mirror of
https://sourceware.org/git/binutils-gdb.git
synced 2024-12-21 04:42:53 +08:00
fd67aa1129
Adds two new external authors to etc/update-copyright.py to cover bfd/ax_tls.m4, and adds gprofng to dirs handled automatically, then updates copyright messages as follows: 1) Update cgen/utils.scm emitted copyrights. 2) Run "etc/update-copyright.py --this-year" with an extra external author I haven't committed, 'Kalray SA.', to cover gas testsuite files (which should have their copyright message removed). 3) Build with --enable-maintainer-mode --enable-cgen-maint=yes. 4) Check out */po/*.pot which we don't update frequently.
519 lines
17 KiB
C++
519 lines
17 KiB
C++
// incremental.cc -- incremental linking test/debug tool
|
|
|
|
// Copyright (C) 2009-2024 Free Software Foundation, Inc.
|
|
// Written by Rafael Avila de Espindola <rafael.espindola@gmail.com>
|
|
|
|
// This file is part of gold.
|
|
|
|
// This program is free software; you can redistribute it and/or modify
|
|
// it under the terms of the GNU General Public License as published by
|
|
// the Free Software Foundation; either version 3 of the License, or
|
|
// (at your option) any later version.
|
|
|
|
// This program is distributed in the hope that it will be useful,
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
// GNU General Public License for more details.
|
|
|
|
// You should have received a copy of the GNU General Public License
|
|
// along with this program; if not, write to the Free Software
|
|
// Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston,
|
|
// MA 02110-1301, USA.
|
|
|
|
|
|
// This file is a (still incomplete) test/debug tool that should display
|
|
// all information available in the incremental linking sections in a
|
|
// format that is easy to read.
|
|
// Once the format is a bit more stable, this should probably be moved to
|
|
// readelf. Because of that, the use of gold's data structures and functions
|
|
// is just a short term convenience and not a design decision.
|
|
|
|
#include "gold.h"
|
|
|
|
#include <stdio.h>
|
|
#include <errno.h>
|
|
#include <time.h>
|
|
|
|
#include "incremental.h"
|
|
|
|
namespace gold
|
|
{
|
|
class Output_file;
|
|
}
|
|
|
|
using namespace gold;
|
|
|
|
template<int size, bool big_endian>
|
|
static typename Incremental_inputs_reader<size, big_endian>::
|
|
Incremental_input_entry_reader
|
|
find_input_containing_global(
|
|
Incremental_inputs_reader<size, big_endian>& incremental_inputs,
|
|
unsigned int offset,
|
|
unsigned int* symndx)
|
|
{
|
|
typedef Incremental_inputs_reader<size, big_endian> Inputs_reader;
|
|
static const unsigned int global_sym_entry_size =
|
|
Incremental_inputs_reader<size, big_endian>::global_sym_entry_size;
|
|
|
|
for (unsigned int i = 0; i < incremental_inputs.input_file_count(); ++i)
|
|
{
|
|
typename Inputs_reader::Incremental_input_entry_reader input_file =
|
|
incremental_inputs.input_file(i);
|
|
if (input_file.type() != INCREMENTAL_INPUT_OBJECT
|
|
&& input_file.type() != INCREMENTAL_INPUT_ARCHIVE_MEMBER)
|
|
continue;
|
|
unsigned int nsyms = input_file.get_global_symbol_count();
|
|
if (offset >= input_file.get_symbol_offset(0)
|
|
&& offset < input_file.get_symbol_offset(nsyms))
|
|
{
|
|
*symndx = ((offset - input_file.get_symbol_offset(0))
|
|
/ global_sym_entry_size);
|
|
return input_file;
|
|
}
|
|
}
|
|
gold_unreachable();
|
|
}
|
|
|
|
template<int size, bool big_endian>
|
|
static void
|
|
dump_incremental_inputs(const char* argv0, const char* filename,
|
|
Sized_incremental_binary<size, big_endian>* inc)
|
|
{
|
|
typedef Incremental_binary::Location Location;
|
|
typedef Incremental_binary::View View;
|
|
typedef Incremental_inputs_reader<size, big_endian> Inputs_reader;
|
|
typedef typename Inputs_reader::Incremental_input_entry_reader Entry_reader;
|
|
|
|
if (!inc->has_incremental_info())
|
|
{
|
|
fprintf(stderr, "%s: %s: no .gnu_incremental_inputs section\n", argv0,
|
|
filename);
|
|
exit(1);
|
|
}
|
|
|
|
// Create a reader object for the .gnu_incremental_inputs section.
|
|
|
|
Incremental_inputs_reader<size, big_endian>
|
|
incremental_inputs(inc->inputs_reader());
|
|
|
|
if (incremental_inputs.version() != 2)
|
|
{
|
|
fprintf(stderr, "%s: %s: unknown incremental version %d\n", argv0,
|
|
filename, incremental_inputs.version());
|
|
exit(1);
|
|
}
|
|
|
|
const char* command_line = incremental_inputs.command_line();
|
|
if (command_line == NULL)
|
|
{
|
|
fprintf(stderr,
|
|
"%s: %s: failed to get link command line\n",
|
|
argv0, filename);
|
|
exit(1);
|
|
}
|
|
printf("Link command line: %s\n", command_line);
|
|
|
|
printf("\nInput files:\n");
|
|
for (unsigned int i = 0; i < incremental_inputs.input_file_count(); ++i)
|
|
{
|
|
Entry_reader input_file = incremental_inputs.input_file(i);
|
|
|
|
const char* objname = input_file.filename();
|
|
if (objname == NULL)
|
|
{
|
|
fprintf(stderr,"%s: %s: failed to get file name for object %u\n",
|
|
argv0, filename, i);
|
|
exit(1);
|
|
}
|
|
printf("[%d] %s\n", i, objname);
|
|
|
|
Timespec mtime = input_file.get_mtime();
|
|
printf(" Timestamp: %llu.%09d %s",
|
|
static_cast<unsigned long long>(mtime.seconds),
|
|
mtime.nanoseconds,
|
|
ctime(&mtime.seconds));
|
|
|
|
printf(" Serial Number: %d\n", input_file.arg_serial());
|
|
printf(" In System Directory: %s\n",
|
|
input_file.is_in_system_directory() ? "true" : "false");
|
|
|
|
Incremental_input_type input_type = input_file.type();
|
|
printf(" Type: ");
|
|
switch (input_type)
|
|
{
|
|
case INCREMENTAL_INPUT_OBJECT:
|
|
case INCREMENTAL_INPUT_ARCHIVE_MEMBER:
|
|
printf("%s\n", (input_type == INCREMENTAL_INPUT_OBJECT
|
|
? "Object" : "Archive member"));
|
|
printf(" Input section count: %d\n",
|
|
input_file.get_input_section_count());
|
|
printf(" Global symbol count: %d\n",
|
|
input_file.get_global_symbol_count());
|
|
printf(" Local symbol offset: %d\n",
|
|
input_file.get_local_symbol_offset());
|
|
printf(" Local symbol count: %d\n",
|
|
input_file.get_local_symbol_count());
|
|
printf(" First dynamic reloc: %d\n",
|
|
input_file.get_first_dyn_reloc());
|
|
printf(" Dynamic reloc count: %d\n",
|
|
input_file.get_dyn_reloc_count());
|
|
printf(" COMDAT group count: %d\n",
|
|
input_file.get_comdat_group_count());
|
|
break;
|
|
case INCREMENTAL_INPUT_ARCHIVE:
|
|
printf("Archive\n");
|
|
printf(" Member count: %d\n", input_file.get_member_count());
|
|
printf(" Unused symbol count: %d\n",
|
|
input_file.get_unused_symbol_count());
|
|
break;
|
|
case INCREMENTAL_INPUT_SHARED_LIBRARY:
|
|
printf("Shared library\n");
|
|
printf(" As needed: %s\n",
|
|
input_file.as_needed() ? "true" : "false");
|
|
printf(" soname: %s\n",
|
|
input_file.get_soname());
|
|
printf(" Symbol count: %d\n",
|
|
input_file.get_global_symbol_count());
|
|
break;
|
|
case INCREMENTAL_INPUT_SCRIPT:
|
|
printf("Linker script\n");
|
|
printf(" Object count: %d\n", input_file.get_object_count());
|
|
break;
|
|
default:
|
|
fprintf(stderr, "%s: invalid file type for object %u: %d\n",
|
|
argv0, i, input_type);
|
|
exit(1);
|
|
}
|
|
}
|
|
|
|
printf("\nInput sections:\n");
|
|
for (unsigned int i = 0; i < incremental_inputs.input_file_count(); ++i)
|
|
{
|
|
Entry_reader input_file(incremental_inputs.input_file(i));
|
|
|
|
if (input_file.type() != INCREMENTAL_INPUT_OBJECT
|
|
&& input_file.type() != INCREMENTAL_INPUT_ARCHIVE_MEMBER)
|
|
continue;
|
|
|
|
const char* objname = input_file.filename();
|
|
if (objname == NULL)
|
|
{
|
|
fprintf(stderr,"%s: %s: failed to get file name for object %u\n",
|
|
argv0, filename, i);
|
|
exit(1);
|
|
}
|
|
|
|
printf("[%d] %s\n", i, objname);
|
|
|
|
printf(" %3s %6s %8s %8s %s\n",
|
|
"n", "outndx", "offset", "size", "name");
|
|
unsigned int nsections = input_file.get_input_section_count();
|
|
for (unsigned int shndx = 0; shndx < nsections; ++shndx)
|
|
{
|
|
typename Entry_reader::Input_section_info info(
|
|
input_file.get_input_section(shndx));
|
|
printf(" %3d %6d %8lld %8lld %s\n", shndx + 1,
|
|
info.output_shndx,
|
|
static_cast<long long>(info.sh_offset),
|
|
static_cast<long long>(info.sh_size),
|
|
info.name);
|
|
}
|
|
|
|
unsigned int ncomdat = input_file.get_comdat_group_count();
|
|
for (unsigned int i = 0; i < ncomdat; ++i)
|
|
printf(" Comdat group: %s\n",
|
|
input_file.get_comdat_group_signature(i));
|
|
}
|
|
|
|
// Get a view of the .symtab section.
|
|
|
|
elfcpp::Elf_file<size, big_endian, Incremental_binary> elf_file(inc);
|
|
|
|
unsigned int symtab_shndx = elf_file.find_section_by_type(elfcpp::SHT_SYMTAB);
|
|
if (symtab_shndx == elfcpp::SHN_UNDEF) // Not found.
|
|
{
|
|
fprintf(stderr, "%s: %s: no symbol table section\n", argv0, filename);
|
|
exit(1);
|
|
}
|
|
Location symtab_location(elf_file.section_contents(symtab_shndx));
|
|
View symtab_view(inc->view(symtab_location));
|
|
|
|
// Get a view of the .strtab section.
|
|
|
|
unsigned int strtab_shndx = elf_file.section_link(symtab_shndx);
|
|
if (strtab_shndx == elfcpp::SHN_UNDEF
|
|
|| strtab_shndx > elf_file.shnum()
|
|
|| elf_file.section_type(strtab_shndx) != elfcpp::SHT_STRTAB)
|
|
{
|
|
fprintf(stderr, "%s: %s: no string table section\n", argv0, filename);
|
|
exit(1);
|
|
}
|
|
Location strtab_location(elf_file.section_contents(strtab_shndx));
|
|
View strtab_view(inc->view(strtab_location));
|
|
elfcpp::Elf_strtab strtab(strtab_view.data(), strtab_location.data_size);
|
|
|
|
// The .gnu_incremental_symtab section contains entries that parallel
|
|
// the global symbols of the main symbol table. The sh_info field
|
|
// of the main symbol table's section header tells us how many global
|
|
// symbols there are, but that count does not include any global
|
|
// symbols that were forced local during the link. Therefore, we
|
|
// use the size of the .gnu_incremental_symtab section to deduce
|
|
// the number of global symbols + forced-local symbols there are
|
|
// in the symbol table.
|
|
Incremental_symtab_reader<big_endian> isymtab(inc->symtab_reader());
|
|
Incremental_relocs_reader<size, big_endian> irelocs(inc->relocs_reader());
|
|
unsigned int sym_size = elfcpp::Elf_sizes<size>::sym_size;
|
|
unsigned int nsyms = symtab_location.data_size / sym_size;
|
|
unsigned int nglobals = isymtab.symbol_count();
|
|
unsigned int first_global = nsyms - nglobals;
|
|
unsigned const char* sym_p;
|
|
|
|
printf("\nGlobal symbols per input file:\n");
|
|
for (unsigned int i = 0; i < incremental_inputs.input_file_count(); ++i)
|
|
{
|
|
Entry_reader input_file(incremental_inputs.input_file(i));
|
|
|
|
if (input_file.type() != INCREMENTAL_INPUT_OBJECT
|
|
&& input_file.type() != INCREMENTAL_INPUT_ARCHIVE_MEMBER
|
|
&& input_file.type() != INCREMENTAL_INPUT_SHARED_LIBRARY)
|
|
continue;
|
|
|
|
const char* objname = input_file.filename();
|
|
if (objname == NULL)
|
|
{
|
|
fprintf(stderr,"%s: %s: failed to get file name for object %u\n",
|
|
argv0, filename, i);
|
|
exit(1);
|
|
}
|
|
|
|
printf("[%d] %s\n", i, objname);
|
|
|
|
unsigned int nsyms = input_file.get_global_symbol_count();
|
|
if (nsyms > 0)
|
|
printf(" %6s %6s %8s %8s %8s %8s\n",
|
|
"outndx", "shndx", "offset", "chain", "#relocs", "rbase");
|
|
if (input_file.type() == INCREMENTAL_INPUT_SHARED_LIBRARY)
|
|
{
|
|
for (unsigned int symndx = 0; symndx < nsyms; ++symndx)
|
|
{
|
|
bool is_def;
|
|
bool is_copy;
|
|
unsigned int output_symndx =
|
|
input_file.get_output_symbol_index(symndx, &is_def, &is_copy);
|
|
sym_p = symtab_view.data() + output_symndx * sym_size;
|
|
elfcpp::Sym<size, big_endian> sym(sym_p);
|
|
const char* symname;
|
|
if (!strtab.get_c_string(sym.get_st_name(), &symname))
|
|
symname = "<unknown>";
|
|
printf(" %6d %6s %8s %8s %8s %8s %-5s %s\n",
|
|
output_symndx,
|
|
"", "", "", "", "",
|
|
is_copy ? "COPY" : (is_def ? "DEF" : "UNDEF"),
|
|
symname);
|
|
}
|
|
}
|
|
else
|
|
{
|
|
for (unsigned int symndx = 0; symndx < nsyms; ++symndx)
|
|
{
|
|
Incremental_global_symbol_reader<big_endian> info(
|
|
input_file.get_global_symbol_reader(symndx));
|
|
unsigned int output_symndx = info.output_symndx();
|
|
sym_p = symtab_view.data() + output_symndx * sym_size;
|
|
elfcpp::Sym<size, big_endian> sym(sym_p);
|
|
const char* symname;
|
|
if (!strtab.get_c_string(sym.get_st_name(), &symname))
|
|
symname = "<unknown>";
|
|
printf(" %6d %6d %8d %8d %8d %8d %-5s %s\n",
|
|
output_symndx,
|
|
info.shndx() == -1U ? -1 : info.shndx(),
|
|
input_file.get_symbol_offset(symndx),
|
|
info.next_offset(),
|
|
info.reloc_count(),
|
|
info.reloc_offset(),
|
|
(info.shndx() == -1U
|
|
? "BASE"
|
|
: info.shndx() == 0 ? "UNDEF" : "DEF"),
|
|
symname);
|
|
}
|
|
}
|
|
}
|
|
|
|
sym_p = symtab_view.data() + first_global * sym_size;
|
|
printf("\nGlobal symbol table:\n");
|
|
for (unsigned int i = 0; i < nglobals; i++)
|
|
{
|
|
elfcpp::Sym<size, big_endian> sym(sym_p);
|
|
const char* symname;
|
|
if (!strtab.get_c_string(sym.get_st_name(), &symname))
|
|
symname = "<unknown>";
|
|
printf("[%d] %s\n", first_global + i, symname);
|
|
unsigned int offset = isymtab.get_list_head(i);
|
|
while (offset > 0)
|
|
{
|
|
unsigned int sym_ndx;
|
|
Entry_reader input_file =
|
|
find_input_containing_global<size, big_endian>(incremental_inputs,
|
|
offset, &sym_ndx);
|
|
Incremental_global_symbol_reader<big_endian> sym_info(
|
|
input_file.get_global_symbol_reader(sym_ndx));
|
|
printf(" %s (first reloc: %d, reloc count: %d)",
|
|
input_file.filename(), sym_info.reloc_offset(),
|
|
sym_info.reloc_count());
|
|
if (sym_info.output_symndx() != first_global + i)
|
|
printf(" ** wrong output symndx (%d) **", sym_info.output_symndx());
|
|
printf("\n");
|
|
// Dump the relocations from this input file for this symbol.
|
|
unsigned int r_off = sym_info.reloc_offset();
|
|
for (unsigned int j = 0; j < sym_info.reloc_count(); j++)
|
|
{
|
|
printf(" %4d relocation type %3d shndx %2d"
|
|
" offset %016llx addend %016llx %s\n",
|
|
r_off,
|
|
irelocs.get_r_type(r_off),
|
|
irelocs.get_r_shndx(r_off),
|
|
static_cast<long long>(irelocs.get_r_offset(r_off)),
|
|
static_cast<long long>(irelocs.get_r_addend(r_off)),
|
|
symname);
|
|
r_off += irelocs.reloc_size;
|
|
}
|
|
offset = sym_info.next_offset();
|
|
}
|
|
sym_p += sym_size;
|
|
}
|
|
|
|
Incremental_got_plt_reader<big_endian> igot_plt(inc->got_plt_reader());
|
|
unsigned int ngot = igot_plt.get_got_entry_count();
|
|
unsigned int nplt = igot_plt.get_plt_entry_count();
|
|
|
|
printf("\nGOT entries:\n");
|
|
for (unsigned int i = 0; i < ngot; ++i)
|
|
{
|
|
unsigned int got_type = igot_plt.get_got_type(i);
|
|
unsigned int got_symndx = igot_plt.get_got_symndx(i);
|
|
unsigned int got_input_index = igot_plt.get_got_input_index(i);
|
|
printf("[%d] type %02x, ", i, got_type & 0x7f);
|
|
if ((got_type & 0x7f) == 0x7f)
|
|
printf("reserved");
|
|
else if (got_type & 0x80)
|
|
{
|
|
Entry_reader input_file =
|
|
incremental_inputs.input_file(got_input_index);
|
|
const char* objname = input_file.filename();
|
|
printf("local: %s (%d)", objname, got_symndx);
|
|
}
|
|
else
|
|
{
|
|
sym_p = symtab_view.data() + got_symndx * sym_size;
|
|
elfcpp::Sym<size, big_endian> sym(sym_p);
|
|
const char* symname;
|
|
if (!strtab.get_c_string(sym.get_st_name(), &symname))
|
|
symname = "<unknown>";
|
|
printf("global %s (%d)", symname, got_symndx);
|
|
}
|
|
printf("\n");
|
|
}
|
|
|
|
printf("\nPLT entries:\n");
|
|
for (unsigned int i = 0; i < nplt; ++i)
|
|
{
|
|
unsigned int plt_desc = igot_plt.get_plt_desc(i);
|
|
printf("[%d] ", i);
|
|
sym_p = symtab_view.data() + plt_desc * sym_size;
|
|
elfcpp::Sym<size, big_endian> sym(sym_p);
|
|
const char* symname;
|
|
if (!strtab.get_c_string(sym.get_st_name(), &symname))
|
|
symname = "<unknown>";
|
|
printf("%s (%d)\n", symname, plt_desc);
|
|
}
|
|
|
|
printf("\nUnused archive symbols:\n");
|
|
for (unsigned int i = 0; i < incremental_inputs.input_file_count(); ++i)
|
|
{
|
|
Entry_reader input_file(incremental_inputs.input_file(i));
|
|
|
|
if (input_file.type() != INCREMENTAL_INPUT_ARCHIVE)
|
|
continue;
|
|
|
|
const char* objname = input_file.filename();
|
|
if (objname == NULL)
|
|
{
|
|
fprintf(stderr,"%s: %s: failed to get file name for object %u\n",
|
|
argv0, filename, i);
|
|
exit(1);
|
|
}
|
|
|
|
printf("[%d] %s\n", i, objname);
|
|
unsigned int nsyms = input_file.get_unused_symbol_count();
|
|
for (unsigned int symndx = 0; symndx < nsyms; ++symndx)
|
|
printf(" %s\n", input_file.get_unused_symbol(symndx));
|
|
}
|
|
|
|
}
|
|
|
|
int
|
|
main(int argc, char** argv)
|
|
{
|
|
if (argc != 2)
|
|
{
|
|
fprintf(stderr, "Usage: %s <file>\n", argv[0]);
|
|
return 1;
|
|
}
|
|
const char* filename = argv[1];
|
|
|
|
Output_file* file = new Output_file(filename);
|
|
|
|
bool t = file->open_base_file(NULL, false);
|
|
if (!t)
|
|
{
|
|
fprintf(stderr, "%s: open_base_file(%s): %s\n", argv[0], filename,
|
|
strerror(errno));
|
|
return 1;
|
|
}
|
|
|
|
Incremental_binary* inc = open_incremental_binary(file);
|
|
|
|
if (inc == NULL)
|
|
{
|
|
fprintf(stderr, "%s: open_incremental_binary(%s): %s\n", argv[0],
|
|
filename, strerror(errno));
|
|
return 1;
|
|
}
|
|
|
|
switch (parameters->size_and_endianness())
|
|
{
|
|
#ifdef HAVE_TARGET_32_LITTLE
|
|
case Parameters::TARGET_32_LITTLE:
|
|
dump_incremental_inputs<32, false>(
|
|
argv[0], filename,
|
|
static_cast<Sized_incremental_binary<32, false>*>(inc));
|
|
break;
|
|
#endif
|
|
#ifdef HAVE_TARGET_32_BIG
|
|
case Parameters::TARGET_32_BIG:
|
|
dump_incremental_inputs<32, true>(
|
|
argv[0], filename,
|
|
static_cast<Sized_incremental_binary<32, true>*>(inc));
|
|
break;
|
|
#endif
|
|
#ifdef HAVE_TARGET_64_LITTLE
|
|
case Parameters::TARGET_64_LITTLE:
|
|
dump_incremental_inputs<64, false>(
|
|
argv[0], filename,
|
|
static_cast<Sized_incremental_binary<64, false>*>(inc));
|
|
break;
|
|
#endif
|
|
#ifdef HAVE_TARGET_64_BIG
|
|
case Parameters::TARGET_64_BIG:
|
|
dump_incremental_inputs<64, true>(
|
|
argv[0], filename,
|
|
static_cast<Sized_incremental_binary<64, true>*>(inc));
|
|
break;
|
|
#endif
|
|
default:
|
|
gold_unreachable();
|
|
}
|
|
|
|
return 0;
|
|
}
|