binutils-gdb/gdb/elfread.c
Jim Kingdon 3c02636b4e * Clean up xcoff relocation.
objfiles.h (struct objfiles): Add section_offsets, num_sections.
	symfile.c (syms_from_objfile), xcoffread.c (xcoff_symfile_offsets):
	Set them.
	symtab.h (struct general_symbol_info): Add section field.
	minsyms.c (prim_record_minimal_symbol{,_and_info}): Set it.
	xcoffread.c: Set section for symbols and msymbols.
	(struct symtab): Add block_line_section field.
	buildsym.c (end_symtab): Set it.
	(end_symtab and callers): Add section parameter.
	objfiles.c (objfile_relocate): New funciton.
	xcoffexec.c (vmap_symtab): Use it.
	xcoffsolib.h (struct vmap): Remove unused fields.
	config/rs6000/tm-rs6000.h, stack.c, xcoffexec.c: Remove
	CORE_NEEDS_RELOCATION, symtab_relocated.
	config/rs6000/tm-rs6000.h: Remove use of loadinfotext.
	rs6000-tdep.c: Make loadinfotext static.
	breakpoint.c (fixup_breakpoints): Doc fix.
	symtab.h (struct symtab), config/rs6000/tm-rs6000.h, buildsym.c
	(end_symtab): primary field replaces nonreloc.
1993-03-26 14:59:43 +00:00

659 lines
20 KiB
C
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

/* Read ELF (Executable and Linking Format) object files for GDB.
Copyright 1991, 1992 Free Software Foundation, Inc.
Written by Fred Fish at Cygnus Support.
This file is part of GDB.
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 2 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., 675 Mass Ave, Cambridge, MA 02139, USA. */
/************************************************************************
* *
* NOTICE *
* *
* This file is still under construction. When it is complete, this *
* notice will be removed. Until then, direct any questions or changes *
* to Fred Fish at Cygnus Support (fnf@cygnus.com) *
* *
* FIXME Still needs support for shared libraries. *
* FIXME Still needs support for core files. *
* FIXME The ".debug" and ".line" section names are hardwired. *
* *
************************************************************************/
#include "defs.h"
#include "elf/common.h"
#include "elf/external.h"
#include "elf/internal.h"
#include "bfd.h"
#include "libbfd.h" /* For bfd_elf_find_section */
#include "symtab.h"
#include "symfile.h"
#include "objfiles.h"
#include "buildsym.h"
#include "gdb-stabs.h"
#include "complaints.h"
#include <string.h>
#include "demangle.h"
/* The struct elfinfo is available only during ELF symbol table and
psymtab reading. It is destroyed at the complation of psymtab-reading.
It's local to elf_symfile_read. */
struct elfinfo {
file_ptr dboffset; /* Offset to dwarf debug section */
unsigned int dbsize; /* Size of dwarf debug section */
file_ptr lnoffset; /* Offset to dwarf line number section */
unsigned int lnsize; /* Size of dwarf line number section */
asection *stabsect; /* Section pointer for .stab section */
asection *stabindexsect; /* Section pointer for .stab.index section */
};
/* Various things we might complain about... */
struct complaint section_info_complaint =
{"elf/stab section information %s without a preceding file symbol", 0, 0};
struct complaint section_info_dup_complaint =
{"duplicated elf/stab section information for %s", 0, 0};
struct complaint stab_info_mismatch_complaint =
{"elf/stab section information missing for %s", 0, 0};
struct complaint stab_info_questionable_complaint =
{"elf/stab section information questionable for %s", 0, 0};
static void
elf_symfile_init PARAMS ((struct objfile *));
static void
elf_new_init PARAMS ((struct objfile *));
static void
elf_symfile_read PARAMS ((struct objfile *, struct section_offsets *, int));
static void
elf_symfile_finish PARAMS ((struct objfile *));
static void
elf_symtab_read PARAMS ((bfd *, CORE_ADDR, struct objfile *));
static void
free_elfinfo PARAMS ((PTR));
static struct section_offsets *
elf_symfile_offsets PARAMS ((struct objfile *, CORE_ADDR));
static void
record_minimal_symbol_and_info PARAMS ((char *, CORE_ADDR,
enum minimal_symbol_type, char *,
struct objfile *));
static void
elf_locate_sections PARAMS ((bfd *, asection *, PTR));
/* We are called once per section from elf_symfile_read. We
need to examine each section we are passed, check to see
if it is something we are interested in processing, and
if so, stash away some access information for the section.
For now we recognize the dwarf debug information sections and
line number sections from matching their section names. The
ELF definition is no real help here since it has no direct
knowledge of DWARF (by design, so any debugging format can be
used).
We also recognize the ".stab" sections used by the Sun compilers
released with Solaris 2.
FIXME: The section names should not be hardwired strings. */
static void
elf_locate_sections (ignore_abfd, sectp, eip)
bfd *ignore_abfd;
asection *sectp;
PTR eip;
{
register struct elfinfo *ei;
ei = (struct elfinfo *) eip;
if (STREQ (sectp -> name, ".debug"))
{
ei -> dboffset = sectp -> filepos;
ei -> dbsize = bfd_get_section_size_before_reloc (sectp);
}
else if (STREQ (sectp -> name, ".line"))
{
ei -> lnoffset = sectp -> filepos;
ei -> lnsize = bfd_get_section_size_before_reloc (sectp);
}
else if (STREQ (sectp -> name, ".stab"))
{
ei -> stabsect = sectp;
}
else if (STREQ (sectp -> name, ".stab.index"))
{
ei -> stabindexsect = sectp;
}
}
#if 0 /* Currently unused */
char *
elf_interpreter (abfd)
bfd *abfd;
{
sec_ptr interp_sec;
unsigned size;
char *interp = NULL;
interp_sec = bfd_get_section_by_name (abfd, ".interp");
if (interp_sec)
{
size = bfd_section_size (abfd, interp_sec);
interp = alloca (size);
if (bfd_get_section_contents (abfd, interp_sec, interp, (file_ptr)0,
size))
{
interp = savestring (interp, size - 1);
}
else
{
interp = NULL;
}
}
return (interp);
}
#endif
static void
record_minimal_symbol_and_info (name, address, ms_type, info, objfile)
char *name;
CORE_ADDR address;
enum minimal_symbol_type ms_type;
char *info; /* FIXME, is this really char *? */
struct objfile *objfile;
{
name = obsavestring (name, strlen (name), &objfile -> symbol_obstack);
prim_record_minimal_symbol_and_info (name, address, ms_type, info, -1);
}
/*
LOCAL FUNCTION
elf_symtab_read -- read the symbol table of an ELF file
SYNOPSIS
void elf_symtab_read (bfd *abfd, CORE_ADDR addr,
struct objfile *objfile)
DESCRIPTION
Given an open bfd, a base address to relocate symbols to, and a
flag that specifies whether or not this bfd is for an executable
or not (may be shared library for example), add all the global
function and data symbols to the minimal symbol table.
In stabs-in-ELF, as implemented by Sun, there are some local symbols
defined in the ELF symbol table, which can be used to locate
the beginnings of sections from each ".o" file that was linked to
form the executable objfile. We gather any such info and record it
in data structures hung off the objfile's private data.
*/
static void
elf_symtab_read (abfd, addr, objfile)
bfd *abfd;
CORE_ADDR addr;
struct objfile *objfile;
{
unsigned int storage_needed;
asymbol *sym;
asymbol **symbol_table;
unsigned int number_of_symbols;
unsigned int i;
int index;
struct cleanup *back_to;
CORE_ADDR symaddr;
enum minimal_symbol_type ms_type;
/* If sectinfo is nonzero, it contains section info that should end up
filed in the objfile. */
struct stab_section_info *sectinfo = 0;
/* If filesym is nonzero, it points to a file symbol, but we haven't
seen any section info for it yet. */
asymbol *filesym = 0;
struct dbx_symfile_info *dbx = (struct dbx_symfile_info *)
objfile->sym_private;
storage_needed = get_symtab_upper_bound (abfd);
if (storage_needed > 0)
{
symbol_table = (asymbol **) xmalloc (storage_needed);
back_to = make_cleanup (free, symbol_table);
number_of_symbols = bfd_canonicalize_symtab (abfd, symbol_table);
for (i = 0; i < number_of_symbols; i++)
{
sym = symbol_table[i];
/* Select global/weak symbols that are defined in a specific section.
Note that bfd now puts abs symbols in their own section, so
all symbols we are interested in will have a section. */
if ((sym -> flags & (BSF_GLOBAL | BSF_WEAK))
&& (sym -> section != NULL))
{
symaddr = sym -> value;
/* Relocate all non-absolute symbols by base address. */
if (sym -> section != &bfd_abs_section)
symaddr += addr;
/* For non-absolute symbols, use the type of the section
they are relative to, to intuit text/data. Bfd provides
no way of figuring this out for absolute symbols. */
if (sym -> section -> flags & SEC_CODE)
{
ms_type = mst_text;
}
else if (sym -> section -> flags & SEC_DATA)
{
ms_type = mst_data;
}
else
{
/* FIXME: Solaris2 shared libraries include lots of
odd "absolute" and "undefined" symbols, that play
hob with actions like finding what function the PC
is in. Ignore them if they aren't text or data. */
/* ms_type = mst_unknown; */
continue; /* Skip this symbol. */
}
/* Pass symbol size field in via BFD. FIXME!!! */
record_minimal_symbol_and_info ((char *) sym -> name,
symaddr, ms_type, sym->udata, objfile);
}
/* See if this is a debugging symbol that helps Solaris
stabs-in-elf debugging. */
else if (sym->flags & BSF_FILE)
{
/* Chain any old one onto the objfile; remember new sym. */
if (sectinfo)
{
sectinfo->next = dbx->stab_section_info;
dbx->stab_section_info = sectinfo;
sectinfo = 0;
}
filesym = sym;
}
else if ((sym->flags & BSF_LOCAL) &&
(sym->section) &&
(sym->section->flags & SEC_DATA) &&
(sym->name))
{
/* Named Local variable in a Data section. Check its name. */
index = -1;
switch (sym->name[1])
{
case 'b':
if (STREQ ("Bbss.bss", sym->name))
index = SECT_OFF_BSS;
break;
case 'd':
if (STREQ ("Ddata.data", sym->name))
index = SECT_OFF_DATA;
break;
case 'r':
if (STREQ ("Drodata.rodata", sym->name))
index = SECT_OFF_RODATA;
break;
}
if (index > 0)
{
/* We have some new info. Allocate a sectinfo, if
needed, and fill it in. */
if (!sectinfo)
{
sectinfo = (struct stab_section_info *)
xmmalloc (objfile -> md,
sizeof (*sectinfo));
memset ((PTR) sectinfo, 0, sizeof (*sectinfo));
if (!filesym)
complain (&section_info_complaint, sym->name);
else
sectinfo->filename = (char *)filesym->name;
}
if (sectinfo->sections[index])
complain (&section_info_dup_complaint, sectinfo->filename);
symaddr = sym -> value;
/* Relocate all non-absolute symbols by base address. */
if (sym -> section != &bfd_abs_section)
symaddr += addr;
sectinfo->sections[index] = symaddr;
}
}
}
do_cleanups (back_to);
}
}
/* Scan and build partial symbols for a symbol file.
We have been initialized by a call to elf_symfile_init, which
currently does nothing.
SECTION_OFFSETS is a set of offsets to apply to relocate the symbols
in each section. We simplify it down to a single offset for all
symbols. FIXME.
MAINLINE is true if we are reading the main symbol
table (as opposed to a shared lib or dynamically loaded file).
This function only does the minimum work necessary for letting the
user "name" things symbolically; it does not read the entire symtab.
Instead, it reads the external and static symbols and puts them in partial
symbol tables. When more extensive information is requested of a
file, the corresponding partial symbol table is mutated into a full
fledged symbol table by going back and reading the symbols
for real.
We look for sections with specific names, to tell us what debug
format to look for: FIXME!!!
dwarf_build_psymtabs() builds psymtabs for DWARF symbols;
elfstab_build_psymtabs() handles STABS symbols.
Note that ELF files have a "minimal" symbol table, which looks a lot
like a COFF symbol table, but has only the minimal information necessary
for linking. We process this also, and use the information to
build gdb's minimal symbol table. This gives us some minimal debugging
capability even for files compiled without -g. */
static void
elf_symfile_read (objfile, section_offsets, mainline)
struct objfile *objfile;
struct section_offsets *section_offsets;
int mainline;
{
bfd *abfd = objfile->obfd;
struct elfinfo ei;
struct cleanup *back_to;
CORE_ADDR offset;
init_minimal_symbol_collection ();
back_to = make_cleanup (discard_minimal_symbols, 0);
memset ((char *) &ei, 0, sizeof (ei));
/* Allocate struct to keep track of the symfile */
objfile->sym_private = (PTR)
xmmalloc (objfile -> md, sizeof (struct dbx_symfile_info));
memset ((char *) objfile->sym_private, 0, sizeof (struct dbx_symfile_info));
make_cleanup (free_elfinfo, (PTR) objfile);
/* Process the normal ELF symbol table first. This may write some
chain of info into the dbx_symfile_info in objfile->sym_private,
which can later be used by elfstab_offset_sections. */
/* FIXME, should take a section_offsets param, not just an offset. */
offset = ANOFFSET (section_offsets, 0);
elf_symtab_read (abfd, offset, objfile);
/* Now process debugging information, which is contained in
special ELF sections. We first have to find them... */
bfd_map_over_sections (abfd, elf_locate_sections, (PTR) &ei);
if (ei.dboffset && ei.lnoffset)
{
/* DWARF sections */
dwarf_build_psymtabs (objfile,
section_offsets, mainline,
ei.dboffset, ei.dbsize,
ei.lnoffset, ei.lnsize);
}
if (ei.stabsect)
{
/* STABS sections */
/* FIXME: Sun didn't really know how to implement this well.
They made .stab sections that don't point to the .stabstr
section with the sh_link field. BFD doesn't make string table
sections visible to the caller. So we have to search the
ELF section table, not the BFD section table, for the string
table. */
struct elf_internal_shdr *elf_sect;
elf_sect = bfd_elf_find_section (abfd, ".stabstr");
if (elf_sect)
elfstab_build_psymtabs (objfile,
section_offsets,
mainline,
ei.stabsect->filepos, /* .stab offset */
bfd_get_section_size_before_reloc (ei.stabsect),/* .stab size */
(file_ptr) elf_sect->sh_offset, /* .stabstr offset */
elf_sect->sh_size); /* .stabstr size */
}
if (!have_partial_symbols ())
{
wrap_here ("");
printf_filtered ("(no debugging symbols found)...");
wrap_here ("");
}
/* Install any minimal symbols that have been collected as the current
minimal symbols for this objfile. */
install_minimal_symbols (objfile);
do_cleanups (back_to);
}
/* This cleans up the objfile's sym_private pointer, and the chain of
stab_section_info's, that might be dangling from it. */
static void
free_elfinfo (objp)
PTR objp;
{
struct objfile *objfile = (struct objfile *)objp;
struct dbx_symfile_info *dbxinfo = (struct dbx_symfile_info *)
objfile->sym_private;
struct stab_section_info *ssi, *nssi;
ssi = dbxinfo->stab_section_info;
while (ssi)
{
nssi = ssi->next;
mfree (objfile->md, ssi);
ssi = nssi;
}
dbxinfo->stab_section_info = 0; /* Just say No mo info about this. */
}
/* Initialize anything that needs initializing when a completely new symbol
file is specified (not just adding some symbols from another file, e.g. a
shared library).
We reinitialize buildsym, since we may be reading stabs from an ELF file. */
static void
elf_new_init (ignore)
struct objfile *ignore;
{
stabsread_new_init ();
buildsym_new_init ();
}
/* Perform any local cleanups required when we are done with a particular
objfile. I.E, we are in the process of discarding all symbol information
for an objfile, freeing up all memory held for it, and unlinking the
objfile struct from the global list of known objfiles. */
static void
elf_symfile_finish (objfile)
struct objfile *objfile;
{
if (objfile -> sym_private != NULL)
{
mfree (objfile -> md, objfile -> sym_private);
}
}
/* ELF specific initialization routine for reading symbols.
It is passed a pointer to a struct sym_fns which contains, among other
things, the BFD for the file whose symbols are being read, and a slot for
a pointer to "private data" which we can fill with goodies.
For now at least, we have nothing in particular to do, so this function is
just a stub. */
static void
elf_symfile_init (ignore)
struct objfile *ignore;
{
}
/* ELF specific parsing routine for section offsets.
Plain and simple for now. */
static
struct section_offsets *
elf_symfile_offsets (objfile, addr)
struct objfile *objfile;
CORE_ADDR addr;
{
struct section_offsets *section_offsets;
int i;
section_offsets = (struct section_offsets *)
obstack_alloc (&objfile -> psymbol_obstack,
sizeof (struct section_offsets) +
sizeof (section_offsets->offsets) * (SECT_OFF_MAX-1));
for (i = 0; i < SECT_OFF_MAX; i++)
ANOFFSET (section_offsets, i) = addr;
return section_offsets;
}
/* When handling an ELF file that contains Sun STABS debug info,
some of the debug info is relative to the particular chunk of the
section that was generated in its individual .o file. E.g.
offsets to static variables are relative to the start of the data
segment *for that module before linking*. This information is
painfully squirreled away in the ELF symbol table as local symbols
with wierd names. Go get 'em when needed. */
void
elfstab_offset_sections (objfile, pst)
struct objfile *objfile;
struct partial_symtab *pst;
{
char *filename = pst->filename;
struct dbx_symfile_info *dbx = (struct dbx_symfile_info *)
objfile->sym_private;
struct stab_section_info *maybe = dbx->stab_section_info;
struct stab_section_info *questionable = 0;
int i;
char *p;
/* The ELF symbol info doesn't include path names, so strip the path
(if any) from the psymtab filename. */
while (0 != (p = strchr (filename, '/')))
filename = p+1;
/* FIXME: This linear search could speed up significantly
if it was chained in the right order to match how we search it,
and if we unchained when we found a match. */
for (; maybe; maybe = maybe->next)
{
if (filename[0] == maybe->filename[0]
&& STREQ (filename, maybe->filename))
{
/* We found a match. But there might be several source files
(from different directories) with the same name. */
if (0 == maybe->found)
break;
questionable = maybe; /* Might use it later. */
}
}
if (maybe == 0 && questionable != 0)
{
complain (&stab_info_questionable_complaint, filename);
maybe = questionable;
}
if (maybe)
{
/* Found it! Allocate a new psymtab struct, and fill it in. */
maybe->found++;
pst->section_offsets = (struct section_offsets *)
obstack_alloc (&objfile -> psymbol_obstack,
sizeof (struct section_offsets) +
sizeof (pst->section_offsets->offsets) * (SECT_OFF_MAX-1));
for (i = 0; i < SECT_OFF_MAX; i++)
ANOFFSET (pst->section_offsets, i) = maybe->sections[i];
return;
}
/* We were unable to find any offsets for this file. Complain. */
if (dbx->stab_section_info) /* If there *is* any info, */
complain (&stab_info_mismatch_complaint, filename);
}
/* Register that we are able to handle ELF object file formats and DWARF
debugging formats.
Unlike other object file formats, where the debugging information format
is implied by the object file format, the ELF object file format and the
DWARF debugging information format are two distinct, and potentially
separate entities. I.E. it is perfectly possible to have ELF objects
with debugging formats other than DWARF. And it is conceivable that the
DWARF debugging format might be used with another object file format,
like COFF, by simply using COFF's custom section feature.
GDB, and to a lesser extent BFD, should support the notion of separate
object file formats and debugging information formats. For now, we just
use "elf" in the same sense as "a.out" or "coff", to imply both the ELF
object file format and the DWARF debugging format. */
static struct sym_fns elf_sym_fns =
{
"elf", /* sym_name: name or name prefix of BFD target type */
3, /* sym_namelen: number of significant sym_name chars */
elf_new_init, /* sym_new_init: init anything gbl to entire symtab */
elf_symfile_init, /* sym_init: read initial info, setup for sym_read() */
elf_symfile_read, /* sym_read: read a symbol file into symtab */
elf_symfile_finish, /* sym_finish: finished with file, cleanup */
elf_symfile_offsets, /* sym_offsets: Translate ext. to int. relocation */
NULL /* next: pointer to next struct sym_fns */
};
void
_initialize_elfread ()
{
add_symtab_fns (&elf_sym_fns);
}