicy::None)
return;
if (Pat.SortOuter == SortSectionPolicy::Default &&
Config->SortSection == SortSectionPolicy::Default) {
// If -symbol-ordering-file was given, sort accordingly.
// Usually, Order is empty.
if (!Order.empty())
sortByOrder(Vec, [&](InputSectionBase *S) { return Order.lookup(S); });
return;
}
if (Pat.SortInner == SortSectionPolicy::Default)
sortSections(Vec, Config->SortSection);
else
@ -275,8 +372,7 @@ static void sortInputSections(
// Compute and remember which sections the InputSectionDescription matches.
std::vector<InputSection *>
LinkerScript::computeInputSections(const InputSectionDescription *Cmd,
const DenseMap<SectionBase *, int> &Order) {
LinkerScript::computeInputSections(const InputSectionDescription *Cmd) {
std::vector<InputSection *> Ret;
// Collects all sections that satisfy constraints of Cmd.
@ -290,8 +386,11 @@ LinkerScript::computeInputSections(const InputSectionDescription *Cmd,
// For -emit-relocs we have to ignore entries like
// .rela.dyn : { *(.rela.data) }
// which are common because they are in the default bfd script.
if (Sec->Type == SHT_REL || Sec->Type == SHT_RELA)
continue;
// We do not ignore SHT_REL[A] linker-synthesized sections here because
// want to support scripts that do custom layout for them.
if (auto *IS = dyn_cast<InputSection>(Sec))
if (IS->getRelocatedSection())
continue;
std::string Filename = getFilename(Sec->File);
if (!Cmd->FilePat.match(Filename) ||
@ -307,7 +406,7 @@ LinkerScript::computeInputSections(const InputSectionDescription *Cmd,
}
sortInputSections(MutableArrayRef<InputSection *>(Ret).slice(SizeBefore),
Pat, Order);
Pat);
}
return Ret;
}
@ -315,22 +414,31 @@ LinkerScript::computeInputSections(const InputSectionDescription *Cmd,
void LinkerScript::discard(ArrayRef<InputSection *> V) {
for (InputSection *S : V) {
if (S == InX::ShStrTab || S == InX::Dynamic || S == InX::DynSymTab ||
S == InX::DynStrTab)
S == InX::DynStrTab || S == InX::RelaPlt || S == InX::RelaDyn ||
S == InX::RelrDyn)
error("discarding " + S->Name + " section is not allowed");
// You can discard .hash and .gnu.hash sections by linker scripts. Since
// they are synthesized sections, we need to handle them differently than
// other regular sections.
if (S == InX::GnuHashTab)
InX::GnuHashTab = nullptr;
if (S == InX::HashTab)
InX::HashTab = nullptr;
S->Assigned = false;
S->Live = false;
discard(S->DependentSections);
}
}
std::vector<InputSection *> LinkerScript::createInputSectionList(
OutputSection &OutCmd, const DenseMap<SectionBase *, int> &Order) {
std::vector<InputSection *>
LinkerScript::createInputSectionList(OutputSection &OutCmd) {
std::vector<InputSection *> Ret;
for (BaseCommand *Base : OutCmd.SectionCommands) {
if (auto *Cmd = dyn_cast<InputSectionDescription>(Base)) {
Cmd->Sections = computeInputSections(Cmd, Order);
Cmd->Sections = computeInputSections(Cmd);
Ret.insert(Ret.end(), Cmd->Sections.begin(), Cmd->Sections.end());
}
}
@ -359,7 +467,6 @@ void LinkerScript::processSectionCommands() {
Ctx->OutSec = Aether;
size_t I = 0;
DenseMap<SectionBase *, int> Order = buildSectionOrder();
// Add input sections to output sections.
for (BaseCommand *Base : SectionCommands) {
// Handle symbol assignments outside of any output section.
@ -369,12 +476,13 @@ void LinkerScript::processSectionCommands() {
}
if (auto *Sec = dyn_cast<OutputSection>(Base)) {
std::vector<InputSection *> V = createInputSectionList(*Sec, Order);
std::vector<InputSection *> V = createInputSectionList(*Sec);
// The output section name `/DISCARD/' is special.
// Any input section assigned to it is discarded.
if (Sec->Name == "/DISCARD/") {
discard(V);
Sec->SectionCommands.clear();
continue;
}
@ -414,6 +522,8 @@ void LinkerScript::processSectionCommands() {
Sec->SectionIndex = I++;
if (Sec->Noload)
Sec->Type = SHT_NOBITS;
if (Sec->NonAlloc)
Sec->Flags &= ~(uint64_t)SHF_ALLOC;
}
}
Ctx = nullptr;
@ -484,7 +594,7 @@ static OutputSection *addInputSec(StringMap<OutputSection *> &Map,
// ignored. We should not have two output .text sections just because one was
// in a group and another was not for example.
//
// It also seems that that wording was a late addition and didn't get the
// It also seems that wording was a late addition and didn't get the
// necessary scrutiny.
//
// Merging sections with different flags is expected by some users. One
@ -529,11 +639,11 @@ static OutputSection *addInputSec(StringMap<OutputSection *> &Map,
void LinkerScript::addOrphanSections() {
unsigned End = SectionCommands.size();
StringMap<OutputSection *> Map;
std::vector<OutputSection *> V;
for (InputSectionBase *S : InputSections) {
auto Add = [&](InputSectionBase *S) {
if (!S->Live || S->Parent)
continue;
return;
StringRef Name = getOutputSectionName(S);
@ -545,12 +655,24 @@ void LinkerScript::addOrphanSections() {
if (OutputSection *Sec =
findByName(makeArrayRef(SectionCommands).slice(0, End), Name)) {
Sec->addSection(cast<InputSection>(S));
continue;
return;
}
if (OutputSection *OS = addInputSec(Map, S, Name))
V.push_back(OS);
assert(S->getOutputSection()->SectionIndex == INT_MAX);
assert(S->getOutputSection()->SectionIndex == UINT32_MAX);
};
// For futher --emit-reloc handling code we need target output section
// to be created before we create relocation output section, so we want
// to create target sections first. We do not want priority handling
// for synthetic sections because them are special.
for (InputSectionBase *IS : InputSections) {
if (auto *Sec = dyn_cast<InputSection>(IS))
if (InputSectionBase *Rel = Sec->getRelocatedSection())
if (auto *RelIS = dyn_cast_or_null<InputSectionBase>(Rel->Parent))
Add(RelIS);
Add(IS);
}
// If no SECTIONS command was given, we should insert sections commands
@ -585,33 +707,15 @@ void LinkerScript::output(InputSection *S) {
// Update output section size after adding each section. This is so that
// SIZEOF works correctly in the case below:
// .foo { *(.aaa) a = SIZEOF(.foo); *(.bbb) }
Ctx->OutSec->Size = Pos - Ctx->OutSec->Addr;
// If there is a memory region associated with this input section, then
// place the section in that region and update the region index.
if (Ctx->LMARegion)
Ctx->LMARegion->CurPos += Pos - Before;
// FIXME: should we also produce overflow errors for LMARegion?
if (Ctx->MemRegion) {
uint64_t &CurOffset = Ctx->MemRegion->CurPos;
CurOffset += Pos - Before;
uint64_t CurSize = CurOffset - Ctx->MemRegion->Origin;
if (CurSize > Ctx->MemRegion->Length) {
uint64_t OverflowAmt = CurSize - Ctx->MemRegion->Length;
error("section '" + Ctx->OutSec->Name + "' will not fit in region '" +
Ctx->MemRegion->Name + "': overflowed by " + Twine(OverflowAmt) +
" bytes");
}
}
expandOutputSection(Pos - Before);
}
void LinkerScript::switchTo(OutputSection *Sec) {
if (Ctx->OutSec == Sec)
return;
Ctx->OutSec = Sec;
uint64_t Before = advance(0, 1);
Ctx->OutSec->Addr = advance(0, Ctx->OutSec->Alignment);
expandMemoryRegions(Ctx->OutSec->Addr - Before);
}
// This function searches for a memory region to place the given output
@ -674,9 +778,8 @@ void LinkerScript::assignOffsets(OutputSection *Sec) {
if (PhdrEntry *L = Ctx->OutSec->PtLoad)
L->LMAOffset = Ctx->LMAOffset;
// The Size previously denoted how many InputSections had been added to this
// section, and was used for sorting SHF_LINK_ORDER sections. Reset it to
// compute the actual size value.
// We can call this method multiple times during the creation of
// thunks and want to start over calculation each time.
Sec->Size = 0;
// We visited SectionsCommands from processSectionCommands to
@ -685,7 +788,9 @@ void LinkerScript::assignOffsets(OutputSection *Sec) {
for (BaseCommand *Base : Sec->SectionCommands) {
// This handles the assignments to symbol or to the dot.
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base)) {
Cmd->Addr = Dot;
assignSymbol(Cmd, true);
Cmd->Size = Dot - Cmd->Addr;
continue;
}
@ -693,17 +798,7 @@ void LinkerScript::assignOffsets(OutputSection *Sec) {
if (auto *Cmd = dyn_cast<ByteCommand>(Base)) {
Cmd->Offset = Dot - Ctx->OutSec->Addr;
Dot += Cmd->Size;
if (Ctx->MemRegion)
Ctx->MemRegion->CurPos += Cmd->Size;
if (Ctx->LMARegion)
Ctx->LMARegion->CurPos += Cmd->Size;
Ctx->OutSec->Size = Dot - Ctx->OutSec->Addr;
continue;
}
// Handle ASSERT().
if (auto *Cmd = dyn_cast<AssertCommand>(Base)) {
Cmd->Expression();
expandOutputSection(Cmd->Size);
continue;
}
@ -728,24 +823,28 @@ void LinkerScript::assignOffsets(OutputSection *Sec) {
}
}
void LinkerScript::removeEmptyCommands() {
// It is common practice to use very generic linker scripts. So for any
// given run some of the output sections in the script will be empty.
// We could create corresponding empty output sections, but that would
// clutter the output.
// We instead remove trivially empty sections. The bfd linker seems even
// more aggressive at removing them.
llvm::erase_if(SectionCommands, [&](BaseCommand *Base) {
if (auto *Sec = dyn_cast<OutputSection>(Base))
return !Sec->Live;
static bool isDiscardable(OutputSection &Sec) {
// We do not remove empty sections that are explicitly
// assigned to any segment.
if (!Sec.Phdrs.empty())
return false;
});
}
static bool isAllSectionDescription(const OutputSection &Cmd) {
for (BaseCommand *Base : Cmd.SectionCommands)
// We do not want to remove sections that reference symbols in address and
// other expressions. We add script symbols as undefined, and want to ensure
// all of them are defined in the output, hence have to keep them.
if (Sec.ExpressionsUseSymbols)
return false;
for (BaseCommand *Base : Sec.SectionCommands) {
if (auto Cmd = dyn_cast<SymbolAssignment>(Base))
// Don't create empty output sections just for unreferenced PROVIDE
// symbols.
if (Cmd->Name != "." && !Cmd->Sym)
continue;
if (!isa<InputSectionDescription>(*Base))
return false;
}
return true;
}
@ -761,7 +860,7 @@ void LinkerScript::adjustSectionsBeforeSorting() {
// Given that we want to create the section, we have to worry what impact
// it will have on the link. For example, if we just create a section with
// 0 for flags, it would change which PT_LOADs are created.
// We could remember that that particular section is dummy and ignore it in
// We could remember that particular section is dummy and ignore it in
// other parts of the linker, but unfortunately there are quite a few places
// that would need to change:
// * The program header creation.
@ -772,29 +871,47 @@ void LinkerScript::adjustSectionsBeforeSorting() {
// the previous sections. Only a few flags are needed to keep the impact low.
uint64_t Flags = SHF_ALLOC;
for (BaseCommand *Cmd : SectionCommands) {
for (BaseCommand *&Cmd : SectionCommands) {
auto *Sec = dyn_cast<OutputSection>(Cmd);
if (!Sec)
continue;
if (Sec->Live) {
Flags = Sec->Flags & (SHF_ALLOC | SHF_WRITE | SHF_EXECINSTR);
continue;
// Handle align (e.g. ".foo : ALIGN(16) { ... }").
if (Sec->AlignExpr)
Sec->Alignment =
std::max<uint32_t>(Sec->Alignment, Sec->AlignExpr().getValue());
// A live output section means that some input section was added to it. It
// might have been removed (if it was empty synthetic section), but we at
// least know the flags.
if (Sec->Live)
Flags = Sec->Flags;
// We do not want to keep any special flags for output section
// in case it is empty.
bool IsEmpty = getInputSections(Sec).empty();
if (IsEmpty)
Sec->Flags = Flags & (SHF_ALLOC | SHF_WRITE | SHF_EXECINSTR);
if (IsEmpty && isDiscardable(*Sec)) {
Sec->Live = false;
Cmd = nullptr;
}
if (isAllSectionDescription(*Sec))
continue;
Sec->Live = true;
Sec->Flags = Flags;
}
// It is common practice to use very generic linker scripts. So for any
// given run some of the output sections in the script will be empty.
// We could create corresponding empty output sections, but that would
// clutter the output.
// We instead remove trivially empty sections. The bfd linker seems even
// more aggressive at removing them.
llvm::erase_if(SectionCommands, [&](BaseCommand *Base) { return !Base; });
}
void LinkerScript::adjustSectionsAfterSorting() {
// Try and find an appropriate memory region to assign offsets in.
for (BaseCommand *Base : SectionCommands) {
if (auto *Sec = dyn_cast<OutputSection>(Base)) {
if (!Sec->Live)
continue;
if (!Sec->LMARegionName.empty()) {
if (MemoryRegion *M = MemoryRegions.lookup(Sec->LMARegionName))
Sec->LMARegion = M;
@ -802,10 +919,6 @@ void LinkerScript::adjustSectionsAfterSorting() {
error("memory region '" + Sec->LMARegionName + "' not declared");
}
Sec->MemRegion = findMemoryRegion(Sec);
// Handle align (e.g. ".foo : ALIGN(16) { ... }").
if (Sec->AlignExpr)
Sec->Alignment =
std::max<uint32_t>(Sec->Alignment, Sec->AlignExpr().getValue());
}
}
@ -816,9 +929,9 @@ void LinkerScript::adjustSectionsAfterSorting() {
// PHDRS { seg PT_LOAD; }
// SECTIONS { .aaa : { *(.aaa) } }
std::vector<StringRef> DefPhdrs;
auto FirstPtLoad =
std::find_if(PhdrsCommands.begin(), PhdrsCommands.end(),
[](const PhdrsCommand &Cmd) { return Cmd.Type == PT_LOAD; });
auto FirstPtLoad = llvm::find_if(PhdrsCommands, [](const PhdrsCommand &Cmd) {
return Cmd.Type == PT_LOAD;
});
if (FirstPtLoad != PhdrsCommands.end())
DefPhdrs.push_back(FirstPtLoad->Name);
@ -847,6 +960,15 @@ static OutputSection *findFirstSection(PhdrEntry *Load) {
return nullptr;
}
static uint64_t computeBase(uint64_t Min, bool AllocateHeaders) {
// If there is no SECTIONS or if the linkerscript is explicit about program
// headers, do our best to allocate them.
if (!Script->HasSectionsCommand || AllocateHeaders)
return 0;
// Otherwise only allocate program headers if that would not add a page.
return alignDown(Min, Config->MaxPageSize);
}
// Try to find an address for the file and program headers output sections,
// which were unconditionally added to the first PT_LOAD segment earlier.
//
@ -870,17 +992,22 @@ void LinkerScript::allocateHeaders(std::vector<PhdrEntry *> &Phdrs) {
return;
PhdrEntry *FirstPTLoad = *It;
bool HasExplicitHeaders =
llvm::any_of(PhdrsCommands, [](const PhdrsCommand &Cmd) {
return Cmd.HasPhdrs || Cmd.HasFilehdr;
});
uint64_t HeaderSize = getHeaderSize();
// When linker script with SECTIONS is being used, don't output headers
// unless there's a space for them.
uint64_t Base = HasSectionsCommand ? alignDown(Min, Config->MaxPageSize) : 0;
if (HeaderSize <= Min - Base || Script->hasPhdrsCommands()) {
if (HeaderSize <= Min - computeBase(Min, HasExplicitHeaders)) {
Min = alignDown(Min - HeaderSize, Config->MaxPageSize);
Out::ElfHeader->Addr = Min;
Out::ProgramHeaders->Addr = Min + Out::ElfHeader->Size;
return;
}
// Error if we were explicitly asked to allocate headers.
if (HasExplicitHeaders)
error("could not allocate headers");
Out::ElfHeader->PtLoad = nullptr;
Out::ProgramHeaders->PtLoad = nullptr;
FirstPTLoad->FirstSec = findFirstSection(FirstPTLoad);
@ -924,15 +1051,11 @@ void LinkerScript::assignAddresses() {
for (BaseCommand *Base : SectionCommands) {
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base)) {
Cmd->Addr = Dot;
assignSymbol(Cmd, false);
Cmd->Size = Dot - Cmd->Addr;
continue;
}
if (auto *Cmd = dyn_cast<AssertCommand>(Base)) {
Cmd->Expression();
continue;
}
assignOffsets(cast<OutputSection>(Base));
}
Ctx = nullptr;
@ -996,9 +1119,9 @@ ExprValue LinkerScript::getSymbolValue(StringRef Name, const Twine &Loc) {
if (Symbol *Sym = Symtab->find(Name)) {
if (auto *DS = dyn_cast<Defined>(Sym))
return {DS->Section, false, DS->Value, Loc};
if (auto *SS = dyn_cast<SharedSymbol>(Sym))
if (!ErrorOnMissingSection || SS->CopyRelSec)
return {SS->CopyRelSec, false, 0, Loc};
if (isa<SharedSymbol>(Sym))
if (!ErrorOnMissingSection)
return {nullptr, false, 0, Loc};
}
error(Loc + ": symbol not found: " + Name);

53
contrib/llvm/tools/lld/ELF/LinkerScript.h
View File

@ -11,9 +11,9 @@
#define LLD_ELF_LINKER_SCRIPT_H
#include "Config.h"
#include "Strings.h"
#include "Writer.h"
#include "lld/Common/LLVM.h"
#include "lld/Common/Strings.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/DenseSet.h"
@ -75,7 +75,6 @@ enum SectionsCommandKind {
AssignmentKind, // . = expr or <sym> = expr
OutputSectionKind,
InputSectionKind,
AssertKind, // ASSERT(expr)
ByteKind // BYTE(expr), SHORT(expr), LONG(expr) or QUAD(expr)
};
@ -106,6 +105,16 @@ struct SymbolAssignment : BaseCommand {
// Holds file name and line number for error reporting.
std::string Location;
// A string representation of this command. We use this for -Map.
std::string CommandString;
// Address of this assignment command.
unsigned Addr;
// Size of this assignment command. This is usually 0, but if
// you move '.' this may be greater than 0.
unsigned Size;
};
// Linker scripts allow additional constraints to be put on ouput sections.
@ -167,24 +176,23 @@ struct InputSectionDescription : BaseCommand {
std::vector<std::pair<ThunkSection *, uint32_t>> ThunkSections;
};
// Represents an ASSERT().
struct AssertCommand : BaseCommand {
AssertCommand(Expr E) : BaseCommand(AssertKind), Expression(E) {}
static bool classof(const BaseCommand *C) { return C->Kind == AssertKind; }
Expr Expression;
};
// Represents BYTE(), SHORT(), LONG(), or QUAD().
struct ByteCommand : BaseCommand {
ByteCommand(Expr E, unsigned Size)
: BaseCommand(ByteKind), Expression(E), Size(Size) {}
ByteCommand(Expr E, unsigned Size, std::string CommandString)
: BaseCommand(ByteKind), CommandString(CommandString), Expression(E),
Size(Size) {}
static bool classof(const BaseCommand *C) { return C->Kind == ByteKind; }
// Keeps string representing the command. Used for -Map" is perhaps better.
std::string CommandString;
Expr Expression;
// This is just an offset of this assignment command in the output section.
unsigned Offset;
// Size of this data command.
unsigned Size;
};
@ -215,14 +223,13 @@ class LinkerScript final {
void addSymbol(SymbolAssignment *Cmd);
void assignSymbol(SymbolAssignment *Cmd, bool InSec);
void setDot(Expr E, const Twine &Loc, bool InSec);
void expandOutputSection(uint64_t Size);
void expandMemoryRegions(uint64_t Size);
std::vector<InputSection *>
computeInputSections(const InputSectionDescription *,
const llvm::DenseMap<SectionBase *, int> &Order);
computeInputSections(const InputSectionDescription *);
std::vector<InputSection *>
createInputSectionList(OutputSection &Cmd,
const llvm::DenseMap<SectionBase *, int> &Order);
std::vector<InputSection *> createInputSectionList(OutputSection &Cmd);
std::vector<size_t> getPhdrIndices(OutputSection *Sec);
@ -257,7 +264,6 @@ class LinkerScript final {
ExprValue getSymbolValue(StringRef Name, const Twine &Loc);
void addOrphanSections();
void removeEmptyCommands();
void adjustSectionsBeforeSorting();
void adjustSectionsAfterSorting();
@ -268,6 +274,10 @@ class LinkerScript final {
void assignAddresses();
void allocateHeaders(std::vector<PhdrEntry *> &Phdrs);
void processSectionCommands();
void declareSymbols();
// Used to handle INSERT AFTER statements.
void processInsertCommands();
// SECTIONS command list.
std::vector<BaseCommand *> SectionCommands;
@ -287,6 +297,11 @@ class LinkerScript final {
// A list of symbols referenced by the script.
std::vector<llvm::StringRef> ReferencedSymbols;
// Used to implement INSERT [AFTER|BEFORE]. Contains commands that need
// to be inserted into SECTIONS commands list.
llvm::DenseMap<StringRef, std::vector<BaseCommand *>> InsertAfterCommands;
llvm::DenseMap<StringRef, std::vector<BaseCommand *>> InsertBeforeCommands;
};
extern LinkerScript *Script;

213
contrib/llvm/tools/lld/ELF/MapFile.cpp
View File

@ -23,11 +23,13 @@
#include "InputFiles.h"
#include "LinkerScript.h"
#include "OutputSections.h"
#include "Strings.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "SyntheticSections.h"
#include "lld/Common/Strings.h"
#include "lld/Common/Threads.h"
#include "llvm/ADT/MapVector.h"
#include "llvm/ADT/SetVector.h"
#include "llvm/Support/raw_ostream.h"
using namespace llvm;
@ -36,57 +38,46 @@ using namespace llvm::object;
using namespace lld;
using namespace lld::elf;
typedef DenseMap<const SectionBase *, SmallVector<Symbol *, 4>> SymbolMapTy;
typedef DenseMap<const SectionBase *, SmallVector<Defined *, 4>> SymbolMapTy;
static const std::string Indent8 = " "; // 8 spaces
static const std::string Indent16 = " "; // 16 spaces
// Print out the first three columns of a line.
static void writeHeader(raw_ostream &OS, uint64_t Addr, uint64_t Size,
uint64_t Align) {
int W = Config->Is64 ? 16 : 8;
OS << format("%0*llx %0*llx %5lld ", W, Addr, W, Size, Align);
static void writeHeader(raw_ostream &OS, uint64_t VMA, uint64_t LMA,
uint64_t Size, uint64_t Align) {
if (Config->Is64)
OS << format("%16llx %16llx %8llx %5lld ", VMA, LMA, Size, Align);
else
OS << format("%8llx %8llx %8llx %5lld ", VMA, LMA, Size, Align);
}
static std::string indent(int Depth) { return std::string(Depth * 8, ' '); }
// Returns a list of all symbols that we want to print out.
static std::vector<Symbol *> getSymbols() {
std::vector<Symbol *> V;
for (InputFile *File : ObjectFiles) {
for (Symbol *B : File->getSymbols()) {
if (auto *SS = dyn_cast<SharedSymbol>(B))
if (SS->CopyRelSec || SS->NeedsPltAddr)
V.push_back(SS);
static std::vector<Defined *> getSymbols() {
std::vector<Defined *> V;
for (InputFile *File : ObjectFiles)
for (Symbol *B : File->getSymbols())
if (auto *DR = dyn_cast<Defined>(B))
if (DR->File == File && !DR->isSection() && DR->Section &&
DR->Section->Live)
if (!DR->isSection() && DR->Section && DR->Section->Live &&
(DR->File == File || DR->NeedsPltAddr || DR->Section->Bss))
V.push_back(DR);
}
}
return V;
}
// Returns a map from sections to their symbols.
static SymbolMapTy getSectionSyms(ArrayRef<Symbol *> Syms) {
static SymbolMapTy getSectionSyms(ArrayRef<Defined *> Syms) {
SymbolMapTy Ret;
for (Symbol *S : Syms) {
if (auto *DR = dyn_cast<Defined>(S)) {
Ret[DR->Section].push_back(S);
continue;
}
SharedSymbol *SS = cast<SharedSymbol>(S);
if (SS->CopyRelSec)
Ret[SS->CopyRelSec].push_back(S);
else
Ret[InX::Plt].push_back(S);
}
for (Defined *DR : Syms)
Ret[DR->Section].push_back(DR);
// Sort symbols by address. We want to print out symbols in the
// order in the output file rather than the order they appeared
// in the input files.
for (auto &It : Ret) {
SmallVectorImpl<Symbol *> &V = It.second;
std::sort(V.begin(), V.end(),
[](Symbol *A, Symbol *B) { return A->getVA() < B->getVA(); });
SmallVectorImpl<Defined *> &V = It.second;
std::stable_sort(V.begin(), V.end(), [](Defined *A, Defined *B) {
return A->getVA() < B->getVA();
});
}
return Ret;
}
@ -95,12 +86,15 @@ static SymbolMapTy getSectionSyms(ArrayRef<Symbol *> Syms) {
// Demangling symbols (which is what toString() does) is slow, so
// we do that in batch using parallel-for.
static DenseMap<Symbol *, std::string>
getSymbolStrings(ArrayRef<Symbol *> Syms) {
getSymbolStrings(ArrayRef<Defined *> Syms) {
std::vector<std::string> Str(Syms.size());
parallelForEachN(0, Syms.size(), [&](size_t I) {
raw_string_ostream OS(Str[I]);
writeHeader(OS, Syms[I]->getVA(), Syms[I]->getSize(), 0);
OS << indent(2) << toString(*Syms[I]);
OutputSection *OSec = Syms[I]->getOutputSection();
uint64_t VMA = Syms[I]->getVA();
uint64_t LMA = OSec ? OSec->getLMA() + VMA - OSec->getVA(0) : 0;
writeHeader(OS, VMA, LMA, Syms[I]->getSize(), 1);
OS << Indent16 << toString(*Syms[I]);
});
DenseMap<Symbol *, std::string> Ret;
@ -109,6 +103,44 @@ getSymbolStrings(ArrayRef<Symbol *> Syms) {
return Ret;
}
// Print .eh_frame contents. Since the section consists of EhSectionPieces,
// we need a specialized printer for that section.
//
// .eh_frame tend to contain a lot of section pieces that are contiguous
// both in input file and output file. Such pieces are squashed before
// being displayed to make output compact.
static void printEhFrame(raw_ostream &OS, OutputSection *OSec) {
std::vector<EhSectionPiece> Pieces;
auto Add = [&](const EhSectionPiece &P) {
// If P is adjacent to Last, squash the two.
if (!Pieces.empty()) {
EhSectionPiece &Last = Pieces.back();
if (Last.Sec == P.Sec && Last.InputOff + Last.Size == P.InputOff &&
Last.OutputOff + Last.Size == P.OutputOff) {
Last.Size += P.Size;
return;
}
}
Pieces.push_back(P);
};
// Gather section pieces.
for (const CieRecord *Rec : InX::EhFrame->getCieRecords()) {
Add(*Rec->Cie);
for (const EhSectionPiece *Fde : Rec->Fdes)
Add(*Fde);
}
// Print out section pieces.
for (EhSectionPiece &P : Pieces) {
writeHeader(OS, OSec->Addr + P.OutputOff, OSec->getLMA() + P.OutputOff,
P.Size, 1);
OS << Indent8 << toString(P.Sec->File) << ":(" << P.Sec->Name << "+0x"
<< Twine::utohexstr(P.InputOff) + ")\n";
}
}
void elf::writeMapFile() {
if (Config->MapFile.empty())
return;
@ -122,32 +154,109 @@ void elf::writeMapFile() {
}
// Collect symbol info that we want to print out.
std::vector<Symbol *> Syms = getSymbols();
std::vector<Defined *> Syms = getSymbols();
SymbolMapTy SectionSyms = getSectionSyms(Syms);
DenseMap<Symbol *, std::string> SymStr = getSymbolStrings(Syms);
// Print out the header line.
int W = Config->Is64 ? 16 : 8;
OS << left_justify("Address", W) << ' ' << left_justify("Size", W)
<< " Align Out In Symbol\n";
OS << right_justify("VMA", W) << ' ' << right_justify("LMA", W)
<< " Size Align Out In Symbol\n";
// Print out file contents.
for (OutputSection *OSec : OutputSections) {
writeHeader(OS, OSec->Addr, OSec->Size, OSec->Alignment);
for (BaseCommand *Base : Script->SectionCommands) {
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base)) {
if (Cmd->Provide && !Cmd->Sym)
continue;
//FIXME: calculate and print LMA.
writeHeader(OS, Cmd->Addr, 0, Cmd->Size, 1);
OS << Cmd->CommandString << '\n';
continue;
}
auto *OSec = cast<OutputSection>(Base);
writeHeader(OS, OSec->Addr, OSec->getLMA(), OSec->Size, OSec->Alignment);
OS << OSec->Name << '\n';
// Dump symbols for each input section.
for (BaseCommand *Base : OSec->SectionCommands) {
auto *ISD = dyn_cast<InputSectionDescription>(Base);
if (!ISD)
if (auto *ISD = dyn_cast<InputSectionDescription>(Base)) {
for (InputSection *IS : ISD->Sections) {
if (IS == InX::EhFrame) {
printEhFrame(OS, OSec);
continue;
}
writeHeader(OS, IS->getVA(0), OSec->getLMA() + IS->getOffset(0),
IS->getSize(), IS->Alignment);
OS << Indent8 << toString(IS) << '\n';
for (Symbol *Sym : SectionSyms[IS])
OS << SymStr[Sym] << '\n';
}
continue;
}
if (auto *Cmd = dyn_cast<ByteCommand>(Base)) {
writeHeader(OS, OSec->Addr + Cmd->Offset, OSec->getLMA() + Cmd->Offset,
Cmd->Size, 1);
OS << Indent8 << Cmd->CommandString << '\n';
continue;
}
if (auto *Cmd = dyn_cast<SymbolAssignment>(Base)) {
if (Cmd->Provide && !Cmd->Sym)
continue;
writeHeader(OS, Cmd->Addr, OSec->getLMA() + Cmd->Addr - OSec->getVA(0),
Cmd->Size, 1);
OS << Indent8 << Cmd->CommandString << '\n';
continue;
for (InputSection *IS : ISD->Sections) {
writeHeader(OS, OSec->Addr + IS->OutSecOff, IS->getSize(),
IS->Alignment);
OS << indent(1) << toString(IS) << '\n';
for (Symbol *Sym : SectionSyms[IS])
OS << SymStr[Sym] << '\n';
}
}
}
}
static void print(StringRef A, StringRef B) {
outs() << left_justify(A, 49) << " " << B << "\n";
}
// Output a cross reference table to stdout. This is for --cref.
//
// For each global symbol, we print out a file that defines the symbol
// followed by files that uses that symbol. Here is an example.
//
// strlen /lib/x86_64-linux-gnu/libc.so.6
// tools/lld/tools/lld/CMakeFiles/lld.dir/lld.cpp.o
// lib/libLLVMSupport.a(PrettyStackTrace.cpp.o)
//
// In this case, strlen is defined by libc.so.6 and used by other two
// files.
void elf::writeCrossReferenceTable() {
if (!Config->Cref)
return;
// Collect symbols and files.
MapVector<Symbol *, SetVector<InputFile *>> Map;
for (InputFile *File : ObjectFiles) {
for (Symbol *Sym : File->getSymbols()) {
if (isa<SharedSymbol>(Sym))
Map[Sym].insert(File);
if (auto *D = dyn_cast<Defined>(Sym))
if (!D->isLocal() && (!D->Section || D->Section->Live))
Map[D].insert(File);
}
}
// Print out a header.
outs() << "Cross Reference Table\n\n";
print("Symbol", "File");
// Print out a table.
for (auto KV : Map) {
Symbol *Sym = KV.first;
SetVector<InputFile *> &Files = KV.second;
print(toString(*Sym), toString(Sym->File));
for (InputFile *File : Files)
if (File != Sym->File)
print("", toString(File));
}
}

1
contrib/llvm/tools/lld/ELF/MapFile.h
View File

@ -13,6 +13,7 @@
namespace lld {
namespace elf {
void writeMapFile();
void writeCrossReferenceTable();
} // namespace elf
} // namespace lld

38
contrib/llvm/tools/lld/ELF/MarkLive.cpp
View File

@ -20,15 +20,15 @@
//
//===----------------------------------------------------------------------===//
#include "MarkLive.h"
#include "InputSection.h"
#include "LinkerScript.h"
#include "OutputSections.h"
#include "Strings.h"
#include "SymbolTable.h"
#include "Symbols.h"
#include "Target.h"
#include "Writer.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Strings.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/Object/ELF.h"
#include <functional>
@ -60,8 +60,9 @@ static typename ELFT::uint getAddend(InputSectionBase &Sec,
static DenseMap<StringRef, std::vector<InputSectionBase *>> CNamedSections;
template <class ELFT, class RelT>
static void resolveReloc(InputSectionBase &Sec, RelT &Rel,
std::function<void(InputSectionBase *, uint64_t)> Fn) {
static void
resolveReloc(InputSectionBase &Sec, RelT &Rel,
llvm::function_ref<void(InputSectionBase *, uint64_t)> Fn) {
Symbol &B = Sec.getFile<ELFT>()->getRelocTargetSym(Rel);
// If a symbol is referenced in a live section, it is used.
@ -90,7 +91,7 @@ static void resolveReloc(InputSectionBase &Sec, RelT &Rel,
template <class ELFT>
static void
forEachSuccessor(InputSection &Sec,
std::function<void(InputSectionBase *, uint64_t)> Fn) {
llvm::function_ref<void(InputSectionBase *, uint64_t)> Fn) {
if (Sec.AreRelocsRela) {
for (const typename ELFT::Rela &Rel : Sec.template relas<ELFT>())
resolveReloc<ELFT>(Sec, Rel, Fn);
@ -120,7 +121,7 @@ forEachSuccessor(InputSection &Sec,
template <class ELFT, class RelTy>
static void
scanEhFrameSection(EhInputSection &EH, ArrayRef<RelTy> Rels,
std::function<void(InputSectionBase *, uint64_t)> Fn) {
llvm::function_ref<void(InputSectionBase *, uint64_t)> Fn) {
const endianness E = ELFT::TargetEndianness;
for (unsigned I = 0, N = EH.Pieces.size(); I < N; ++I) {
@ -155,14 +156,10 @@ scanEhFrameSection(EhInputSection &EH, ArrayRef<RelTy> Rels,
template <class ELFT>
static void
scanEhFrameSection(EhInputSection &EH,
std::function<void(InputSectionBase *, uint64_t)> Fn) {
llvm::function_ref<void(InputSectionBase *, uint64_t)> Fn) {
if (!EH.NumRelocations)
return;
// Unfortunately we need to split .eh_frame early since some relocations in
// .eh_frame keep other section alive and some don't.
EH.split<ELFT>();
if (EH.AreRelocsRela)
scanEhFrameSection<ELFT>(EH, EH.template relas<ELFT>(), Fn);
else
@ -207,7 +204,7 @@ template <class ELFT> static void doGcSections() {
// (splittable) sections, each piece of data has independent liveness bit.
// So we explicitly tell it which offset is in use.
if (auto *MS = dyn_cast<MergeInputSection>(Sec))
MS->markLiveAt(Offset);
MS->getSectionPiece(Offset)->Live = true;
if (Sec->Live)
return;
@ -279,13 +276,18 @@ template <class ELFT> void elf::markLive() {
// The -gc-sections option works only for SHF_ALLOC sections
// (sections that are memory-mapped at runtime). So we can
// unconditionally make non-SHF_ALLOC sections alive.
// unconditionally make non-SHF_ALLOC sections alive except
// SHF_LINK_ORDER and SHT_REL/SHT_RELA sections.
//
// Non SHF_ALLOC sections are not removed even if they are
// Usually, SHF_ALLOC sections are not removed even if they are
// unreachable through relocations because reachability is not
// a good signal whether they are garbage or not (e.g. there is
// usually no section referring to a .comment section, but we
// want to keep it.)
// want to keep it.).
//
// Note on SHF_LINK_ORDER: Such sections contain metadata and they
// have a reverse dependency on the InputSection they are linked with.
// We are able to garbage collect them.
//
// Note on SHF_REL{,A}: Such sections reach here only when -r
// or -emit-reloc were given. And they are subject of garbage
@ -293,8 +295,9 @@ template <class ELFT> void elf::markLive() {
// remove its relocation section.
for (InputSectionBase *Sec : InputSections) {
bool IsAlloc = (Sec->Flags & SHF_ALLOC);
bool IsLinkOrder = (Sec->Flags & SHF_LINK_ORDER);
bool IsRel = (Sec->Type == SHT_REL || Sec->Type == SHT_RELA);
if (!IsAlloc && !IsRel)
if (!IsAlloc && !IsLinkOrder && !IsRel)
Sec->Live = true;
}
@ -305,8 +308,7 @@ template <class ELFT> void elf::markLive() {
if (Config->PrintGcSections)
for (InputSectionBase *Sec : InputSections)
if (!Sec->Live)
message("removing unused section from '" + Sec->Name + "' in file '" +
Sec->File->getName() + "'");
message("removing unused section " + toString(Sec));
}
template void elf::markLive<ELF32LE>();

21
contrib/llvm/tools/lld/ELF/MarkLive.h Normal file
View File

@ -0,0 +1,21 @@
//===- MarkLive.h -----------------------------------------------*- C++ -*-===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#ifndef LLD_ELF_MARKLIVE_H
#define LLD_ELF_MARKLIVE_H
namespace lld {
namespace elf {
template <class ELFT> void markLive();
} // namespace elf
} // namespace lld
#endif // LLD_ELF_MARKLIVE_H

520
contrib/llvm/tools/lld/ELF/Options.td
View File

@ -4,66 +4,91 @@ include "llvm/Option/OptParser.td"
// two can precede the option name except those that start with 'o'.
class F<string name>: Flag<["--", "-"], name>;
class J<string name>: Joined<["--", "-"], name>;
class S<string name>: Separate<["--", "-"], name>;
multiclass Eq<string name> {
def "": Separate<["--", "-"], name>;
def _eq: Joined<["--", "-"], name # "=">, Alias<!cast<Separate>(NAME)>;
multiclass Eq<string name, string help> {
def NAME: Separate<["--", "-"], name>;
def NAME # _eq: Joined<["--", "-"], name # "=">, Alias<!cast<Separate>(NAME)>,
HelpText<help>;
}
def auxiliary: S<"auxiliary">, HelpText<"Set DT_AUXILIARY field to the specified name">;
multiclass B<string name, string help1, string help2> {
def NAME: Flag<["--", "-"], name>, HelpText<help1>;
def no_ # NAME: Flag<["--", "-"], "no-" # name>, HelpText<help2>;
}
defm auxiliary: Eq<"auxiliary", "Set DT_AUXILIARY field to the specified name">;
def Bsymbolic: F<"Bsymbolic">, HelpText<"Bind defined symbols locally">;
def Bsymbolic_functions: F<"Bsymbolic-functions">,
HelpText<"Bind defined function symbols locally">;
def Bdynamic: F<"Bdynamic">, HelpText<"Link against shared libraries">;
def Bdynamic: F<"Bdynamic">, HelpText<"Link against shared libraries (default)">;
def Bstatic: F<"Bstatic">, HelpText<"Do not link against shared libraries">;
def build_id: F<"build-id">, HelpText<"Generate build ID note">;
def build_id: F<"build-id">, HelpText<"Alias for --build-id=fast">;
def build_id_eq: J<"build-id=">, HelpText<"Generate build ID note">;
def build_id_eq: J<"build-id=">, HelpText<"Generate build ID note">,
MetaVarName<"[fast,md5,sha,uuid,0x<hexstring>]">;
defm compress_debug_sections : Eq<"compress-debug-sections">,
HelpText<"Compress DWARF debug sections">;
defm check_sections: B<"check-sections",
"Check section addresses for overlaps (default)",
"Do not check section addresses for overlaps">;
defm defsym: Eq<"defsym">, HelpText<"Define a symbol alias">;
defm compress_debug_sections:
Eq<"compress-debug-sections", "Compress DWARF debug sections">,
MetaVarName<"[none,zlib]">;
defm library_path: Eq<"library-path">,
HelpText<"Add a directory to the library search path">, MetaVarName<"<dir>">;
defm defsym: Eq<"defsym", "Define a symbol alias">, MetaVarName<"<symbol>=<value>">;
defm library_path:
Eq<"library-path", "Add a directory to the library search path">, MetaVarName<"<dir>">;
def O: JoinedOrSeparate<["-"], "O">, HelpText<"Optimize output file size">;
defm Tbss: Eq<"Tbss">,
HelpText<"Same as --section-start with .bss as the sectionname">;
defm Tbss: Eq<"Tbss", "Same as --section-start with .bss as the sectionname">;
defm Tdata: Eq<"Tdata">,
HelpText<"Same as --section-start with .data as the sectionname">;
defm Tdata: Eq<"Tdata", "Same as --section-start with .data as the sectionname">;
defm Ttext: Eq<"Ttext">,
HelpText<"Same as --section-start with .text as the sectionname">;
defm Ttext: Eq<"Ttext", "Same as --section-start with .text as the sectionname">;
def allow_multiple_definition: F<"allow-multiple-definition">,
HelpText<"Allow multiple definitions">;
defm allow_multiple_definition: B<"allow-multiple-definition",
"Allow multiple definitions",
"Do not allow multiple definitions (default)">;
def as_needed: F<"as-needed">,
HelpText<"Only set DT_NEEDED for shared libraries if used">;
defm apply_dynamic_relocs: B<"apply-dynamic-relocs",
"Apply dynamic relocations to place",
"Do not apply dynamic relocations to place">;
defm as_needed: B<"as-needed",
"Only set DT_NEEDED for shared libraries if used",
"Always set DT_NEEDED for shared libraries (default)">;
defm call_graph_ordering_file:
Eq<"call-graph-ordering-file", "Layout sections to optimize the given callgraph">;
// -chroot doesn't have a help text because it is an internal option.
def chroot: S<"chroot">;
def chroot: Separate<["--", "-"], "chroot">;
def color_diagnostics: F<"color-diagnostics">,
HelpText<"Use colors in diagnostics">;
HelpText<"Alias for --color-diagnostics=always">;
def color_diagnostics_eq: J<"color-diagnostics=">,
HelpText<"Use colors in diagnostics">;
HelpText<"Use colors in diagnostics">,
MetaVarName<"[auto,always,never]">;
def define_common: F<"define-common">,
HelpText<"Assign space to common symbols">;
defm cref: B<"cref",
"Output cross reference table",
"Do not output cross reference table">;
def demangle: F<"demangle">, HelpText<"Demangle symbol names">;
defm define_common: B<"define-common",
"Assign space to common symbols",
"Do not assign space to common symbols">;
defm demangle: B<"demangle",
"Demangle symbol names (default)",
"Do not demangle symbol names">;
def disable_new_dtags: F<"disable-new-dtags">,
HelpText<"Disable new dynamic tags">;
@ -76,158 +101,126 @@ def discard_locals: F<"discard-locals">,
def discard_none: F<"discard-none">,
HelpText<"Keep all symbols in the symbol table">;
def dynamic_linker: S<"dynamic-linker">,
HelpText<"Which dynamic linker to use">;
defm dynamic_linker: Eq<"dynamic-linker", "Which dynamic linker to use">;
defm dynamic_list: Eq<"dynamic-list">,
HelpText<"Read a list of dynamic symbols">;
defm dynamic_list: Eq<"dynamic-list", "Read a list of dynamic symbols">;
def eh_frame_hdr: F<"eh-frame-hdr">,
HelpText<"Request creation of .eh_frame_hdr section and PT_GNU_EH_FRAME segment header">;
defm eh_frame_hdr: B<"eh-frame-hdr",
"Request creation of .eh_frame_hdr section and PT_GNU_EH_FRAME segment header",
"Do not create .eh_frame_hdr section">;
def emit_relocs: F<"emit-relocs">, HelpText<"Generate relocations in output">;
def enable_new_dtags: F<"enable-new-dtags">,
HelpText<"Enable new dynamic tags">;
HelpText<"Enable new dynamic tags (default)">;
def end_group: F<"end-group">,
HelpText<"Ignored for compatibility with GNU unless you pass --warn-backrefs">;
def end_lib: F<"end-lib">,
HelpText<"End a grouping of objects that should be treated as if they were together in an archive">;
defm entry: Eq<"entry">, HelpText<"Name of entry point symbol">,
defm entry: Eq<"entry", "Name of entry point symbol">,
MetaVarName<"<entry>">;
defm error_limit: Eq<"error-limit">,
HelpText<"Maximum number of errors to emit before stopping (0 = no limit)">;
defm error_limit:
Eq<"error-limit", "Maximum number of errors to emit before stopping (0 = no limit)">;
def error_unresolved_symbols: F<"error-unresolved-symbols">,
HelpText<"Report unresolved symbols as errors">;
defm exclude_libs: Eq<"exclude-libs">,
HelpText<"Exclude static libraries from automatic export">;
defm exclude_libs: Eq<"exclude-libs", "Exclude static libraries from automatic export">;
def export_dynamic: F<"export-dynamic">,
HelpText<"Put symbols in the dynamic symbol table">;
defm export_dynamic: B<"export-dynamic",
"Put symbols in the dynamic symbol table",
"Do not put symbols in the dynamic symbol table (default)">;
defm export_dynamic_symbol: Eq<"export-dynamic-symbol">,
HelpText<"Put a symbol in the dynamic symbol table">;
defm export_dynamic_symbol:
Eq<"export-dynamic-symbol", "Put a symbol in the dynamic symbol table">;
def fatal_warnings: F<"fatal-warnings">,
HelpText<"Treat warnings as errors">;
defm fatal_warnings: B<"fatal-warnings",
"Treat warnings as errors",
"Do not treat warnings as errors (default)">;
defm filter: Eq<"filter">,
HelpText<"Set DT_FILTER field to the specified name">;
defm filter: Eq<"filter", "Set DT_FILTER field to the specified name">;
defm fini: Eq<"fini">,
HelpText<"Specify a finalizer function">, MetaVarName<"<symbol>">;
defm fini: Eq<"fini", "Specify a finalizer function">, MetaVarName<"<symbol>">;
def fix_cortex_a53_843419: F<"fix-cortex-a53-843419">,
HelpText<"Apply fixes for AArch64 Cortex-A53 erratum 843419">;
def full_shutdown : F<"full-shutdown">,
HelpText<"Perform a full shutdown instead of calling _exit">;
defm format: Eq<"format", "Change the input format of the inputs following this option">,
MetaVarName<"[default,elf,binary]">;
defm format: Eq<"format">,
HelpText<"Change the input format of the inputs following this option">,
MetaVarName<"<input-format>">;
defm gc_sections: B<"gc-sections",
"Enable garbage collection of unused sections",
"Disable garbage collection of unused sections (default)">;
def gc_sections: F<"gc-sections">,
HelpText<"Enable garbage collection of unused sections">;
defm gdb_index: B<"gdb-index",
"Generate .gdb_index section",
"Do not generate .gdb_index section (default)">;
def gdb_index: F<"gdb-index">,
HelpText<"Generate .gdb_index section">;
defm gnu_unique: B<"gnu-unique",
"Enable STB_GNU_UNIQUE symbol binding (default)",
"Disable STB_GNU_UNIQUE symbol binding">;
defm hash_style: Eq<"hash-style">,
HelpText<"Specify hash style (sysv, gnu or both)">;
defm hash_style: Eq<"hash-style", "Specify hash style (sysv, gnu or both)">;
def help: F<"help">, HelpText<"Print option help">;
def icf_all: F<"icf=all">, HelpText<"Enable identical code folding">;
def icf_data: F<"icf-data">,
HelpText<"Enable ICF to also fold identical read only data">;
def icf_safe: F<"icf=safe">, HelpText<"Enable safe identical code folding">;
def icf_none: F<"icf=none">, HelpText<"Disable identical code folding">;
def icf_none: F<"icf=none">, HelpText<"Disable identical code folding (default)">;
defm image_base : Eq<"image-base">, HelpText<"Set the base address">;
def ignore_function_address_equality: F<"ignore-function-address-equality">,
HelpText<"lld can break the address equality of functions">;
defm init: Eq<"init">, HelpText<"Specify an initializer function">,
def ignore_data_address_equality: F<"ignore-data-address-equality">,
HelpText<"lld can break the address equality of data">;
defm image_base: Eq<"image-base", "Set the base address">;
defm init: Eq<"init", "Specify an initializer function">,
MetaVarName<"<symbol>">;
defm library: Eq<"library">, HelpText<"Root name of library to use">,
MetaVarName<"<libName>">;
defm just_symbols: Eq<"just-symbols", "Just link symbols">;
def lto_O: J<"lto-O">, MetaVarName<"<opt-level>">,
HelpText<"Optimization level for LTO">;
defm keep_unique: Eq<"keep-unique", "Do not fold this symbol during ICF">;
defm library: Eq<"library", "Root name of library to use">,
MetaVarName<"<libName>">;
def m: JoinedOrSeparate<["-"], "m">, HelpText<"Set target emulation">;
defm Map: Eq<"Map">, HelpText<"Print a link map to the specified file">;
defm Map: Eq<"Map", "Print a link map to the specified file">;
def merge_exidx_entries: F<"merge-exidx-entries">,
HelpText<"Enable merging .ARM.exidx entries">;
defm merge_exidx_entries: B<"merge-exidx-entries",
"Enable merging .ARM.exidx entries (default)",
"Disable merging .ARM.exidx entries">;
def nostdlib: F<"nostdlib">,
HelpText<"Only search directories specified on the command line">;
def no_as_needed: F<"no-as-needed">,
HelpText<"Always DT_NEEDED for shared libraries">;
def no_color_diagnostics: F<"no-color-diagnostics">,
HelpText<"Do not use colors in diagnostics">;
def no_define_common: F<"no-define-common">,
HelpText<"Do not assign space to common symbols">;
def no_demangle: F<"no-demangle">,
HelpText<"Do not demangle symbol names">;
def no_dynamic_linker: F<"no-dynamic-linker">,
HelpText<"Inhibit output of .interp section">;
def no_eh_frame_hdr: F<"no-eh-frame-hdr">,
HelpText<"Do not create .eh_frame_hdr section">;
def no_export_dynamic: F<"no-export-dynamic">;
def no_fatal_warnings: F<"no-fatal-warnings">;
def no_gc_sections: F<"no-gc-sections">,
HelpText<"Disable garbage collection of unused sections">;
def no_gdb_index: F<"no-gdb-index">,
HelpText<"Do not generate .gdb_index section">;
def no_gnu_unique: F<"no-gnu-unique">,
HelpText<"Disable STB_GNU_UNIQUE symbol binding">;
def no_merge_exidx_entries: F<"no-merge-exidx-entries">,
HelpText<"Disable merging .ARM.exidx entries">;
def no_threads: F<"no-threads">,
HelpText<"Do not run the linker multi-threaded">;
def no_whole_archive: F<"no-whole-archive">,
HelpText<"Restores the default behavior of loading archive members">;
def noinhibit_exec: F<"noinhibit-exec">,
HelpText<"Retain the executable output file whenever it is still usable">;
def nopie: F<"nopie">, HelpText<"Do not create a position independent executable">;
def no_omagic: Flag<["--"], "no-omagic">, MetaVarName<"<magic>">,
def no_omagic: F<"no-omagic">, MetaVarName<"<magic>">,
HelpText<"Do not set the text data sections to be writable">;
def no_print_gc_sections: F<"no-print-gc-sections">,
HelpText<"Do not list removed unused sections">;
def no_rosegment: F<"no-rosegment">,
HelpText<"Do not put read-only non-executable sections in their own segment">;
def no_undefined: F<"no-undefined">,
HelpText<"Report unresolved symbols even if the linker is creating a shared library">;
def no_undefined_version: F<"no-undefined-version">,
HelpText<"Report version scripts that refer undefined symbols">;
def o: JoinedOrSeparate<["-"], "o">, MetaVarName<"<path>">,
HelpText<"Path to file to write output">;
@ -237,42 +230,62 @@ def oformat: Separate<["--"], "oformat">, MetaVarName<"<format>">,
def omagic: Flag<["--"], "omagic">, MetaVarName<"<magic>">,
HelpText<"Set the text and data sections to be readable and writable">;
defm orphan_handling: Eq<"orphan-handling">,
HelpText<"Control how orphan sections are handled when linker script used">;
defm orphan_handling:
Eq<"orphan-handling", "Control how orphan sections are handled when linker script used">;
def pack_dyn_relocs_eq: J<"pack-dyn-relocs=">, MetaVarName<"<format>">,
HelpText<"Pack dynamic relocations in the given format (none or android)">;
defm pack_dyn_relocs:
Eq<"pack-dyn-relocs", "Pack dynamic relocations in the given format">,
MetaVarName<"[none,android,relr,android+relr]">;
def pie: F<"pie">, HelpText<"Create a position independent executable">;
defm use_android_relr_tags: B<"use-android-relr-tags",
"Use SHT_ANDROID_RELR / DT_ANDROID_RELR* tags instead of SHT_RELR / DT_RELR*",
"Use SHT_RELR / DT_RELR* tags (default)">;
def print_gc_sections: F<"print-gc-sections">,
HelpText<"List removed unused sections">;
defm pie: B<"pie",
"Create a position independent executable",
"Do not create a position independent executable (default)">;
defm print_gc_sections: B<"print-gc-sections",
"List removed unused sections",
"Do not list removed unused sections (default)">;
defm print_icf_sections: B<"print-icf-sections",
"List identical folded sections",
"Do not list identical folded sections (default)">;
def pop_state: F<"pop-state">,
HelpText<"Undo the effect of -push-state">;
def push_state: F<"push-state">,
HelpText<"Save the current state of -as-needed, -static and -whole-archive">;
def print_map: F<"print-map">,
HelpText<"Print a link map to the standard output">;
defm reproduce: Eq<"reproduce">,
HelpText<"Dump linker invocation and input files for debugging">;
defm reproduce: Eq<"reproduce", "Dump linker invocation and input files for debugging">;
defm rpath: Eq<"rpath">, HelpText<"Add a DT_RUNPATH to the output">;
defm rpath: Eq<"rpath", "Add a DT_RUNPATH to the output">;
def relocatable: F<"relocatable">, HelpText<"Create relocatable object file">;
defm retain_symbols_file: Eq<"retain-symbols-file">,
HelpText<"Retain only the symbols listed in the file">,
defm retain_symbols_file:
Eq<"retain-symbols-file", "Retain only the symbols listed in the file">,
MetaVarName<"<file>">;
defm script: Eq<"script">, HelpText<"Read linker script">;
defm script: Eq<"script", "Read linker script">;
def section_start: S<"section-start">, MetaVarName<"<address>">,
HelpText<"Set address of section">;
defm section_start: Eq<"section-start", "Set address of section">,
MetaVarName<"<address>">;
def shared: F<"shared">, HelpText<"Build a shared object">;
defm soname: Eq<"soname">, HelpText<"Set DT_SONAME">;
defm soname: Eq<"soname", "Set DT_SONAME">;
defm sort_section: Eq<"sort-section">,
HelpText<"Specifies sections sorting rule when linkerscript is used">;
defm sort_section:
Eq<"sort-section", "Specifies sections sorting rule when linkerscript is used">;
def start_group: F<"start-group">,
HelpText<"Ignored for compatibility with GNU unless you pass --warn-backrefs">;
def start_lib: F<"start-lib">,
HelpText<"Start a grouping of objects that should be treated as if they were together in an archive">;
@ -281,33 +294,39 @@ def strip_all: F<"strip-all">, HelpText<"Strip all symbols">;
def strip_debug: F<"strip-debug">, HelpText<"Strip debugging information">;
def symbol_ordering_file: S<"symbol-ordering-file">,
HelpText<"Layout sections in the order specified by symbol file">;
defm symbol_ordering_file:
Eq<"symbol-ordering-file", "Layout sections to place symbols in the order specified by symbol ordering file">;
defm sysroot: Eq<"sysroot">, HelpText<"Set the system root">;
defm sysroot: Eq<"sysroot", "Set the system root">;
def target1_rel: F<"target1-rel">, HelpText<"Interpret R_ARM_TARGET1 as R_ARM_REL32">;
def target1_abs: F<"target1-abs">, HelpText<"Interpret R_ARM_TARGET1 as R_ARM_ABS32">;
def target1_abs: F<"target1-abs">, HelpText<"Interpret R_ARM_TARGET1 as R_ARM_ABS32 (default)">;
defm target2: Eq<"target2">,
HelpText<"Interpret R_ARM_TARGET2 as <type>, where <type> is one of rel, abs, or got-rel">,
defm target2:
Eq<"target2", "Interpret R_ARM_TARGET2 as <type>, where <type> is one of rel, abs, or got-rel">,
MetaVarName<"<type>">;
def threads: F<"threads">, HelpText<"Run the linker multi-threaded">;
defm threads: B<"threads",
"Run the linker multi-threaded (default)",
"Do not run the linker multi-threaded">;
def trace: F<"trace">, HelpText<"Print the names of the input files">;
defm trace_symbol : Eq<"trace-symbol">, HelpText<"Trace references to symbols">;
defm trace_symbol: Eq<"trace-symbol", "Trace references to symbols">;
defm undefined: Eq<"undefined">,
HelpText<"Force undefined symbol during linking">;
defm undefined: Eq<"undefined", "Force undefined symbol during linking">,
MetaVarName<"<symbol>">;
defm unresolved_symbols: Eq<"unresolved-symbols">,
HelpText<"Determine how to handle unresolved symbols">;
defm unresolved_symbols:
Eq<"unresolved-symbols", "Determine how to handle unresolved symbols">;
defm rsp_quoting: Eq<"rsp-quoting">,
HelpText<"Quoting style for response files. Values supported: windows|posix">;
defm undefined_version: B<"undefined-version",
"Allow unused version in version script (default)",
"Report version scripts that refer undefined symbols">;
defm rsp_quoting: Eq<"rsp-quoting", "Quoting style for response files">,
MetaVarName<"[posix,windows]">;
def v: Flag<["-"], "v">, HelpText<"Display the version number">;
@ -315,91 +334,123 @@ def verbose: F<"verbose">, HelpText<"Verbose mode">;
def version: F<"version">, HelpText<"Display the version number and exit">;
defm version_script: Eq<"version-script">, HelpText<"Read a version script">;
defm version_script: Eq<"version-script", "Read a version script">;
def warn_common: F<"warn-common">,
HelpText<"Warn about duplicate common symbols">;
defm warn_backrefs: B<"warn-backrefs",
"Warn about backward symbol references to fetch archive members",
"Do not warn about backward symbol references to fetch archive members (default)">;
defm warn_common: B<"warn-common",
"Warn about duplicate common symbols",
"Do not warn about duplicate common symbols (default)">;
defm warn_symbol_ordering: B<"warn-symbol-ordering",
"Warn about problems with the symbol ordering file (default)",
"Do not warn about problems with the symbol ordering file">;
def warn_unresolved_symbols: F<"warn-unresolved-symbols">,
HelpText<"Report unresolved symbols as warnings">;
def whole_archive: F<"whole-archive">,
HelpText<"Force load of all members in a static library">;
defm whole_archive: B<"whole-archive",
"Force load of all members in a static library",
"Do not force load of all members in a static library (default)">;
defm wrap: Eq<"wrap">, HelpText<"Use wrapper functions for symbol">,
MetaVarName<"<symbol>">;
defm wrap: Eq<"wrap", "Use wrapper functions for symbol">,
MetaVarName<"<symbol>=<symbol>">;
def z: JoinedOrSeparate<["-"], "z">, MetaVarName<"<option>">,
HelpText<"Linker option extensions">;
// Aliases
def alias_auxiliary: Separate<["-"], "f">, Alias<auxiliary>;
def alias_Bdynamic_call_shared: F<"call_shared">, Alias<Bdynamic>;
def alias_Bdynamic_dy: F<"dy">, Alias<Bdynamic>;
def alias_Bstatic_dn: F<"dn">, Alias<Bstatic>;
def alias_Bstatic_non_shared: F<"non_shared">, Alias<Bstatic>;
def alias_Bstatic_static: F<"static">, Alias<Bstatic>;
def alias_define_common_d: Flag<["-"], "d">, Alias<define_common>;
def alias_define_common_dc: F<"dc">, Alias<define_common>;
def alias_define_common_dp: F<"dp">, Alias<define_common>;
def alias_discard_all_x: Flag<["-"], "x">, Alias<discard_all>;
def alias_discard_locals_X: Flag<["-"], "X">, Alias<discard_locals>;
def alias_emit_relocs: Flag<["-"], "q">, Alias<emit_relocs>;
def alias_entry_e: JoinedOrSeparate<["-"], "e">, Alias<entry>;
def alias_export_dynamic_E: Flag<["-"], "E">, Alias<export_dynamic>;
def alias_filter: Separate<["-"], "F">, Alias<filter>;
def alias_format_b: S<"b">, Alias<format>;
def alias_library: JoinedOrSeparate<["-"], "l">, Alias<library>;
def alias_library_path: JoinedOrSeparate<["-"], "L">, Alias<library_path>;
def alias_omagic: Flag<["-"], "N">, Alias<omagic>;
def alias_o_output: Joined<["--"], "output=">, Alias<o>;
def alias_o_output2 : Separate<["--"], "output">, Alias<o>;
def alias_pie_pic_executable: F<"pic-executable">, Alias<pie>;
def alias_print_map_M: Flag<["-"], "M">, Alias<print_map>;
def alias_relocatable_r: Flag<["-"], "r">, Alias<relocatable>;
def alias_rpath_R: JoinedOrSeparate<["-"], "R">, Alias<rpath>;
def alias_script_T: JoinedOrSeparate<["-"], "T">, Alias<script>;
def alias_shared_Bshareable: F<"Bshareable">, Alias<shared>;
def alias_soname_h: JoinedOrSeparate<["-"], "h">, Alias<soname>;
def alias_strip_all: Flag<["-"], "s">, Alias<strip_all>;
def alias_strip_debug_S: Flag<["-"], "S">, Alias<strip_debug>;
def alias_trace: Flag<["-"], "t">, Alias<trace>;
def alias_trace_symbol_y : JoinedOrSeparate<["-"], "y">, Alias<trace_symbol>;
def alias_Ttext_segment: S<"Ttext-segment">, Alias<Ttext>;
def alias_Ttext_segment_eq: J<"Ttext-segment=">, Alias<Ttext>;
def alias_undefined_u: JoinedOrSeparate<["-"], "u">, Alias<undefined>;
def alias_version_V: Flag<["-"], "V">, Alias<version>;
// Our symbol resolution algorithm handles symbols in archive files differently
// than traditional linkers, so we don't need --start-group and --end-group.
// These options are recongized for compatibility but ignored.
def end_group: F<"end-group">;
def end_group_paren: Flag<["-"], ")">;
def start_group: F<"start-group">;
def start_group_paren: Flag<["-"], "(">;
def: Separate<["-"], "f">, Alias<auxiliary>, HelpText<"Alias for --auxiliary">;
def: F<"call_shared">, Alias<Bdynamic>, HelpText<"Alias for --Bdynamic">;
def: F<"dy">, Alias<Bdynamic>, HelpText<"Alias for --Bdynamic">;
def: F<"dn">, Alias<Bstatic>, HelpText<"Alias for --Bstatic">;
def: F<"non_shared">, Alias<Bstatic>, HelpText<"Alias for --Bstatic">;
def: F<"static">, Alias<Bstatic>, HelpText<"Alias for --Bstatic">;
def: Flag<["-"], "d">, Alias<define_common>, HelpText<"Alias for --define-common">;
def: F<"dc">, Alias<define_common>, HelpText<"Alias for --define-common">;
def: F<"dp">, Alias<define_common>, HelpText<"Alias for --define-common">;
def: Flag<["-"], "x">, Alias<discard_all>, HelpText<"Alias for --discard-all">;
def: Flag<["-"], "X">, Alias<discard_locals>, HelpText<"Alias for --discard-locals">;
def: Flag<["-"], "q">, Alias<emit_relocs>, HelpText<"Alias for --emit-relocs">;
def: Flag<["-"], ")">, Alias<end_group>, HelpText<"Alias for --end-group">;
def: JoinedOrSeparate<["-"], "e">, Alias<entry>, HelpText<"Alias for --entry">;
def: Flag<["-"], "E">, Alias<export_dynamic>, HelpText<"Alias for --export-dynamic">;
def: Separate<["-"], "F">, Alias<filter>, HelpText<"Alias for --filter">;
def: Separate<["-"], "b">, Alias<format>, HelpText<"Alias for --format">;
def: JoinedOrSeparate<["-"], "l">, Alias<library>, HelpText<"Alias for --library">;
def: JoinedOrSeparate<["-"], "L">, Alias<library_path>, HelpText<"Alias for --library-path">;
def: F<"no-pic-executable">, Alias<no_pie>, HelpText<"Alias for --no-pie">;
def: Flag<["-"], "N">, Alias<omagic>, HelpText<"Alias for --omagic">;
def: Joined<["--"], "output=">, Alias<o>, HelpText<"Alias for -o">;
def: Separate<["--"], "output">, Alias<o>, HelpText<"Alias for -o">;
def: F<"pic-executable">, Alias<pie>, HelpText<"Alias for --pie">;
def: Flag<["-"], "M">, Alias<print_map>, HelpText<"Alias for --print-map">;
def: Flag<["-"], "r">, Alias<relocatable>, HelpText<"Alias for --relocatable">;
def: JoinedOrSeparate<["-"], "R">, Alias<rpath>, HelpText<"Alias for --rpath">;
def: JoinedOrSeparate<["-"], "T">, Alias<script>, HelpText<"Alias for --script">;
def: F<"Bshareable">, Alias<shared>, HelpText<"Alias for --shared">;
def: JoinedOrSeparate<["-"], "h">, Alias<soname>, HelpText<"Alias for --soname">;
def: Flag<["-"], "(">, Alias<start_group>, HelpText<"Alias for --start-group">;
def: Flag<["-"], "s">, Alias<strip_all>, HelpText<"Alias for --strip-all">;
def: Flag<["-"], "S">, Alias<strip_debug>, HelpText<"Alias for --strip-debug">;
def: Flag<["-"], "t">, Alias<trace>, HelpText<"Alias for --trace">;
def: JoinedOrSeparate<["-"], "y">, Alias<trace_symbol>, HelpText<"Alias for --trace-symbol">;
def: Separate<["-", "--"], "Ttext-segment">, Alias<Ttext>, HelpText<"Alias for --Ttext">;
def: Joined<["-", "--"], "Ttext-segment=">, Alias<Ttext>, HelpText<"Alias for --Ttext">;
def: JoinedOrSeparate<["-"], "u">, Alias<undefined>, HelpText<"Alias for --undefined">;
def: Flag<["-"], "V">, Alias<version>, HelpText<"Alias for --version">;
// LTO-related options.
def lto_aa_pipeline: J<"lto-aa-pipeline=">,
HelpText<"AA pipeline to run during LTO. Used in conjunction with -lto-newpm-passes">;
def lto_debug_pass_manager: F<"lto-debug-pass-manager">,
HelpText<"Debug new pass manager">;
def lto_new_pass_manager: F<"lto-new-pass-manager">,
HelpText<"Use new pass manager">;
def lto_newpm_passes: J<"lto-newpm-passes=">,
HelpText<"Passes to run during LTO">;
def lto_O: J<"lto-O">, MetaVarName<"<opt-level>">,
HelpText<"Optimization level for LTO">;
def lto_partitions: J<"lto-partitions=">,
HelpText<"Number of LTO codegen partitions">;
def lto_sample_profile: J<"lto-sample-profile=">,
HelpText<"Sample profile file path">;
def disable_verify: F<"disable-verify">;
def mllvm: S<"mllvm">;
defm mllvm: Eq<"mllvm", "Additional arguments to forward to LLVM's option processing">;
def opt_remarks_filename: Separate<["--"], "opt-remarks-filename">,
HelpText<"YAML output file for optimization remarks">;
def opt_remarks_with_hotness: Flag<["--"], "opt-remarks-with-hotness">,
HelpText<"Include hotness informations in the optimization remarks file">;
defm plugin_opt: Eq<"plugin-opt">,
HelpText<"specifies LTO options for compatibility with GNU linkers">;
HelpText<"Include hotness information in the optimization remarks file">;
defm plugin_opt: Eq<"plugin-opt", "specifies LTO options for compatibility with GNU linkers">;
def save_temps: F<"save-temps">;
def thinlto_cache_dir: J<"thinlto-cache-dir=">,
HelpText<"Path to ThinLTO cached object file directory">;
def thinlto_cache_policy: S<"thinlto-cache-policy">,
HelpText<"Pruning policy for the ThinLTO cache">;
defm thinlto_cache_policy: Eq<"thinlto-cache-policy", "Pruning policy for the ThinLTO cache">;
def thinlto_jobs: J<"thinlto-jobs=">, HelpText<"Number of ThinLTO jobs">;
def: J<"plugin-opt=O">, Alias<lto_O>, HelpText<"Alias for -lto-O">;
def: F<"plugin-opt=debug-pass-manager">,
Alias<lto_debug_pass_manager>, HelpText<"Alias for -lto-debug-pass-manager">;
def: F<"plugin-opt=disable-verify">, Alias<disable_verify>, HelpText<"Alias for -disable-verify">;
def plugin_opt_dwo_dir_eq: J<"plugin-opt=dwo_dir=">,
HelpText<"Directory to store .dwo files when LTO and debug fission are used">;
def: J<"plugin-opt=jobs=">, Alias<thinlto_jobs>, HelpText<"Alias for -thinlto-jobs">;
def: J<"plugin-opt=lto-partitions=">, Alias<lto_partitions>, HelpText<"Alias for -lto-partitions">;
def plugin_opt_mcpu_eq: J<"plugin-opt=mcpu=">;
def: F<"plugin-opt=new-pass-manager">,
Alias<lto_new_pass_manager>, HelpText<"Alias for -lto-new-pass-manager">;
def plugin_opt_obj_path_eq: J<"plugin-opt=obj-path=">;
def: J<"plugin-opt=sample-profile=">,
Alias<lto_sample_profile>, HelpText<"Alias for -lto-sample-profile">;
def: F<"plugin-opt=save-temps">, Alias<save_temps>, HelpText<"Alias for -save-temps">;
def plugin_opt_thinlto_emit_imports_files: F<"plugin-opt=thinlto-emit-imports-files">;
def plugin_opt_thinlto_index_only: F<"plugin-opt=thinlto-index-only">;
def plugin_opt_thinlto_index_only_eq: J<"plugin-opt=thinlto-index-only=">;
def plugin_opt_thinlto_object_suffix_replace_eq: J<"plugin-opt=thinlto-object-suffix-replace=">;
def plugin_opt_thinlto_prefix_replace_eq: J<"plugin-opt=thinlto-prefix-replace=">;
// Ignore LTO plugin-related options.
// clang -flto passes -plugin and -plugin-opt to the linker. This is required
// for ld.gold and ld.bfd to get LTO working. But it's not for lld which doesn't
@ -407,31 +458,38 @@ def thinlto_jobs: J<"thinlto-jobs=">, HelpText<"Number of ThinLTO jobs">;
// just ignore the option on lld side as it's easier. In fact, the linker could
// be called 'ld' and understanding which linker is used would require parsing of
// --version output.
def plugin: S<"plugin">;
def plugin_eq: J<"plugin=">;
defm plugin: Eq<"plugin", "Ignored for compatibility with GNU linkers">;
def plugin_opt_fresolution_eq: J<"plugin-opt=-fresolution=">;
def plugin_opt_pass_through_eq: J<"plugin-opt=-pass-through=">;
def plugin_opt_thinlto: J<"plugin-opt=thinlto">;
def plugin_opt_slash: J<"plugin-opt=/">;
// Options listed below are silently ignored for now for compatibility.
def allow_shlib_undefined: F<"allow-shlib-undefined">;
def cref: F<"cref">;
def detect_odr_violations: F<"detect-odr-violations">;
def g: Flag<["-"], "g">;
def long_plt: F<"long-plt">;
def no_add_needed: F<"no-add-needed">;
def no_allow_shlib_undefined: F<"no-allow-shlib-undefined">;
def no_copy_dt_needed_entries: F<"no-copy-dt-needed-entries">;
def no_ctors_in_init_array: F<"no-ctors-in-init-array">;
def no_keep_memory: F<"no-keep-memory">;
def no_mmap_output_file: F<"no-mmap-output-file">;
def no_warn_common: F<"no-warn-common">;
def no_warn_mismatch: F<"no-warn-mismatch">;
def rpath_link: S<"rpath-link">;
def rpath_link_eq: J<"rpath-link=">;
def sort_common: F<"sort-common">;
def stats: F<"stats">;
def warn_execstack: F<"warn-execstack">;
def warn_once: F<"warn-once">;
def warn_shared_textrel: F<"warn-shared-textrel">;
def EB : F<"EB">;
def EL : F<"EL">;
def G: JoinedOrSeparate<["-"], "G">;
def Qy : F<"Qy">;
def: F<"allow-shlib-undefined">;
def: F<"detect-odr-violations">;
def: Flag<["-"], "g">;
def: F<"long-plt">;
def: F<"no-add-needed">;
def: F<"no-allow-shlib-undefined">;
def: F<"no-copy-dt-needed-entries">;
def: F<"no-ctors-in-init-array">;
def: F<"no-keep-memory">;
def: F<"no-mmap-output-file">;
def: F<"no-warn-mismatch">;
def: Separate<["--", "-"], "rpath-link">;
def: J<"rpath-link=">;
def: F<"sort-common">;
def: F<"stats">;
def: F<"warn-execstack">;
def: F<"warn-once">;
def: F<"warn-shared-textrel">;
def: F<"EB">;
def: F<"EL">;
def: JoinedOrSeparate<["-"], "G">;
def: F<"Qy">;
// Hidden option used for testing MIPS multi-GOT implementation.
defm mips_got_size:
Eq<"mips-got-size", "Max size of a single MIPS GOT. 0x10000 by default.">,
Flags<[HelpHidden]>;

81
contrib/llvm/tools/lld/ELF/OutputSections.cpp
View File

@ -10,11 +10,11 @@
#include "OutputSections.h"
#include "Config.h"
#include "LinkerScript.h"
#include "Strings.h"
#include "SymbolTable.h"
#include "SyntheticSections.h"
#include "Target.h"
#include "lld/Common/Memory.h"
#include "lld/Common/Strings.h"
#include "lld/Common/Threads.h"
#include "llvm/BinaryFormat/Dwarf.h"
#include "llvm/Support/Compression.h"
@ -25,15 +25,12 @@
using namespace llvm;
using namespace llvm::dwarf;
using namespace llvm::object;
using namespace llvm::support::endian;
using namespace llvm::ELF;
using namespace lld;
using namespace lld::elf;
uint8_t Out::First;
OutputSection *Out::Opd;
uint8_t *Out::OpdBuf;
PhdrEntry *Out::TlsPhdr;
OutputSection *Out::DebugInfo;
OutputSection *Out::ElfHeader;
@ -45,7 +42,9 @@ OutputSection *Out::FiniArray;
std::vector<OutputSection *> elf::OutputSections;
uint32_t OutputSection::getPhdrFlags() const {
uint32_t Ret = PF_R;
uint32_t Ret = 0;
if (Config->EMachine != EM_ARM || !(Flags & SHF_ARM_PURECODE))
Ret |= PF_R;
if (Flags & SHF_WRITE)
Ret |= PF_W;
if (Flags & SHF_EXECINSTR)
@ -70,9 +69,7 @@ void OutputSection::writeHeaderTo(typename ELFT::Shdr *Shdr) {
OutputSection::OutputSection(StringRef Name, uint32_t Type, uint64_t Flags)
: BaseCommand(OutputSectionKind),
SectionBase(Output, Name, Flags, /*Entsize*/ 0, /*Alignment*/ 1, Type,
/*Info*/ 0,
/*Link*/ 0),
SectionIndex(INT_MAX) {
/*Info*/ 0, /*Link*/ 0) {
Live = false;
}
@ -91,13 +88,15 @@ static bool canMergeToProgbits(unsigned Type) {
void OutputSection::addSection(InputSection *IS) {
if (!Live) {
// If IS is the first section to be added to this section,
// initialize Type and Entsize from IS.
// initialize Type, Entsize and flags from IS.
Live = true;
Type = IS->Type;
Entsize = IS->Entsize;
Flags = IS->Flags;
} else {
// Otherwise, check if new type or flags are compatible with existing ones.
if ((Flags & (SHF_ALLOC | SHF_TLS)) != (IS->Flags & (SHF_ALLOC | SHF_TLS)))
unsigned Mask = SHF_ALLOC | SHF_TLS | SHF_LINK_ORDER;
if ((Flags & Mask) != (IS->Flags & Mask))
error("incompatible section flags for " + Name + "\n>>> " + toString(IS) +
": 0x" + utohexstr(IS->Flags) + "\n>>> output section " + Name +
": 0x" + utohexstr(Flags));
@ -114,9 +113,14 @@ void OutputSection::addSection(InputSection *IS) {
}
IS->Parent = this;
Flags |= IS->Flags;
uint64_t AndMask =
Config->EMachine == EM_ARM ? (uint64_t)SHF_ARM_PURECODE : 0;
uint64_t OrMask = ~AndMask;
uint64_t AndFlags = (Flags & IS->Flags) & AndMask;
uint64_t OrFlags = (Flags | IS->Flags) & OrMask;
Flags = AndFlags | OrFlags;
Alignment = std::max(Alignment, IS->Alignment);
IS->OutSecOff = Size++;
// If this section contains a table of fixed-size entries, sh_entsize
// holds the element size. If it contains elements of different size we
@ -134,8 +138,8 @@ void OutputSection::addSection(InputSection *IS) {
}
}
void elf::sortByOrder(MutableArrayRef<InputSection *> In,
std::function<int(InputSectionBase *S)> Order) {
static void sortByOrder(MutableArrayRef<InputSection *> In,
llvm::function_ref<int(InputSectionBase *S)> Order) {
typedef std::pair<int, InputSection *> Pair;
auto Comp = [](const Pair &A, const Pair &B) { return A.first < B.first; };
@ -158,11 +162,11 @@ bool OutputSection::classof(const BaseCommand *C) {
return C->Kind == OutputSectionKind;
}
void OutputSection::sort(std::function<int(InputSectionBase *S)> Order) {
void OutputSection::sort(llvm::function_ref<int(InputSectionBase *S)> Order) {
assert(Live);
assert(SectionCommands.size() == 1);
sortByOrder(cast<InputSectionDescription>(SectionCommands[0])->Sections,
Order);
for (BaseCommand *B : SectionCommands)
if (auto *ISD = dyn_cast<InputSectionDescription>(B))
sortByOrder(ISD->Sections, Order);
}
// Fill [Buf, Buf + Size) with Filler.
@ -205,11 +209,11 @@ static void writeInt(uint8_t *Buf, uint64_t Data, uint64_t Size) {
if (Size == 1)
*Buf = Data;
else if (Size == 2)
write16(Buf, Data, Config->Endianness);
write16(Buf, Data);
else if (Size == 4)
write32(Buf, Data, Config->Endianness);
write32(Buf, Data);
else if (Size == 8)
write64(Buf, Data, Config->Endianness);
write64(Buf, Data);
else
llvm_unreachable("unsupported Size argument");
}
@ -231,12 +235,7 @@ template <class ELFT> void OutputSection::writeTo(uint8_t *Buf) {
}
// Write leading padding.
std::vector<InputSection *> Sections;
for (BaseCommand *Cmd : SectionCommands)
if (auto *ISD = dyn_cast<InputSectionDescription>(Cmd))
for (InputSection *IS : ISD->Sections)
if (IS->Live)
Sections.push_back(IS);
std::vector<InputSection *> Sections = getInputSections(this);
uint32_t Filler = getFiller();
if (Filler)
fill(Buf, Sections.empty() ? Size : Sections[0]->OutSecOff, Filler);
@ -281,17 +280,13 @@ static void finalizeShtGroup(OutputSection *OS,
}
template <class ELFT> void OutputSection::finalize() {
InputSection *First = nullptr;
for (BaseCommand *Base : SectionCommands) {
if (auto *ISD = dyn_cast<InputSectionDescription>(Base)) {
if (ISD->Sections.empty())
continue;
if (First == nullptr)
First = ISD->Sections.front();
}
if (isa<ByteCommand>(Base) && Type == SHT_NOBITS)
Type = SHT_PROGBITS;
}
if (Type == SHT_NOBITS)
for (BaseCommand *Base : SectionCommands)
if (isa<ByteCommand>(Base))
Type = SHT_PROGBITS;
std::vector<InputSection *> V = getInputSections(this);
InputSection *First = V.empty() ? nullptr : V[0];
if (Flags & SHF_LINK_ORDER) {
// We must preserve the link order dependency of sections with the
@ -367,8 +362,6 @@ static bool compCtors(const InputSection *A, const InputSection *B) {
assert(Y.startswith(".ctors") || Y.startswith(".dtors"));
X = X.substr(6);
Y = Y.substr(6);
if (X.empty() && Y.empty())
return false;
return X < Y;
}
@ -394,6 +387,14 @@ int elf::getPriority(StringRef S) {
return V;
}
std::vector<InputSection *> elf::getInputSections(OutputSection *OS) {
std::vector<InputSection *> Ret;
for (BaseCommand *Base : OS->SectionCommands)
if (auto *ISD = dyn_cast<InputSectionDescription>(Base))
Ret.insert(Ret.end(), ISD->Sections.begin(), ISD->Sections.end());
return Ret;
}
// Sorts input sections by section name suffixes, so that .foo.N comes
// before .foo.M if N < M. Used to sort .{init,fini}_array.N sections.
// We want to keep the original order if the priorities are the same

14
contrib/llvm/tools/lld/ELF/OutputSections.h
View File

@ -14,7 +14,6 @@
#include "InputSection.h"
#include "LinkerScript.h"
#include "Relocations.h"
#include "lld/Common/LLVM.h"
#include "llvm/MC/StringTableBuilder.h"
#include "llvm/Object/ELF.h"
@ -52,7 +51,7 @@ class OutputSection final : public BaseCommand, public SectionBase {
uint64_t getLMA() const { return PtLoad ? Addr + PtLoad->LMAOffset : Addr; }
template <typename ELFT> void writeHeaderTo(typename ELFT::Shdr *SHdr);
unsigned SectionIndex;
uint32_t SectionIndex = UINT32_MAX;
unsigned SortRank;
uint32_t getPhdrFlags() const;
@ -100,13 +99,16 @@ class OutputSection final : public BaseCommand, public SectionBase {
std::string Location;
std::string MemoryRegionName;
std::string LMARegionName;
bool NonAlloc = false;
bool Noload = false;
bool ExpressionsUseSymbols = false;
bool InOverlay = false;
template <class ELFT> void finalize();
template <class ELFT> void writeTo(uint8_t *Buf);
template <class ELFT> void maybeCompress();
void sort(std::function<int(InputSectionBase *S)> Order);
void sort(llvm::function_ref<int(InputSectionBase *S)> Order);
void sortInitFini();
void sortCtorsDtors();
@ -120,13 +122,13 @@ class OutputSection final : public BaseCommand, public SectionBase {
int getPriority(StringRef S);
std::vector<InputSection *> getInputSections(OutputSection* OS);
// All output sections that are handled by the linker specially are
// globally accessible. Writer initializes them, so don't use them
// until Writer is initialized.
struct Out {
static uint8_t First;
static OutputSection *Opd;
static uint8_t *OpdBuf;
static PhdrEntry *TlsPhdr;
static OutputSection *DebugInfo;
static OutputSection *ElfHeader;
@ -143,8 +145,6 @@ namespace lld {
namespace elf {
uint64_t getHeaderSize();
void sortByOrder(llvm::MutableArrayRef<InputSection *> In,
std::function<int(InputSectionBase *S)> Order);
extern std::vector<OutputSection *> OutputSections;
} // namespace elf

826
contrib/llvm/tools/lld/ELF/Relocations.cpp
View File

File diff suppressed because it is too large Load Diff

21
contrib/llvm/tools/lld/ELF/Relocations.h
View File

@ -21,7 +21,7 @@ class Symbol;
class InputSection;
class InputSectionBase;
class OutputSection;
class OutputSection;
class SectionBase;
// Represents a relocation type, such as R_X86_64_PC32 or R_ARM_THM_CALL.
typedef uint32_t RelType;
@ -32,6 +32,7 @@ typedef uint32_t RelType;
enum RelExpr {
R_INVALID,
R_ABS,
R_ADDEND,
R_ARM_SBREL,
R_GOT,
R_GOTONLY_PC,
@ -58,27 +59,33 @@ enum RelExpr {
R_PLT,
R_PLT_PAGE_PC,
R_PLT_PC,
R_PPC_OPD,
R_PPC_PLT_OPD,
R_PPC_CALL,
R_PPC_CALL_PLT,
R_PPC_TOC,
R_RELAX_GOT_PC,
R_RELAX_GOT_PC_NOPIC,
R_RELAX_TLS_GD_TO_IE,
R_RELAX_TLS_GD_TO_IE_ABS,
R_RELAX_TLS_GD_TO_IE_END,
R_RELAX_TLS_GD_TO_IE_GOT_OFF,
R_RELAX_TLS_GD_TO_IE_PAGE_PC,
R_RELAX_TLS_GD_TO_LE,
R_RELAX_TLS_GD_TO_LE_NEG,
R_RELAX_TLS_IE_TO_LE,
R_RELAX_TLS_LD_TO_LE,
R_RELAX_TLS_LD_TO_LE_ABS,
R_SIZE,
R_TLS,
R_TLSDESC,
R_TLSDESC_CALL,
R_TLSDESC_PAGE,
R_TLSGD,
R_TLSGD_GOT,
R_TLSGD_GOT_FROM_END,
R_TLSGD_PC,
R_TLSLD,
R_TLSLD_GOT,
R_TLSLD_GOT_FROM_END,
R_TLSLD_GOT_OFF,
R_TLSLD_HINT,
R_TLSLD_PC,
};
@ -150,7 +157,7 @@ class ThunkCreator {
void forEachInputSectionDescription(
ArrayRef<OutputSection *> OutputSections,
std::function<void(OutputSection *, InputSectionDescription *)> Fn);
llvm::function_ref<void(OutputSection *, InputSectionDescription *)> Fn);
std::pair<Thunk *, bool> getThunk(Symbol &Sym, RelType Type, uint64_t Src);
@ -160,6 +167,8 @@ class ThunkCreator {
bool normalizeExistingThunk(Relocation &Rel, uint64_t Src);
// Record all the available Thunks for a Symbol
llvm::DenseMap<std::pair<SectionBase *, uint64_t>, std::vector<Thunk *>>
ThunkedSymbolsBySection;
llvm::DenseMap<Symbol *, std::vector<Thunk *>> ThunkedSymbols;
// Find a Thunk from the Thunks symbol definition, we can use this to find

10
contrib/llvm/tools/lld/ELF/ScriptLexer.cpp
View File

@ -66,8 +66,6 @@ size_t ScriptLexer::getColumnNumber() {
std::string ScriptLexer::getCurrentLocation() {
std::string Filename = getCurrentMB().getBufferIdentifier();
if (!Pos)
return Filename;
return (Filename + ":" + Twine(getLineNumber())).str();
}
@ -116,8 +114,9 @@ void ScriptLexer::tokenize(MemoryBufferRef MB) {
}
// ">foo" is parsed to ">" and "foo", but ">>" is parsed to ">>".
// "|", "||", "&" and "&&" are different operators.
if (S.startswith("<<") || S.startswith("<=") || S.startswith(">>") ||
S.startswith(">=")) {
S.startswith(">=") || S.startswith("||") || S.startswith("&&")) {
Vec.push_back(S.substr(0, 2));
S = S.substr(2);
continue;
@ -282,10 +281,7 @@ static bool encloses(StringRef S, StringRef T) {
MemoryBufferRef ScriptLexer::getCurrentMB() {
// Find input buffer containing the current token.
assert(!MBs.empty());
if (!Pos)
return MBs[0];
assert(!MBs.empty() && Pos > 0);
for (MemoryBufferRef MB : MBs)
if (encloses(MB.getBuffer(), Tokens[Pos - 1]))
return MB;

291
contrib/llvm/tools/lld/ELF/ScriptParser.cpp
View File

@ -63,6 +63,7 @@ class ScriptParser final : ScriptLexer {
void readExtern();
void readGroup();
void readInclude();
void readInput();
void readMemory();
void readOutput();
void readOutputArch();
@ -74,12 +75,14 @@ class ScriptParser final : ScriptLexer {
void readVersion();
void readVersionScriptCommand();
SymbolAssignment *readAssignment(StringRef Name);
SymbolAssignment *readSymbolAssignment(StringRef Name);
ByteCommand *readByteCommand(StringRef Tok);
uint32_t readFill();
uint32_t parseFill(StringRef Tok);
void readSectionAddressType(OutputSection *Cmd);
OutputSection *readOverlaySectionDescription();
OutputSection *readOutputSectionDescription(StringRef OutSec);
std::vector<BaseCommand *> readOverlay();
std::vector<StringRef> readOutputSectionPhdrs();
InputSectionDescription *readInputSectionDescription(StringRef Tok);
StringMatcher readFilePatterns();
@ -88,16 +91,16 @@ class ScriptParser final : ScriptLexer {
unsigned readPhdrType();
SortSectionPolicy readSortKind();
SymbolAssignment *readProvideHidden(bool Provide, bool Hidden);
SymbolAssignment *readProvideOrAssignment(StringRef Tok);
SymbolAssignment *readAssignment(StringRef Tok);
void readSort();
AssertCommand *readAssert();
Expr readAssertExpr();
Expr readAssert();
Expr readConstant();
Expr getPageSize();
uint64_t readMemoryAssignment(StringRef, StringRef, StringRef);
std::pair<uint32_t, uint32_t> readMemoryAttributes();
Expr combine(StringRef Op, Expr L, Expr R);
Expr readExpr();
Expr readExpr1(Expr Lhs, int MinPrec);
StringRef readParenLiteral();
@ -157,17 +160,6 @@ static ExprValue sub(ExprValue A, ExprValue B) {
return {A.Sec, false, A.getSectionOffset() - B.getValue(), A.Loc};
}
static ExprValue mul(ExprValue A, ExprValue B) {
return A.getValue() * B.getValue();
}
static ExprValue div(ExprValue A, ExprValue B) {
if (uint64_t BV = B.getValue())
return A.getValue() / BV;
error("division by zero");
return 0;
}
static ExprValue bitAnd(ExprValue A, ExprValue B) {
moveAbsRight(A, B);
return {A.Sec, A.ForceAbsolute,
@ -237,16 +229,16 @@ void ScriptParser::readLinkerScript() {
if (Tok == ";")
continue;
if (Tok == "ASSERT") {
Script->SectionCommands.push_back(readAssert());
} else if (Tok == "ENTRY") {
if (Tok == "ENTRY") {
readEntry();
} else if (Tok == "EXTERN") {
readExtern();
} else if (Tok == "GROUP" || Tok == "INPUT") {
} else if (Tok == "GROUP") {
readGroup();
} else if (Tok == "INCLUDE") {
readInclude();
} else if (Tok == "INPUT") {
readInput();
} else if (Tok == "MEMORY") {
readMemory();
} else if (Tok == "OUTPUT") {
@ -265,7 +257,7 @@ void ScriptParser::readLinkerScript() {
readSections();
} else if (Tok == "VERSION") {
readVersion();
} else if (SymbolAssignment *Cmd = readProvideOrAssignment(Tok)) {
} else if (SymbolAssignment *Cmd = readAssignment(Tok)) {
Script->SectionCommands.push_back(Cmd);
} else {
setError("unknown directive: " + Tok);
@ -336,13 +328,12 @@ void ScriptParser::readExtern() {
}
void ScriptParser::readGroup() {
expect("(");
while (!errorCount() && !consume(")")) {
if (consume("AS_NEEDED"))
readAsNeeded();
else
addFile(unquote(next()));
}
bool Orig = InputFile::IsInGroup;
InputFile::IsInGroup = true;
readInput();
InputFile::IsInGroup = Orig;
if (!Orig)
++InputFile::NextGroupId;
}
void ScriptParser::readInclude() {
@ -353,7 +344,7 @@ void ScriptParser::readInclude() {
return;
}
if (Optional<std::string> Path = searchLinkerScript(Tok)) {
if (Optional<std::string> Path = searchScript(Tok)) {
if (Optional<MemoryBufferRef> MB = readFile(*Path))
tokenize(*MB);
return;
@ -361,6 +352,16 @@ void ScriptParser::readInclude() {
setError("cannot find linker script " + Tok);
}
void ScriptParser::readInput() {
expect("(");
while (!errorCount() && !consume(")")) {
if (consume("AS_NEEDED"))
readAsNeeded();
else
addFile(unquote(next()));
}
}
void ScriptParser::readOutput() {
// -o <file> takes predecence over OUTPUT(<file>).
expect("(");
@ -437,6 +438,49 @@ void ScriptParser::readSearchDir() {
expect(")");
}
// This reads an overlay description. Overlays are used to describe output
// sections that use the same virtual memory range and normally would trigger
// linker's sections sanity check failures.
// https://sourceware.org/binutils/docs/ld/Overlay-Description.html#Overlay-Description
std::vector<BaseCommand *> ScriptParser::readOverlay() {
// VA and LMA expressions are optional, though for simplicity of
// implementation we assume they are not. That is what OVERLAY was designed
// for first of all: to allow sections with overlapping VAs at different LMAs.
Expr AddrExpr = readExpr();
expect(":");
expect("AT");
Expr LMAExpr = readParenExpr();
expect("{");
std::vector<BaseCommand *> V;
OutputSection *Prev = nullptr;
while (!errorCount() && !consume("}")) {
// VA is the same for all sections. The LMAs are consecutive in memory
// starting from the base load address specified.
OutputSection *OS = readOverlaySectionDescription();
OS->AddrExpr = AddrExpr;
if (Prev)
OS->LMAExpr = [=] { return Prev->getLMA() + Prev->Size; };
else
OS->LMAExpr = LMAExpr;
V.push_back(OS);
Prev = OS;
}
// According to the specification, at the end of the overlay, the location
// counter should be equal to the overlay base address plus size of the
// largest section seen in the overlay.
// Here we want to create the Dot assignment command to achieve that.
Expr MoveDot = [=] {
uint64_t Max = 0;
for (BaseCommand *Cmd : V)
Max = std::max(Max, cast<OutputSection>(Cmd)->Size);
return AddrExpr().getValue() + Max;
};
V.push_back(make<SymbolAssignment>(".", MoveDot, getCurrentLocation()));
return V;
}
void ScriptParser::readSections() {
Script->HasSectionsCommand = true;
@ -446,26 +490,48 @@ void ScriptParser::readSections() {
Config->SingleRoRx = true;
expect("{");
std::vector<BaseCommand *> V;
while (!errorCount() && !consume("}")) {
StringRef Tok = next();
BaseCommand *Cmd = readProvideOrAssignment(Tok);
if (!Cmd) {
if (Tok == "ASSERT")
Cmd = readAssert();
else
Cmd = readOutputSectionDescription(Tok);
if (Tok == "OVERLAY") {
for (BaseCommand *Cmd : readOverlay())
V.push_back(Cmd);
continue;
}
Script->SectionCommands.push_back(Cmd);
if (BaseCommand *Cmd = readAssignment(Tok))
V.push_back(Cmd);
else
V.push_back(readOutputSectionDescription(Tok));
}
if (!atEOF() && consume("INSERT")) {
std::vector<BaseCommand *> *Dest = nullptr;
if (consume("AFTER"))
Dest = &Script->InsertAfterCommands[next()];
else if (consume("BEFORE"))
Dest = &Script->InsertBeforeCommands[next()];
else
setError("expected AFTER/BEFORE, but got '" + next() + "'");
if (Dest)
Dest->insert(Dest->end(), V.begin(), V.end());
return;
}
Script->SectionCommands.insert(Script->SectionCommands.end(), V.begin(),
V.end());
}
static int precedence(StringRef Op) {
return StringSwitch<int>(Op)
.Cases("*", "/", 5)
.Cases("+", "-", 4)
.Cases("<<", ">>", 3)
.Cases("<", "<=", ">", ">=", "==", "!=", 2)
.Cases("&", "|", 1)
.Cases("*", "/", "%", 8)
.Cases("+", "-", 7)
.Cases("<<", ">>", 6)
.Cases("<", "<=", ">", ">=", "==", "!=", 5)
.Case("&", 4)
.Case("|", 3)
.Case("&&", 2)
.Case("||", 1)
.Default(-1);
}
@ -588,11 +654,7 @@ void ScriptParser::readSort() {
expect(")");
}
AssertCommand *ScriptParser::readAssert() {
return make<AssertCommand>(readAssertExpr());
}
Expr ScriptParser::readAssertExpr() {
Expr ScriptParser::readAssert() {
expect("(");
Expr E = readExpr();
expect(",");
@ -617,12 +679,14 @@ uint32_t ScriptParser::readFill() {
return V;
}
// Reads an expression and/or the special directive "(NOLOAD)" for an
// output section definition.
// Reads an expression and/or the special directive for an output
// section definition. Directive is one of following: "(NOLOAD)",
// "(COPY)", "(INFO)" or "(OVERLAY)".
//
// An output section name can be followed by an address expression
// and/or by "(NOLOAD)". This grammar is not LL(1) because "(" can be
// interpreted as either the beginning of some expression or "(NOLOAD)".
// and/or directive. This grammar is not LL(1) because "(" can be
// interpreted as either the beginning of some expression or beginning
// of directive.
//
// https://sourceware.org/binutils/docs/ld/Output-Section-Address.html
// https://sourceware.org/binutils/docs/ld/Output-Section-Type.html
@ -633,6 +697,11 @@ void ScriptParser::readSectionAddressType(OutputSection *Cmd) {
Cmd->Noload = true;
return;
}
if (consume("COPY") || consume("INFO") || consume("OVERLAY")) {
expect(")");
Cmd->NonAlloc = true;
return;
}
Cmd->AddrExpr = readExpr();
expect(")");
} else {
@ -657,10 +726,23 @@ static Expr checkAlignment(Expr E, std::string &Loc) {
};
}
OutputSection *ScriptParser::readOverlaySectionDescription() {
OutputSection *Cmd =
Script->createOutputSection(next(), getCurrentLocation());
Cmd->InOverlay = true;
expect("{");
while (!errorCount() && !consume("}"))
Cmd->SectionCommands.push_back(readInputSectionRules(next()));
Cmd->Phdrs = readOutputSectionPhdrs();
return Cmd;
}
OutputSection *ScriptParser::readOutputSectionDescription(StringRef OutSec) {
OutputSection *Cmd =
Script->createOutputSection(OutSec, getCurrentLocation());
size_t SymbolsReferenced = Script->ReferencedSymbols.size();
if (peek() != ":")
readSectionAddressType(Cmd);
expect(":");
@ -684,13 +766,10 @@ OutputSection *ScriptParser::readOutputSectionDescription(StringRef OutSec) {
StringRef Tok = next();
if (Tok == ";") {
// Empty commands are allowed. Do nothing here.
} else if (SymbolAssignment *Assign = readProvideOrAssignment(Tok)) {
} else if (SymbolAssignment *Assign = readAssignment(Tok)) {
Cmd->SectionCommands.push_back(Assign);
} else if (ByteCommand *Data = readByteCommand(Tok)) {
Cmd->SectionCommands.push_back(Data);
} else if (Tok == "ASSERT") {
Cmd->SectionCommands.push_back(readAssert());
expect(";");
} else if (Tok == "CONSTRUCTORS") {
// CONSTRUCTORS is a keyword to make the linker recognize C++ ctors/dtors
// by name. This is for very old file formats such as ECOFF/XCOFF.
@ -727,6 +806,8 @@ OutputSection *ScriptParser::readOutputSectionDescription(StringRef OutSec) {
// Consume optional comma following output section command.
consume(",");
if (Script->ReferencedSymbols.size() > SymbolsReferenced)
Cmd->ExpressionsUseSymbols = true;
return Cmd;
}
@ -749,30 +830,39 @@ uint32_t ScriptParser::parseFill(StringRef Tok) {
SymbolAssignment *ScriptParser::readProvideHidden(bool Provide, bool Hidden) {
expect("(");
SymbolAssignment *Cmd = readAssignment(next());
SymbolAssignment *Cmd = readSymbolAssignment(next());
Cmd->Provide = Provide;
Cmd->Hidden = Hidden;
expect(")");
expect(";");
return Cmd;
}
SymbolAssignment *ScriptParser::readProvideOrAssignment(StringRef Tok) {
SymbolAssignment *ScriptParser::readAssignment(StringRef Tok) {
// Assert expression returns Dot, so this is equal to ".=."
if (Tok == "ASSERT")
return make<SymbolAssignment>(".", readAssert(), getCurrentLocation());
size_t OldPos = Pos;
SymbolAssignment *Cmd = nullptr;
if (peek() == "=" || peek() == "+=") {
Cmd = readAssignment(Tok);
expect(";");
} else if (Tok == "PROVIDE") {
if (peek() == "=" || peek() == "+=")
Cmd = readSymbolAssignment(Tok);
else if (Tok == "PROVIDE")
Cmd = readProvideHidden(true, false);
} else if (Tok == "HIDDEN") {
else if (Tok == "HIDDEN")
Cmd = readProvideHidden(false, true);
} else if (Tok == "PROVIDE_HIDDEN") {
else if (Tok == "PROVIDE_HIDDEN")
Cmd = readProvideHidden(true, true);
if (Cmd) {
Cmd->CommandString =
Tok.str() + " " +
llvm::join(Tokens.begin() + OldPos, Tokens.begin() + Pos, " ");
expect(";");
}
return Cmd;
}
SymbolAssignment *ScriptParser::readAssignment(StringRef Name) {
SymbolAssignment *ScriptParser::readSymbolAssignment(StringRef Name) {
StringRef Op = next();
assert(Op == "=" || Op == "+=");
Expr E = readExpr();
@ -795,15 +885,31 @@ Expr ScriptParser::readExpr() {
return E;
}
static Expr combine(StringRef Op, Expr L, Expr R) {
Expr ScriptParser::combine(StringRef Op, Expr L, Expr R) {
if (Op == "+")
return [=] { return add(L(), R()); };
if (Op == "-")
return [=] { return sub(L(), R()); };
if (Op == "*")
return [=] { return mul(L(), R()); };
if (Op == "/")
return [=] { return div(L(), R()); };
return [=] { return L().getValue() * R().getValue(); };
if (Op == "/") {
std::string Loc = getCurrentLocation();
return [=]() -> uint64_t {
if (uint64_t RV = R().getValue())
return L().getValue() / RV;
error(Loc + ": division by zero");
return 0;
};
}
if (Op == "%") {
std::string Loc = getCurrentLocation();
return [=]() -> uint64_t {
if (uint64_t RV = R().getValue())
return L().getValue() % RV;
error(Loc + ": modulo by zero");
return 0;
};
}
if (Op == "<<")
return [=] { return L().getValue() << R().getValue(); };
if (Op == ">>")
@ -820,6 +926,10 @@ static Expr combine(StringRef Op, Expr L, Expr R) {
return [=] { return L().getValue() == R().getValue(); };
if (Op == "!=")
return [=] { return L().getValue() != R().getValue(); };
if (Op == "||")
return [=] { return L().getValue() || R().getValue(); };
if (Op == "&&")
return [=] { return L().getValue() && R().getValue(); };
if (Op == "&")
return [=] { return bitAnd(L(), R()); };
if (Op == "|")
@ -873,20 +983,13 @@ Expr ScriptParser::readConstant() {
if (S == "MAXPAGESIZE")
return [] { return Config->MaxPageSize; };
setError("unknown constant: " + S);
return {};
return [] { return 0; };
}
// Parses Tok as an integer. It recognizes hexadecimal (prefixed with
// "0x" or suffixed with "H") and decimal numbers. Decimal numbers may
// have "K" (Ki) or "M" (Mi) suffixes.
static Optional<uint64_t> parseInt(StringRef Tok) {
// Negative number
if (Tok.startswith("-")) {
if (Optional<uint64_t> Val = parseInt(Tok.substr(1)))
return -*Val;
return None;
}
// Hexadecimal
uint64_t Val;
if (Tok.startswith_lower("0x")) {
@ -925,7 +1028,13 @@ ByteCommand *ScriptParser::readByteCommand(StringRef Tok) {
.Default(-1);
if (Size == -1)
return nullptr;
return make<ByteCommand>(readParenExpr(), Size);
size_t OldPos = Pos;
Expr E = readParenExpr();
std::string CommandString =
Tok.str() + " " +
llvm::join(Tokens.begin() + OldPos, Tokens.begin() + Pos, " ");
return make<ByteCommand>(E, Size, CommandString);
}
StringRef ScriptParser::readParenLiteral() {
@ -1006,7 +1115,7 @@ Expr ScriptParser::readPrimary() {
};
}
if (Tok == "ASSERT")
return readAssertExpr();
return readAssert();
if (Tok == "CONSTANT")
return readConstant();
if (Tok == "DATA_SEGMENT_ALIGN") {
@ -1043,8 +1152,10 @@ Expr ScriptParser::readPrimary() {
}
if (Tok == "LENGTH") {
StringRef Name = readParenLiteral();
if (Script->MemoryRegions.count(Name) == 0)
if (Script->MemoryRegions.count(Name) == 0) {
setError("memory region not defined: " + Name);
return [] { return 0; };
}
return [=] { return Script->MemoryRegions[Name]->Length; };
}
if (Tok == "LOADADDR") {
@ -1055,10 +1166,22 @@ Expr ScriptParser::readPrimary() {
return Cmd->getLMA();
};
}
if (Tok == "MAX" || Tok == "MIN") {
expect("(");
Expr A = readExpr();
expect(",");
Expr B = readExpr();
expect(")");
if (Tok == "MIN")
return [=] { return std::min(A().getValue(), B().getValue()); };
return [=] { return std::max(A().getValue(), B().getValue()); };
}
if (Tok == "ORIGIN") {
StringRef Name = readParenLiteral();
if (Script->MemoryRegions.count(Name) == 0)
if (Script->MemoryRegions.count(Name) == 0) {
setError("memory region not defined: " + Name);
return [] { return 0; };
}
return [=] { return Script->MemoryRegions[Name]->Origin; };
}
if (Tok == "SEGMENT_START") {
@ -1240,6 +1363,9 @@ ScriptParser::readSymbols() {
// Reads an "extern C++" directive, e.g.,
// "extern "C++" { ns::*; "f(int, double)"; };"
//
// The last semicolon is optional. E.g. this is OK:
// "extern "C++" { ns::*; "f(int, double)" };"
std::vector<SymbolVersion> ScriptParser::readVersionExtern() {
StringRef Tok = next();
bool IsCXX = Tok == "\"C++\"";
@ -1252,6 +1378,8 @@ std::vector<SymbolVersion> ScriptParser::readVersionExtern() {
StringRef Tok = next();
bool HasWildcard = !Tok.startswith("\"") && hasWildcard(Tok);
Ret.push_back({unquote(Tok), IsCXX, HasWildcard});
if (consume("}"))
return Ret;
expect(";");
}
@ -1291,11 +1419,10 @@ void ScriptParser::readMemory() {
uint64_t Length = readMemoryAssignment("LENGTH", "len", "l");
// Add the memory region to the region map.
if (Script->MemoryRegions.count(Name))
setError("region '" + Name + "' already defined");
MemoryRegion *MR =
make<MemoryRegion>(Name, Origin, Length, Flags, NegFlags);
Script->MemoryRegions[Name] = MR;
if (!Script->MemoryRegions.insert({Name, MR}).second)
setError("region '" + Name + "' already defined");
}
}

62
contrib/llvm/tools/lld/ELF/Strings.cpp
View File

@ -1,62 +0,0 @@
//===- Strings.cpp -------------------------------------------------------===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "Strings.h"
#include "Config.h"
#include "lld/Common/ErrorHandler.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/Twine.h"
#include "llvm/Demangle/Demangle.h"
#include <algorithm>
#include <cstring>
using namespace llvm;
using namespace lld;
using namespace lld::elf;
StringMatcher::StringMatcher(ArrayRef<StringRef> Pat) {
for (StringRef S : Pat) {
Expected<GlobPattern> Pat = GlobPattern::create(S);
if (!Pat)
error(toString(Pat.takeError()));
else
Patterns.push_back(*Pat);
}
}
bool StringMatcher::match(StringRef S) const {
for (const GlobPattern &Pat : Patterns)
if (Pat.match(S))
return true;
return false;
}
// Converts a hex string (e.g. "deadbeef") to a vector.
std::vector<uint8_t> elf::parseHex(StringRef S) {
std::vector<uint8_t> Hex;
while (!S.empty()) {
StringRef B = S.substr(0, 2);
S = S.substr(2);
uint8_t H;
if (!to_integer(B, H, 16)) {
error("not a hexadecimal value: " + B);
return {};
}
Hex.push_back(H);
}
return Hex;
}
// Returns true if S is valid as a C language identifier.
bool elf::isValidCIdentifier(StringRef S) {
return !S.empty() && (isAlpha(S[0]) || S[0] == '_') &&
std::all_of(S.begin() + 1, S.end(),
[](char C) { return C == '_' || isAlnum(C); });
}

75
contrib/llvm/tools/lld/ELF/Strings.h
View File

@ -1,75 +0,0 @@
//===- Strings.h ------------------------------------------------*- C++ -*-===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#ifndef LLD_ELF_STRINGS_H
#define LLD_ELF_STRINGS_H
#include "lld/Common/LLVM.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/BitVector.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/Support/GlobPattern.h"
#include <vector>
namespace lld {
namespace elf {
std::vector<uint8_t> parseHex(StringRef S);
bool isValidCIdentifier(StringRef S);
// This is a lazy version of StringRef. String size is computed lazily
// when it is needed. It is more efficient than StringRef to instantiate
// if you have a string whose size is unknown.
//
// ELF string tables contain a lot of null-terminated strings.
// Most of them are not necessary for the linker because they are names
// of local symbols and the linker doesn't use local symbol names for
// name resolution. So, we use this class to represents strings read
// from string tables.
class StringRefZ {
public:
StringRefZ() : Start(nullptr), Size(0) {}
StringRefZ(const char *S, size_t Size) : Start(S), Size(Size) {}
/*implicit*/ StringRefZ(const char *S) : Start(S), Size(-1) {}
/*implicit*/ StringRefZ(llvm::StringRef S)
: Start(S.data()), Size(S.size()) {}
operator llvm::StringRef() const {
if (Size == (size_t)-1)
Size = strlen(Start);
return {Start, Size};
}
private:
const char *Start;
mutable size_t Size;
};
// This class represents multiple glob patterns.
class StringMatcher {
public:
StringMatcher() = default;
explicit StringMatcher(ArrayRef<StringRef> Pat);
bool match(StringRef S) const;
private:
std::vector<llvm::GlobPattern> Patterns;
};
inline ArrayRef<uint8_t> toArrayRef(StringRef S) {
return {(const uint8_t *)S.data(), S.size()};
}
} // namespace elf
} // namespace lld
#endif

290
contrib/llvm/tools/lld/ELF/SymbolTable.cpp
View File

@ -38,7 +38,7 @@ static InputFile *getFirstElf() {
return ObjectFiles[0];
if (!SharedFiles.empty())
return SharedFiles[0];
return nullptr;
return BitcodeFiles[0];
}
// All input object files must be for the same architecture
@ -82,6 +82,7 @@ template <class ELFT> void SymbolTable::addFile(InputFile *File) {
// Lazy object file
if (auto *F = dyn_cast<LazyObjFile>(File)) {
LazyObjFiles.push_back(F);
F->parse<ELFT>();
return;
}
@ -117,7 +118,7 @@ template <class ELFT> void SymbolTable::addFile(InputFile *File) {
// not in native object file format but in the LLVM bitcode format.
// This function compiles bitcode files into a few big native files
// using LLVM functions and replaces bitcode symbols with the results.
// Because all bitcode files that consist of a program are passed
// Because all bitcode files that the program consists of are passed
// to the compiler at once, it can do whole-program optimization.
template <class ELFT> void SymbolTable::addCombinedLTOObject() {
if (BitcodeFiles.empty())
@ -157,6 +158,12 @@ template <class ELFT> void SymbolTable::addSymbolWrap(StringRef Name) {
Symbol *Sym = find(Name);
if (!Sym)
return;
// Do not wrap the same symbol twice.
for (const WrappedSymbol &S : WrappedSymbols)
if (S.Sym == Sym)
return;
Symbol *Real = addUndefined<ELFT>(Saver.save("__real_" + Name));
Symbol *Wrap = addUndefined<ELFT>(Saver.save("__wrap_" + Name));
WrappedSymbols.push_back({Sym, Real, Wrap});
@ -189,7 +196,7 @@ void SymbolTable::applySymbolWrap() {
// First, make a copy of __real_sym.
Symbol *Real = nullptr;
if (W.Real->isDefined()) {
Real = (Symbol *)make<SymbolUnion>();
Real = reinterpret_cast<Symbol *>(make<SymbolUnion>());
memcpy(Real, W.Real, sizeof(SymbolUnion));
}
@ -237,8 +244,7 @@ std::pair<Symbol *, bool> SymbolTable::insert(StringRef Name) {
Symbol *Sym;
if (IsNew) {
Sym = (Symbol *)make<SymbolUnion>();
Sym->InVersionScript = false;
Sym = reinterpret_cast<Symbol *>(make<SymbolUnion>());
Sym->Visibility = STV_DEFAULT;
Sym->IsUsedInRegularObj = false;
Sym->ExportDynamic = false;
@ -297,26 +303,88 @@ Symbol *SymbolTable::addUndefined(StringRef Name, uint8_t Binding,
uint8_t Visibility = getVisibility(StOther);
std::tie(S, WasInserted) =
insert(Name, Type, Visibility, CanOmitFromDynSym, File);
// An undefined symbol with non default visibility must be satisfied
// in the same DSO.
if (WasInserted || (isa<SharedSymbol>(S) && Visibility != STV_DEFAULT)) {
replaceSymbol<Undefined>(S, File, Name, Binding, StOther, Type);
return S;
}
if (S->isShared() || S->isLazy() || (S->isUndefined() && Binding != STB_WEAK))
S->Binding = Binding;
if (Binding != STB_WEAK) {
if (!Config->GcSections && Binding != STB_WEAK)
if (auto *SS = dyn_cast<SharedSymbol>(S))
if (!Config->GcSections)
SS->getFile<ELFT>().IsNeeded = true;
}
if (auto *L = dyn_cast<Lazy>(S)) {
SS->getFile<ELFT>().IsNeeded = true;
if (S->isLazy()) {
// An undefined weak will not fetch archive members. See comment on Lazy in
// Symbols.h for the details.
if (Binding == STB_WEAK)
L->Type = Type;
else if (InputFile *F = L->fetch())
addFile<ELFT>(F);
if (Binding == STB_WEAK) {
S->Type = Type;
return S;
}
// Do extra check for --warn-backrefs.
//
// --warn-backrefs is an option to prevent an undefined reference from
// fetching an archive member written earlier in the command line. It can be
// used to keep compatibility with GNU linkers to some degree.
// I'll explain the feature and why you may find it useful in this comment.
//
// lld's symbol resolution semantics is more relaxed than traditional Unix
// linkers. For example,
//
// ld.lld foo.a bar.o
//
// succeeds even if bar.o contains an undefined symbol that has to be
// resolved by some object file in foo.a. Traditional Unix linkers don't
// allow this kind of backward reference, as they visit each file only once
// from left to right in the command line while resolving all undefined
// symbols at the moment of visiting.
//
// In the above case, since there's no undefined symbol when a linker visits
// foo.a, no files are pulled out from foo.a, and because the linker forgets
// about foo.a after visiting, it can't resolve undefined symbols in bar.o
// that could have been resolved otherwise.
//
// That lld accepts more relaxed form means that (besides it'd make more
// sense) you can accidentally write a command line or a build file that
// works only with lld, even if you have a plan to distribute it to wider
// users who may be using GNU linkers. With --warn-backrefs, you can detect
// a library order that doesn't work with other Unix linkers.
//
// The option is also useful to detect cyclic dependencies between static
// archives. Again, lld accepts
//
// ld.lld foo.a bar.a
//
// even if foo.a and bar.a depend on each other. With --warn-backrefs, it is
// handled as an error.
//
// Here is how the option works. We assign a group ID to each file. A file
// with a smaller group ID can pull out object files from an archive file
// with an equal or greater group ID. Otherwise, it is a reverse dependency
// and an error.
//
// A file outside --{start,end}-group gets a fresh ID when instantiated. All
// files within the same --{start,end}-group get the same group ID. E.g.
//
// ld.lld A B --start-group C D --end-group E
//
// A forms group 0. B form group 1. C and D (including their member object
// files) form group 2. E forms group 3. I think that you can see how this
// group assignment rule simulates the traditional linker's semantics.
bool Backref =
Config->WarnBackrefs && File && S->File->GroupId < File->GroupId;
fetchLazy<ELFT>(S);
// We don't report backward references to weak symbols as they can be
// overridden later.
if (Backref && S->Binding != STB_WEAK)
warn("backward reference detected: " + Name + " in " + toString(File) +
" refers to " + toString(S->File));
}
return S;
}
@ -384,7 +452,11 @@ Symbol *SymbolTable::addCommon(StringRef N, uint64_t Size, uint32_t Alignment,
bool WasInserted;
std::tie(S, WasInserted) = insert(N, Type, getVisibility(StOther),
/*CanOmitFromDynSym*/ false, &File);
int Cmp = compareDefined(S, WasInserted, Binding, N);
if (Cmp < 0)
return S;
if (Cmp > 0) {
auto *Bss = make<BssSection>("COMMON", Size, Alignment);
Bss->File = &File;
@ -392,45 +464,43 @@ Symbol *SymbolTable::addCommon(StringRef N, uint64_t Size, uint32_t Alignment,
InputSections.push_back(Bss);
replaceSymbol<Defined>(S, &File, N, Binding, StOther, Type, 0, Size, Bss);
} else if (Cmp == 0) {
auto *D = cast<Defined>(S);
auto *Bss = dyn_cast_or_null<BssSection>(D->Section);
if (!Bss) {
// Non-common symbols take precedence over common symbols.
if (Config->WarnCommon)
warn("common " + S->getName() + " is overridden");
return S;
}
return S;
}
auto *D = cast<Defined>(S);
auto *Bss = dyn_cast_or_null<BssSection>(D->Section);
if (!Bss) {
// Non-common symbols take precedence over common symbols.
if (Config->WarnCommon)
warn("multiple common of " + D->getName());
warn("common " + S->getName() + " is overridden");
return S;
}
Bss->Alignment = std::max(Bss->Alignment, Alignment);
if (Size > Bss->Size) {
D->File = Bss->File = &File;
D->Size = Bss->Size = Size;
}
if (Config->WarnCommon)
warn("multiple common of " + D->getName());
Bss->Alignment = std::max(Bss->Alignment, Alignment);
if (Size > Bss->Size) {
D->File = Bss->File = &File;
D->Size = Bss->Size = Size;
}
return S;
}
static void warnOrError(const Twine &Msg) {
if (Config->AllowMultipleDefinition)
warn(Msg);
else
error(Msg);
}
static void reportDuplicate(Symbol *Sym, InputFile *NewFile) {
warnOrError("duplicate symbol: " + toString(*Sym) + "\n>>> defined in " +
toString(Sym->File) + "\n>>> defined in " + toString(NewFile));
if (!Config->AllowMultipleDefinition)
error("duplicate symbol: " + toString(*Sym) + "\n>>> defined in " +
toString(Sym->File) + "\n>>> defined in " + toString(NewFile));
}
static void reportDuplicate(Symbol *Sym, InputSectionBase *ErrSec,
uint64_t ErrOffset) {
static void reportDuplicate(Symbol *Sym, InputFile *NewFile,
InputSectionBase *ErrSec, uint64_t ErrOffset) {
if (Config->AllowMultipleDefinition)
return;
Defined *D = cast<Defined>(Sym);
if (!D->Section || !ErrSec) {
reportDuplicate(Sym, ErrSec ? ErrSec->File : nullptr);
reportDuplicate(Sym, NewFile);
return;
}
@ -454,7 +524,7 @@ static void reportDuplicate(Symbol *Sym, InputSectionBase *ErrSec,
if (!Src2.empty())
Msg += Src2 + "\n>>> ";
Msg += Obj2;
warnOrError(Msg);
error(Msg);
}
Symbol *SymbolTable::addRegular(StringRef Name, uint8_t StOther, uint8_t Type,
@ -470,7 +540,8 @@ Symbol *SymbolTable::addRegular(StringRef Name, uint8_t StOther, uint8_t Type,
replaceSymbol<Defined>(S, File, Name, Binding, StOther, Type, Value, Size,
Section);
else if (Cmp == 0)
reportDuplicate(S, dyn_cast_or_null<InputSectionBase>(Section), Value);
reportDuplicate(S, File, dyn_cast_or_null<InputSectionBase>(Section),
Value);
return S;
}
@ -491,8 +562,8 @@ void SymbolTable::addShared(StringRef Name, SharedFile<ELFT> &File,
// An undefined symbol with non default visibility must be satisfied
// in the same DSO.
if (WasInserted || ((S->isUndefined() || S->isLazy()) &&
S->getVisibility() == STV_DEFAULT)) {
if (WasInserted ||
((S->isUndefined() || S->isLazy()) && S->Visibility == STV_DEFAULT)) {
uint8_t Binding = S->Binding;
bool WasUndefined = S->isUndefined();
replaceSymbol<SharedSymbol>(S, File, Name, Sym.getBinding(), Sym.st_other,
@ -531,85 +602,58 @@ Symbol *SymbolTable::find(StringRef Name) {
return SymVector[It->second];
}
template <class ELFT>
Symbol *SymbolTable::addLazyArchive(StringRef Name, ArchiveFile &F,
const object::Archive::Symbol Sym) {
// This is used to handle lazy symbols. May replace existent
// symbol with lazy version or request to Fetch it.
template <class ELFT, typename LazyT, typename... ArgT>
static void replaceOrFetchLazy(StringRef Name, InputFile &File,
llvm::function_ref<InputFile *()> Fetch,
ArgT &&... Arg) {
Symbol *S;
bool WasInserted;
std::tie(S, WasInserted) = insert(Name);
std::tie(S, WasInserted) = Symtab->insert(Name);
if (WasInserted) {
replaceSymbol<LazyArchive>(S, F, Sym, Symbol::UnknownType);
return S;
replaceSymbol<LazyT>(S, File, Symbol::UnknownType,
std::forward<ArgT>(Arg)...);
return;
}
if (!S->isUndefined())
return S;
return;
// An undefined weak will not fetch archive members. See comment on Lazy in
// Symbols.h for the details.
if (S->isWeak()) {
replaceSymbol<LazyArchive>(S, F, Sym, S->Type);
replaceSymbol<LazyT>(S, File, S->Type, std::forward<ArgT>(Arg)...);
S->Binding = STB_WEAK;
return S;
return;
}
std::pair<MemoryBufferRef, uint64_t> MBInfo = F.getMember(&Sym);
if (!MBInfo.first.getBuffer().empty())
addFile<ELFT>(createObjectFile(MBInfo.first, F.getName(), MBInfo.second));
return S;
if (InputFile *F = Fetch())
Symtab->addFile<ELFT>(F);
}
template <class ELFT>
void SymbolTable::addLazyArchive(StringRef Name, ArchiveFile &F,
const object::Archive::Symbol Sym) {
replaceOrFetchLazy<ELFT, LazyArchive>(Name, F, [&]() { return F.fetch(Sym); },
Sym);
}
template <class ELFT>
void SymbolTable::addLazyObject(StringRef Name, LazyObjFile &Obj) {
Symbol *S;
bool WasInserted;
std::tie(S, WasInserted) = insert(Name);
if (WasInserted) {
replaceSymbol<LazyObject>(S, Obj, Name, Symbol::UnknownType);
return;
}
if (!S->isUndefined())
return;
// See comment for addLazyArchive above.
if (S->isWeak())
replaceSymbol<LazyObject>(S, Obj, Name, S->Type);
else if (InputFile *F = Obj.fetch())
addFile<ELFT>(F);
replaceOrFetchLazy<ELFT, LazyObject>(Name, Obj, [&]() { return Obj.fetch(); },
Name);
}
// If we already saw this symbol, force loading its file.
template <class ELFT> void SymbolTable::fetchIfLazy(StringRef Name) {
if (Symbol *B = find(Name)) {
// Mark the symbol not to be eliminated by LTO
// even if it is a bitcode symbol.
B->IsUsedInRegularObj = true;
if (auto *L = dyn_cast<Lazy>(B))
if (InputFile *File = L->fetch())
addFile<ELFT>(File);
template <class ELFT> void SymbolTable::fetchLazy(Symbol *Sym) {
if (auto *S = dyn_cast<LazyArchive>(Sym)) {
if (InputFile *File = S->fetch())
addFile<ELFT>(File);
return;
}
}
// This function takes care of the case in which shared libraries depend on
// the user program (not the other way, which is usual). Shared libraries
// may have undefined symbols, expecting that the user program provides
// the definitions for them. An example is BSD's __progname symbol.
// We need to put such symbols to the main program's .dynsym so that
// shared libraries can find them.
// Except this, we ignore undefined symbols in DSOs.
template <class ELFT> void SymbolTable::scanShlibUndefined() {
for (InputFile *F : SharedFiles) {
for (StringRef U : cast<SharedFile<ELFT>>(F)->getUndefinedSymbols()) {
Symbol *Sym = find(U);
if (!Sym || !Sym->isDefined())
continue;
Sym->ExportDynamic = true;
// If -dynamic-list is given, the default version is set to
// VER_NDX_LOCAL, which prevents a symbol to be exported via .dynsym.
// Set to VER_NDX_GLOBAL so the symbol will be handled as if it were
// specified by -dynamic-list.
Sym->VersionId = VER_NDX_GLOBAL;
}
}
auto *S = cast<LazyObject>(Sym);
if (InputFile *File = cast<LazyObjFile>(S->File)->fetch())
addFile<ELFT>(File);
}
// Initialize DemangledSyms with a map from demangled symbols to symbol
@ -708,7 +752,7 @@ void SymbolTable::assignExactVersion(SymbolVersion Ver, uint16_t VersionId,
// Get a list of symbols which we need to assign the version to.
std::vector<Symbol *> Syms = findByVersion(Ver);
if (Syms.empty()) {
if (Config->NoUndefinedVersion)
if (!Config->UndefinedVersion)
error("version script assignment of '" + VersionName + "' to symbol '" +
Ver.Name + "' failed: symbol not defined");
return;
@ -722,10 +766,10 @@ void SymbolTable::assignExactVersion(SymbolVersion Ver, uint16_t VersionId,
if (Sym->getName().contains('@'))
continue;
if (Sym->InVersionScript)
warn("duplicate symbol '" + Ver.Name + "' in version script");
if (Sym->VersionId != Config->DefaultSymbolVersion &&
Sym->VersionId != VersionId)
error("duplicate symbol '" + Ver.Name + "' in version script");
Sym->VersionId = VersionId;
Sym->InVersionScript = true;
}
}
@ -773,6 +817,11 @@ void SymbolTable::scanVersionScript() {
Sym->parseSymbolVersion();
}
template void SymbolTable::addFile<ELF32LE>(InputFile *);
template void SymbolTable::addFile<ELF32BE>(InputFile *);
template void SymbolTable::addFile<ELF64LE>(InputFile *);
template void SymbolTable::addFile<ELF64BE>(InputFile *);
template void SymbolTable::addSymbolWrap<ELF32LE>(StringRef);
template void SymbolTable::addSymbolWrap<ELF32BE>(StringRef);
template void SymbolTable::addSymbolWrap<ELF64LE>(StringRef);
@ -797,16 +846,16 @@ template void SymbolTable::addCombinedLTOObject<ELF32BE>();
template void SymbolTable::addCombinedLTOObject<ELF64LE>();
template void SymbolTable::addCombinedLTOObject<ELF64BE>();
template Symbol *
template void
SymbolTable::addLazyArchive<ELF32LE>(StringRef, ArchiveFile &,
const object::Archive::Symbol);
template Symbol *
template void
SymbolTable::addLazyArchive<ELF32BE>(StringRef, ArchiveFile &,
const object::Archive::Symbol);
template Symbol *
template void
SymbolTable::addLazyArchive<ELF64LE>(StringRef, ArchiveFile &,
const object::Archive::Symbol);
template Symbol *
template void
SymbolTable::addLazyArchive<ELF64BE>(StringRef, ArchiveFile &,
const object::Archive::Symbol);
@ -815,6 +864,11 @@ template void SymbolTable::addLazyObject<ELF32BE>(StringRef, LazyObjFile &);
template void SymbolTable::addLazyObject<ELF64LE>(StringRef, LazyObjFile &);
template void SymbolTable::addLazyObject<ELF64BE>(StringRef, LazyObjFile &);
template void SymbolTable::fetchLazy<ELF32LE>(Symbol *);
template void SymbolTable::fetchLazy<ELF32BE>(Symbol *);
template void SymbolTable::fetchLazy<ELF64LE>(Symbol *);
template void SymbolTable::fetchLazy<ELF64BE>(Symbol *);
template void SymbolTable::addShared<ELF32LE>(StringRef, SharedFile<ELF32LE> &,
const typename ELF32LE::Sym &,
uint32_t Alignment, uint32_t);
@ -827,13 +881,3 @@ template void SymbolTable::addShared<ELF64LE>(StringRef, SharedFile<ELF64LE> &,
template void SymbolTable::addShared<ELF64BE>(StringRef, SharedFile<ELF64BE> &,
const typename ELF64BE::Sym &,
uint32_t Alignment, uint32_t);
template void SymbolTable::fetchIfLazy<ELF32LE>(StringRef);
template void SymbolTable::fetchIfLazy<ELF32BE>(StringRef);
template void SymbolTable::fetchIfLazy<ELF64LE>(StringRef);
template void SymbolTable::fetchIfLazy<ELF64BE>(StringRef);
template void SymbolTable::scanShlibUndefined<ELF32LE>();
template void SymbolTable::scanShlibUndefined<ELF32BE>();
template void SymbolTable::scanShlibUndefined<ELF64LE>();
template void SymbolTable::scanShlibUndefined<ELF64BE>();

11
contrib/llvm/tools/lld/ELF/SymbolTable.h
View File

@ -12,7 +12,7 @@
#include "InputFiles.h"
#include "LTO.h"
#include "Strings.h"
#include "lld/Common/Strings.h"
#include "llvm/ADT/CachedHashString.h"
#include "llvm/ADT/DenseMap.h"
@ -60,8 +60,8 @@ class SymbolTable {
uint32_t VerdefIndex);
template <class ELFT>
Symbol *addLazyArchive(StringRef Name, ArchiveFile &F,
const llvm::object::Archive::Symbol S);
void addLazyArchive(StringRef Name, ArchiveFile &F,
const llvm::object::Archive::Symbol S);
template <class ELFT> void addLazyObject(StringRef Name, LazyObjFile &Obj);
@ -77,8 +77,8 @@ class SymbolTable {
uint8_t Visibility, bool CanOmitFromDynSym,
InputFile *File);
template <class ELFT> void fetchIfLazy(StringRef Name);
template <class ELFT> void scanShlibUndefined();
template <class ELFT> void fetchLazy(Symbol *Sym);
void scanVersionScript();
Symbol *find(StringRef Name);
@ -90,7 +90,6 @@ class SymbolTable {
private:
std::vector<Symbol *> findByVersion(SymbolVersion Ver);
std::vector<Symbol *> findAllByVersion(SymbolVersion Ver);
void defsym(Symbol *Dst, Symbol *Src);
llvm::StringMap<std::vector<Symbol *>> &getDemangledSyms();
void handleAnonymousVersion();

95
contrib/llvm/tools/lld/ELF/Symbols.cpp
View File

@ -14,7 +14,6 @@
#include "SyntheticSections.h"
#include "Target.h"
#include "Writer.h"
#include "lld/Common/ErrorHandler.h"
#include "lld/Common/Strings.h"
#include "llvm/ADT/STLExtras.h"
@ -39,6 +38,7 @@ Defined *ElfSym::GlobalOffsetTable;
Defined *ElfSym::MipsGp;
Defined *ElfSym::MipsGpDisp;
Defined *ElfSym::MipsLocalGp;
Defined *ElfSym::RelaIpltEnd;
static uint64_t getSymVA(const Symbol &Sym, int64_t &Addend) {
switch (Sym.kind()) {
@ -58,6 +58,7 @@ static uint64_t getSymVA(const Symbol &Sym, int64_t &Addend) {
return D.Value;
IS = IS->Repl;
uint64_t Offset = D.Value;
// An object in an SHF_MERGE section might be referenced via a
@ -76,8 +77,6 @@ static uint64_t getSymVA(const Symbol &Sym, int64_t &Addend) {
Addend = 0;
}
const OutputSection *OutSec = IS->getOutputSection();
// In the typical case, this is actually very simple and boils
// down to adding together 3 numbers:
// 1. The address of the output section.
@ -88,7 +87,7 @@ static uint64_t getSymVA(const Symbol &Sym, int64_t &Addend) {
// If you understand the data structures involved with this next
// line (and how they get built), then you have a pretty good
// understanding of the linker.
uint64_t VA = (OutSec ? OutSec->Addr : 0) + IS->getOffset(Offset);
uint64_t VA = IS->getVA(Offset);
if (D.isTls() && !Config->Relocatable) {
if (!Out::TlsPhdr)
@ -98,20 +97,12 @@ static uint64_t getSymVA(const Symbol &Sym, int64_t &Addend) {
}
return VA;
}
case Symbol::SharedKind: {
auto &SS = cast<SharedSymbol>(Sym);
if (SS.CopyRelSec)
return SS.CopyRelSec->getParent()->Addr + SS.CopyRelSec->OutSecOff;
if (SS.NeedsPltAddr)
return Sym.getPltVA();
return 0;
}
case Symbol::SharedKind:
case Symbol::UndefinedKind:
return 0;
case Symbol::LazyArchiveKind:
case Symbol::LazyObjectKind:
assert(Sym.IsUsedInRegularObj && "lazy symbol reached writer");
return 0;
llvm_unreachable("lazy symbol reached writer");
}
llvm_unreachable("invalid symbol kind");
}
@ -134,22 +125,26 @@ uint64_t Symbol::getGotPltVA() const {
}
uint64_t Symbol::getGotPltOffset() const {
return GotPltIndex * Target->GotPltEntrySize;
if (IsInIgot)
return PltIndex * Target->GotPltEntrySize;
return (PltIndex + Target->GotPltHeaderEntriesNum) * Target->GotPltEntrySize;
}
uint64_t Symbol::getPltVA() const {
if (this->IsInIplt)
return InX::Iplt->getVA() + PltIndex * Target->PltEntrySize;
return InX::Plt->getVA() + Target->PltHeaderSize +
PltIndex * Target->PltEntrySize;
return InX::Plt->getVA() + Target->getPltEntryOffset(PltIndex);
}
uint64_t Symbol::getPltOffset() const {
assert(!this->IsInIplt);
return Target->getPltEntryOffset(PltIndex);
}
uint64_t Symbol::getSize() const {
if (const auto *DR = dyn_cast<Defined>(this))
return DR->Size;
if (const auto *S = dyn_cast<SharedSymbol>(this))
return S->Size;
return 0;
return cast<SharedSymbol>(this)->Size;
}
OutputSection *Symbol::getOutputSection() const {
@ -158,13 +153,6 @@ OutputSection *Symbol::getOutputSection() const {
return Sec->Repl->getOutputSection();
return nullptr;
}
if (auto *S = dyn_cast<SharedSymbol>(this)) {
if (S->CopyRelSec)
return S->CopyRelSec->getParent();
return nullptr;
}
return nullptr;
}
@ -180,7 +168,7 @@ void Symbol::parseSymbolVersion() {
return;
// Truncate the symbol name so that it doesn't include the version string.
Name = {S.data(), Pos};
NameSize = Pos;
// If this is not in this DSO, it is not a definition.
if (!isDefined())
@ -206,33 +194,15 @@ void Symbol::parseSymbolVersion() {
// It is an error if the specified version is not defined.
// Usually version script is not provided when linking executable,
// but we may still want to override a versioned symbol from DSO,
// so we do not report error in this case.
if (Config->Shared)
// so we do not report error in this case. We also do not error
// if the symbol has a local version as it won't be in the dynamic
// symbol table.
if (Config->Shared && VersionId != VER_NDX_LOCAL)
error(toString(File) + ": symbol " + S + " has undefined version " +
Verstr);
}
InputFile *Lazy::fetch() {
if (auto *S = dyn_cast<LazyArchive>(this))
return S->fetch();
return cast<LazyObject>(this)->fetch();
}
ArchiveFile &LazyArchive::getFile() { return *cast<ArchiveFile>(File); }
InputFile *LazyArchive::fetch() {
std::pair<MemoryBufferRef, uint64_t> MBInfo = getFile().getMember(&Sym);
// getMember returns an empty buffer if the member was already
// read from the library.
if (MBInfo.first.getBuffer().empty())
return nullptr;
return createObjectFile(MBInfo.first, getFile().getName(), MBInfo.second);
}
LazyObjFile &LazyObject::getFile() { return *cast<LazyObjFile>(File); }
InputFile *LazyObject::fetch() { return getFile().fetch(); }
InputFile *LazyArchive::fetch() { return cast<ArchiveFile>(File)->fetch(Sym); }
uint8_t Symbol::computeBinding() const {
if (Config->Relocatable)
@ -241,7 +211,7 @@ uint8_t Symbol::computeBinding() const {
return STB_LOCAL;
if (VersionId == VER_NDX_LOCAL && isDefined())
return STB_LOCAL;
if (Config->NoGnuUnique && Binding == STB_GNU_UNIQUE)
if (!Config->GnuUnique && Binding == STB_GNU_UNIQUE)
return STB_GLOBAL;
return Binding;
}
@ -273,6 +243,27 @@ void elf::printTraceSymbol(Symbol *Sym) {
message(toString(Sym->File) + S + Sym->getName());
}
void elf::warnUnorderableSymbol(const Symbol *Sym) {
if (!Config->WarnSymbolOrdering)
return;
const InputFile *File = Sym->File;
auto *D = dyn_cast<Defined>(Sym);
auto Warn = [&](StringRef S) { warn(toString(File) + S + Sym->getName()); };
if (Sym->isUndefined())
Warn(": unable to order undefined symbol: ");
else if (Sym->isShared())
Warn(": unable to order shared symbol: ");
else if (D && !D->Section)
Warn(": unable to order absolute symbol: ");
else if (D && isa<OutputSection>(D->Section))
Warn(": unable to order synthetic symbol: ");
else if (D && !D->Section->Repl->Live)
Warn(": unable to order discarded symbol: ");
}
// Returns a symbol for an error message.
std::string lld::toString(const Symbol &B) {
if (Config->Demangle)

159
contrib/llvm/tools/lld/ELF/Symbols.h
View File

@ -7,8 +7,7 @@
//
//===----------------------------------------------------------------------===//
//
// All symbols are handled as SymbolBodies regardless of their types.
// This file defines various types of SymbolBodies.
// This file defines various types of Symbols.
//
//===----------------------------------------------------------------------===//
@ -16,9 +15,8 @@
#define LLD_ELF_SYMBOLS_H
#include "InputSection.h"
#include "Strings.h"
#include "lld/Common/LLVM.h"
#include "lld/Common/Strings.h"
#include "llvm/Object/Archive.h"
#include "llvm/Object/ELF.h"
@ -34,6 +32,20 @@ template <class ELFT> class ObjFile;
class OutputSection;
template <class ELFT> class SharedFile;
// This is a StringRef-like container that doesn't run strlen().
//
// ELF string tables contain a lot of null-terminated strings. Most of them
// are not necessary for the linker because they are names of local symbols,
// and the linker doesn't use local symbol names for name resolution. So, we
// use this class to represents strings read from string tables.
struct StringRefZ {
StringRefZ(const char *S) : Data(S), Size(-1) {}
StringRefZ(StringRef S) : Data(S.data()), Size(S.size()) {}
const char *Data;
const uint32_t Size;
};
// The base class for real symbol classes.
class Symbol {
public:
@ -47,6 +59,25 @@ class Symbol {
Kind kind() const { return static_cast<Kind>(SymbolKind); }
// The file from which this symbol was created.
InputFile *File;
protected:
const char *NameData;
mutable uint32_t NameSize;
public:
uint32_t DynsymIndex = 0;
uint32_t GotIndex = -1;
uint32_t PltIndex = -1;
uint32_t GlobalDynIndex = -1;
// This field is a index to the symbol's version definition.
uint32_t VerdefIndex = -1;
// Version definition index.
uint16_t VersionId;
// Symbol binding. This is not overwritten by replaceSymbol to track
// changes during resolution. In particular:
// - An undefined weak is still weak when it resolves to a shared library.
@ -54,8 +85,11 @@ class Symbol {
// remember it is weak.
uint8_t Binding;
// Version definition index.
uint16_t VersionId;
// The following fields have the same meaning as the ELF symbol attributes.
uint8_t Type; // symbol type
uint8_t StOther; // st_other field value
const uint8_t SymbolKind;
// Symbol visibility. This is the computed minimum visibility of all
// observed non-DSO symbols.
@ -81,12 +115,6 @@ class Symbol {
// True if this symbol is specified by --trace-symbol option.
unsigned Traced : 1;
// This symbol version was found in a version script.
unsigned InVersionScript : 1;
// The file from which this symbol was created.
InputFile *File;
bool includeInDynsym() const;
uint8_t computeBinding() const;
bool isWeak() const { return Binding == llvm::ELF::STB_WEAK; }
@ -100,15 +128,15 @@ class Symbol {
return SymbolKind == LazyArchiveKind || SymbolKind == LazyObjectKind;
}
// True is this is an undefined weak symbol. This only works once
// all input files have been added.
bool isUndefWeak() const {
// See comment on Lazy the details.
return isWeak() && (isUndefined() || isLazy());
// True if this is an undefined weak symbol.
bool isUndefWeak() const { return isWeak() && isUndefined(); }
StringRef getName() const {
if (NameSize == (uint32_t)-1)
NameSize = strlen(NameData);
return {NameData, NameSize};
}
StringRef getName() const { return Name; }
uint8_t getVisibility() const { return StOther & 0x3; }
void parseSymbolVersion();
bool isInGot() const { return GotIndex != -1U; }
@ -121,34 +149,22 @@ class Symbol {
uint64_t getGotPltOffset() const;
uint64_t getGotPltVA() const;
uint64_t getPltVA() const;
uint64_t getPltOffset() const;
uint64_t getSize() const;
OutputSection *getOutputSection() const;
uint32_t DynsymIndex = 0;
uint32_t GotIndex = -1;
uint32_t GotPltIndex = -1;
uint32_t PltIndex = -1;
uint32_t GlobalDynIndex = -1;
protected:
Symbol(Kind K, InputFile *File, StringRefZ Name, uint8_t Binding,
uint8_t StOther, uint8_t Type)
: Binding(Binding), File(File), SymbolKind(K), NeedsPltAddr(false),
IsInGlobalMipsGot(false), Is32BitMipsGot(false), IsInIplt(false),
IsInIgot(false), IsPreemptible(false), Used(!Config->GcSections),
Type(Type), StOther(StOther), Name(Name) {}
const unsigned SymbolKind : 8;
: File(File), NameData(Name.Data), NameSize(Name.Size), Binding(Binding),
Type(Type), StOther(StOther), SymbolKind(K), NeedsPltAddr(false),
IsInIplt(false), IsInIgot(false), IsPreemptible(false),
Used(!Config->GcSections), NeedsTocRestore(false) {}
public:
// True the symbol should point to its PLT entry.
// For SharedSymbol only.
unsigned NeedsPltAddr : 1;
// True if this symbol has an entry in the global part of MIPS GOT.
unsigned IsInGlobalMipsGot : 1;
// True if this symbol is referenced by 32-bit GOT relocations.
unsigned Is32BitMipsGot : 1;
// True if this symbol is in the Iplt sub-section of the Plt.
unsigned IsInIplt : 1;
@ -156,14 +172,15 @@ class Symbol {
// True if this symbol is in the Igot sub-section of the .got.plt or .got.
unsigned IsInIgot : 1;
// True if this symbol is preemptible at load time.
unsigned IsPreemptible : 1;
// True if an undefined or shared symbol is used from a live section.
unsigned Used : 1;
// The following fields have the same meaning as the ELF symbol attributes.
uint8_t Type; // symbol type
uint8_t StOther; // st_other field value
// True if a call to this symbol needs to be followed by a restore of the
// PPC64 toc pointer.
unsigned NeedsTocRestore : 1;
// The Type field may also have this value. It means that we have not yet seen
// a non-Lazy symbol with this name, so we don't know what its type is. The
@ -178,9 +195,6 @@ class Symbol {
bool isGnuIFunc() const { return Type == llvm::ELF::STT_GNU_IFUNC; }
bool isObject() const { return Type == llvm::ELF::STT_OBJECT; }
bool isFile() const { return Type == llvm::ELF::STT_FILE; }
protected:
StringRefZ Name;
};
// Represents a symbol that is defined in the current output file.
@ -214,8 +228,9 @@ class SharedSymbol : public Symbol {
SharedSymbol(InputFile &File, StringRef Name, uint8_t Binding,
uint8_t StOther, uint8_t Type, uint64_t Value, uint64_t Size,
uint32_t Alignment, uint32_t VerdefIndex)
: Symbol(SharedKind, &File, Name, Binding, StOther, Type), Value(Value),
Size(Size), VerdefIndex(VerdefIndex), Alignment(Alignment) {
: Symbol(SharedKind, &File, Name, Binding, StOther, Type),
Alignment(Alignment), Value(Value), Size(Size) {
this->VerdefIndex = VerdefIndex;
// GNU ifunc is a mechanism to allow user-supplied functions to
// resolve PLT slot values at load-time. This is contrary to the
// regular symbol resolution scheme in which symbols are resolved just
@ -240,54 +255,36 @@ class SharedSymbol : public Symbol {
return *cast<SharedFile<ELFT>>(File);
}
// If not null, there is a copy relocation to this section.
InputSection *CopyRelSec = nullptr;
uint32_t Alignment;
uint64_t Value; // st_value
uint64_t Size; // st_size
// This field is a index to the symbol's version definition.
uint32_t VerdefIndex;
uint32_t Alignment;
};
// This represents a symbol that is not yet in the link, but we know where to
// find it if needed. If the resolver finds both Undefined and Lazy for the same
// name, it will ask the Lazy to load a file.
// LazyArchive and LazyObject represent a symbols that is not yet in the link,
// but we know where to find it if needed. If the resolver finds both Undefined
// and Lazy for the same name, it will ask the Lazy to load a file.
//
// A special complication is the handling of weak undefined symbols. They should
// not load a file, but we have to remember we have seen both the weak undefined
// and the lazy. We represent that with a lazy symbol with a weak binding. This
// means that code looking for undefined symbols normally also has to take lazy
// symbols into consideration.
class Lazy : public Symbol {
public:
static bool classof(const Symbol *S) { return S->isLazy(); }
// Returns an object file for this symbol, or a nullptr if the file
// was already returned.
InputFile *fetch();
protected:
Lazy(Kind K, InputFile &File, StringRef Name, uint8_t Type)
: Symbol(K, &File, Name, llvm::ELF::STB_GLOBAL, llvm::ELF::STV_DEFAULT,
Type) {}
};
// This class represents a symbol defined in an archive file. It is
// created from an archive file header, and it knows how to load an
// object file from an archive to replace itself with a defined
// symbol.
class LazyArchive : public Lazy {
class LazyArchive : public Symbol {
public:
LazyArchive(InputFile &File, const llvm::object::Archive::Symbol S,
uint8_t Type)
: Lazy(LazyArchiveKind, File, S.getName(), Type), Sym(S) {}
LazyArchive(InputFile &File, uint8_t Type,
const llvm::object::Archive::Symbol S)
: Symbol(LazyArchiveKind, &File, S.getName(), llvm::ELF::STB_GLOBAL,
llvm::ELF::STV_DEFAULT, Type),
Sym(S) {}
static bool classof(const Symbol *S) { return S->kind() == LazyArchiveKind; }
ArchiveFile &getFile();
InputFile *fetch();
private:
@ -296,15 +293,13 @@ class LazyArchive : public Lazy {
// LazyObject symbols represents symbols in object files between
// --start-lib and --end-lib options.
class LazyObject : public Lazy {
class LazyObject : public Symbol {
public:
LazyObject(InputFile &File, StringRef Name, uint8_t Type)
: Lazy(LazyObjectKind, File, Name, Type) {}
LazyObject(InputFile &File, uint8_t Type, StringRef Name)
: Symbol(LazyObjectKind, &File, Name, llvm::ELF::STB_GLOBAL,
llvm::ELF::STV_DEFAULT, Type) {}
static bool classof(const Symbol *S) { return S->kind() == LazyObjectKind; }
LazyObjFile &getFile();
InputFile *fetch();
};
// Some linker-generated symbols need to be created as
@ -334,6 +329,9 @@ struct ElfSym {
static Defined *MipsGp;
static Defined *MipsGpDisp;
static Defined *MipsLocalGp;
// __rela_iplt_end or __rel_iplt_end
static Defined *RelaIpltEnd;
};
// A buffer class that is large enough to hold any Symbol-derived
@ -351,6 +349,8 @@ void printTraceSymbol(Symbol *Sym);
template <typename T, typename... ArgT>
void replaceSymbol(Symbol *S, ArgT &&... Arg) {
static_assert(std::is_trivially_destructible<T>(),
"Symbol types must be trivially destructible");
static_assert(sizeof(T) <= sizeof(SymbolUnion), "SymbolUnion too small");
static_assert(alignof(T) <= alignof(SymbolUnion),
"SymbolUnion not aligned enough");
@ -367,13 +367,14 @@ void replaceSymbol(Symbol *S, ArgT &&... Arg) {
S->ExportDynamic = Sym.ExportDynamic;
S->CanInline = Sym.CanInline;
S->Traced = Sym.Traced;
S->InVersionScript = Sym.InVersionScript;
// Print out a log message if --trace-symbol was specified.
// This is for debugging.
if (S->Traced)
printTraceSymbol(S);
}
void warnUnorderableSymbol(const Symbol *Sym);
} // namespace elf
std::string toString(const elf::Symbol &B);

1498
contrib/llvm/tools/lld/ELF/SyntheticSections.cpp
View File

File diff suppressed because it is too large Load Diff

337
contrib/llvm/tools/lld/ELF/SyntheticSections.h
View File

@ -26,10 +26,12 @@
#include "InputSection.h"
#include "llvm/ADT/MapVector.h"
#include "llvm/MC/StringTableBuilder.h"
#include "llvm/Support/Endian.h"
#include <functional>
namespace lld {
namespace elf {
class Defined;
class SharedSymbol;
class SyntheticSection : public InputSection {
@ -51,7 +53,6 @@ class SyntheticSection : public InputSection {
// If any additional finalization of contents are needed post thunk creation.
virtual void postThunkContents() {}
virtual bool empty() const { return false; }
uint64_t getVA() const;
static bool classof(const SectionBase *D) {
return D->kind() == InputSectionBase::Synthetic;
@ -78,13 +79,18 @@ class EhFrameSection final : public SyntheticSection {
size_t NumFdes = 0;
struct FdeData {
uint32_t Pc;
uint32_t FdeVA;
uint32_t PcRel;
uint32_t FdeVARel;
};
std::vector<FdeData> getFdeData() const;
ArrayRef<CieRecord *> getCieRecords() const { return CieRecords; }
private:
// This is used only when parsing EhInputSection. We keep it here to avoid
// allocating one for each EhInputSection.
llvm::DenseMap<size_t, CieRecord *> OffsetToCie;
uint64_t Size = 0;
template <class ELFT, class RelTy>
@ -173,12 +179,21 @@ class MipsGotSection final : public SyntheticSection {
bool updateAllocSize() override;
void finalizeContents() override;
bool empty() const override;
void addEntry(Symbol &Sym, int64_t Addend, RelExpr Expr);
bool addDynTlsEntry(Symbol &Sym);
bool addTlsIndex();
uint64_t getPageEntryOffset(const Symbol &B, int64_t Addend) const;
uint64_t getSymEntryOffset(const Symbol &B, int64_t Addend) const;
uint64_t getGlobalDynOffset(const Symbol &B) const;
// Join separate GOTs built for each input file to generate
// primary and optional multiple secondary GOTs.
template <class ELFT> void build();
void addEntry(InputFile &File, Symbol &Sym, int64_t Addend, RelExpr Expr);
void addDynTlsEntry(InputFile &File, Symbol &Sym);
void addTlsIndex(InputFile &File);
uint64_t getPageEntryOffset(const InputFile *F, const Symbol &S,
int64_t Addend) const;
uint64_t getSymEntryOffset(const InputFile *F, const Symbol &S,
int64_t Addend) const;
uint64_t getGlobalDynOffset(const InputFile *F, const Symbol &S) const;
uint64_t getTlsIndexOffset(const InputFile *F) const;
// Returns the symbol which corresponds to the first entry of the global part
// of GOT on MIPS platform. It is required to fill up MIPS-specific dynamic
@ -190,13 +205,8 @@ class MipsGotSection final : public SyntheticSection {
// the number of reserved entries.
unsigned getLocalEntriesNum() const;
// Returns offset of TLS part of the MIPS GOT table. This part goes
// after 'local' and 'global' entries.
uint64_t getTlsOffset() const;
uint32_t getTlsIndexOff() const { return TlsIndexOff; }
uint64_t getGp() const;
// Return _gp value for primary GOT (nullptr) or particular input file.
uint64_t getGp(const InputFile *F = nullptr) const;
private:
// MIPS GOT consists of three parts: local, global and tls. Each part
@ -235,32 +245,110 @@ class MipsGotSection final : public SyntheticSection {
// addressing, but MIPS ABI requires that these entries be present in GOT.
// TLS entries:
// Entries created by TLS relocations.
//
// If the sum of local, global and tls entries is less than 64K only single
// got is enough. Otherwise, multi-got is created. Series of primary and
// multiple secondary GOTs have the following layout:
// - Primary GOT
// Header
// Local entries
// Global entries
// Relocation only entries
// TLS entries
//
// - Secondary GOT
// Local entries
// Global entries
// TLS entries
// ...
//
// All GOT entries required by relocations from a single input file entirely
// belong to either primary or one of secondary GOTs. To reference GOT entries
// each GOT has its own _gp value points to the "middle" of the GOT.
// In the code this value loaded to the register which is used for GOT access.
//
// MIPS 32 function's prologue:
// lui v0,0x0
// 0: R_MIPS_HI16 _gp_disp
// addiu v0,v0,0
// 4: R_MIPS_LO16 _gp_disp
//
// MIPS 64:
// lui at,0x0
// 14: R_MIPS_GPREL16 main
//
// Dynamic linker does not know anything about secondary GOTs and cannot
// use a regular MIPS mechanism for GOT entries initialization. So we have
// to use an approach accepted by other architectures and create dynamic
// relocations R_MIPS_REL32 to initialize global entries (and local in case
// of PIC code) in secondary GOTs. But ironically MIPS dynamic linker
// requires GOT entries and correspondingly ordered dynamic symbol table
// entries to deal with dynamic relocations. To handle this problem
// relocation-only section in the primary GOT contains entries for all
// symbols referenced in global parts of secondary GOTs. Although the sum
// of local and normal global entries of the primary got should be less
// than 64K, the size of the primary got (including relocation-only entries
// can be greater than 64K, because parts of the primary got that overflow
// the 64K limit are used only by the dynamic linker at dynamic link-time
// and not by 16-bit gp-relative addressing at run-time.
//
// For complete multi-GOT description see the following link
// https://dmz-portal.mips.com/wiki/MIPS_Multi_GOT
// Number of "Header" entries.
static const unsigned HeaderEntriesNum = 2;
// Number of allocated "Page" entries.
uint32_t PageEntriesNum = 0;
// Map output sections referenced by MIPS GOT relocations
// to the first index of "Page" entries allocated for this section.
llvm::SmallMapVector<const OutputSection *, size_t, 16> PageIndexMap;
typedef std::pair<const Symbol *, uint64_t> GotEntry;
typedef std::vector<GotEntry> GotEntries;
// Map from Symbol-Addend pair to the GOT index.
llvm::DenseMap<GotEntry, size_t> EntryIndexMap;
// Local entries (16-bit access).
GotEntries LocalEntries;
// Local entries (32-bit access).
GotEntries LocalEntries32;
// Normal and reloc-only global entries.
GotEntries GlobalEntries;
// TLS entries.
std::vector<const Symbol *> TlsEntries;
uint32_t TlsIndexOff = -1;
uint64_t Size = 0;
size_t LocalEntriesNum = 0;
// Symbol and addend.
typedef std::pair<Symbol *, int64_t> GotEntry;
struct FileGot {
InputFile *File = nullptr;
size_t StartIndex = 0;
struct PageBlock {
size_t FirstIndex = 0;
size_t Count = 0;
};
// Map output sections referenced by MIPS GOT relocations
// to the description (index/count) "page" entries allocated
// for this section.
llvm::SmallMapVector<const OutputSection *, PageBlock, 16> PagesMap;
// Maps from Symbol+Addend pair or just Symbol to the GOT entry index.
llvm::MapVector<GotEntry, size_t> Local16;
llvm::MapVector<GotEntry, size_t> Local32;
llvm::MapVector<Symbol *, size_t> Global;
llvm::MapVector<Symbol *, size_t> Relocs;
llvm::MapVector<Symbol *, size_t> Tls;
// Set of symbols referenced by dynamic TLS relocations.
llvm::MapVector<Symbol *, size_t> DynTlsSymbols;
// Total number of all entries.
size_t getEntriesNum() const;
// Number of "page" entries.
size_t getPageEntriesNum() const;
// Number of entries require 16-bit index to access.
size_t getIndexedEntriesNum() const;
bool isOverflow() const;
};
// Container of GOT created for each input file.
// After building a final series of GOTs this container
// holds primary and secondary GOT's.
std::vector<FileGot> Gots;
// Return (and create if necessary) `FileGot`.
FileGot &getGot(InputFile &F);
// Try to merge two GOTs. In case of success the `Dst` contains
// result of merging and the function returns true. In case of
// ovwerflow the `Dst` is unchanged and the function returns false.
bool tryMergeGots(FileGot & Dst, FileGot & Src, bool IsPrimary);
};
class GotPltSection final : public SyntheticSection {
@ -269,7 +357,7 @@ class GotPltSection final : public SyntheticSection {
void addEntry(Symbol &Sym);
size_t getSize() const override;
void writeTo(uint8_t *Buf) override;
bool empty() const override { return Entries.empty(); }
bool empty() const override;
private:
std::vector<const Symbol *> Entries;
@ -310,30 +398,48 @@ class StringTableSection final : public SyntheticSection {
class DynamicReloc {
public:
DynamicReloc(uint32_t Type, const InputSectionBase *InputSec,
DynamicReloc(RelType Type, const InputSectionBase *InputSec,
uint64_t OffsetInSec, bool UseSymVA, Symbol *Sym, int64_t Addend)
: Type(Type), Sym(Sym), InputSec(InputSec), OffsetInSec(OffsetInSec),
UseSymVA(UseSymVA), Addend(Addend) {}
UseSymVA(UseSymVA), Addend(Addend), OutputSec(nullptr) {}
// This constructor records dynamic relocation settings used by MIPS
// multi-GOT implementation. It's to relocate addresses of 64kb pages
// lie inside the output section.
DynamicReloc(RelType Type, const InputSectionBase *InputSec,
uint64_t OffsetInSec, const OutputSection *OutputSec,
int64_t Addend)
: Type(Type), Sym(nullptr), InputSec(InputSec), OffsetInSec(OffsetInSec),
UseSymVA(false), Addend(Addend), OutputSec(OutputSec) {}
uint64_t getOffset() const;
int64_t getAddend() const;
uint32_t getSymIndex() const;
const InputSectionBase *getInputSec() const { return InputSec; }
uint32_t Type;
// Computes the addend of the dynamic relocation. Note that this is not the
// same as the Addend member variable as it also includes the symbol address
// if UseSymVA is true.
int64_t computeAddend() const;
RelType Type;
private:
Symbol *Sym;
const InputSectionBase *InputSec = nullptr;
uint64_t OffsetInSec;
// If this member is true, the dynamic relocation will not be against the
// symbol but will instead be a relative relocation that simply adds the
// load address. This means we need to write the symbol virtual address
// plus the original addend as the final relocation addend.
bool UseSymVA;
int64_t Addend;
const OutputSection *OutputSec;
};
template <class ELFT> class DynamicSection final : public SyntheticSection {
typedef typename ELFT::Dyn Elf_Dyn;
typedef typename ELFT::Rel Elf_Rel;
typedef typename ELFT::Rela Elf_Rela;
typedef typename ELFT::Relr Elf_Relr;
typedef typename ELFT::Shdr Elf_Shdr;
typedef typename ELFT::Sym Elf_Sym;
@ -350,6 +456,7 @@ template <class ELFT> class DynamicSection final : public SyntheticSection {
void add(int32_t Tag, std::function<uint64_t()> Fn);
void addInt(int32_t Tag, uint64_t Val);
void addInSec(int32_t Tag, InputSection *Sec);
void addInSecRelative(int32_t Tag, InputSection *Sec);
void addOutSec(int32_t Tag, OutputSection *Sec);
void addSize(int32_t Tag, OutputSection *Sec);
void addSym(int32_t Tag, Symbol *Sym);
@ -361,6 +468,13 @@ class RelocationBaseSection : public SyntheticSection {
public:
RelocationBaseSection(StringRef Name, uint32_t Type, int32_t DynamicTag,
int32_t SizeDynamicTag);
void addReloc(RelType DynType, InputSectionBase *IS, uint64_t OffsetInSec,
Symbol *Sym);
// Add a dynamic relocation that might need an addend. This takes care of
// writing the addend to the output section if needed.
void addReloc(RelType DynType, InputSectionBase *InputSec,
uint64_t OffsetInSec, Symbol *Sym, int64_t Addend, RelExpr Expr,
RelType Type);
void addReloc(const DynamicReloc &Reloc);
bool empty() const override { return Relocs.empty(); }
size_t getSize() const override { return Relocs.size() * this->Entsize; }
@ -405,6 +519,39 @@ class AndroidPackedRelocationSection final : public RelocationBaseSection {
SmallVector<char, 0> RelocData;
};
struct RelativeReloc {
uint64_t getOffset() const { return InputSec->getVA(OffsetInSec); }
const InputSectionBase *InputSec;
uint64_t OffsetInSec;
};
class RelrBaseSection : public SyntheticSection {
public:
RelrBaseSection();
std::vector<RelativeReloc> Relocs;
};
// RelrSection is used to encode offsets for relative relocations.
// Proposal for adding SHT_RELR sections to generic-abi is here:
// https://groups.google.com/forum/#!topic/generic-abi/bX460iggiKg
// For more details, see the comment in RelrSection::updateAllocSize().
template <class ELFT> class RelrSection final : public RelrBaseSection {
typedef typename ELFT::Relr Elf_Relr;
public:
RelrSection();
bool updateAllocSize() override;
size_t getSize() const override { return RelrRelocs.size() * this->Entsize; }
void writeTo(uint8_t *Buf) override {
memcpy(Buf, RelrRelocs.data(), getSize());
}
private:
std::vector<Elf_Relr> RelrRelocs;
};
struct SymbolTableEntry {
Symbol *Sym;
size_t StrTabOffset;
@ -455,7 +602,7 @@ class GnuHashTableSection final : public SyntheticSection {
void addSymbols(std::vector<SymbolTableEntry> &Symbols);
private:
size_t getShift2() const { return Config->Is64 ? 6 : 5; }
enum { Shift2 = 6 };
void writeBloomFilter(uint8_t *Buf);
void writeHashTable(uint8_t *Buf);
@ -484,13 +631,13 @@ class HashTableSection final : public SyntheticSection {
size_t Size = 0;
};
// The PltSection is used for both the Plt and Iplt. The former always has a
// The PltSection is used for both the Plt and Iplt. The former usually has a
// header as its first entry that is used at run-time to resolve lazy binding.
// The latter is used for GNU Ifunc symbols, that will be subject to a
// Target->IRelativeRel.
class PltSection : public SyntheticSection {
public:
PltSection(size_t HeaderSize);
PltSection(bool IsIplt);
void writeTo(uint8_t *Buf) override;
size_t getSize() const override;
bool empty() const override { return Entries.empty(); }
@ -501,13 +648,12 @@ class PltSection : public SyntheticSection {
private:
unsigned getPltRelocOff() const;
std::vector<std::pair<const Symbol *, unsigned>> Entries;
// Iplt always has HeaderSize of 0, the Plt HeaderSize is always non-zero
size_t HeaderSize;
bool IsIplt;
};
// GdbIndexChunk is created for each .debug_info section and contains
// information to create a part of .gdb_index for a given input section.
struct GdbIndexChunk {
class GdbIndexSection final : public SyntheticSection {
public:
struct AddressEntry {
InputSection *Section;
uint64_t LowAddress;
@ -522,59 +668,51 @@ struct GdbIndexChunk {
struct NameTypeEntry {
llvm::CachedHashStringRef Name;
uint8_t Type;
uint32_t Type;
};
InputSection *DebugInfoSec;
std::vector<AddressEntry> AddressAreas;
std::vector<CuEntry> CompilationUnits;
std::vector<NameTypeEntry> NamesAndTypes;
};
struct GdbChunk {
InputSection *Sec;
std::vector<AddressEntry> AddressAreas;
std::vector<CuEntry> CompilationUnits;
};
// The symbol type for the .gdb_index section.
struct GdbSymbol {
uint32_t NameHash;
size_t NameOffset;
size_t CuVectorIndex;
};
struct GdbSymbol {
llvm::CachedHashStringRef Name;
std::vector<uint32_t> CuVector;
uint32_t NameOff;
uint32_t CuVectorOff;
};
class GdbIndexSection final : public SyntheticSection {
public:
GdbIndexSection(std::vector<GdbIndexChunk> &&Chunks);
GdbIndexSection();
template <typename ELFT> static GdbIndexSection *create();
void writeTo(uint8_t *Buf) override;
size_t getSize() const override;
size_t getSize() const override { return Size; }
bool empty() const override;
private:
void fixCuIndex();
std::vector<std::vector<uint32_t>> createCuVectors();
std::vector<GdbSymbol *> createGdbSymtab();
struct GdbIndexHeader {
llvm::support::ulittle32_t Version;
llvm::support::ulittle32_t CuListOff;
llvm::support::ulittle32_t CuTypesOff;
llvm::support::ulittle32_t AddressAreaOff;
llvm::support::ulittle32_t SymtabOff;
llvm::support::ulittle32_t ConstantPoolOff;
};
void initOutputSize();
size_t computeSymtabSize() const;
// Each chunk contains information gathered from debug sections of a
// single object file.
std::vector<GdbChunk> Chunks;
// A symbol table for this .gdb_index section.
std::vector<GdbSymbol *> GdbSymtab;
std::vector<GdbSymbol> Symbols;
// CU vector is a part of constant pool area of section.
std::vector<std::vector<uint32_t>> CuVectors;
// Symbol table contents.
llvm::DenseMap<llvm::CachedHashStringRef, GdbSymbol *> Symbols;
// Each chunk contains information gathered from a debug sections of single
// object and used to build different areas of gdb index.
std::vector<GdbIndexChunk> Chunks;
static constexpr uint32_t CuListOffset = 24;
uint32_t CuTypesOffset;
uint32_t SymtabOffset;
uint32_t ConstantPoolOffset;
uint32_t StringPoolOffset;
uint32_t StringPoolSize;
std::vector<size_t> CuVectorOffsets;
size_t Size;
};
template <class ELFT> GdbIndexSection *createGdbIndex();
// --eh-frame-hdr option tells linker to construct a header for all the
// .eh_frame sections. This header is placed to a section named .eh_frame_hdr
// and also to a PT_GNU_EH_FRAME segment.
@ -653,7 +791,7 @@ template <class ELFT> class VersionNeedSection final : public SyntheticSection {
public:
VersionNeedSection();
void addSymbol(SharedSymbol *SS);
void addSymbol(Symbol *Sym);
void finalizeContents() override;
void writeTo(uint8_t *Buf) override;
size_t getSize() const override;
@ -668,13 +806,12 @@ template <class ELFT> class VersionNeedSection final : public SyntheticSection {
class MergeSyntheticSection : public SyntheticSection {
public:
void addSection(MergeInputSection *MS);
std::vector<MergeInputSection *> Sections;
protected:
MergeSyntheticSection(StringRef Name, uint32_t Type, uint64_t Flags,
uint32_t Alignment)
: SyntheticSection(Flags, Type, Alignment, Name) {}
std::vector<MergeInputSection *> Sections;
};
class MergeTailSection final : public MergeSyntheticSection {
@ -787,7 +924,12 @@ class ARMExidxSentinelSection : public SyntheticSection {
void writeTo(uint8_t *Buf) override;
bool empty() const override;
InputSection *Highest = 0;
static bool classof(const SectionBase *D);
// The last section referenced by a regular .ARM.exidx section.
// It is found and filled in Writer<ELFT>::resolveShfLinkOrder().
// The sentinel points at the end of that section.
InputSection *Highest = nullptr;
};
// A container for one or more linker generated thunks. Instances of these
@ -805,19 +947,21 @@ class ThunkSection : public SyntheticSection {
size_t getSize() const override { return Size; }
void writeTo(uint8_t *Buf) override;
InputSection *getTargetInputSection() const;
bool assignOffsets();
private:
std::vector<const Thunk *> Thunks;
std::vector<Thunk *> Thunks;
size_t Size = 0;
};
InputSection *createInterpSection();
MergeInputSection *createCommentSection();
void decompressSections();
template <class ELFT> void splitSections();
void mergeSections();
Symbol *addSyntheticLocal(StringRef Name, uint8_t Type, uint64_t Value,
uint64_t Size, InputSectionBase &Section);
Defined *addSyntheticLocal(StringRef Name, uint8_t Type, uint64_t Value,
uint64_t Size, InputSectionBase &Section);
// Linker generated sections which can be used as inputs.
struct InX {
@ -842,6 +986,7 @@ struct InX {
static PltSection *Plt;
static PltSection *Iplt;
static RelocationBaseSection *RelaDyn;
static RelrBaseSection *RelrDyn;
static RelocationBaseSection *RelaPlt;
static RelocationBaseSection *RelaIplt;
static StringTableSection *ShStrTab;

28
contrib/llvm/tools/lld/ELF/Target.cpp
View File

@ -60,6 +60,8 @@ TargetInfo *elf::getTarget() {
return getARMTargetInfo();
case EM_AVR:
return getAVRTargetInfo();
case EM_HEXAGON:
return getHexagonTargetInfo();
case EM_MIPS:
switch (Config->EKind) {
case ELF32LEKind:
@ -87,29 +89,29 @@ TargetInfo *elf::getTarget() {
fatal("unknown target machine");
}
template <class ELFT> static std::string getErrorLoc(const uint8_t *Loc) {
template <class ELFT> static ErrorPlace getErrPlace(const uint8_t *Loc) {
for (InputSectionBase *D : InputSections) {
auto *IS = dyn_cast<InputSection>(D);
if (!IS || !IS->getParent())
auto *IS = cast<InputSection>(D);
if (!IS->getParent())
continue;
uint8_t *ISLoc = IS->getParent()->Loc + IS->OutSecOff;
if (ISLoc <= Loc && Loc < ISLoc + IS->getSize())
return IS->template getLocation<ELFT>(Loc - ISLoc) + ": ";
return {IS, IS->template getLocation<ELFT>(Loc - ISLoc) + ": "};
}
return "";
return {};
}
std::string elf::getErrorLocation(const uint8_t *Loc) {
ErrorPlace elf::getErrorPlace(const uint8_t *Loc) {
switch (Config->EKind) {
case ELF32LEKind:
return getErrorLoc<ELF32LE>(Loc);
return getErrPlace<ELF32LE>(Loc);
case ELF32BEKind:
return getErrorLoc<ELF32BE>(Loc);
return getErrPlace<ELF32BE>(Loc);
case ELF64LEKind:
return getErrorLoc<ELF64LE>(Loc);
return getErrPlace<ELF64LE>(Loc);
case ELF64BEKind:
return getErrorLoc<ELF64BE>(Loc);
return getErrPlace<ELF64BE>(Loc);
default:
llvm_unreachable("unknown ELF type");
}
@ -128,6 +130,12 @@ bool TargetInfo::needsThunk(RelExpr Expr, RelType Type, const InputFile *File,
return false;
}
bool TargetInfo::adjustPrologueForCrossSplitStack(uint8_t *Loc,
uint8_t *End) const {
llvm_unreachable("Target doesn't support split stacks.");
}
bool TargetInfo::inBranchRange(RelType Type, uint64_t Src, uint64_t Dst) const {
return true;
}

113
contrib/llvm/tools/lld/ELF/Target.h
View File

@ -25,9 +25,9 @@ class Symbol;
class TargetInfo {
public:
virtual uint32_t calcEFlags() const { return 0; }
virtual bool isPicRel(RelType Type) const { return true; }
virtual RelType getDynRel(RelType Type) const { return Type; }
virtual void writeGotPltHeader(uint8_t *Buf) const {}
virtual void writeGotHeader(uint8_t *Buf) const {}
virtual void writeGotPlt(uint8_t *Buf, const Symbol &S) const {};
virtual void writeIgotPlt(uint8_t *Buf, const Symbol &S) const;
virtual int64_t getImplicitAddend(const uint8_t *Buf, RelType Type) const;
@ -43,8 +43,12 @@ class TargetInfo {
virtual void addPltHeaderSymbols(InputSection &IS) const {}
virtual void addPltSymbols(InputSection &IS, uint64_t Off) const {}
unsigned getPltEntryOffset(unsigned Index) const {
return Index * PltEntrySize + PltHeaderSize;
}
// Returns true if a relocation only uses the low bits of a value such that
// all those bits are in in the same page. For example, if the relocation
// all those bits are in the same page. For example, if the relocation
// only uses the low 12 bits in a system with 4k pages. If this is true, the
// bits will always have the same value at runtime and we don't have to emit
// a dynamic relocation.
@ -55,6 +59,13 @@ class TargetInfo {
virtual bool needsThunk(RelExpr Expr, RelType RelocType,
const InputFile *File, uint64_t BranchAddr,
const Symbol &S) const;
// The function with a prologue starting at Loc was compiled with
// -fsplit-stack and it calls a function compiled without. Adjust the prologue
// to do the right thing. See https://gcc.gnu.org/wiki/SplitStacks.
virtual bool adjustPrologueForCrossSplitStack(uint8_t *Loc,
uint8_t *End) const;
// Return true if we can reach Dst from Src with Relocation RelocType
virtual bool inBranchRange(RelType Type, uint64_t Src,
uint64_t Dst) const;
@ -71,9 +82,10 @@ class TargetInfo {
uint64_t getImageBase();
// Offset of _GLOBAL_OFFSET_TABLE_ from base of .got section. Use -1 for
// end of .got
// Offset of _GLOBAL_OFFSET_TABLE_ from base of .got or .got.plt section.
uint64_t GotBaseSymOff = 0;
// True if _GLOBAL_OFFSET_TABLE_ is relative to .got.plt, false if .got.
bool GotBaseSymInGotPlt = true;
// On systems with range extensions we place collections of Thunks at
// regular spacings that enable the majority of branches reach the Thunks.
@ -97,9 +109,18 @@ class TargetInfo {
// to support lazy loading.
unsigned GotPltHeaderEntriesNum = 3;
// Set to 0 for variant 2
// On PPC ELF V2 abi, the first entry in the .got is the .TOC.
unsigned GotHeaderEntriesNum = 0;
// For TLS variant 1, the TCB is a fixed size specified by the Target.
// For variant 2, the TCB is an unspecified size.
// Set to 0 for variant 2.
unsigned TcbSize = 0;
// Set to the offset (in bytes) that the thread pointer is initialized to
// point to, relative to the start of the thread local storage.
unsigned TlsTpOffset = 0;
bool NeedsThunks = false;
// A 4-byte field corresponding to one or more trap instructions, used to pad
@ -126,6 +147,7 @@ TargetInfo *getAArch64TargetInfo();
TargetInfo *getAMDGPUTargetInfo();
TargetInfo *getARMTargetInfo();
TargetInfo *getAVRTargetInfo();
TargetInfo *getHexagonTargetInfo();
TargetInfo *getPPC64TargetInfo();
TargetInfo *getPPCTargetInfo();
TargetInfo *getSPARCV9TargetInfo();
@ -134,7 +156,17 @@ TargetInfo *getX86TargetInfo();
TargetInfo *getX86_64TargetInfo();
template <class ELFT> TargetInfo *getMipsTargetInfo();
std::string getErrorLocation(const uint8_t *Loc);
struct ErrorPlace {
InputSectionBase *IS;
std::string Loc;
};
// Returns input section and corresponding source string for the given location.
ErrorPlace getErrorPlace(const uint8_t *Loc);
static inline std::string getErrorLocation(const uint8_t *Loc) {
return getErrorPlace(Loc).Loc;
}
uint64_t getPPC64TocBase();
uint64_t getAArch64Page(uint64_t Expr);
@ -146,39 +178,74 @@ template <class ELFT> bool isMipsPIC(const Defined *Sym);
static inline void reportRangeError(uint8_t *Loc, RelType Type, const Twine &V,
int64_t Min, uint64_t Max) {
error(getErrorLocation(Loc) + "relocation " + lld::toString(Type) +
" out of range: " + V + " is not in [" + Twine(Min) + ", " +
Twine(Max) + "]");
ErrorPlace ErrPlace = getErrorPlace(Loc);
StringRef Hint;
if (ErrPlace.IS && ErrPlace.IS->Name.startswith(".debug"))
Hint = "; consider recompiling with -fdebug-types-section to reduce size "
"of debug sections";
error(ErrPlace.Loc + "relocation " + lld::toString(Type) +
" out of range: " + V.str() + " is not in [" + Twine(Min).str() + ", " +
Twine(Max).str() + "]" + Hint);
}
template <unsigned N>
static void checkInt(uint8_t *Loc, int64_t V, RelType Type) {
if (!llvm::isInt<N>(V))
// Sign-extend Nth bit all the way to MSB.
inline int64_t signExtend(uint64_t V, int N) {
return int64_t(V << (64 - N)) >> (64 - N);
}
// Make sure that V can be represented as an N bit signed integer.
inline void checkInt(uint8_t *Loc, int64_t V, int N, RelType Type) {
if (V != signExtend(V, N))
reportRangeError(Loc, Type, Twine(V), llvm::minIntN(N), llvm::maxIntN(N));
}
template <unsigned N>
static void checkUInt(uint8_t *Loc, uint64_t V, RelType Type) {
if (!llvm::isUInt<N>(V))
// Make sure that V can be represented as an N bit unsigned integer.
inline void checkUInt(uint8_t *Loc, uint64_t V, int N, RelType Type) {
if ((V >> N) != 0)
reportRangeError(Loc, Type, Twine(V), 0, llvm::maxUIntN(N));
}
template <unsigned N>
static void checkIntUInt(uint8_t *Loc, uint64_t V, RelType Type) {
if (!llvm::isInt<N>(V) && !llvm::isUInt<N>(V))
// For the error message we should cast V to a signed integer so that error
// messages show a small negative value rather than an extremely large one
// Make sure that V can be represented as an N bit signed or unsigned integer.
inline void checkIntUInt(uint8_t *Loc, uint64_t V, int N, RelType Type) {
// For the error message we should cast V to a signed integer so that error
// messages show a small negative value rather than an extremely large one
if (V != (uint64_t)signExtend(V, N) && (V >> N) != 0)
reportRangeError(Loc, Type, Twine((int64_t)V), llvm::minIntN(N),
llvm::maxUIntN(N));
llvm::maxIntN(N));
}
template <unsigned N>
static void checkAlignment(uint8_t *Loc, uint64_t V, RelType Type) {
inline void checkAlignment(uint8_t *Loc, uint64_t V, int N, RelType Type) {
if ((V & (N - 1)) != 0)
error(getErrorLocation(Loc) + "improper alignment for relocation " +
lld::toString(Type) + ": 0x" + llvm::utohexstr(V) +
" is not aligned to " + Twine(N) + " bytes");
}
// Endianness-aware read/write.
inline uint16_t read16(const void *P) {
return llvm::support::endian::read16(P, Config->Endianness);
}
inline uint32_t read32(const void *P) {
return llvm::support::endian::read32(P, Config->Endianness);
}
inline uint64_t read64(const void *P) {
return llvm::support::endian::read64(P, Config->Endianness);
}
inline void write16(void *P, uint16_t V) {
llvm::support::endian::write16(P, V, Config->Endianness);
}
inline void write32(void *P, uint32_t V) {
llvm::support::endian::write32(P, V, Config->Endianness);
}
inline void write64(void *P, uint64_t V) {
llvm::support::endian::write64(P, V, Config->Endianness);
}
} // namespace elf
} // namespace lld

390
contrib/llvm/tools/lld/ELF/Thunks.cpp
View File

@ -40,7 +40,6 @@
using namespace llvm;
using namespace llvm::object;
using namespace llvm::support::endian;
using namespace llvm::ELF;
namespace lld {
@ -52,59 +51,112 @@ namespace {
class AArch64ABSLongThunk final : public Thunk {
public:
AArch64ABSLongThunk(Symbol &Dest) : Thunk(Dest) {}
uint32_t size() const override { return 16; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t size() override { return 16; }
void writeTo(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
};
class AArch64ADRPThunk final : public Thunk {
public:
AArch64ADRPThunk(Symbol &Dest) : Thunk(Dest) {}
uint32_t size() const override { return 12; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t size() override { return 12; }
void writeTo(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
};
// Base class for ARM thunks.
//
// An ARM thunk may be either short or long. A short thunk is simply a branch
// (B) instruction, and it may be used to call ARM functions when the distance
// from the thunk to the target is less than 32MB. Long thunks can branch to any
// virtual address and can switch between ARM and Thumb, and they are
// implemented in the derived classes. This class tries to create a short thunk
// if the target is in range, otherwise it creates a long thunk.
class ARMThunk : public Thunk {
public:
ARMThunk(Symbol &Dest) : Thunk(Dest) {}
bool mayUseShortThunk();
uint32_t size() override { return mayUseShortThunk() ? 4 : sizeLong(); }
void writeTo(uint8_t *Buf) override;
bool isCompatibleWith(RelType Type) const override;
// Returns the size of a long thunk.
virtual uint32_t sizeLong() = 0;
// Writes a long thunk to Buf.
virtual void writeLong(uint8_t *Buf) = 0;
private:
// This field tracks whether all previously considered layouts would allow
// this thunk to be short. If we have ever needed a long thunk, we always
// create a long thunk, even if the thunk may be short given the current
// distance to the target. We do this because transitioning from long to short
// can create layout oscillations in certain corner cases which would prevent
// the layout from converging.
bool MayUseShortThunk = true;
};
// Base class for Thumb-2 thunks.
//
// This class is similar to ARMThunk, but it uses the Thumb-2 B.W instruction
// which has a range of 16MB.
class ThumbThunk : public Thunk {
public:
ThumbThunk(Symbol &Dest) : Thunk(Dest) { Alignment = 2; }
bool mayUseShortThunk();
uint32_t size() override { return mayUseShortThunk() ? 4 : sizeLong(); }
void writeTo(uint8_t *Buf) override;
bool isCompatibleWith(RelType Type) const override;
// Returns the size of a long thunk.
virtual uint32_t sizeLong() = 0;
// Writes a long thunk to Buf.
virtual void writeLong(uint8_t *Buf) = 0;
private:
// See comment in ARMThunk above.
bool MayUseShortThunk = true;
};
// Specific ARM Thunk implementations. The naming convention is:
// Source State, TargetState, Target Requirement, ABS or PI, Range
class ARMV7ABSLongThunk final : public Thunk {
class ARMV7ABSLongThunk final : public ARMThunk {
public:
ARMV7ABSLongThunk(Symbol &Dest) : Thunk(Dest) {}
ARMV7ABSLongThunk(Symbol &Dest) : ARMThunk(Dest) {}
uint32_t size() const override { return 12; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t sizeLong() override { return 12; }
void writeLong(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
bool isCompatibleWith(RelType Type) const override;
};
class ARMV7PILongThunk final : public Thunk {
class ARMV7PILongThunk final : public ARMThunk {
public:
ARMV7PILongThunk(Symbol &Dest) : Thunk(Dest) {}
ARMV7PILongThunk(Symbol &Dest) : ARMThunk(Dest) {}
uint32_t size() const override { return 16; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t sizeLong() override { return 16; }
void writeLong(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
bool isCompatibleWith(RelType Type) const override;
};
class ThumbV7ABSLongThunk final : public Thunk {
class ThumbV7ABSLongThunk final : public ThumbThunk {
public:
ThumbV7ABSLongThunk(Symbol &Dest) : Thunk(Dest) { Alignment = 2; }
ThumbV7ABSLongThunk(Symbol &Dest) : ThumbThunk(Dest) {}
uint32_t size() const override { return 10; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t sizeLong() override { return 10; }
void writeLong(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
bool isCompatibleWith(RelType Type) const override;
};
class ThumbV7PILongThunk final : public Thunk {
class ThumbV7PILongThunk final : public ThumbThunk {
public:
ThumbV7PILongThunk(Symbol &Dest) : Thunk(Dest) { Alignment = 2; }
ThumbV7PILongThunk(Symbol &Dest) : ThumbThunk(Dest) {}
uint32_t size() const override { return 12; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t sizeLong() override { return 12; }
void writeLong(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
bool isCompatibleWith(RelType Type) const override;
};
// MIPS LA25 thunk
@ -112,8 +164,8 @@ class MipsThunk final : public Thunk {
public:
MipsThunk(Symbol &Dest) : Thunk(Dest) {}
uint32_t size() const override { return 16; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t size() override { return 16; }
void writeTo(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
InputSection *getTargetInputSection() const override;
};
@ -123,8 +175,8 @@ class MicroMipsThunk final : public Thunk {
public:
MicroMipsThunk(Symbol &Dest) : Thunk(Dest) {}
uint32_t size() const override { return 14; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t size() override { return 14; }
void writeTo(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
InputSection *getTargetInputSection() const override;
};
@ -134,14 +186,44 @@ class MicroMipsR6Thunk final : public Thunk {
public:
MicroMipsR6Thunk(Symbol &Dest) : Thunk(Dest) {}
uint32_t size() const override { return 12; }
void writeTo(uint8_t *Buf, ThunkSection &IS) const override;
uint32_t size() override { return 12; }
void writeTo(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
InputSection *getTargetInputSection() const override;
};
// PPC64 Plt call stubs.
// Any call site that needs to call through a plt entry needs a call stub in
// the .text section. The call stub is responsible for:
// 1) Saving the toc-pointer to the stack.
// 2) Loading the target functions address from the procedure linkage table into
// r12 for use by the target functions global entry point, and into the count
// register.
// 3) Transfering control to the target function through an indirect branch.
class PPC64PltCallStub final : public Thunk {
public:
PPC64PltCallStub(Symbol &Dest) : Thunk(Dest) {}
uint32_t size() override { return 20; }
void writeTo(uint8_t *Buf) override;
void addSymbols(ThunkSection &IS) override;
};
} // end anonymous namespace
Defined *Thunk::addSymbol(StringRef Name, uint8_t Type, uint64_t Value,
InputSectionBase &Section) {
Defined *D = addSyntheticLocal(Name, Type, Value, /*Size=*/0, Section);
Syms.push_back(D);
return D;
}
void Thunk::setOffset(uint64_t NewOffset) {
for (Defined *D : Syms)
D->Value = D->Value - Offset + NewOffset;
Offset = NewOffset;
}
// AArch64 long range Thunks
static uint64_t getAArch64ThunkDestVA(const Symbol &S) {
@ -149,7 +231,7 @@ static uint64_t getAArch64ThunkDestVA(const Symbol &S) {
return V;
}
void AArch64ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
void AArch64ABSLongThunk::writeTo(uint8_t *Buf) {
const uint8_t Data[] = {
0x50, 0x00, 0x00, 0x58, // ldr x16, L0
0x00, 0x02, 0x1f, 0xd6, // br x16
@ -162,11 +244,10 @@ void AArch64ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
}
void AArch64ABSLongThunk::addSymbols(ThunkSection &IS) {
ThunkSym = addSyntheticLocal(
Saver.save("__AArch64AbsLongThunk_" + Destination.getName()), STT_FUNC,
Offset, size(), IS);
addSyntheticLocal("$x", STT_NOTYPE, Offset, 0, IS);
addSyntheticLocal("$d", STT_NOTYPE, Offset + 8, 0, IS);
addSymbol(Saver.save("__AArch64AbsLongThunk_" + Destination.getName()),
STT_FUNC, 0, IS);
addSymbol("$x", STT_NOTYPE, 0, IS);
addSymbol("$d", STT_NOTYPE, 8, IS);
}
// This Thunk has a maximum range of 4Gb, this is sufficient for all programs
@ -174,26 +255,24 @@ void AArch64ABSLongThunk::addSymbols(ThunkSection &IS) {
// clang and gcc do not support the large code model for position independent
// code so it is safe to use this for position independent thunks without
// worrying about the destination being more than 4Gb away.
void AArch64ADRPThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
void AArch64ADRPThunk::writeTo(uint8_t *Buf) {
const uint8_t Data[] = {
0x10, 0x00, 0x00, 0x90, // adrp x16, Dest R_AARCH64_ADR_PREL_PG_HI21(Dest)
0x10, 0x02, 0x00, 0x91, // add x16, x16, R_AARCH64_ADD_ABS_LO12_NC(Dest)
0x00, 0x02, 0x1f, 0xd6, // br x16
};
uint64_t S = getAArch64ThunkDestVA(Destination);
uint64_t P = ThunkSym->getVA();
uint64_t P = getThunkTargetSym()->getVA();
memcpy(Buf, Data, sizeof(Data));
Target->relocateOne(Buf, R_AARCH64_ADR_PREL_PG_HI21,
getAArch64Page(S) - getAArch64Page(P));
Target->relocateOne(Buf + 4, R_AARCH64_ADD_ABS_LO12_NC, S);
}
void AArch64ADRPThunk::addSymbols(ThunkSection &IS)
{
ThunkSym = addSyntheticLocal(
Saver.save("__AArch64ADRPThunk_" + Destination.getName()), STT_FUNC,
Offset, size(), IS);
addSyntheticLocal("$x", STT_NOTYPE, Offset, 0, IS);
void AArch64ADRPThunk::addSymbols(ThunkSection &IS) {
addSymbol(Saver.save("__AArch64ADRPThunk_" + Destination.getName()), STT_FUNC,
0, IS);
addSymbol("$x", STT_NOTYPE, 0, IS);
}
// ARM Target Thunks
@ -202,7 +281,81 @@ static uint64_t getARMThunkDestVA(const Symbol &S) {
return SignExtend64<32>(V);
}
void ARMV7ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
// This function returns true if the target is not Thumb and is within 2^26, and
// it has not previously returned false (see comment for MayUseShortThunk).
bool ARMThunk::mayUseShortThunk() {
if (!MayUseShortThunk)
return false;
uint64_t S = getARMThunkDestVA(Destination);
if (S & 1) {
MayUseShortThunk = false;
return false;
}
uint64_t P = getThunkTargetSym()->getVA();
int64_t Offset = S - P - 8;
MayUseShortThunk = llvm::isInt<26>(Offset);
return MayUseShortThunk;
}
void ARMThunk::writeTo(uint8_t *Buf) {
if (!mayUseShortThunk()) {
writeLong(Buf);
return;
}
uint64_t S = getARMThunkDestVA(Destination);
uint64_t P = getThunkTargetSym()->getVA();
int64_t Offset = S - P - 8;
const uint8_t Data[] = {
0x00, 0x00, 0x00, 0xea, // b S
};
memcpy(Buf, Data, sizeof(Data));
Target->relocateOne(Buf, R_ARM_JUMP24, Offset);
}
bool ARMThunk::isCompatibleWith(RelType Type) const {
// Thumb branch relocations can't use BLX
return Type != R_ARM_THM_JUMP19 && Type != R_ARM_THM_JUMP24;
}
// This function returns true if the target is Thumb and is within 2^25, and
// it has not previously returned false (see comment for MayUseShortThunk).
bool ThumbThunk::mayUseShortThunk() {
if (!MayUseShortThunk)
return false;
uint64_t S = getARMThunkDestVA(Destination);
if ((S & 1) == 0) {
MayUseShortThunk = false;
return false;
}
uint64_t P = getThunkTargetSym()->getVA() & ~1;
int64_t Offset = S - P - 4;
MayUseShortThunk = llvm::isInt<25>(Offset);
return MayUseShortThunk;
}
void ThumbThunk::writeTo(uint8_t *Buf) {
if (!mayUseShortThunk()) {
writeLong(Buf);
return;
}
uint64_t S = getARMThunkDestVA(Destination);
uint64_t P = getThunkTargetSym()->getVA();
int64_t Offset = S - P - 4;
const uint8_t Data[] = {
0x00, 0xf0, 0x00, 0xb0, // b.w S
};
memcpy(Buf, Data, sizeof(Data));
Target->relocateOne(Buf, R_ARM_THM_JUMP24, Offset);
}
bool ThumbThunk::isCompatibleWith(RelType Type) const {
// ARM branch relocations can't use BLX
return Type != R_ARM_JUMP24 && Type != R_ARM_PC24 && Type != R_ARM_PLT32;
}
void ARMV7ABSLongThunk::writeLong(uint8_t *Buf) {
const uint8_t Data[] = {
0x00, 0xc0, 0x00, 0xe3, // movw ip,:lower16:S
0x00, 0xc0, 0x40, 0xe3, // movt ip,:upper16:S
@ -215,18 +368,12 @@ void ARMV7ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
}
void ARMV7ABSLongThunk::addSymbols(ThunkSection &IS) {
ThunkSym = addSyntheticLocal(
Saver.save("__ARMv7ABSLongThunk_" + Destination.getName()), STT_FUNC,
Offset, size(), IS);
addSyntheticLocal("$a", STT_NOTYPE, Offset, 0, IS);
addSymbol(Saver.save("__ARMv7ABSLongThunk_" + Destination.getName()),
STT_FUNC, 0, IS);
addSymbol("$a", STT_NOTYPE, 0, IS);
}
bool ARMV7ABSLongThunk::isCompatibleWith(RelType Type) const {
// Thumb branch relocations can't use BLX
return Type != R_ARM_THM_JUMP19 && Type != R_ARM_THM_JUMP24;
}
void ThumbV7ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
void ThumbV7ABSLongThunk::writeLong(uint8_t *Buf) {
const uint8_t Data[] = {
0x40, 0xf2, 0x00, 0x0c, // movw ip, :lower16:S
0xc0, 0xf2, 0x00, 0x0c, // movt ip, :upper16:S
@ -239,18 +386,12 @@ void ThumbV7ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
}
void ThumbV7ABSLongThunk::addSymbols(ThunkSection &IS) {
ThunkSym = addSyntheticLocal(
Saver.save("__Thumbv7ABSLongThunk_" + Destination.getName()), STT_FUNC,
Offset | 0x1, size(), IS);
addSyntheticLocal("$t", STT_NOTYPE, Offset, 0, IS);
addSymbol(Saver.save("__Thumbv7ABSLongThunk_" + Destination.getName()),
STT_FUNC, 1, IS);
addSymbol("$t", STT_NOTYPE, 0, IS);
}
bool ThumbV7ABSLongThunk::isCompatibleWith(RelType Type) const {
// ARM branch relocations can't use BLX
return Type != R_ARM_JUMP24 && Type != R_ARM_PC24 && Type != R_ARM_PLT32;
}
void ARMV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
void ARMV7PILongThunk::writeLong(uint8_t *Buf) {
const uint8_t Data[] = {
0xf0, 0xcf, 0x0f, 0xe3, // P: movw ip,:lower16:S - (P + (L1-P) + 8)
0x00, 0xc0, 0x40, 0xe3, // movt ip,:upper16:S - (P + (L1-P) + 8)
@ -258,7 +399,7 @@ void ARMV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
0x1c, 0xff, 0x2f, 0xe1, // bx r12
};
uint64_t S = getARMThunkDestVA(Destination);
uint64_t P = ThunkSym->getVA();
uint64_t P = getThunkTargetSym()->getVA();
uint64_t Offset = S - P - 16;
memcpy(Buf, Data, sizeof(Data));
Target->relocateOne(Buf, R_ARM_MOVW_PREL_NC, Offset);
@ -266,18 +407,12 @@ void ARMV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
}
void ARMV7PILongThunk::addSymbols(ThunkSection &IS) {
ThunkSym = addSyntheticLocal(
Saver.save("__ARMV7PILongThunk_" + Destination.getName()), STT_FUNC,
Offset, size(), IS);
addSyntheticLocal("$a", STT_NOTYPE, Offset, 0, IS);
addSymbol(Saver.save("__ARMV7PILongThunk_" + Destination.getName()), STT_FUNC,
0, IS);
addSymbol("$a", STT_NOTYPE, 0, IS);
}
bool ARMV7PILongThunk::isCompatibleWith(RelType Type) const {
// Thumb branch relocations can't use BLX
return Type != R_ARM_THM_JUMP19 && Type != R_ARM_THM_JUMP24;
}
void ThumbV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
void ThumbV7PILongThunk::writeLong(uint8_t *Buf) {
const uint8_t Data[] = {
0x4f, 0xf6, 0xf4, 0x7c, // P: movw ip,:lower16:S - (P + (L1-P) + 4)
0xc0, 0xf2, 0x00, 0x0c, // movt ip,:upper16:S - (P + (L1-P) + 4)
@ -285,7 +420,7 @@ void ThumbV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
0x60, 0x47, // bx r12
};
uint64_t S = getARMThunkDestVA(Destination);
uint64_t P = ThunkSym->getVA() & ~0x1;
uint64_t P = getThunkTargetSym()->getVA() & ~0x1;
uint64_t Offset = S - P - 12;
memcpy(Buf, Data, sizeof(Data));
Target->relocateOne(Buf, R_ARM_THM_MOVW_PREL_NC, Offset);
@ -293,32 +428,25 @@ void ThumbV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const {
}
void ThumbV7PILongThunk::addSymbols(ThunkSection &IS) {
ThunkSym = addSyntheticLocal(
Saver.save("__ThumbV7PILongThunk_" + Destination.getName()), STT_FUNC,
Offset | 0x1, size(), IS);
addSyntheticLocal("$t", STT_NOTYPE, Offset, 0, IS);
}
bool ThumbV7PILongThunk::isCompatibleWith(RelType Type) const {
// ARM branch relocations can't use BLX
return Type != R_ARM_JUMP24 && Type != R_ARM_PC24 && Type != R_ARM_PLT32;
addSymbol(Saver.save("__ThumbV7PILongThunk_" + Destination.getName()),
STT_FUNC, 1, IS);
addSymbol("$t", STT_NOTYPE, 0, IS);
}
// Write MIPS LA25 thunk code to call PIC function from the non-PIC one.
void MipsThunk::writeTo(uint8_t *Buf, ThunkSection &) const {
void MipsThunk::writeTo(uint8_t *Buf) {
uint64_t S = Destination.getVA();
write32(Buf, 0x3c190000, Config->Endianness); // lui $25, %hi(func)
write32(Buf + 4, 0x08000000 | (S >> 2), Config->Endianness); // j func
write32(Buf + 8, 0x27390000, Config->Endianness); // addiu $25, $25, %lo(func)
write32(Buf + 12, 0x00000000, Config->Endianness); // nop
write32(Buf, 0x3c190000); // lui $25, %hi(func)
write32(Buf + 4, 0x08000000 | (S >> 2)); // j func
write32(Buf + 8, 0x27390000); // addiu $25, $25, %lo(func)
write32(Buf + 12, 0x00000000); // nop
Target->relocateOne(Buf, R_MIPS_HI16, S);
Target->relocateOne(Buf + 8, R_MIPS_LO16, S);
}
void MipsThunk::addSymbols(ThunkSection &IS) {
ThunkSym =
addSyntheticLocal(Saver.save("__LA25Thunk_" + Destination.getName()),
STT_FUNC, Offset, size(), IS);
addSymbol(Saver.save("__LA25Thunk_" + Destination.getName()), STT_FUNC, 0,
IS);
}
InputSection *MipsThunk::getTargetInputSection() const {
@ -328,22 +456,21 @@ InputSection *MipsThunk::getTargetInputSection() const {
// Write microMIPS R2-R5 LA25 thunk code
// to call PIC function from the non-PIC one.
void MicroMipsThunk::writeTo(uint8_t *Buf, ThunkSection &) const {
void MicroMipsThunk::writeTo(uint8_t *Buf) {
uint64_t S = Destination.getVA() | 1;
write16(Buf, 0x41b9, Config->Endianness); // lui $25, %hi(func)
write16(Buf + 4, 0xd400, Config->Endianness); // j func
write16(Buf + 8, 0x3339, Config->Endianness); // addiu $25, $25, %lo(func)
write16(Buf + 12, 0x0c00, Config->Endianness); // nop
write16(Buf, 0x41b9); // lui $25, %hi(func)
write16(Buf + 4, 0xd400); // j func
write16(Buf + 8, 0x3339); // addiu $25, $25, %lo(func)
write16(Buf + 12, 0x0c00); // nop
Target->relocateOne(Buf, R_MICROMIPS_HI16, S);
Target->relocateOne(Buf + 4, R_MICROMIPS_26_S1, S);
Target->relocateOne(Buf + 8, R_MICROMIPS_LO16, S);
}
void MicroMipsThunk::addSymbols(ThunkSection &IS) {
ThunkSym =
addSyntheticLocal(Saver.save("__microLA25Thunk_" + Destination.getName()),
STT_FUNC, Offset, size(), IS);
ThunkSym->StOther |= STO_MIPS_MICROMIPS;
Defined *D = addSymbol(
Saver.save("__microLA25Thunk_" + Destination.getName()), STT_FUNC, 0, IS);
D->StOther |= STO_MIPS_MICROMIPS;
}
InputSection *MicroMipsThunk::getTargetInputSection() const {
@ -353,22 +480,21 @@ InputSection *MicroMipsThunk::getTargetInputSection() const {
// Write microMIPS R6 LA25 thunk code
// to call PIC function from the non-PIC one.
void MicroMipsR6Thunk::writeTo(uint8_t *Buf, ThunkSection &) const {
void MicroMipsR6Thunk::writeTo(uint8_t *Buf) {
uint64_t S = Destination.getVA() | 1;
uint64_t P = ThunkSym->getVA();
write16(Buf, 0x1320, Config->Endianness); // lui $25, %hi(func)
write16(Buf + 4, 0x3339, Config->Endianness); // addiu $25, $25, %lo(func)
write16(Buf + 8, 0x9400, Config->Endianness); // bc func
uint64_t P = getThunkTargetSym()->getVA();
write16(Buf, 0x1320); // lui $25, %hi(func)
write16(Buf + 4, 0x3339); // addiu $25, $25, %lo(func)
write16(Buf + 8, 0x9400); // bc func
Target->relocateOne(Buf, R_MICROMIPS_HI16, S);
Target->relocateOne(Buf + 4, R_MICROMIPS_LO16, S);
Target->relocateOne(Buf + 8, R_MICROMIPS_PC26_S1, S - P - 12);
}
void MicroMipsR6Thunk::addSymbols(ThunkSection &IS) {
ThunkSym =
addSyntheticLocal(Saver.save("__microLA25Thunk_" + Destination.getName()),
STT_FUNC, Offset, size(), IS);
ThunkSym->StOther |= STO_MIPS_MICROMIPS;
Defined *D = addSymbol(
Saver.save("__microLA25Thunk_" + Destination.getName()), STT_FUNC, 0, IS);
D->StOther |= STO_MIPS_MICROMIPS;
}
InputSection *MicroMipsR6Thunk::getTargetInputSection() const {
@ -376,6 +502,25 @@ InputSection *MicroMipsR6Thunk::getTargetInputSection() const {
return dyn_cast<InputSection>(DR.Section);
}
void PPC64PltCallStub::writeTo(uint8_t *Buf) {
int64_t Off = Destination.getGotPltVA() - getPPC64TocBase();
// Need to add 0x8000 to offset to account for the low bits being signed.
uint16_t OffHa = (Off + 0x8000) >> 16;
uint16_t OffLo = Off;
write32(Buf + 0, 0xf8410018); // std r2,24(r1)
write32(Buf + 4, 0x3d820000 | OffHa); // addis r12,r2, X@plt@to@ha
write32(Buf + 8, 0xe98c0000 | OffLo); // ld r12,X@plt@toc@l(r12)
write32(Buf + 12, 0x7d8903a6); // mtctr r12
write32(Buf + 16, 0x4e800420); // bctr
}
void PPC64PltCallStub::addSymbols(ThunkSection &IS) {
Defined *S = addSymbol(Saver.save("__plt_" + Destination.getName()), STT_FUNC,
0, IS);
S->NeedsTocRestore = true;
}
Thunk::Thunk(Symbol &D) : Destination(D), Offset(0) {}
Thunk::~Thunk() = default;
@ -419,15 +564,26 @@ static Thunk *addThunkMips(RelType Type, Symbol &S) {
return make<MipsThunk>(S);
}
static Thunk *addThunkPPC64(RelType Type, Symbol &S) {
if (Type == R_PPC64_REL24)
return make<PPC64PltCallStub>(S);
fatal("unexpected relocation type");
}
Thunk *addThunk(RelType Type, Symbol &S) {
if (Config->EMachine == EM_AARCH64)
return addThunkAArch64(Type, S);
else if (Config->EMachine == EM_ARM)
if (Config->EMachine == EM_ARM)
return addThunkArm(Type, S);
else if (Config->EMachine == EM_MIPS)
if (Config->EMachine == EM_MIPS)
return addThunkMips(Type, S);
llvm_unreachable("add Thunk only supported for ARM and Mips");
return nullptr;
if (Config->EMachine == EM_PPC64)
return addThunkPPC64(Type, S);
llvm_unreachable("add Thunk only supported for ARM, Mips and PowerPC");
}
} // end namespace elf

20
contrib/llvm/tools/lld/ELF/Thunks.h
View File

@ -14,6 +14,7 @@
namespace lld {
namespace elf {
class Defined;
class Symbol;
class ThunkSection;
// Class to describe an instance of a Thunk.
@ -30,12 +31,17 @@ class Thunk {
Thunk(Symbol &Destination);
virtual ~Thunk();
virtual uint32_t size() const { return 0; }
virtual void writeTo(uint8_t *Buf, ThunkSection &IS) const {}
virtual uint32_t size() = 0;
virtual void writeTo(uint8_t *Buf) = 0;
// All Thunks must define at least one symbol ThunkSym so that we can
// redirect relocations to it.
virtual void addSymbols(ThunkSection &IS) {}
// All Thunks must define at least one symbol, known as the thunk target
// symbol, so that we can redirect relocations to it. The thunk may define
// additional symbols, but these are never targets for relocations.
virtual void addSymbols(ThunkSection &IS) = 0;
void setOffset(uint64_t Offset);
Defined *addSymbol(StringRef Name, uint8_t Type, uint64_t Value,
InputSectionBase &Section);
// Some Thunks must be placed immediately before their Target as they elide
// a branch and fall through to the first Symbol in the Target.
@ -45,10 +51,12 @@ class Thunk {
// compatible with it.
virtual bool isCompatibleWith(RelType Type) const { return true; }
Defined *getThunkTargetSym() const { return Syms[0]; }
// The alignment requirement for this Thunk, defaults to the size of the
// typical code section alignment.
Symbol &Destination;
Symbol *ThunkSym;
llvm::SmallVector<Defined *, 3> Syms;
uint64_t Offset = 0;
uint32_t Alignment = 4;
};

719
contrib/llvm/tools/lld/ELF/Writer.cpp
View File

File diff suppressed because it is too large Load Diff

8
contrib/llvm/tools/lld/ELF/Writer.h
View File

@ -23,7 +23,6 @@ class InputSectionBase;
template <class ELFT> class ObjFile;
class SymbolTable;
template <class ELFT> void writeResult();
template <class ELFT> void markLive();
// This describes a program header entry.
// Each contains type, access flags and range of output sections that will be
@ -45,16 +44,11 @@ struct PhdrEntry {
OutputSection *LastSec = nullptr;
bool HasLMA = false;
// True if one of the sections in this program header has a LMA specified via
// linker script: AT(addr). We never allow 2 or more sections with LMA in the
// same program header.
bool ASectionHasLMA = false;
uint64_t LMAOffset = 0;
};
void addReservedSymbols();
llvm::StringRef getOutputSectionName(InputSectionBase *S);
llvm::StringRef getOutputSectionName(const InputSectionBase *S);
template <class ELFT> uint32_t calcMipsEFlags();

2
contrib/llvm/tools/lld/LICENSE.TXT
View File

@ -4,7 +4,7 @@ lld License
University of Illinois/NCSA
Open Source License
Copyright (c) 2011-2016 by the contributors listed in CREDITS.TXT
Copyright (c) 2011-2018 by the contributors listed in CREDITS.TXT
All rights reserved.
Developed by:

2
contrib/llvm/tools/lld/include/lld/Common/Driver.h
View File

@ -30,7 +30,7 @@ bool link(llvm::ArrayRef<const char *> Args, bool CanExitEarly,
}
namespace mach_o {
bool link(llvm::ArrayRef<const char *> Args,
bool link(llvm::ArrayRef<const char *> Args, bool CanExitEarly,
llvm::raw_ostream &Diag = llvm::errs());
}

70
contrib/llvm/tools/lld/include/lld/Common/ErrorHandler.h
View File

@ -7,21 +7,62 @@
//
//===----------------------------------------------------------------------===//
//
// In LLD, we have three levels of errors: fatal, error or warn.
// We designed lld's error handlers with the following goals in mind:
//
// Fatal makes the program exit immediately with an error message.
// You shouldn't use it except for reporting a corrupted input file.
// - Errors can occur at any place where we handle user input, but we don't
// want them to affect the normal execution path too much. Ideally,
// handling errors should be as simple as reporting them and exit (but
// without actually doing exit).
//
// Error prints out an error message and increment a global variable
// ErrorCount to record the fact that we met an error condition. It does
// not exit, so it is safe for a lld-as-a-library use case. It is generally
// useful because it can report more than one error in a single run.
// In particular, the design to wrap all functions that could fail with
// ErrorOr<T> is rejected because otherwise we would have to wrap a large
// number of functions in lld with ErrorOr. With that approach, if some
// function F can fail, not only F but all functions that transitively call
// F have to be wrapped with ErrorOr. That seemed too much.
//
// Warn doesn't do anything but printing out a given message.
// - Finding only one error at a time is not sufficient. We want to find as
// many errors as possible with one execution of the linker. That means the
// linker needs to keep running after a first error and give up at some
// checkpoint (beyond which it would find cascading, false errors caused by
// the previous errors).
//
// It is not recommended to use llvm::outs() or llvm::errs() directly
// in LLD because they are not thread-safe. The functions declared in
// this file are mutually excluded, so you want to use them instead.
// - We want a simple interface to report errors. Unlike Clang, the data we
// handle is compiled binary, so we don't need an error reporting mechanism
// that's as sophisticated as the one that Clang has.
//
// The current lld's error handling mechanism is simple:
//
// - When you find an error, report it using error() and continue as far as
// you can. An internal error counter is incremented by one every time you
// call error().
//
// A common idiom to handle an error is calling error() and then returning
// a reasonable default value. For example, if your function handles a
// user-supplied alignment value, and if you find an invalid alignment
// (e.g. 17 which is not 2^n), you may report it using error() and continue
// as if it were alignment 1 (which is the simplest reasonable value).
//
// Note that you should not continue with an invalid value; that breaks the
// internal consistency. You need to maintain all variables have some sane
// value even after an error occurred. So, when you have to continue with
// some value, always use a dummy value.
//
// - Find a reasonable checkpoint at where you want to stop the linker, and
// add code to return from the function if errorCount() > 0. In most cases,
// a checkpoint already exists, so you don't need to do anything for this.
//
// This interface satisfies all the goals that we mentioned above.
//
// You should never call fatal() except for reporting a corrupted input file.
// fatal() immediately terminates the linker, so the function is not desirable
// if you are using lld as a subroutine in other program, and with that you
// can find only one error at a time.
//
// warn() doesn't do anything but printing out a given message.
//
// It is not recommended to use llvm::outs() or llvm::errs() directly in lld
// because they are not thread-safe. The functions declared in this file are
// thread-safe.
//
//===----------------------------------------------------------------------===//
@ -34,6 +75,10 @@
#include "llvm/Support/Error.h"
#include "llvm/Support/FileOutputBuffer.h"
namespace llvm {
class DiagnosticInfo;
}
namespace lld {
class ErrorHandler {
@ -74,6 +119,9 @@ inline uint64_t errorCount() { return errorHandler().ErrorCount; }
LLVM_ATTRIBUTE_NORETURN void exitLld(int Val);
void diagnosticHandler(const llvm::DiagnosticInfo &DI);
void checkError(Error E);
// check functions are convenient functions to strip errors
// from error-or-value objects.
template <class T> T check(ErrorOr<T> E) {

26
contrib/llvm/tools/lld/include/lld/Common/Strings.h
View File

@ -10,14 +10,40 @@
#ifndef LLD_STRINGS_H
#define LLD_STRINGS_H
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/Support/GlobPattern.h"
#include <string>
#include <vector>
namespace lld {
// Returns a demangled C++ symbol name. If Name is not a mangled
// name, it returns Optional::None.
llvm::Optional<std::string> demangleItanium(llvm::StringRef Name);
llvm::Optional<std::string> demangleMSVC(llvm::StringRef S);
std::vector<uint8_t> parseHex(llvm::StringRef S);
bool isValidCIdentifier(llvm::StringRef S);
// Write the contents of the a buffer to a file
void saveBuffer(llvm::StringRef Buffer, const llvm::Twine &Path);
// This class represents multiple glob patterns.
class StringMatcher {
public:
StringMatcher() = default;
explicit StringMatcher(llvm::ArrayRef<llvm::StringRef> Pat);
bool match(llvm::StringRef S) const;
private:
std::vector<llvm::GlobPattern> Patterns;
};
inline llvm::ArrayRef<uint8_t> toArrayRef(llvm::StringRef S) {
return {reinterpret_cast<const uint8_t *>(S.data()), S.size()};
}
} // namespace lld
#endif

1
contrib/llvm/tools/lld/include/lld/Common/TargetOptionsCommandFlags.h
View File

@ -18,4 +18,5 @@
namespace lld {
llvm::TargetOptions InitTargetOptionsFromCodeGenFlags();
llvm::Optional<llvm::CodeModel::Model> GetCodeModelFromCMModel();
std::string GetCPUStr();
}

59
contrib/llvm/tools/lld/include/lld/Common/Timer.h Normal file
View File

@ -0,0 +1,59 @@
//===- Timer.h ----------------------------------------------*- C++ -*-===//
//
// The LLVM Linker
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#ifndef LLD_COMMON_TIMER_H
#define LLD_COMMON_TIMER_H
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/StringRef.h"
#include <assert.h>
#include <chrono>
#include <map>
#include <memory>
namespace lld {
class Timer;
struct ScopedTimer {
explicit ScopedTimer(Timer &T);
~ScopedTimer();
void stop();
Timer *T = nullptr;
};
class Timer {
public:
Timer(llvm::StringRef Name, Timer &Parent);
static Timer &root();
void start();
void stop();
void print();
double millis() const;
private:
explicit Timer(llvm::StringRef Name);
void print(int Depth, double TotalDuration, bool Recurse = true) const;
std::chrono::time_point<std::chrono::high_resolution_clock> StartTime;
std::chrono::nanoseconds Total;
std::vector<Timer *> Children;
std::string Name;
Timer *Parent;
};
} // namespace lld
#endif

Some files were not shown because too many files have changed in this diff Show More