mirror of
https://sourceware.org/git/binutils-gdb.git
synced 2024-12-09 04:21:49 +08:00
49864cfa6c
expr.c (__): Undefine before defining. as.c (got_sig): Don't return anything; return type might be void. Whitespace/comment cleanup in frags.c. Some patches for `-pedantic' or `-fno-common' compilation. (Some of these changes are from Michael Meissner; see change log.)
1190 lines
30 KiB
C
1190 lines
30 KiB
C
/* ELF object file format
|
|
Copyright (C) 1992, 1993 Free Software Foundation, Inc.
|
|
|
|
This file is part of GAS, the GNU Assembler.
|
|
|
|
GAS 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,
|
|
or (at your option) any later version.
|
|
|
|
GAS 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 GAS; see the file COPYING. If not, write
|
|
to the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */
|
|
|
|
/* HP PA-RISC support was contributed by the Center for Software Science
|
|
at the University of Utah. */
|
|
|
|
#include "as.h"
|
|
#include "aout/stab_gnu.h"
|
|
#include "obstack.h"
|
|
|
|
static void obj_elf_stab PARAMS ((int what));
|
|
static void obj_elf_line PARAMS ((void));
|
|
static void obj_elf_desc PARAMS ((void));
|
|
static void obj_elf_version PARAMS ((void));
|
|
static void obj_elf_section PARAMS ((int));
|
|
static void obj_elf_size PARAMS ((void));
|
|
static void obj_elf_type PARAMS ((void));
|
|
static void obj_elf_ident PARAMS ((void));
|
|
|
|
const pseudo_typeS obj_pseudo_table[] = {
|
|
{ "ident", obj_elf_ident, 0 },
|
|
{ "section", obj_elf_section, 0 },
|
|
{ "size", obj_elf_size, 0 },
|
|
{ "type", obj_elf_type, 0 },
|
|
{ "version", obj_elf_version, 0 },
|
|
|
|
/* These are used for stabs-in-elf configurations. */
|
|
{ "desc", obj_elf_desc, 0 },
|
|
{ "line", obj_elf_line, 0 },
|
|
{ "stabd", obj_elf_stab, 'd' },
|
|
{ "stabn", obj_elf_stab, 'n' },
|
|
{ "stabs", obj_elf_stab, 's' },
|
|
/* This is used on Solaris 2.x on SPARC, but not supported yet. */
|
|
{ "xstabs", s_ignore, 0 },
|
|
|
|
{ NULL} /* end sentinel */
|
|
};
|
|
|
|
static void
|
|
obj_elf_section (xxx)
|
|
int xxx;
|
|
{
|
|
char *string;
|
|
asection *sec;
|
|
|
|
/* Initialize this with inclusive-or of all flags that can be cleared
|
|
by attributes, but not set by them. Also include flags that won't
|
|
get set properly in the assembler, but which the user/compiler
|
|
shouldn't be expected to set. */
|
|
flagword flags = SEC_READONLY | SEC_ALLOC | SEC_RELOC;
|
|
/* Initialize this with the default flags to be used if none are
|
|
specified. */
|
|
flagword default_flags = SEC_ALLOC | SEC_RELOC;
|
|
|
|
string = demand_copy_C_string (&xxx);
|
|
SKIP_WHITESPACE ();
|
|
if (*input_line_pointer != ',')
|
|
flags = default_flags;
|
|
while (*input_line_pointer == ',')
|
|
{
|
|
flagword bit;
|
|
int len, inv;
|
|
char *p, oldp;
|
|
|
|
input_line_pointer++;
|
|
if (*input_line_pointer != '#')
|
|
{
|
|
as_bad ("unrecognized syntax in .section command");
|
|
ignore_rest_of_line ();
|
|
break;
|
|
}
|
|
input_line_pointer++;
|
|
|
|
#define CHECK(X,BIT,NEG) \
|
|
if (!strncmp(X,input_line_pointer,len = sizeof(X) - 1)) { \
|
|
bit = BIT; inv = NEG; goto match; }
|
|
|
|
CHECK ("write", SEC_READONLY, 1);
|
|
CHECK ("alloc", SEC_ALLOC, 0);
|
|
#undef CHECK
|
|
|
|
p = input_line_pointer;
|
|
while (!is_end_of_line[*p] && *p != 0 && *p != ',')
|
|
p++;
|
|
*p = 0;
|
|
oldp = *p;
|
|
as_bad ("unrecognized section attribute `%s' ignored",
|
|
input_line_pointer);
|
|
*p = oldp;
|
|
continue;
|
|
|
|
match:
|
|
if (inv)
|
|
flags &= ~bit;
|
|
else
|
|
flags |= bit;
|
|
input_line_pointer += len;
|
|
}
|
|
demand_empty_rest_of_line ();
|
|
|
|
sec = bfd_get_section_by_name (stdoutput, string);
|
|
if (sec == 0)
|
|
{
|
|
sec = bfd_make_section_old_way (stdoutput, string);
|
|
bfd_set_section_flags (stdoutput, sec,
|
|
/* @@ What should the flags be?? */
|
|
flags);
|
|
sec->output_section = sec;
|
|
}
|
|
subseg_change (sec, 0);
|
|
}
|
|
|
|
#if 0
|
|
/* pa-spaces.c -- Space/subspace support for the HP PA-RISC version of GAS */
|
|
|
|
/* for space, subspace, and symbol maintenance on HP 9000 Series 800 */
|
|
|
|
space_dict_chainS *space_dict_root;
|
|
space_dict_chainS *space_dict_last;
|
|
|
|
space_dict_chainS *current_space;
|
|
subspace_dict_chainS *current_subspace;
|
|
#endif
|
|
|
|
symbolS *start_symbol_root;
|
|
symbolS *start_symbol_last;
|
|
|
|
#if 0 /* I really don't think this belongs in this file. */
|
|
void pa_spaces_begin()
|
|
{
|
|
space_dict_chainS *space;
|
|
int i;
|
|
subsegT now_subseg = now_subseg;
|
|
|
|
space_dict_root = NULL;
|
|
space_dict_last = NULL;
|
|
|
|
start_symbol_root = NULL;
|
|
start_symbol_last = NULL;
|
|
|
|
/* create default space and subspace dictionaries */
|
|
|
|
i = 0;
|
|
while ( pa_def_spaces[i].name ) {
|
|
if ( pa_def_spaces[i].alias )
|
|
pa_def_spaces[i].segment = subseg_new(pa_def_spaces[i].alias,0);
|
|
else
|
|
pa_def_spaces[i].segment = bfd_make_section_old_way(stdoutput,pa_def_spaces[i].name);
|
|
|
|
create_new_space(pa_def_spaces[i].name,pa_def_spaces[i].spnum,
|
|
pa_def_spaces[i].loadable,pa_def_spaces[i].defined,
|
|
pa_def_spaces[i].private,pa_def_spaces[i].sort,0,
|
|
pa_def_spaces[i].segment);
|
|
i++;
|
|
}
|
|
|
|
i = 0;
|
|
while ( pa_def_subspaces[i].name ) {
|
|
space = pa_segment_to_space(pa_def_spaces[pa_def_subspaces[i].def_space_index].segment);
|
|
if ( space ) {
|
|
create_new_subspace(space,
|
|
pa_def_subspaces[i].name,pa_def_subspaces[i].defined,
|
|
pa_def_subspaces[i].loadable,
|
|
pa_def_subspaces[i].code_only,pa_def_subspaces[i].common,
|
|
pa_def_subspaces[i].dup_common,pa_def_subspaces[i].zero,
|
|
pa_def_subspaces[i].sort,pa_def_subspaces[i].access,
|
|
pa_def_subspaces[i].space_index,
|
|
pa_def_subspaces[i].alignment,
|
|
pa_def_subspaces[i].quadrant,
|
|
pa_def_spaces[pa_def_subspaces[i].def_space_index].segment);
|
|
subseg_new(pa_def_subspaces[i].name,pa_def_subspaces[i].subsegment);
|
|
}
|
|
else
|
|
as_fatal("Internal error: space missing for subspace \"%s\"\n",
|
|
pa_def_subspaces[i].name);
|
|
i++;
|
|
}
|
|
}
|
|
|
|
space_dict_chainS *create_new_space(name,spnum,loadable,defined,private,sort,defined_in_file,seg)
|
|
char *name;
|
|
int spnum;
|
|
char loadable;
|
|
char defined;
|
|
char private;
|
|
char sort;
|
|
char defined_in_file;
|
|
asection * seg;
|
|
|
|
{
|
|
Elf_Internal_Shdr *new_space;
|
|
space_dict_chainS *chain_entry;
|
|
|
|
new_space = (Elf_Internal_Shdr *)xmalloc(sizeof(Elf_Internal_Shdr));
|
|
if ( !new_space )
|
|
as_fatal("Out of memory: could not allocate new Elf_Internal_Shdr: %s\n",name);
|
|
|
|
/*
|
|
new_space->space_number = spnum;
|
|
new_space->is_loadable = loadable & 1;
|
|
new_space->is_defined = defined & 1;
|
|
new_space->is_private = private & 1;
|
|
new_space->sort_key = sort & 0xff;
|
|
|
|
new_space->loader_fix_index = ~0;
|
|
new_space->loader_fix_quantity = 0;
|
|
new_space->init_pointer_index = ~0;
|
|
new_space->init_pointer_quantity = 0;
|
|
new_space->subspace_quantity = 0;
|
|
*/
|
|
|
|
chain_entry = (space_dict_chainS *)xmalloc(sizeof(space_dict_chainS));
|
|
if ( !chain_entry )
|
|
as_fatal("Out of memory: could not allocate new space chain entry: %s\n",name);
|
|
|
|
SPACE_NAME(chain_entry) = (char *)xmalloc(strlen(name)+1);
|
|
strcpy(SPACE_NAME(chain_entry),name);
|
|
|
|
chain_entry->sd_entry = new_space;
|
|
chain_entry->sd_defined = defined_in_file;
|
|
chain_entry->sd_seg = seg;
|
|
chain_entry->sd_last_subseg = -1;
|
|
chain_entry->sd_next = NULL;
|
|
|
|
SPACE_SPNUM(chain_entry) = spnum;
|
|
SPACE_LOADABLE(chain_entry) = loadable & 1;
|
|
SPACE_DEFINED(chain_entry) = defined & 1;
|
|
SPACE_PRIVATE(chain_entry) = private & 1;
|
|
SPACE_SORT(chain_entry) = sort & 0xff;
|
|
|
|
/* find spot for the new space based on its sort key */
|
|
|
|
if ( !space_dict_last )
|
|
space_dict_last = chain_entry;
|
|
|
|
if ( space_dict_root == NULL ) /* if root is null, it is very easy */
|
|
space_dict_root = chain_entry;
|
|
else
|
|
{
|
|
space_dict_chainS *sdcP;
|
|
space_dict_chainS *last_sdcP;
|
|
|
|
sdcP = space_dict_root;
|
|
last_sdcP = NULL;
|
|
|
|
while ( sdcP ) {
|
|
if ( SPACE_SORT(sdcP) < SPACE_SORT(chain_entry) ) {
|
|
last_sdcP = sdcP;
|
|
sdcP = sdcP->sd_next;
|
|
}
|
|
else if ( SPACE_SORT(sdcP) == SPACE_SORT(chain_entry) ) {
|
|
last_sdcP = sdcP;
|
|
sdcP = sdcP->sd_next;
|
|
}
|
|
else if ( SPACE_SORT(sdcP) > SPACE_SORT(chain_entry) ) {
|
|
break;
|
|
}
|
|
}
|
|
|
|
if ( last_sdcP ) {
|
|
chain_entry->sd_next = sdcP;
|
|
last_sdcP->sd_next = chain_entry;
|
|
}
|
|
else {
|
|
space_dict_root = chain_entry;
|
|
chain_entry->sd_next = sdcP;
|
|
}
|
|
|
|
if ( chain_entry->sd_next == NULL )
|
|
space_dict_last = chain_entry;
|
|
}
|
|
|
|
return chain_entry;
|
|
}
|
|
|
|
subspace_dict_chainS
|
|
*create_new_subspace(space,name,defined,loadable,code_only,common,dup_common,
|
|
is_zero,sort,access,space_index,alignment,quadrant,seg)
|
|
space_dict_chainS *space;
|
|
char *name;
|
|
char defined,loadable,code_only,common,dup_common,is_zero;
|
|
char sort;
|
|
int access;
|
|
int space_index;
|
|
int alignment;
|
|
int quadrant;
|
|
asection * seg;
|
|
{
|
|
Elf_Internal_Shdr * new_subspace;
|
|
subspace_dict_chainS * chain_entry;
|
|
symbolS * start_symbol;
|
|
|
|
new_subspace = (Elf_Internal_Shdr *)xmalloc(sizeof(Elf_Internal_Shdr));
|
|
if ( !new_subspace )
|
|
as_fatal("Out of memory: could not allocate new Elf_Internal_Shdr: %s\n",
|
|
name);
|
|
|
|
/*
|
|
new_subspace->space_index = space_index;
|
|
new_subspace->fixup_request_index = ~0;
|
|
*/
|
|
|
|
chain_entry = (subspace_dict_chainS *)xmalloc(sizeof(subspace_dict_chainS));
|
|
if ( !chain_entry )
|
|
as_fatal("Out of memory: could not allocate new subspace chain entry: %s\n",name);
|
|
|
|
chain_entry->ssd_entry = new_subspace;
|
|
SUBSPACE_NAME(chain_entry) = (char *)xmalloc(strlen(name)+1);
|
|
strcpy(SUBSPACE_NAME(chain_entry),name);
|
|
|
|
SUBSPACE_ACCESS(chain_entry) = access & 0x7f;
|
|
SUBSPACE_LOADABLE(chain_entry) = loadable & 1;
|
|
SUBSPACE_COMMON(chain_entry) = common & 1;
|
|
SUBSPACE_DUP_COMM(chain_entry) = dup_common & 1;
|
|
SUBSPACE_SORT(chain_entry) = sort & 0xff;
|
|
SET_SUBSPACE_CODE_ONLY(chain_entry,code_only & 1);
|
|
SUBSPACE_ALIGN(chain_entry) = alignment & 0xffff;
|
|
SUBSPACE_QUADRANT(chain_entry) = quadrant & 0x3;
|
|
SUBSPACE_SUBSPACE_START(chain_entry) = pa_subspace_start(space,quadrant);
|
|
|
|
chain_entry->ssd_defined = defined;
|
|
chain_entry->ssd_space_number = space_index;
|
|
chain_entry->ssd_subseg = pa_next_subseg(space);
|
|
chain_entry->ssd_seg = seg;
|
|
SUBSPACE_ZERO(chain_entry) = is_zero;
|
|
chain_entry->ssd_last_align = 1;
|
|
chain_entry->ssd_next = NULL;
|
|
|
|
/* find spot for the new subspace based on its sort key */
|
|
|
|
if ( space->sd_subspaces == NULL ) /* if root is null, it is very easy */
|
|
space->sd_subspaces = chain_entry;
|
|
else
|
|
{
|
|
subspace_dict_chainS *ssdcP;
|
|
subspace_dict_chainS *last_ssdcP;
|
|
|
|
ssdcP = space->sd_subspaces;
|
|
last_ssdcP = NULL;
|
|
|
|
while ( ssdcP ) {
|
|
if ( SUBSPACE_SORT(ssdcP) < SUBSPACE_SORT(chain_entry) ) {
|
|
last_ssdcP = ssdcP;
|
|
ssdcP = ssdcP->ssd_next;
|
|
}
|
|
else if ( SUBSPACE_SORT(ssdcP) == SUBSPACE_SORT(chain_entry) ) {
|
|
last_ssdcP = ssdcP;
|
|
ssdcP = ssdcP->ssd_next;
|
|
}
|
|
else if ( SUBSPACE_SORT(ssdcP) > SUBSPACE_SORT(chain_entry) ) {
|
|
break;
|
|
}
|
|
}
|
|
|
|
if ( last_ssdcP ) {
|
|
chain_entry->ssd_next = ssdcP;
|
|
last_ssdcP->ssd_next = chain_entry;
|
|
}
|
|
else {
|
|
space->sd_subspaces = chain_entry;
|
|
chain_entry->ssd_next = ssdcP;
|
|
}
|
|
}
|
|
|
|
start_symbol = pa_set_start_symbol(seg,space->sd_last_subseg);
|
|
chain_entry->ssd_start_sym = start_symbol;
|
|
return chain_entry;
|
|
|
|
}
|
|
|
|
subspace_dict_chainS
|
|
*update_subspace(name,defined,loadable,code_only,common,dup_common,sort,zero,
|
|
access,space_index,alignment,quadrant,subseg)
|
|
char *name;
|
|
char defined,loadable,code_only,common,dup_common,zero;
|
|
char sort;
|
|
int access;
|
|
int space_index;
|
|
int alignment;
|
|
int quadrant;
|
|
subsegT subseg;
|
|
{
|
|
subspace_dict_chainS *chain_entry;
|
|
subspace_dict_chainS *is_defined_subspace();
|
|
|
|
if ( (chain_entry = is_defined_subspace(name,subseg)) ) {
|
|
|
|
SUBSPACE_ACCESS(chain_entry) = access & 0x7f;
|
|
SUBSPACE_LOADABLE(chain_entry) = loadable & 1;
|
|
SUBSPACE_COMMON(chain_entry) = common & 1;
|
|
SUBSPACE_DUP_COMM(chain_entry) = dup_common & 1;
|
|
SET_SUBSPACE_CODE_ONLY(chain_entry,code_only & 1);
|
|
SUBSPACE_SORT(chain_entry) = sort & 0xff;
|
|
/* chain_entry->ssd_entry->space_index = space_index; */
|
|
SUBSPACE_ALIGN(chain_entry) = alignment & 0xffff;
|
|
SUBSPACE_QUADRANT(chain_entry) = quadrant & 0x3;
|
|
|
|
chain_entry->ssd_defined = defined;
|
|
chain_entry->ssd_space_number = space_index;
|
|
SUBSPACE_ZERO(chain_entry) = zero;
|
|
}
|
|
else
|
|
chain_entry = NULL;
|
|
|
|
return chain_entry;
|
|
|
|
}
|
|
|
|
space_dict_chainS *is_defined_space(name)
|
|
char *name;
|
|
{
|
|
space_dict_chainS *spaceCh;
|
|
|
|
for (spaceCh = space_dict_root;spaceCh;spaceCh=spaceCh->sd_next ) {
|
|
if ( strcmp(SPACE_NAME(spaceCh),name) == 0 ) {
|
|
return spaceCh;
|
|
}
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
|
|
space_dict_chainS *pa_segment_to_space(seg)
|
|
asection * seg;
|
|
{
|
|
space_dict_chainS *spaceCh;
|
|
|
|
for (spaceCh = space_dict_root;spaceCh;spaceCh=spaceCh->sd_next ) {
|
|
if ( spaceCh->sd_seg == seg ) {
|
|
return spaceCh;
|
|
}
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
|
|
subspace_dict_chainS *is_defined_subspace(name,subseg)
|
|
char *name;
|
|
subsegT subseg;
|
|
{
|
|
space_dict_chainS *spaceCh;
|
|
subspace_dict_chainS *subspCh;
|
|
|
|
for ( spaceCh = space_dict_root; spaceCh; spaceCh = spaceCh->sd_next ) {
|
|
for ( subspCh = spaceCh->sd_subspaces; subspCh; subspCh=subspCh->ssd_next ) {
|
|
/*
|
|
if ( strcmp(SUBSPACE_NAME(subspCh),name) == 0 &&
|
|
subspCh->ssd_subseg == subseg ) {
|
|
*/
|
|
if ( strcmp(SUBSPACE_NAME(subspCh),name) == 0 ) {
|
|
return subspCh;
|
|
}
|
|
}
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
subspace_dict_chainS *pa_subsegment_to_subspace(seg,subseg)
|
|
asection * seg;
|
|
subsegT subseg;
|
|
{
|
|
space_dict_chainS *spaceCh;
|
|
subspace_dict_chainS *subspCh;
|
|
|
|
for ( spaceCh = space_dict_root; spaceCh; spaceCh = spaceCh->sd_next ) {
|
|
if ( spaceCh->sd_seg == seg ) {
|
|
for (subspCh = spaceCh->sd_subspaces;subspCh;subspCh=subspCh->ssd_next ) {
|
|
if ( subspCh->ssd_subseg == (int)subseg ) {
|
|
return subspCh;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
|
|
space_dict_chainS *pa_find_space_by_number(number)
|
|
int number;
|
|
{
|
|
space_dict_chainS *spaceCh;
|
|
|
|
for (spaceCh = space_dict_root;spaceCh;spaceCh=spaceCh->sd_next ) {
|
|
if ( SPACE_SPNUM(spaceCh) == number ) {
|
|
return spaceCh;
|
|
}
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
|
|
unsigned int pa_subspace_start(space,quadrant)
|
|
space_dict_chainS *space;
|
|
int quadrant;
|
|
{
|
|
if ( (strcasecmp(SPACE_NAME(space),"$PRIVATE$") == 0) &&
|
|
quadrant == 1 ) {
|
|
return 0x40000000;
|
|
}
|
|
else if ( space->sd_seg == data_section && quadrant == 1 ) { /* in case name is */
|
|
/* already converted */
|
|
/* to a space dict- */
|
|
/* ionary index */
|
|
return 0x40000000;
|
|
}
|
|
else
|
|
return 0;
|
|
}
|
|
|
|
int pa_next_subseg(space)
|
|
space_dict_chainS *space;
|
|
{
|
|
|
|
space->sd_last_subseg++;
|
|
return space->sd_last_subseg;
|
|
}
|
|
|
|
int is_last_defined_subspace(ssd)
|
|
subspace_dict_chainS *ssd;
|
|
{
|
|
|
|
for ( ;ssd; ssd = ssd->ssd_next ) {
|
|
if ( ssd->ssd_defined )
|
|
return FALSE;
|
|
}
|
|
|
|
return TRUE;
|
|
}
|
|
|
|
symbolS *pa_get_start_symbol(seg,subseg)
|
|
asection * seg;
|
|
subsegT subseg;
|
|
{
|
|
symbolS *start_symbol;
|
|
subspace_dict_chainS *ssd;
|
|
|
|
start_symbol = NULL;
|
|
|
|
/* each time a new space is created, build a symbol called LS$START_seg_subseg$ */
|
|
/* where <space-name> is the name of the space */
|
|
/* the start symbol will be SS_LOCAL and ST_CODE */
|
|
|
|
if ( seg == bfd_make_section_old_way ( stdoutput, ".text" ) ||
|
|
seg == bfd_make_section_old_way ( stdoutput, ".data" ) ||
|
|
seg == bfd_make_section_old_way ( stdoutput, GDB_DEBUG_SPACE_NAME ) ) {
|
|
ssd = pa_subsegment_to_subspace(seg,subseg);
|
|
if ( ssd ) {
|
|
start_symbol = ssd->ssd_start_sym;
|
|
}
|
|
else
|
|
as_fatal("Internal error: missing subspace for (seg,subseg)=('%s',%d)",
|
|
seg->name,subseg);
|
|
}
|
|
else
|
|
as_fatal("Internal error: attempt to find start symbol for unloadable segment: '%s'",
|
|
seg->name);
|
|
|
|
return start_symbol;
|
|
}
|
|
|
|
/*
|
|
Function to define a symbol whose address is the beginning of a subspace.
|
|
This function assumes the symbol is to be defined for the current subspace.
|
|
*/
|
|
|
|
symbolS *pa_set_start_symbol(seg,subseg)
|
|
asection * seg;
|
|
subsegT subseg;
|
|
{
|
|
symbolS *start_symbol;
|
|
subspace_dict_chainS *ssd;
|
|
char *symbol_name;
|
|
|
|
symbol_name = (char *)xmalloc(strlen("LS$START__000000$")+strlen(seg->name)+1);
|
|
|
|
sprintf(symbol_name,"LS$START_%s_%03d$",seg->name,subseg);
|
|
|
|
start_symbol
|
|
= symbol_new(symbol_name,seg,0,frag_now); /* XXX: not sure if value s.b. 0 or frag s.b. NULL */
|
|
|
|
start_symbol->bsym->flags = BSF_LOCAL; /* XXX: isn't there a macro defined for this? */
|
|
|
|
/* each time a new space is created, build a symbol called LS$START_seg_subseg$ */
|
|
/* where <space-name> is the name of the space */
|
|
/* the start symbol will be SS_LOCAL and ST_CODE */
|
|
/* This function assumes that (seg,subseg) is a new subsegment(subspace) */
|
|
|
|
if ( seg == bfd_make_section_old_way ( stdoutput, ".text" ) ||
|
|
seg == bfd_make_section_old_way ( stdoutput, ".data" ) ||
|
|
seg == bfd_make_section_old_way ( stdoutput, GDB_DEBUG_SPACE_NAME ) ) {
|
|
ssd = pa_subsegment_to_subspace(seg,subseg);
|
|
if ( ssd ) {
|
|
ssd->ssd_start_sym = start_symbol;
|
|
}
|
|
else
|
|
as_fatal("Internal error: missing subspace for (seg,subseg)=('%s',%d)",
|
|
seg,subseg);
|
|
}
|
|
else
|
|
as_fatal("Internal error: attempt to define start symbol for unloadable segment: '%s'",
|
|
seg->name);
|
|
|
|
return start_symbol;
|
|
}
|
|
#endif
|
|
|
|
int
|
|
obj_elf_frob_symbol (sym, punt)
|
|
symbolS *sym;
|
|
int *punt;
|
|
{
|
|
|
|
/* If this is a local symbol, are there any relocations for */
|
|
/* which need this symbol? */
|
|
|
|
/* To find this out, we examine all relocations in all bfd */
|
|
/* sections that have relocations. If there is one that */
|
|
/* references this symbol, we need to keep this symbol. In */
|
|
/* this case, we return a true status. In all other cases, we */
|
|
/* return a false status. */
|
|
|
|
if ( S_IS_LOCAL(sym) ) {
|
|
asymbol *bsym = sym->bsym;
|
|
bfd *abfd = bsym->the_bfd;
|
|
asection *bsec;
|
|
|
|
for ( bsec = abfd->sections; bsec; bsec = bsec->next ) {
|
|
struct reloc_cache_entry **rlocs = bsec->orelocation;
|
|
int rcnt = bsec->reloc_count;
|
|
|
|
if ( rlocs ) {
|
|
int i;
|
|
|
|
for ( i = 0; i < rcnt; i++ ) {
|
|
if ( rlocs[i]->sym_ptr_ptr
|
|
&& rlocs[i]->sym_ptr_ptr[0] == bsym )
|
|
return 1;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static void obj_elf_line() {
|
|
/* Assume delimiter is part of expression.
|
|
BSD4.2 as fails with delightful bug, so we
|
|
are not being incompatible here. */
|
|
new_logical_line((char *)NULL, (int)(get_absolute_expression()));
|
|
demand_empty_rest_of_line();
|
|
} /* obj_elf_line() */
|
|
|
|
/*
|
|
* stab()
|
|
*
|
|
* Handle .stabX directives, which used to be open-coded.
|
|
* So much creeping featurism overloaded the semantics that we decided
|
|
* to put all .stabX thinking in one place. Here.
|
|
*
|
|
* We try to make any .stabX directive legal. Other people's AS will often
|
|
* do assembly-time consistency checks: eg assigning meaning to n_type bits
|
|
* and "protecting" you from setting them to certain values. (They also zero
|
|
* certain bits before emitting symbols. Tut tut.)
|
|
*
|
|
* If an expression is not absolute we either gripe or use the relocation
|
|
* information. Other people's assemblers silently forget information they
|
|
* don't need and invent information they need that you didn't supply.
|
|
*
|
|
* .stabX directives always make a symbol table entry. It may be junk if
|
|
* the rest of your .stabX directive is malformed.
|
|
*/
|
|
|
|
/*
|
|
* pa_stab_symbol_string()
|
|
*
|
|
* Build a string dictionary entry for a .stabX symbol.
|
|
* The symbol is added to the .stabstr section.
|
|
*
|
|
*/
|
|
|
|
static unsigned int gdb_string_index = 0;
|
|
|
|
static unsigned int
|
|
elf_stab_symbol_string(string)
|
|
char *string;
|
|
{
|
|
asection *save_seg;
|
|
asection *seg;
|
|
subsegT save_subseg;
|
|
unsigned int length;
|
|
unsigned int old_gdb_string_index;
|
|
char *clengthP;
|
|
int i;
|
|
char c;
|
|
|
|
old_gdb_string_index = 0;
|
|
length = strlen(string);
|
|
clengthP = (char *)&length;
|
|
if ( length > 0 ) { /* Ordinary case. */
|
|
save_seg = now_seg;
|
|
save_subseg = now_subseg;
|
|
|
|
/* Create the stab sections, if they are not already created. */
|
|
seg = bfd_get_section_by_name(stdoutput,".stabstr");
|
|
if ( seg == 0 ) {
|
|
seg = bfd_make_section_old_way(stdoutput,".stabstr");
|
|
bfd_set_section_flags (stdoutput,
|
|
seg,
|
|
SEC_READONLY | SEC_ALLOC | SEC_LOAD );
|
|
}
|
|
subseg_new((char *) seg->name,save_subseg);
|
|
old_gdb_string_index = gdb_string_index;
|
|
i = 0;
|
|
while ( (c = *string++) )
|
|
{
|
|
i++;
|
|
gdb_string_index++;
|
|
FRAG_APPEND_1_CHAR( c );
|
|
}
|
|
{
|
|
FRAG_APPEND_1_CHAR( (char)0 );
|
|
i++;
|
|
gdb_string_index++;
|
|
}
|
|
while ( i%4 != 0 ) {
|
|
FRAG_APPEND_1_CHAR( (char)0 );
|
|
i++;
|
|
gdb_string_index++;
|
|
}
|
|
subseg_new((char *) save_seg->name,save_subseg);
|
|
}
|
|
|
|
return old_gdb_string_index;
|
|
}
|
|
|
|
unsigned int
|
|
elf_stab_symbol(symbolP,stab_type)
|
|
symbolS *symbolP;
|
|
int stab_type;
|
|
{
|
|
unsigned int length;
|
|
int i;
|
|
char c;
|
|
char *toP;
|
|
|
|
/* the string index portion of the stab */
|
|
|
|
toP = frag_more( sizeof(long) +
|
|
sizeof(S_GET_TYPE(symbolP)) +
|
|
sizeof(S_GET_OTHER(symbolP)) +
|
|
sizeof(S_GET_DESC(symbolP)));
|
|
md_number_to_chars(toP,symbolP->sy_name_offset,sizeof(long));
|
|
md_number_to_chars(toP,
|
|
S_GET_TYPE(symbolP),
|
|
sizeof(S_GET_TYPE(symbolP)));
|
|
md_number_to_chars(toP + sizeof(S_GET_TYPE(symbolP)),
|
|
S_GET_OTHER(symbolP),
|
|
sizeof(S_GET_OTHER(symbolP)));
|
|
md_number_to_chars(toP + sizeof(S_GET_TYPE(symbolP))
|
|
+ sizeof(S_GET_OTHER(symbolP)),
|
|
S_GET_DESC(symbolP),
|
|
sizeof(S_GET_DESC(symbolP)));
|
|
|
|
/* n_value has to be relocated */
|
|
|
|
/* Don't bother relocating if we're only adding in a constant. */
|
|
|
|
if ((stab_type == 's' || stab_type == 'n') && symbolP->sy_forward == 0)
|
|
S_SET_VALUE(symbolP,S_GET_VALUE(symbolP->sy_forward));
|
|
|
|
toP = frag_more(4);
|
|
md_number_to_chars(toP, S_GET_VALUE(symbolP),
|
|
sizeof(S_GET_VALUE(symbolP)));
|
|
|
|
#if 0
|
|
if ( (stab_type == 's' || stab_type == 'n') && symbolP->sy_forward )
|
|
{
|
|
i = S_GET_TYPE(symbolP) & N_TYPE;
|
|
fix_new_hppa(frag_now, /* which frag */
|
|
toP-frag_now->fr_literal, /* where */
|
|
4, /* size */
|
|
symbolP->sy_forward, /* addr of symbol for this stab */
|
|
(asymbol *)NULL,
|
|
0,
|
|
i == N_UNDF || i == N_ABS, /* 1 if internal relocation */
|
|
R_HPPA, /* reloc type */
|
|
e_fsel, /* fixup field = F% */
|
|
32,
|
|
0, /* arg_reloc descriptor */
|
|
(char *)0
|
|
);
|
|
}
|
|
else if ( stab_type == 'd' )
|
|
{
|
|
fix_new_hppa (frag_now, /* which frag */
|
|
toP-frag_now->fr_literal, /* where */
|
|
4, /* size */
|
|
symbolP, /* addr of symbol for this stab */
|
|
(asymbol *)NULL,
|
|
0,
|
|
0,
|
|
R_HPPA, /* reloc type */
|
|
e_fsel, /* fixup field = F% */
|
|
32,
|
|
0, /* arg_reloc descriptor */
|
|
(char *)0
|
|
);
|
|
}
|
|
#else
|
|
/* What needs to replace the above code? */
|
|
abort ();
|
|
#endif
|
|
}
|
|
|
|
static void obj_elf_stab(what)
|
|
int what;
|
|
{
|
|
extern int listing;
|
|
|
|
symbolS * symbolP = 0;
|
|
char * string;
|
|
int saved_type = 0;
|
|
int length;
|
|
int goof = 0;
|
|
long longint;
|
|
asection *saved_seg = now_seg;
|
|
asection *seg;
|
|
subsegT subseg = now_subseg;
|
|
|
|
seg = bfd_get_section_by_name(stdoutput,".stab");
|
|
if ( seg == 0 )
|
|
{
|
|
seg = bfd_make_section_old_way(stdoutput,".stab");
|
|
bfd_set_section_flags (stdoutput,
|
|
seg,
|
|
SEC_READONLY | SEC_ALLOC | SEC_LOAD | SEC_RELOC);
|
|
}
|
|
|
|
/*
|
|
* Enter with input_line_pointer pointing past .stabX and any following
|
|
* whitespace.
|
|
*/
|
|
if (what == 's') {
|
|
string = demand_copy_C_string(& length);
|
|
SKIP_WHITESPACE();
|
|
if (* input_line_pointer == ',')
|
|
input_line_pointer ++;
|
|
else
|
|
{
|
|
as_bad("I need a comma after symbol's name");
|
|
goof = 1;
|
|
}
|
|
}
|
|
else
|
|
string = "";
|
|
|
|
/*
|
|
* Input_line_pointer->after ','. String->symbol name.
|
|
*/
|
|
if (! goof)
|
|
{
|
|
symbolP = symbol_new(string, &bfd_und_section, 0, (struct frag *)0);
|
|
|
|
/* enter the string in the .stab string table (section .stabstr) */
|
|
symbolP->sy_name_offset = elf_stab_symbol_string(string);
|
|
|
|
switch (what)
|
|
{
|
|
case 'd':
|
|
S_SET_NAME(symbolP, NULL); /* .stabd feature. */
|
|
S_SET_VALUE(symbolP, obstack_next_free(&frags) - frag_now->fr_literal);
|
|
S_SET_SEGMENT(symbolP, now_seg);
|
|
symbolP->sy_frag = frag_now;
|
|
break;
|
|
|
|
case 'n':
|
|
symbolP->sy_frag = &zero_address_frag;
|
|
break;
|
|
|
|
case 's':
|
|
symbolP->sy_frag = & zero_address_frag;
|
|
break;
|
|
|
|
default:
|
|
BAD_CASE(what);
|
|
break;
|
|
}
|
|
|
|
if (get_absolute_expression_and_terminator(&longint) == ',')
|
|
{
|
|
saved_type = longint;
|
|
S_SET_TYPE (symbolP, saved_type);
|
|
}
|
|
else
|
|
{
|
|
as_bad("I want a comma after the n_type expression");
|
|
goof = 1;
|
|
input_line_pointer --; /* Backup over a non-',' char. */
|
|
}
|
|
}
|
|
|
|
if (!goof)
|
|
{
|
|
if (get_absolute_expression_and_terminator(&longint) == ',')
|
|
S_SET_OTHER(symbolP, longint);
|
|
else
|
|
{
|
|
as_bad("I want a comma after the n_other expression");
|
|
goof = 1;
|
|
input_line_pointer--; /* Backup over a non-',' char. */
|
|
}
|
|
}
|
|
|
|
if (!goof)
|
|
{
|
|
S_SET_DESC(symbolP, get_absolute_expression());
|
|
if (what == 's' || what == 'n')
|
|
{
|
|
if (*input_line_pointer != ',')
|
|
{
|
|
as_bad("I want a comma after the n_desc expression");
|
|
goof = 1;
|
|
}
|
|
else
|
|
{
|
|
input_line_pointer++;
|
|
}
|
|
}
|
|
}
|
|
|
|
if ( !goof ) {
|
|
if ( what=='s' || what=='n' ) {
|
|
pseudo_set(symbolP);
|
|
S_SET_TYPE (symbolP, saved_type);
|
|
}
|
|
/* Emit the stab symbol. */
|
|
subseg_new((char *) seg->name,subseg);
|
|
elf_stab_symbol(symbolP, what);
|
|
subseg_new((char *) saved_seg->name,subseg);
|
|
|
|
if ( what=='s' || what=='n' && symbolP->sy_forward == NULL ) {
|
|
/* symbol is not needed in the regular symbol table */
|
|
symbol_remove(symbolP,&symbol_rootP,&symbol_lastP);
|
|
}
|
|
|
|
}
|
|
|
|
#ifndef NO_LISTING
|
|
if (listing && !goof)
|
|
switch (S_GET_TYPE (symbolP))
|
|
{
|
|
case N_SLINE:
|
|
listing_source_line (S_GET_DESC (symbolP));
|
|
break;
|
|
case N_SO:
|
|
case N_SOL:
|
|
listing_source_file (string);
|
|
break;
|
|
}
|
|
#endif
|
|
|
|
if (goof)
|
|
ignore_rest_of_line();
|
|
else
|
|
demand_empty_rest_of_line ();
|
|
} /* obj_elf_stab() */
|
|
|
|
static void obj_elf_desc() {
|
|
char *name;
|
|
char c;
|
|
char *p;
|
|
symbolS *symbolP;
|
|
int temp;
|
|
|
|
/*
|
|
* Frob invented at RMS' request. Set the n_desc of a symbol.
|
|
*/
|
|
name = input_line_pointer;
|
|
c = get_symbol_end();
|
|
p = input_line_pointer;
|
|
* p = c;
|
|
SKIP_WHITESPACE();
|
|
if (*input_line_pointer != ',') {
|
|
*p = 0;
|
|
as_bad("Expected comma after name \"%s\"", name);
|
|
*p = c;
|
|
ignore_rest_of_line();
|
|
} else {
|
|
input_line_pointer ++;
|
|
temp = get_absolute_expression();
|
|
*p = 0;
|
|
symbolP = symbol_find_or_make(name);
|
|
*p = c;
|
|
S_SET_DESC(symbolP,temp);
|
|
}
|
|
demand_empty_rest_of_line();
|
|
} /* obj_elf_desc() */
|
|
|
|
void obj_read_begin_hook()
|
|
{
|
|
}
|
|
|
|
void obj_symbol_new_hook(symbolP)
|
|
symbolS *symbolP;
|
|
{
|
|
elf_symbol_type *esym = (elf_symbol_type *)symbolP;
|
|
|
|
/* There is an Elf_Internal_Sym and an Elf_External_Sym. For now,
|
|
just zero them out. */
|
|
|
|
bzero((char *) &esym->internal_elf_sym,sizeof(Elf_Internal_Sym));
|
|
bzero((char *) &esym->native_elf_sym,sizeof(Elf_External_Sym));
|
|
}
|
|
|
|
static void obj_elf_version()
|
|
{
|
|
char *name;
|
|
unsigned int c;
|
|
char ch;
|
|
char *p;
|
|
int temp;
|
|
symbolS * symbolP;
|
|
asection *seg = now_seg;
|
|
subsegT subseg = now_subseg;
|
|
Elf_Internal_Note i_note;
|
|
Elf_External_Note e_note;
|
|
asection *note_secp = (asection *)NULL;
|
|
int i, len;
|
|
|
|
SKIP_WHITESPACE();
|
|
if (* input_line_pointer == '\"') {
|
|
++input_line_pointer; /* -> 1st char of string. */
|
|
name = input_line_pointer;
|
|
|
|
while ( is_a_char(c = next_char_of_string()) )
|
|
;
|
|
c = *input_line_pointer;
|
|
*input_line_pointer = '\0';
|
|
*(input_line_pointer-1) = '\0';
|
|
*input_line_pointer = c;
|
|
|
|
/* create the .note section if this is the first version string */
|
|
|
|
note_secp = bfd_get_section_by_name(stdoutput,".note");
|
|
if ( note_secp == (asection *)NULL ) {
|
|
note_secp = bfd_make_section_old_way(stdoutput,".note");
|
|
bfd_set_section_flags(stdoutput,
|
|
note_secp,
|
|
SEC_LOAD | SEC_ALLOC | SEC_HAS_CONTENTS);
|
|
}
|
|
|
|
/* process the version string */
|
|
|
|
subseg_new((char *)note_secp->name, 0);
|
|
len = strlen(name);
|
|
|
|
i_note.namesz = ((len + 1) + 3) & ~3; /* round this to word boundary */
|
|
i_note.descsz = 0; /* no description */
|
|
i_note.type = NT_VERSION;
|
|
p = frag_more(sizeof(e_note.namesz));
|
|
md_number_to_chars(p, i_note.namesz, 4);
|
|
p = frag_more(sizeof(e_note.descsz));
|
|
md_number_to_chars(p, i_note.descsz, 4);
|
|
p = frag_more(sizeof(e_note.type));
|
|
md_number_to_chars(p, i_note.type, 4);
|
|
|
|
for ( i = 0; i < len; i++ ) {
|
|
ch = *(name + i);
|
|
{
|
|
FRAG_APPEND_1_CHAR( ch );
|
|
}
|
|
}
|
|
frag_align(2,0);
|
|
|
|
subseg_new((char *)seg->name,subseg);
|
|
}
|
|
else {
|
|
as_bad( "Expected \"-ed string" );
|
|
}
|
|
demand_empty_rest_of_line();
|
|
}
|
|
|
|
static void
|
|
obj_elf_size ()
|
|
{
|
|
char *name = input_line_pointer;
|
|
char c = get_symbol_end ();
|
|
char *p;
|
|
expressionS exp;
|
|
segT sec;
|
|
symbolS *sym;
|
|
|
|
p = input_line_pointer;
|
|
*p = c;
|
|
SKIP_WHITESPACE ();
|
|
if (*input_line_pointer != ',')
|
|
{
|
|
*p = 0;
|
|
as_bad ("expected comma after name `%s' in .size directive", name);
|
|
*p = c;
|
|
ignore_rest_of_line ();
|
|
return;
|
|
}
|
|
input_line_pointer++;
|
|
sec = expression (&exp);
|
|
if (sec == absent_section)
|
|
{
|
|
as_bad ("missing expression in .size directive");
|
|
exp.X_seg = absolute_section;
|
|
exp.X_add_number = 0;
|
|
}
|
|
*p = 0;
|
|
sym = symbol_find_or_make (name);
|
|
*p = c;
|
|
if (sec == absolute_section)
|
|
S_SET_SIZE (sym, exp.X_add_number);
|
|
else
|
|
as_tsktsk (".size not yet supported, ignored");
|
|
demand_empty_rest_of_line ();
|
|
}
|
|
|
|
static void
|
|
obj_elf_type ()
|
|
{
|
|
char *name = input_line_pointer;
|
|
char c = get_symbol_end ();
|
|
char *p;
|
|
int type;
|
|
symbolS *sym;
|
|
|
|
p = input_line_pointer;
|
|
*p = c;
|
|
SKIP_WHITESPACE ();
|
|
if (*input_line_pointer != ',')
|
|
{
|
|
as_bad ("expected comma after name in .type directive");
|
|
egress:
|
|
ignore_rest_of_line ();
|
|
return;
|
|
}
|
|
input_line_pointer++;
|
|
SKIP_WHITESPACE ();
|
|
if (*input_line_pointer != '#')
|
|
{
|
|
as_bad ("expected `#' after comma in .type directive");
|
|
goto egress;
|
|
}
|
|
input_line_pointer++;
|
|
if (!strncmp ("function", input_line_pointer, sizeof ("function") - 1))
|
|
{
|
|
type = BSF_FUNCTION;
|
|
input_line_pointer += sizeof ("function") - 1;
|
|
}
|
|
else
|
|
{
|
|
as_bad ("unrecognized symbol type, ignored");
|
|
goto egress;
|
|
}
|
|
demand_empty_rest_of_line ();
|
|
*p = 0;
|
|
sym = symbol_find_or_make (name);
|
|
sym->bsym->flags = type;
|
|
}
|
|
|
|
static void
|
|
obj_elf_ident ()
|
|
{
|
|
int xxx;
|
|
char * string;
|
|
|
|
string = demand_copy_C_string (&xxx);
|
|
as_tsktsk (".ident not supported, ignoring");
|
|
}
|