rui314/mold
1
1
Fork 0
mirror of https://github.com/rui314/mold.git synced 2024-09-22 02:20:51 +03:00
Code Issues Projects Releases Wiki Activity
439ace2597
mold/input_sections.cc
Rui Ueyama 826055a9c6 wip
2021-02-28 11:11:49 +09:00

643 lines
19 KiB
C++
Raw Blame History

#include "mold.h"
#include <limits>
#include <zlib.h>
static u64 read64be(u8 *buf) {
return ((u64)buf[0] << 56) | ((u64)buf[1] << 48) |
((u64)buf[2] << 40) | ((u64)buf[3] << 32) |
((u64)buf[4] << 24) | ((u64)buf[5] << 16) |
((u64)buf[6] << 8) | (u64)buf[7];
}
InputChunk::InputChunk(ObjectFile *file, const ElfShdr *shdr,
std::string_view name)
: file(file), shdr(shdr), name(name),
output_section(OutputSection::get_instance(name, shdr->sh_type, shdr->sh_flags)) {
auto do_uncompress = [&](std::string_view data, u64 size) {
u8 *buf = new u8[size];
unsigned long size2 = size;
if (uncompress(buf, &size2, (u8 *)&data[0], data.size()) != Z_OK)
Fatal() << *this << ": uncompress failed";
if (size != size2)
Fatal() << *this << ": uncompress: invalid size";
ElfShdr *shdr2 = new ElfShdr;
*shdr2 = *shdr;
shdr2->sh_size = size;
shdr2->sh_flags &= ~(u64)SHF_COMPRESSED;
shdr = shdr2;
return std::string_view((char *)buf, size);
};
if (name.starts_with(".zdebug")) {
// Old-style compressed section
std::string_view data = file->get_string(*shdr);
if (!data.starts_with("ZLIB") || data.size() <= 12)
Fatal() << *this << ": corrupted compressed section";
u64 size = read64be((u8 *)&data[4]);
contents = do_uncompress(data.substr(12), size);
// Rename .zdebug -> .debug
name = *new std::string("." + std::string(name.substr(2)));
} else if (shdr->sh_flags & SHF_COMPRESSED) {
// New-style compressed section
std::string_view data = file->get_string(*shdr);
if (data.size() < sizeof(ElfChdr))
Fatal() << *this << ": corrupted compressed section";
ElfChdr &hdr = *(ElfChdr *)&data[0];
if (hdr.ch_type != ELFCOMPRESS_ZLIB)
Fatal() << *this << ": unsupported compression type";
contents = do_uncompress(data.substr(sizeof(ElfChdr)), hdr.ch_size);
} else if (shdr->sh_type != SHT_NOBITS) {
contents = file->get_string(*shdr);
}
}
static std::string rel_to_string(u64 r_type) {
switch (r_type) {
case R_X86_64_NONE: return "R_X86_64_NONE";
case R_X86_64_8: return "R_X86_64_8";
case R_X86_64_16: return "R_X86_64_16";
case R_X86_64_32: return "R_X86_64_32";
case R_X86_64_32S: return "R_X86_64_32S";
case R_X86_64_64: return "R_X86_64_64";
case R_X86_64_PC8: return "R_X86_64_PC8";
case R_X86_64_PC16: return "R_X86_64_PC16";
case R_X86_64_PC32: return "R_X86_64_PC32";
case R_X86_64_PC64: return "R_X86_64_PC64";
case R_X86_64_GOT32: return "R_X86_64_GOT32";
case R_X86_64_GOTPC32: return "R_X86_64_GOTPC32";
case R_X86_64_GOTPCREL: return "R_X86_64_GOTPCREL";
case R_X86_64_GOTPCRELX: return "R_X86_64_GOTPCRELX";
case R_X86_64_REX_GOTPCRELX: return "R_X86_64_REX_GOTPCRELX";
case R_X86_64_PLT32: return "R_X86_64_PLT32";
case R_X86_64_TLSGD: return "R_X86_64_TLSGD";
case R_X86_64_TLSLD: return "R_X86_64_TLSLD";
case R_X86_64_TPOFF32: return "R_X86_64_TPOFF32";
case R_X86_64_DTPOFF32: return "R_X86_64_DTPOFF32";
case R_X86_64_TPOFF64: return "R_X86_64_TPOFF64";
case R_X86_64_DTPOFF64: return "R_X86_64_DTPOFF64";
case R_X86_64_GOTTPOFF: return "R_X86_64_GOTTPOFF";
}
unreachable();
}
static void overflow_check(InputSection *sec, Symbol &sym, u64 r_type, u64 val) {
switch (r_type) {
case R_X86_64_8:
if (val != (u8)val)
Error() << *sec << ": relocation R_X86_64_8 against " << sym
<< " out of range: " << val << " is not in [0, 255]";
return;
case R_X86_64_PC8:
if (val != (i8)val)
Error() << *sec << ": relocation R_X86_64_PC8 against " << sym
<< " out of range: " << (i64)val << " is not in [-128, 127]";
return;
case R_X86_64_16:
if (val != (u16)val)
Error() << *sec << ": relocation R_X86_64_16 against " << sym
<< " out of range: " << val << " is not in [0, 65535]";
return;
case R_X86_64_PC16:
if (val != (i16)val)
Error() << *sec << ": relocation R_X86_64_PC16 against " << sym
<< " out of range: " << (i64)val << " is not in [-32768, 32767]";
return;
case R_X86_64_32:
if (val != (u32)val)
Error() << *sec << ": relocation R_X86_64_32 against " << sym
<< " out of range: " << val << " is not in [0, 4294967296]";
return;
case R_X86_64_32S:
case R_X86_64_PC32:
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:
case R_X86_64_PLT32:
case R_X86_64_TLSGD:
case R_X86_64_TLSLD:
case R_X86_64_TPOFF32:
case R_X86_64_DTPOFF32:
case R_X86_64_GOTTPOFF:
if (val != (i32)val)
Error() << *sec << ": relocation " << rel_to_string(r_type)
<< " against " << sym << " out of range: " << (i64)val
<< " is not in [-2147483648, 2147483647]";
return;
case R_X86_64_NONE:
case R_X86_64_64:
case R_X86_64_PC64:
case R_X86_64_TPOFF64:
case R_X86_64_DTPOFF64:
return;
}
unreachable();
}
static void write_val(u64 r_type, u8 *loc, u64 val) {
switch (r_type) {
case R_X86_64_NONE:
return;
case R_X86_64_8:
case R_X86_64_PC8:
*loc = val;
return;
case R_X86_64_16:
case R_X86_64_PC16:
*(u16 *)loc = val;
return;
case R_X86_64_32:
case R_X86_64_32S:
case R_X86_64_PC32:
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:
case R_X86_64_PLT32:
case R_X86_64_TLSGD:
case R_X86_64_TLSLD:
case R_X86_64_TPOFF32:
case R_X86_64_DTPOFF32:
case R_X86_64_GOTTPOFF:
*(u32 *)loc = val;
return;
case R_X86_64_64:
case R_X86_64_PC64:
case R_X86_64_TPOFF64:
case R_X86_64_DTPOFF64:
*(u64 *)loc = val;
return;
}
unreachable();
}
void InputSection::copy_buf() {
if (shdr->sh_type == SHT_NOBITS || shdr->sh_size == 0)
return;
// Copy data
u8 *base = out::buf + output_section->shdr.sh_offset + offset;
memcpy(base, contents.data(), contents.size());
// Apply relocations
if (shdr->sh_flags & SHF_ALLOC)
apply_reloc_alloc(base);
else
apply_reloc_nonalloc(base);
}
// Apply relocations to SHF_ALLOC sections (i.e. sections that are
// mapped to memory at runtime) based on the result of
// scan_relocations().
void InputSection::apply_reloc_alloc(u8 *base) {
i64 ref_idx = 0;
ElfRela *dynrel = nullptr;
if (out::reldyn)
dynrel = (ElfRela *)(out::buf + out::reldyn->shdr.sh_offset +
file->reldyn_offset + reldyn_offset);
for (i64 i = 0; i < rels.size(); i++) {
const ElfRela &rel = rels[i];
Symbol &sym = *file->symbols[rel.r_sym];
u8 *loc = base + rel.r_offset;
const SectionFragmentRef *ref = nullptr;
if (has_fragments[i])
ref = &rel_fragments[ref_idx++];
auto write = [&](u64 val) {
overflow_check(this, sym, rel.r_type, val);
write_val(rel.r_type, loc, val);
};
#define S (ref ? ref->frag->get_addr() \
: (sym.plt_idx == -1 ? sym.get_addr() : sym.get_plt_addr()))
#define A (ref ? ref->addend : rel.r_addend)
#define P (output_section->shdr.sh_addr + offset + rel.r_offset)
#define G (sym.get_got_addr() - out::got->shdr.sh_addr)
#define GOT out::got->shdr.sh_addr
switch (rel_types[i]) {
case R_NONE:
break;
case R_ABS:
write(S + A);
break;
case R_BASEREL:
write(S + A);
*dynrel++ = {P, R_X86_64_RELATIVE, 0, (i64)(S + A)};
break;
case R_DYN:
*dynrel++ = {P, R_X86_64_64, sym.dynsym_idx, A};
break;
case R_PC:
write(S + A - P);
break;
case R_GOT:
write(G + A);
break;
case R_GOTPC:
write(GOT + A - P);
break;
case R_GOTPCREL:
write(G + GOT + A - P);
break;
case R_TLSGD:
write(sym.get_tlsgd_addr() + A - P);
break;
case R_TLSGD_RELAX_LE: {
// Relax GD to LE
static const u8 insn[] = {
0x64, 0x48, 0x8b, 0x04, 0x25, 0, 0, 0, 0, // mov %fs:0, %rax
0x48, 0x8d, 0x80, 0, 0, 0, 0, // lea x@tpoff, %rax
};
memcpy(loc - 4, insn, sizeof(insn));
*(u32 *)(loc + 8) = S - out::tls_end + A + 4;
i++;
break;
}
case R_TLSLD:
write(out::got->get_tlsld_addr() + A - P);
break;
case R_TLSLD_RELAX_LE: {
// Relax LD to LE
static const u8 insn[] = {
// mov %fs:0, %rax
0x66, 0x66, 0x66, 0x64, 0x48, 0x8b, 0x04, 0x25, 0, 0, 0, 0,
};
memcpy(loc - 3, insn, sizeof(insn));
i++;
break;
}
case R_DTPOFF:
write(S + A - out::tls_begin);
break;
case R_TPOFF:
write(S + A - out::tls_end);
break;
case R_GOTTPOFF:
write(sym.get_gottpoff_addr() + A - P);
break;
default:
unreachable();
}
#undef S
#undef A
#undef P
#undef G
#undef GOT
}
}
// This function is responsible for applying relocations against
// non-SHF_ALLOC sections (i.e. sections that are not mapped to memory
// at runtime).
//
// Relocations against non-SHF_ALLOC sections are much easier to
// handle than that against SHF_ALLOC sections. It is because, since
// they are not mapped to memory, they don't contain any variable or
// function and never need PLT or GOT. Non-SHF_ALLOC sections are
// mostly debug info sections.
//
// Relocations against non-SHF_ALLOC sections are not scanned by
// scan_relocations.
void InputSection::apply_reloc_nonalloc(u8 *base) {
static Counter counter("reloc_nonalloc");
counter += rels.size();
i64 ref_idx = 0;
for (i64 i = 0; i < rels.size(); i++) {
const ElfRela &rel = rels[i];
Symbol &sym = *file->symbols[rel.r_sym];
if (!sym.file || sym.is_placeholder) {
Error() << "undefined symbol: " << *file << ": " << sym;
continue;
}
const SectionFragmentRef *ref = nullptr;
if (has_fragments[i])
ref = &rel_fragments[ref_idx++];
u8 *loc = base + rel.r_offset;
switch (rel.r_type) {
case R_X86_64_NONE:
break;
case R_X86_64_8:
case R_X86_64_16:
case R_X86_64_32:
case R_X86_64_32S:
case R_X86_64_64: {
u64 val = ref ? ref->frag->get_addr() : sym.get_addr();
overflow_check(this, sym, rel.r_type, val);
write_val(rel.r_type, loc, val);
break;
}
case R_X86_64_DTPOFF64:
write_val(rel.r_type, loc, sym.get_addr() + rel.r_addend - out::tls_begin);
break;
case R_X86_64_PC8:
case R_X86_64_PC16:
case R_X86_64_PC32:
case R_X86_64_PC64:
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:
case R_X86_64_PLT32:
case R_X86_64_TLSGD:
case R_X86_64_TLSLD:
case R_X86_64_DTPOFF32:
case R_X86_64_TPOFF32:
case R_X86_64_TPOFF64:
case R_X86_64_GOTTPOFF:
Error() << *this << ": invalid relocation for non-allocated sections: "
<< rel.r_type;
break;
default:
Error() << *this << ": unknown relocation: " << rel.r_type;
}
}
}
static int get_sym_type(Symbol &sym) {
if (sym.is_absolute())
return 0;
if (!sym.is_imported())
return 1;
if (sym.st_type != STT_FUNC)
return 2;
return 3;
}
// Linker has to create data structures in an output file to apply
// some type of relocations. For example, if a relocation refers a GOT
// or a PLT entry of a symbol, linker has to create an entry in .got
// or in .plt for that symbol. In order to fix the file layout, we
// need to scan relocations.
void InputSection::scan_relocations() {
if (!(shdr->sh_flags & SHF_ALLOC))
return;
static Counter counter("reloc_alloc");
counter += rels.size();
this->reldyn_offset = file->num_dynrel * sizeof(ElfRela);
bool is_readonly = !(shdr->sh_flags & SHF_WRITE);
i64 output_type = config.shared ? 2 : (config.pie ? 1 : 0);
// Scan relocations
for (i64 i = 0; i < rels.size(); i++) {
const ElfRela &rel = rels[i];
Symbol &sym = *file->symbols[rel.r_sym];
bool is_code = (sym.st_type == STT_FUNC);
if (!sym.file || sym.is_placeholder) {
Error() << "undefined symbol: " << *file << ": " << sym;
continue;
}
auto none = []() {};
auto error = [&]() {
Error() << *this << ": " << rel_to_string(rel.r_type)
<< " relocation against symbol `" << sym
<< "' can not be used; recompile with -fPIE";
};
auto copyrel = [&]() {
sym.flags |= NEEDS_COPYREL;
};
auto plt = [&]() {
sym.flags |= NEEDS_PLT;
};
auto dynrel = [&]() {
if (is_readonly)
error();
sym.flags |= NEEDS_DYNSYM;
rel_types[i] = R_DYN;
file->num_dynrel++;
};
auto baserel = [&]() {
if (is_readonly)
error();
rel_types[i] = R_BASEREL;
file->num_dynrel++;
};
switch (rel.r_type) {
case R_X86_64_NONE:
rel_types[i] = R_NONE;
break;
case R_X86_64_8:
case R_X86_64_16:
case R_X86_64_32:
case R_X86_64_32S: {
std::function<void()> table[][4] = {
// Absolute Local Imported data Imported code
{ none, none, copyrel, plt }, // PDE
{ none, error, error, error }, // PIE
{ none, error, error, error }, // DSO
};
rel_types[i] = R_ABS;
table[output_type][get_sym_type(sym)]();
break;
}
case R_X86_64_64: {
std::function<void()> table[][4] = {
// Absolute Local Imported data Imported code
{ none, none, copyrel, plt }, // PDE
{ none, baserel, dynrel, dynrel }, // PIE
{ none, baserel, dynrel, dynrel }, // DSO
};
rel_types[i] = R_ABS;
table[output_type][get_sym_type(sym)]();
break;
}
case R_X86_64_PC8:
case R_X86_64_PC16:
case R_X86_64_PC32: {
std::function<void()> table[][4] = {
// Absolute Local Imported data Imported code
{ none, none, copyrel, plt }, // PDE
{ error, none, copyrel, plt }, // PIE
{ error, none, error, error }, // DSO
};
rel_types[i] = R_PC;
table[output_type][get_sym_type(sym)]();
break;
}
case R_X86_64_PC64: {
std::function<void()> table[][4] = {
// Absolute Local Imported data Imported code
{ none, none, copyrel, plt }, // PDE
{ baserel, none, copyrel, plt }, // PIE
{ baserel, none, error, error }, // DSO
};
rel_types[i] = R_PC;
table[output_type][get_sym_type(sym)]();
break;
}
case R_X86_64_GOT32:
sym.flags |= NEEDS_GOT;
rel_types[i] = R_GOT;
break;
case R_X86_64_GOTPC32:
sym.flags |= NEEDS_GOT;
rel_types[i] = R_GOTPC;
break;
case R_X86_64_GOTPCREL:
case R_X86_64_GOTPCRELX:
case R_X86_64_REX_GOTPCRELX:
sym.flags |= NEEDS_GOT;
rel_types[i] = R_GOTPCREL;
break;
case R_X86_64_PLT32:
if (sym.is_imported() || sym.st_type == STT_GNU_IFUNC)
sym.flags |= NEEDS_PLT;
rel_types[i] = R_PC;
break;
case R_X86_64_TLSGD:
if (i + 1 == rels.size() || rels[i + 1].r_type != R_X86_64_PLT32)
Error() << *this << ": TLSGD reloc not followed by PLT32";
if (config.relax && !sym.is_imported()) {
rel_types[i] = R_TLSGD_RELAX_LE;
i++;
} else {
sym.flags |= NEEDS_TLSGD;
sym.flags |= NEEDS_DYNSYM;
rel_types[i] = R_TLSGD;
}
break;
case R_X86_64_TLSLD:
if (i + 1 == rels.size() || rels[i + 1].r_type != R_X86_64_PLT32)
Error() << *this << ": TLSLD reloc not followed by PLT32";
if (sym.is_imported())
Error() << *this << ": TLSLD reloc refers external symbol " << sym;
if (config.relax) {
rel_types[i] = R_TLSLD_RELAX_LE;
i++;
} else {
sym.flags |= NEEDS_TLSLD;
rel_types[i] = R_TLSLD;
}
break;
case R_X86_64_DTPOFF32:
case R_X86_64_DTPOFF64:
if (sym.is_imported())
Error() << *this << ": DTPOFF reloc refers external symbol " << sym;
rel_types[i] = config.relax ? R_TPOFF : R_DTPOFF;
break;
case R_X86_64_TPOFF32:
case R_X86_64_TPOFF64:
rel_types[i] = R_TPOFF;
break;
case R_X86_64_GOTTPOFF:
sym.flags |= NEEDS_GOTTPOFF;
rel_types[i] = R_GOTTPOFF;
break;
default:
Error() << *this << ": unknown relocation: " << rel.r_type;
}
}
}
void InputSection::kill() {
is_alive = false;
for (FdeRecord &fde : fdes)
fde.is_alive = false;
file->sections[section_idx] = nullptr;
}
static size_t find_null(std::string_view data, u64 entsize) {
if (entsize == 1)
return data.find('\0');
for (i64 i = 0; i <= data.size() - entsize; i += entsize)
if (data.substr(i, i + entsize).find_first_not_of('\0') ==
std::string_view::npos)
return i;
return std::string_view::npos;
}
// Mergeable sections (sections with SHF_MERGE bit) typically contain
// string literals. Linker is expected to split the section contents
// into null-terminated strings, merge them with mergeable strings
// from other object files, and emit uniquified strings to an output
// file.
//
// This mechanism reduces the size of an output file. If two source
// files happen to contain the same string literal, the output will
// contain only a single copy of it.
//
// It is less common than string literals, but mergeable sections can
// contain fixed-sized read-only records too.
//
// This function splits the section contents into small pieces that we
// call "section fragments". Section fragment is a unit of merging.
//
// We do not support mergeable sections that have relocations.
MergeableSection::MergeableSection(InputSection *isec)
: InputChunk(isec->file, isec->shdr, isec->name),
parent(*MergedSection::get_instance(isec->name, isec->shdr->sh_type,
isec->shdr->sh_flags)) {
std::string_view data = isec->contents;
const char *begin = data.data();
u64 entsize = isec->shdr->sh_entsize;
static_assert(sizeof(SectionFragment::alignment) == 2);
if (isec->shdr->sh_addralign >= (1 << 16))
Fatal() << *isec << ": alignment too large";
if (isec->shdr->sh_flags & SHF_STRINGS) {
while (!data.empty()) {
size_t end = find_null(data, entsize);
if (end == std::string_view::npos)
Error() << *this << ": string is not null terminated";
std::string_view substr = data.substr(0, end + entsize);
data = data.substr(end + entsize);
SectionFragment *frag = parent.insert(substr, isec->shdr->sh_addralign);
fragments.push_back(frag);
frag_offsets.push_back(substr.data() - begin);
}
} else {
if (data.size() % entsize)
Fatal() << *isec << ": section size is not multiple of sh_entsize";
while (!data.empty()) {
std::string_view substr = data.substr(0, entsize);
data = data.substr(entsize);
SectionFragment *frag = parent.insert(substr, isec->shdr->sh_addralign);
fragments.push_back(frag);
frag_offsets.push_back(substr.data() - begin);
}
}
static Counter counter("string_fragments");
counter += fragments.size();
}
Reference in New Issue View Git Blame Copy Permalink