mirror of
https://sourceware.org/git/binutils-gdb.git
synced 2024-12-15 04:31:49 +08:00
1ff6de0312
This is quite complicated because the CTF section's contents depend on the final contents of the symtab and strtab, because it has two sections whose contents are shuffled to be in 1:1 correspondence with the symtab, and an internal strtab that gets deduplicated against the ELF strtab (with offsets adjusted to point into the ELF strtab instead). It is also compressed if large enough, so its size depends on its contents! So we cannot construct it as early as most sections: we cannot even *begin* construction until after the symtab and strtab are finalized. Thankfully there is already one section treated similarly: compressed debugging sections: the only differences are that compressed debugging sections have extra handling to deal with their changing name if compressed (CTF sections are always called ".ctf" for now, though we have reserved ".ctf.*" against future use), and that compressed debugging sections have previously-uncompressed content which has to be stashed away for later compression, while CTF sections have no content at all until we generate it (very late). BFD also cannot do the link itself: libctf knows how to do it, and BFD cannot call libctf directly because libctf already depends on bfd for file I/O. So we have to use a pair of callbacks, one, examine_strtab, which allows a caller to examine the symtab and strtab after finalization (called from elf_link_swap_symbols_out(), right before the symtabs are written, and after the strtab has been finalized), and one which actually does the emission (called emit_ctf simply because it is grouped with a bunch of section-specific late-emission function calls at the bottom of bfd_elf_final_link, and a section-specific name seems best for that). emit_ctf is actually called *twice*: once from lang_process if the emulation suggests that this bfd target does not examine the symtab or strtab, and once via a bfd callback if it does. (This means that non-ELF targets still get CTF emitted, even though the late CTF emission stage is never called for them). v2: merged with non-ELF support patch: slight commit message adjustments. v3: do not spend time merging CTF, or crash, if the CTF section is explicitly discarded. Do not try to merge or compress CTF unless linking. v4: add CTF_COMPRESSION_THRESHOLD. Annul the freed input ctf_file_t's after writeout: set SEC_IN_MEMORY on the output contents so a future bfd enhancement knows it could free it. Add SEC_LINKER_CREATED | SEC_KEEP to avoid having to add .ctf to the linker script. Drop now-unnecessary ldlang.h-level elf-bfd.h include and hackery around it. Adapt to elf32.em->elf.em and elf-generic.em->ldelf*.c changes. v5: fix tabdamage. Drop #inclusions in .h files: include in .c files, .em files, and use struct forwards instead. Use bfd_section_is_ctf inline function rather than SECTION_IS_CTF macro. Move a few comments. * Makefile.def (dependencies): all-ld depends on all-libctf. * Makefile.in: Regenerated. include/ * bfdlink.h (elf_strtab_hash): New forward. (elf_sym_strtab): Likewise. (struct bfd_link_callbacks <examine_strtab>): New. (struct bfd_link_callbacks <emit_ctf>): Likewise. bfd/ * elf-bfd.h (bfd_section_is_ctf): New inline function. * elf.c (special_sections_c): Add ".ctf". (assign_file_positions_for_non_load_sections): Note that compressed debugging sections etc are not assigned here. Treat CTF sections like SEC_ELF_COMPRESS sections when is_linker_output: sh_offset -1. (assign_file_positions_except_relocs): Likewise. (find_section_in_list): Note that debugging and CTF sections, as well as reloc sections, are assigned later. (_bfd_elf_assign_file_positions_for_non_load): CTF sections get their size and contents updated. (_bfd_elf_set_section_contents): Skip CTF sections: unlike compressed sections, they have no uncompressed content to copy at this stage. * elflink.c (elf_link_swap_symbols_out): Call the examine_strtab callback right before the strtab is written out. (bfd_elf_final_link): Don't cache the section contents of CTF sections: they are not populated yet. Call the emit_ctf callback right at the end, after all the symbols and strings are flushed out. ld/ * ldlang.h: (struct lang_input_statement_struct): Add the_ctf. (struct elf_sym_strtab): Add forward. (struct elf_strtab_hash): Likewise. (ldlang_ctf_apply_strsym): Declare. (ldlang_write_ctf_late): Likewise. * ldemul.h (ldemul_emit_ctf_early): New. (ldemul_examine_strtab_for_ctf): Likewise. (ld_emulation_xfer_type) <emit_ctf_early>: Likewise. (ld_emulation_xfer_type) <examine_strtab_for_ctf>: Likewise. * ldemul.c (ldemul_emit_ctf_early): New. (ldemul_examine_strtab_for_ctf): Likewise. * ldlang.c: Include ctf-api.h. (CTF_COMPRESSION_THRESHOLD): New. (ctf_output): New. Initialized in... (ldlang_open_ctf): ... this new function. Open all the CTF sections in the input files: mark them non-loaded and empty so as not to copy their contents to the output, but linker-created so the section gets created in the target. (ldlang_merge_ctf): New, merge types via ctf_link_add_ctf and ctf_link. (ldlang_ctf_apply_strsym): New, an examine_strtab callback: wrap ldemul_examine_strtab_for_ctf. (lang_write_ctf): New, write out the CTF section. (ldlang_write_ctf_late): New, late call via bfd's emit_ctf hook. (lang_process): Call ldlang_open_ctf, ldlang_merge_ctf, and lang_write_ctf. * ldmain.c (link_callbacks): Add ldlang_ctf_apply_strsym, ldlang_write_ctf_late. * emultempl/aix.em: Add ctf-api.h. * emultempl/armcoff.em: Likewise. * emultempl/beos.em: Likewise. * emultempl/elf.em: Likewise. * emultempl/generic.em: Likewise. * emultempl/linux.em: Likewise. * emultempl/msp430.em: Likewise. * emultempl/pe.em: Likewise. * emultempl/pep.em: Likewise. * emultempl/ticoff.em: Likewise. * emultempl/vanilla.em: Likewise. * ldcref.c: Likewise. * ldctor.c: Likewise. * ldelf.c: Likewise. * ldelfgen.c: Likewise. * ldemul.c: Likewise. * ldexp.c: Likewise. * ldfile.c: Likewise. * ldgram.c: Likewise. * ldlex.l: Likewise. * ldmain.c: Likewise. * ldmisc.c: Likewise. * ldver.c: Likewise. * ldwrite.c: Likewise. * lexsup.c: Likewise. * mri.c: Likewise. * pe-dll.c: Likewise. * plugin.c: Likewise. * ldelfgen.c (ldelf_emit_ctf_early): New. (ldelf_examine_strtab_for_ctf): tell libctf about the symtab and strtab. (struct ctf_strsym_iter_cb_arg): New, state to do so. (ldelf_ctf_strtab_iter_cb): New: tell libctf about each string in the strtab in turn. (ldelf_ctf_symbols_iter_cb): New, tell libctf about each symbol in the symtab in turn. * ldelfgen.h (struct elf_sym_strtab): Add forward. (struct elf_strtab_hash): Likewise. (struct ctf_file): Likewise. (ldelf_emit_ctf_early): Declare. (ldelf_examine_strtab_for_ctf): Likewise. * emultempl/elf-generic.em (LDEMUL_EMIT_CTF_EARLY): Set it. (LDEMUL_EXAMINE_STRTAB_FOR_CTF): Likewise. * emultempl/aix.em (ld_${EMULATION_NAME}_emulation): Add emit_ctf_early and examine_strtab_for_ctf, NULL by default. * emultempl/armcoff.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/beos.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/elf.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/generic.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/linux.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/msp430.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/pe.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/pep.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/ticoff.em (ld_${EMULATION_NAME}_emulation): Likewise. * emultempl/vanilla.em (ld_vanilla_emulation): Likewise. * Makefile.am: Pull in libctf (and zlib, a transitive requirement for compressed CTF section emission). Pass it on to DejaGNU. * configure.ac: Add AM_ZLIB. * aclocal.m4: Added zlib.m4. * Makefile.in: Regenerated. * testsuite/ld-bootstrap/bootstrap.exp: Use it when relinking ld.
385 lines
9.7 KiB
C
385 lines
9.7 KiB
C
/* ldctor.c -- constructor support routines
|
|
Copyright (C) 1991-2019 Free Software Foundation, Inc.
|
|
By Steve Chamberlain <sac@cygnus.com>
|
|
|
|
This file is part of the GNU Binutils.
|
|
|
|
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. */
|
|
|
|
#include "sysdep.h"
|
|
#include "bfd.h"
|
|
#include "bfdlink.h"
|
|
#include "safe-ctype.h"
|
|
#include "ctf-api.h"
|
|
|
|
#include "ld.h"
|
|
#include "ldexp.h"
|
|
#include "ldlang.h"
|
|
#include "ldmisc.h"
|
|
#include <ldgram.h>
|
|
#include "ldmain.h"
|
|
#include "ldctor.h"
|
|
|
|
/* The list of statements needed to handle constructors. These are
|
|
invoked by the command CONSTRUCTORS in the linker script. */
|
|
lang_statement_list_type constructor_list;
|
|
|
|
/* Whether the constructors should be sorted. Note that this is
|
|
global for the entire link; we assume that there is only a single
|
|
CONSTRUCTORS command in the linker script. */
|
|
bfd_boolean constructors_sorted;
|
|
|
|
/* The sets we have seen. */
|
|
struct set_info *sets;
|
|
|
|
/* Add an entry to a set. H is the entry in the linker hash table.
|
|
RELOC is the relocation to use for an entry in the set. SECTION
|
|
and VALUE are the value to add. This is called during the first
|
|
phase of the link, when we are still gathering symbols together.
|
|
We just record the information now. The ldctor_build_sets
|
|
function will construct the sets. */
|
|
|
|
void
|
|
ldctor_add_set_entry (struct bfd_link_hash_entry *h,
|
|
bfd_reloc_code_real_type reloc,
|
|
const char *name,
|
|
asection *section,
|
|
bfd_vma value)
|
|
{
|
|
struct set_info *p;
|
|
struct set_element *e;
|
|
struct set_element **epp;
|
|
|
|
for (p = sets; p != NULL; p = p->next)
|
|
if (p->h == h)
|
|
break;
|
|
|
|
if (p == NULL)
|
|
{
|
|
p = (struct set_info *) xmalloc (sizeof (struct set_info));
|
|
p->next = sets;
|
|
sets = p;
|
|
p->h = h;
|
|
p->reloc = reloc;
|
|
p->count = 0;
|
|
p->elements = NULL;
|
|
}
|
|
else
|
|
{
|
|
if (p->reloc != reloc)
|
|
{
|
|
einfo (_("%X%P: different relocs used in set %s\n"),
|
|
h->root.string);
|
|
return;
|
|
}
|
|
|
|
/* Don't permit a set to be constructed from different object
|
|
file formats. The same reloc may have different results. We
|
|
actually could sometimes handle this, but the case is
|
|
unlikely to ever arise. Sometimes constructor symbols are in
|
|
unusual sections, such as the absolute section--this appears
|
|
to be the case in Linux a.out--and in such cases we just
|
|
assume everything is OK. */
|
|
if (p->elements != NULL
|
|
&& section->owner != NULL
|
|
&& p->elements->section->owner != NULL
|
|
&& strcmp (bfd_get_target (section->owner),
|
|
bfd_get_target (p->elements->section->owner)) != 0)
|
|
{
|
|
einfo (_("%X%P: different object file formats composing set %s\n"),
|
|
h->root.string);
|
|
return;
|
|
}
|
|
}
|
|
|
|
e = (struct set_element *) xmalloc (sizeof (struct set_element));
|
|
e->next = NULL;
|
|
e->name = name;
|
|
e->section = section;
|
|
e->value = value;
|
|
|
|
for (epp = &p->elements; *epp != NULL; epp = &(*epp)->next)
|
|
;
|
|
*epp = e;
|
|
|
|
++p->count;
|
|
}
|
|
|
|
/* Get the priority of a g++ global constructor or destructor from the
|
|
symbol name. */
|
|
|
|
static int
|
|
ctor_prio (const char *name)
|
|
{
|
|
/* The name will look something like _GLOBAL_$I$65535$test02__Fv.
|
|
There might be extra leading underscores, and the $ characters
|
|
might be something else. The I might be a D. */
|
|
|
|
while (*name == '_')
|
|
++name;
|
|
|
|
if (!CONST_STRNEQ (name, "GLOBAL_"))
|
|
return -1;
|
|
|
|
name += sizeof "GLOBAL_" - 1;
|
|
|
|
if (name[0] != name[2])
|
|
return -1;
|
|
if (name[1] != 'I' && name[1] != 'D')
|
|
return -1;
|
|
if (!ISDIGIT (name[3]))
|
|
return -1;
|
|
|
|
return atoi (name + 3);
|
|
}
|
|
|
|
/* This function is used to sort constructor elements by priority. It
|
|
is called via qsort. */
|
|
|
|
static int
|
|
ctor_cmp (const void *p1, const void *p2)
|
|
{
|
|
const struct set_element *const *pe1
|
|
= (const struct set_element *const *) p1;
|
|
const struct set_element *const *pe2
|
|
= (const struct set_element *const *) p2;
|
|
const char *n1;
|
|
const char *n2;
|
|
int prio1;
|
|
int prio2;
|
|
|
|
n1 = (*pe1)->name;
|
|
if (n1 == NULL)
|
|
n1 = "";
|
|
n2 = (*pe2)->name;
|
|
if (n2 == NULL)
|
|
n2 = "";
|
|
|
|
/* We need to sort in reverse order by priority. When two
|
|
constructors have the same priority, we should maintain their
|
|
current relative position. */
|
|
|
|
prio1 = ctor_prio (n1);
|
|
prio2 = ctor_prio (n2);
|
|
|
|
/* We sort in reverse order because that is what g++ expects. */
|
|
if (prio1 < prio2)
|
|
return 1;
|
|
else if (prio1 > prio2)
|
|
return -1;
|
|
|
|
/* Force a stable sort. */
|
|
|
|
if (pe1 < pe2)
|
|
return -1;
|
|
else if (pe1 > pe2)
|
|
return 1;
|
|
else
|
|
return 0;
|
|
}
|
|
|
|
/* This function is called after the first phase of the link and
|
|
before the second phase. At this point all set information has
|
|
been gathered. We now put the statements to build the sets
|
|
themselves into constructor_list. */
|
|
|
|
void
|
|
ldctor_build_sets (void)
|
|
{
|
|
static bfd_boolean called;
|
|
bfd_boolean header_printed;
|
|
struct set_info *p;
|
|
|
|
/* The emulation code may call us directly, but we only want to do
|
|
this once. */
|
|
if (called)
|
|
return;
|
|
called = TRUE;
|
|
|
|
if (constructors_sorted)
|
|
{
|
|
for (p = sets; p != NULL; p = p->next)
|
|
{
|
|
int c, i;
|
|
struct set_element *e;
|
|
struct set_element **array;
|
|
|
|
if (p->elements == NULL)
|
|
continue;
|
|
|
|
c = 0;
|
|
for (e = p->elements; e != NULL; e = e->next)
|
|
++c;
|
|
|
|
array = (struct set_element **) xmalloc (c * sizeof *array);
|
|
|
|
i = 0;
|
|
for (e = p->elements; e != NULL; e = e->next)
|
|
{
|
|
array[i] = e;
|
|
++i;
|
|
}
|
|
|
|
qsort (array, c, sizeof *array, ctor_cmp);
|
|
|
|
e = array[0];
|
|
p->elements = e;
|
|
for (i = 0; i < c - 1; i++)
|
|
array[i]->next = array[i + 1];
|
|
array[i]->next = NULL;
|
|
|
|
free (array);
|
|
}
|
|
}
|
|
|
|
lang_list_init (&constructor_list);
|
|
push_stat_ptr (&constructor_list);
|
|
|
|
header_printed = FALSE;
|
|
for (p = sets; p != NULL; p = p->next)
|
|
{
|
|
struct set_element *e;
|
|
reloc_howto_type *howto;
|
|
int reloc_size, size;
|
|
|
|
/* If the symbol is defined, we may have been invoked from
|
|
collect, and the sets may already have been built, so we do
|
|
not do anything. */
|
|
if (p->h->type == bfd_link_hash_defined
|
|
|| p->h->type == bfd_link_hash_defweak)
|
|
continue;
|
|
|
|
/* For each set we build:
|
|
set:
|
|
.long number_of_elements
|
|
.long element0
|
|
...
|
|
.long elementN
|
|
.long 0
|
|
except that we use the right size instead of .long. When
|
|
generating relocatable output, we generate relocs instead of
|
|
addresses. */
|
|
howto = bfd_reloc_type_lookup (link_info.output_bfd, p->reloc);
|
|
if (howto == NULL)
|
|
{
|
|
if (bfd_link_relocatable (&link_info))
|
|
{
|
|
einfo (_("%X%P: %s does not support reloc %s for set %s\n"),
|
|
bfd_get_target (link_info.output_bfd),
|
|
bfd_get_reloc_code_name (p->reloc),
|
|
p->h->root.string);
|
|
continue;
|
|
}
|
|
|
|
/* If this is not a relocatable link, all we need is the
|
|
size, which we can get from the input BFD. */
|
|
if (p->elements->section->owner != NULL)
|
|
howto = bfd_reloc_type_lookup (p->elements->section->owner,
|
|
p->reloc);
|
|
if (howto == NULL)
|
|
{
|
|
/* See PR 20911 for a reproducer. */
|
|
if (p->elements->section->owner == NULL)
|
|
einfo (_("%X%P: special section %s does not support reloc %s for set %s\n"),
|
|
bfd_section_name (p->elements->section),
|
|
bfd_get_reloc_code_name (p->reloc),
|
|
p->h->root.string);
|
|
else
|
|
einfo (_("%X%P: %s does not support reloc %s for set %s\n"),
|
|
bfd_get_target (p->elements->section->owner),
|
|
bfd_get_reloc_code_name (p->reloc),
|
|
p->h->root.string);
|
|
continue;
|
|
}
|
|
}
|
|
|
|
reloc_size = bfd_get_reloc_size (howto);
|
|
switch (reloc_size)
|
|
{
|
|
case 1: size = BYTE; break;
|
|
case 2: size = SHORT; break;
|
|
case 4: size = LONG; break;
|
|
case 8:
|
|
if (howto->complain_on_overflow == complain_overflow_signed)
|
|
size = SQUAD;
|
|
else
|
|
size = QUAD;
|
|
break;
|
|
default:
|
|
einfo (_("%X%P: unsupported size %d for set %s\n"),
|
|
bfd_get_reloc_size (howto), p->h->root.string);
|
|
size = LONG;
|
|
break;
|
|
}
|
|
|
|
lang_add_assignment (exp_assign (".",
|
|
exp_unop (ALIGN_K,
|
|
exp_intop (reloc_size)),
|
|
FALSE));
|
|
lang_add_assignment (exp_assign (p->h->root.string,
|
|
exp_nameop (NAME, "."),
|
|
FALSE));
|
|
lang_add_data (size, exp_intop (p->count));
|
|
|
|
for (e = p->elements; e != NULL; e = e->next)
|
|
{
|
|
if (config.map_file != NULL)
|
|
{
|
|
int len;
|
|
|
|
if (!header_printed)
|
|
{
|
|
minfo (_("\nSet Symbol\n\n"));
|
|
header_printed = TRUE;
|
|
}
|
|
|
|
minfo ("%s", p->h->root.string);
|
|
len = strlen (p->h->root.string);
|
|
|
|
if (len >= 19)
|
|
{
|
|
print_nl ();
|
|
len = 0;
|
|
}
|
|
while (len < 20)
|
|
{
|
|
print_space ();
|
|
++len;
|
|
}
|
|
|
|
if (e->name != NULL)
|
|
minfo ("%pT\n", e->name);
|
|
else
|
|
minfo ("%G\n", e->section->owner, e->section, e->value);
|
|
}
|
|
|
|
/* Need SEC_KEEP for --gc-sections. */
|
|
if (!bfd_is_abs_section (e->section))
|
|
e->section->flags |= SEC_KEEP;
|
|
|
|
if (bfd_link_relocatable (&link_info))
|
|
lang_add_reloc (p->reloc, howto, e->section, e->name,
|
|
exp_intop (e->value));
|
|
else
|
|
lang_add_data (size, exp_relop (e->section, e->value));
|
|
}
|
|
|
|
lang_add_data (size, exp_intop (0));
|
|
}
|
|
|
|
pop_stat_ptr ();
|
|
}
|