mirror of
https://sourceware.org/git/binutils-gdb.git
synced 2024-11-28 04:25:10 +08:00
6b8505468e
GDB computes structure byte offsets using a 32 bit integer. And, first it computes the offset in bits and then converts to bytes. The result is that any offset that if 512K bytes or larger overflows. This patch changes GDB to use LONGEST for such calculations. PR gdb/17520 Structure offset wrong when 1/4 GB or greater. * c-lang.h: Change all parameters, variables, and struct or union members used as struct or union fie3ld offsets from int to LONGEST. * c-valprint.c: Likewise. * cp-abi.c: Likewise. * cp-abi.h: Likewise. * cp-valprint.c: Likewise. * d-valprint.c: Likewise. * dwarf2loc.c: Likewise. * eval.c: Likewise. * extension-priv.h: Likewise. * extension.c: Likewise. * extension.h: Likewise. * findvar.c: Likewise. * gdbtypes.h: Likewise. * gnu-v2-abi.c: Likewise. * gnu-v3-abi.c: Likewise. * go-valprint.c: Likewise. * guile/guile-internal.h: Likewise. * guile/scm-pretty-print.c: Likewise. * jv-valprint.c Likewise. * opencl-lang.c: Likewise. * p-lang.h: Likewise. * python/py-prettyprint.c: Likewise. * python/python-internal.h: Likewise. * spu-tdep.c: Likewise. * typeprint.c: Likewise. * valarith.c: Likewise. * valops.c: Likewise. * valprint.c: Likewise. * valprint.h: Likewise. * value.c: Likewise. * value.h: Likewise. * p-valprint.c: Likewise. * c-typeprint.c (c_type_print_base): When printing offset, use plongest, not %d. * gdbtypes.c (recursive_dump_type): Ditto.
5223 lines
151 KiB
C
5223 lines
151 KiB
C
/* Support routines for manipulating internal types for GDB.
|
||
|
||
Copyright (C) 1992-2016 Free Software Foundation, Inc.
|
||
|
||
Contributed by Cygnus Support, using pieces from other GDB modules.
|
||
|
||
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 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, see <http://www.gnu.org/licenses/>. */
|
||
|
||
#include "defs.h"
|
||
#include "bfd.h"
|
||
#include "symtab.h"
|
||
#include "symfile.h"
|
||
#include "objfiles.h"
|
||
#include "gdbtypes.h"
|
||
#include "expression.h"
|
||
#include "language.h"
|
||
#include "target.h"
|
||
#include "value.h"
|
||
#include "demangle.h"
|
||
#include "complaints.h"
|
||
#include "gdbcmd.h"
|
||
#include "cp-abi.h"
|
||
#include "hashtab.h"
|
||
#include "cp-support.h"
|
||
#include "bcache.h"
|
||
#include "dwarf2loc.h"
|
||
#include "gdbcore.h"
|
||
|
||
/* Initialize BADNESS constants. */
|
||
|
||
const struct rank LENGTH_MISMATCH_BADNESS = {100,0};
|
||
|
||
const struct rank TOO_FEW_PARAMS_BADNESS = {100,0};
|
||
const struct rank INCOMPATIBLE_TYPE_BADNESS = {100,0};
|
||
|
||
const struct rank EXACT_MATCH_BADNESS = {0,0};
|
||
|
||
const struct rank INTEGER_PROMOTION_BADNESS = {1,0};
|
||
const struct rank FLOAT_PROMOTION_BADNESS = {1,0};
|
||
const struct rank BASE_PTR_CONVERSION_BADNESS = {1,0};
|
||
const struct rank INTEGER_CONVERSION_BADNESS = {2,0};
|
||
const struct rank FLOAT_CONVERSION_BADNESS = {2,0};
|
||
const struct rank INT_FLOAT_CONVERSION_BADNESS = {2,0};
|
||
const struct rank VOID_PTR_CONVERSION_BADNESS = {2,0};
|
||
const struct rank BOOL_CONVERSION_BADNESS = {3,0};
|
||
const struct rank BASE_CONVERSION_BADNESS = {2,0};
|
||
const struct rank REFERENCE_CONVERSION_BADNESS = {2,0};
|
||
const struct rank NULL_POINTER_CONVERSION_BADNESS = {2,0};
|
||
const struct rank NS_POINTER_CONVERSION_BADNESS = {10,0};
|
||
const struct rank NS_INTEGER_POINTER_CONVERSION_BADNESS = {3,0};
|
||
|
||
/* Floatformat pairs. */
|
||
const struct floatformat *floatformats_ieee_half[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ieee_half_big,
|
||
&floatformat_ieee_half_little
|
||
};
|
||
const struct floatformat *floatformats_ieee_single[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ieee_single_big,
|
||
&floatformat_ieee_single_little
|
||
};
|
||
const struct floatformat *floatformats_ieee_double[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ieee_double_big,
|
||
&floatformat_ieee_double_little
|
||
};
|
||
const struct floatformat *floatformats_ieee_double_littlebyte_bigword[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ieee_double_big,
|
||
&floatformat_ieee_double_littlebyte_bigword
|
||
};
|
||
const struct floatformat *floatformats_i387_ext[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_i387_ext,
|
||
&floatformat_i387_ext
|
||
};
|
||
const struct floatformat *floatformats_m68881_ext[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_m68881_ext,
|
||
&floatformat_m68881_ext
|
||
};
|
||
const struct floatformat *floatformats_arm_ext[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_arm_ext_big,
|
||
&floatformat_arm_ext_littlebyte_bigword
|
||
};
|
||
const struct floatformat *floatformats_ia64_spill[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ia64_spill_big,
|
||
&floatformat_ia64_spill_little
|
||
};
|
||
const struct floatformat *floatformats_ia64_quad[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ia64_quad_big,
|
||
&floatformat_ia64_quad_little
|
||
};
|
||
const struct floatformat *floatformats_vax_f[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_vax_f,
|
||
&floatformat_vax_f
|
||
};
|
||
const struct floatformat *floatformats_vax_d[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_vax_d,
|
||
&floatformat_vax_d
|
||
};
|
||
const struct floatformat *floatformats_ibm_long_double[BFD_ENDIAN_UNKNOWN] = {
|
||
&floatformat_ibm_long_double_big,
|
||
&floatformat_ibm_long_double_little
|
||
};
|
||
|
||
/* Should opaque types be resolved? */
|
||
|
||
static int opaque_type_resolution = 1;
|
||
|
||
/* A flag to enable printing of debugging information of C++
|
||
overloading. */
|
||
|
||
unsigned int overload_debug = 0;
|
||
|
||
/* A flag to enable strict type checking. */
|
||
|
||
static int strict_type_checking = 1;
|
||
|
||
/* A function to show whether opaque types are resolved. */
|
||
|
||
static void
|
||
show_opaque_type_resolution (struct ui_file *file, int from_tty,
|
||
struct cmd_list_element *c,
|
||
const char *value)
|
||
{
|
||
fprintf_filtered (file, _("Resolution of opaque struct/class/union types "
|
||
"(if set before loading symbols) is %s.\n"),
|
||
value);
|
||
}
|
||
|
||
/* A function to show whether C++ overload debugging is enabled. */
|
||
|
||
static void
|
||
show_overload_debug (struct ui_file *file, int from_tty,
|
||
struct cmd_list_element *c, const char *value)
|
||
{
|
||
fprintf_filtered (file, _("Debugging of C++ overloading is %s.\n"),
|
||
value);
|
||
}
|
||
|
||
/* A function to show the status of strict type checking. */
|
||
|
||
static void
|
||
show_strict_type_checking (struct ui_file *file, int from_tty,
|
||
struct cmd_list_element *c, const char *value)
|
||
{
|
||
fprintf_filtered (file, _("Strict type checking is %s.\n"), value);
|
||
}
|
||
|
||
|
||
/* Allocate a new OBJFILE-associated type structure and fill it
|
||
with some defaults. Space for the type structure is allocated
|
||
on the objfile's objfile_obstack. */
|
||
|
||
struct type *
|
||
alloc_type (struct objfile *objfile)
|
||
{
|
||
struct type *type;
|
||
|
||
gdb_assert (objfile != NULL);
|
||
|
||
/* Alloc the structure and start off with all fields zeroed. */
|
||
type = OBSTACK_ZALLOC (&objfile->objfile_obstack, struct type);
|
||
TYPE_MAIN_TYPE (type) = OBSTACK_ZALLOC (&objfile->objfile_obstack,
|
||
struct main_type);
|
||
OBJSTAT (objfile, n_types++);
|
||
|
||
TYPE_OBJFILE_OWNED (type) = 1;
|
||
TYPE_OWNER (type).objfile = objfile;
|
||
|
||
/* Initialize the fields that might not be zero. */
|
||
|
||
TYPE_CODE (type) = TYPE_CODE_UNDEF;
|
||
TYPE_CHAIN (type) = type; /* Chain back to itself. */
|
||
|
||
return type;
|
||
}
|
||
|
||
/* Allocate a new GDBARCH-associated type structure and fill it
|
||
with some defaults. Space for the type structure is allocated
|
||
on the obstack associated with GDBARCH. */
|
||
|
||
struct type *
|
||
alloc_type_arch (struct gdbarch *gdbarch)
|
||
{
|
||
struct type *type;
|
||
|
||
gdb_assert (gdbarch != NULL);
|
||
|
||
/* Alloc the structure and start off with all fields zeroed. */
|
||
|
||
type = GDBARCH_OBSTACK_ZALLOC (gdbarch, struct type);
|
||
TYPE_MAIN_TYPE (type) = GDBARCH_OBSTACK_ZALLOC (gdbarch, struct main_type);
|
||
|
||
TYPE_OBJFILE_OWNED (type) = 0;
|
||
TYPE_OWNER (type).gdbarch = gdbarch;
|
||
|
||
/* Initialize the fields that might not be zero. */
|
||
|
||
TYPE_CODE (type) = TYPE_CODE_UNDEF;
|
||
TYPE_CHAIN (type) = type; /* Chain back to itself. */
|
||
|
||
return type;
|
||
}
|
||
|
||
/* If TYPE is objfile-associated, allocate a new type structure
|
||
associated with the same objfile. If TYPE is gdbarch-associated,
|
||
allocate a new type structure associated with the same gdbarch. */
|
||
|
||
struct type *
|
||
alloc_type_copy (const struct type *type)
|
||
{
|
||
if (TYPE_OBJFILE_OWNED (type))
|
||
return alloc_type (TYPE_OWNER (type).objfile);
|
||
else
|
||
return alloc_type_arch (TYPE_OWNER (type).gdbarch);
|
||
}
|
||
|
||
/* If TYPE is gdbarch-associated, return that architecture.
|
||
If TYPE is objfile-associated, return that objfile's architecture. */
|
||
|
||
struct gdbarch *
|
||
get_type_arch (const struct type *type)
|
||
{
|
||
if (TYPE_OBJFILE_OWNED (type))
|
||
return get_objfile_arch (TYPE_OWNER (type).objfile);
|
||
else
|
||
return TYPE_OWNER (type).gdbarch;
|
||
}
|
||
|
||
/* See gdbtypes.h. */
|
||
|
||
struct type *
|
||
get_target_type (struct type *type)
|
||
{
|
||
if (type != NULL)
|
||
{
|
||
type = TYPE_TARGET_TYPE (type);
|
||
if (type != NULL)
|
||
type = check_typedef (type);
|
||
}
|
||
|
||
return type;
|
||
}
|
||
|
||
/* See gdbtypes.h. */
|
||
|
||
unsigned int
|
||
type_length_units (struct type *type)
|
||
{
|
||
struct gdbarch *arch = get_type_arch (type);
|
||
int unit_size = gdbarch_addressable_memory_unit_size (arch);
|
||
|
||
return TYPE_LENGTH (type) / unit_size;
|
||
}
|
||
|
||
/* Alloc a new type instance structure, fill it with some defaults,
|
||
and point it at OLDTYPE. Allocate the new type instance from the
|
||
same place as OLDTYPE. */
|
||
|
||
static struct type *
|
||
alloc_type_instance (struct type *oldtype)
|
||
{
|
||
struct type *type;
|
||
|
||
/* Allocate the structure. */
|
||
|
||
if (! TYPE_OBJFILE_OWNED (oldtype))
|
||
type = XCNEW (struct type);
|
||
else
|
||
type = OBSTACK_ZALLOC (&TYPE_OBJFILE (oldtype)->objfile_obstack,
|
||
struct type);
|
||
|
||
TYPE_MAIN_TYPE (type) = TYPE_MAIN_TYPE (oldtype);
|
||
|
||
TYPE_CHAIN (type) = type; /* Chain back to itself for now. */
|
||
|
||
return type;
|
||
}
|
||
|
||
/* Clear all remnants of the previous type at TYPE, in preparation for
|
||
replacing it with something else. Preserve owner information. */
|
||
|
||
static void
|
||
smash_type (struct type *type)
|
||
{
|
||
int objfile_owned = TYPE_OBJFILE_OWNED (type);
|
||
union type_owner owner = TYPE_OWNER (type);
|
||
|
||
memset (TYPE_MAIN_TYPE (type), 0, sizeof (struct main_type));
|
||
|
||
/* Restore owner information. */
|
||
TYPE_OBJFILE_OWNED (type) = objfile_owned;
|
||
TYPE_OWNER (type) = owner;
|
||
|
||
/* For now, delete the rings. */
|
||
TYPE_CHAIN (type) = type;
|
||
|
||
/* For now, leave the pointer/reference types alone. */
|
||
}
|
||
|
||
/* Lookup a pointer to a type TYPE. TYPEPTR, if nonzero, points
|
||
to a pointer to memory where the pointer type should be stored.
|
||
If *TYPEPTR is zero, update it to point to the pointer type we return.
|
||
We allocate new memory if needed. */
|
||
|
||
struct type *
|
||
make_pointer_type (struct type *type, struct type **typeptr)
|
||
{
|
||
struct type *ntype; /* New type */
|
||
struct type *chain;
|
||
|
||
ntype = TYPE_POINTER_TYPE (type);
|
||
|
||
if (ntype)
|
||
{
|
||
if (typeptr == 0)
|
||
return ntype; /* Don't care about alloc,
|
||
and have new type. */
|
||
else if (*typeptr == 0)
|
||
{
|
||
*typeptr = ntype; /* Tracking alloc, and have new type. */
|
||
return ntype;
|
||
}
|
||
}
|
||
|
||
if (typeptr == 0 || *typeptr == 0) /* We'll need to allocate one. */
|
||
{
|
||
ntype = alloc_type_copy (type);
|
||
if (typeptr)
|
||
*typeptr = ntype;
|
||
}
|
||
else /* We have storage, but need to reset it. */
|
||
{
|
||
ntype = *typeptr;
|
||
chain = TYPE_CHAIN (ntype);
|
||
smash_type (ntype);
|
||
TYPE_CHAIN (ntype) = chain;
|
||
}
|
||
|
||
TYPE_TARGET_TYPE (ntype) = type;
|
||
TYPE_POINTER_TYPE (type) = ntype;
|
||
|
||
/* FIXME! Assumes the machine has only one representation for pointers! */
|
||
|
||
TYPE_LENGTH (ntype)
|
||
= gdbarch_ptr_bit (get_type_arch (type)) / TARGET_CHAR_BIT;
|
||
TYPE_CODE (ntype) = TYPE_CODE_PTR;
|
||
|
||
/* Mark pointers as unsigned. The target converts between pointers
|
||
and addresses (CORE_ADDRs) using gdbarch_pointer_to_address and
|
||
gdbarch_address_to_pointer. */
|
||
TYPE_UNSIGNED (ntype) = 1;
|
||
|
||
/* Update the length of all the other variants of this type. */
|
||
chain = TYPE_CHAIN (ntype);
|
||
while (chain != ntype)
|
||
{
|
||
TYPE_LENGTH (chain) = TYPE_LENGTH (ntype);
|
||
chain = TYPE_CHAIN (chain);
|
||
}
|
||
|
||
return ntype;
|
||
}
|
||
|
||
/* Given a type TYPE, return a type of pointers to that type.
|
||
May need to construct such a type if this is the first use. */
|
||
|
||
struct type *
|
||
lookup_pointer_type (struct type *type)
|
||
{
|
||
return make_pointer_type (type, (struct type **) 0);
|
||
}
|
||
|
||
/* Lookup a C++ `reference' to a type TYPE. TYPEPTR, if nonzero,
|
||
points to a pointer to memory where the reference type should be
|
||
stored. If *TYPEPTR is zero, update it to point to the reference
|
||
type we return. We allocate new memory if needed. */
|
||
|
||
struct type *
|
||
make_reference_type (struct type *type, struct type **typeptr)
|
||
{
|
||
struct type *ntype; /* New type */
|
||
struct type *chain;
|
||
|
||
ntype = TYPE_REFERENCE_TYPE (type);
|
||
|
||
if (ntype)
|
||
{
|
||
if (typeptr == 0)
|
||
return ntype; /* Don't care about alloc,
|
||
and have new type. */
|
||
else if (*typeptr == 0)
|
||
{
|
||
*typeptr = ntype; /* Tracking alloc, and have new type. */
|
||
return ntype;
|
||
}
|
||
}
|
||
|
||
if (typeptr == 0 || *typeptr == 0) /* We'll need to allocate one. */
|
||
{
|
||
ntype = alloc_type_copy (type);
|
||
if (typeptr)
|
||
*typeptr = ntype;
|
||
}
|
||
else /* We have storage, but need to reset it. */
|
||
{
|
||
ntype = *typeptr;
|
||
chain = TYPE_CHAIN (ntype);
|
||
smash_type (ntype);
|
||
TYPE_CHAIN (ntype) = chain;
|
||
}
|
||
|
||
TYPE_TARGET_TYPE (ntype) = type;
|
||
TYPE_REFERENCE_TYPE (type) = ntype;
|
||
|
||
/* FIXME! Assume the machine has only one representation for
|
||
references, and that it matches the (only) representation for
|
||
pointers! */
|
||
|
||
TYPE_LENGTH (ntype) =
|
||
gdbarch_ptr_bit (get_type_arch (type)) / TARGET_CHAR_BIT;
|
||
TYPE_CODE (ntype) = TYPE_CODE_REF;
|
||
|
||
if (!TYPE_REFERENCE_TYPE (type)) /* Remember it, if don't have one. */
|
||
TYPE_REFERENCE_TYPE (type) = ntype;
|
||
|
||
/* Update the length of all the other variants of this type. */
|
||
chain = TYPE_CHAIN (ntype);
|
||
while (chain != ntype)
|
||
{
|
||
TYPE_LENGTH (chain) = TYPE_LENGTH (ntype);
|
||
chain = TYPE_CHAIN (chain);
|
||
}
|
||
|
||
return ntype;
|
||
}
|
||
|
||
/* Same as above, but caller doesn't care about memory allocation
|
||
details. */
|
||
|
||
struct type *
|
||
lookup_reference_type (struct type *type)
|
||
{
|
||
return make_reference_type (type, (struct type **) 0);
|
||
}
|
||
|
||
/* Lookup a function type that returns type TYPE. TYPEPTR, if
|
||
nonzero, points to a pointer to memory where the function type
|
||
should be stored. If *TYPEPTR is zero, update it to point to the
|
||
function type we return. We allocate new memory if needed. */
|
||
|
||
struct type *
|
||
make_function_type (struct type *type, struct type **typeptr)
|
||
{
|
||
struct type *ntype; /* New type */
|
||
|
||
if (typeptr == 0 || *typeptr == 0) /* We'll need to allocate one. */
|
||
{
|
||
ntype = alloc_type_copy (type);
|
||
if (typeptr)
|
||
*typeptr = ntype;
|
||
}
|
||
else /* We have storage, but need to reset it. */
|
||
{
|
||
ntype = *typeptr;
|
||
smash_type (ntype);
|
||
}
|
||
|
||
TYPE_TARGET_TYPE (ntype) = type;
|
||
|
||
TYPE_LENGTH (ntype) = 1;
|
||
TYPE_CODE (ntype) = TYPE_CODE_FUNC;
|
||
|
||
INIT_FUNC_SPECIFIC (ntype);
|
||
|
||
return ntype;
|
||
}
|
||
|
||
/* Given a type TYPE, return a type of functions that return that type.
|
||
May need to construct such a type if this is the first use. */
|
||
|
||
struct type *
|
||
lookup_function_type (struct type *type)
|
||
{
|
||
return make_function_type (type, (struct type **) 0);
|
||
}
|
||
|
||
/* Given a type TYPE and argument types, return the appropriate
|
||
function type. If the final type in PARAM_TYPES is NULL, make a
|
||
varargs function. */
|
||
|
||
struct type *
|
||
lookup_function_type_with_arguments (struct type *type,
|
||
int nparams,
|
||
struct type **param_types)
|
||
{
|
||
struct type *fn = make_function_type (type, (struct type **) 0);
|
||
int i;
|
||
|
||
if (nparams > 0)
|
||
{
|
||
if (param_types[nparams - 1] == NULL)
|
||
{
|
||
--nparams;
|
||
TYPE_VARARGS (fn) = 1;
|
||
}
|
||
else if (TYPE_CODE (check_typedef (param_types[nparams - 1]))
|
||
== TYPE_CODE_VOID)
|
||
{
|
||
--nparams;
|
||
/* Caller should have ensured this. */
|
||
gdb_assert (nparams == 0);
|
||
TYPE_PROTOTYPED (fn) = 1;
|
||
}
|
||
}
|
||
|
||
TYPE_NFIELDS (fn) = nparams;
|
||
TYPE_FIELDS (fn)
|
||
= (struct field *) TYPE_ZALLOC (fn, nparams * sizeof (struct field));
|
||
for (i = 0; i < nparams; ++i)
|
||
TYPE_FIELD_TYPE (fn, i) = param_types[i];
|
||
|
||
return fn;
|
||
}
|
||
|
||
/* Identify address space identifier by name --
|
||
return the integer flag defined in gdbtypes.h. */
|
||
|
||
int
|
||
address_space_name_to_int (struct gdbarch *gdbarch, char *space_identifier)
|
||
{
|
||
int type_flags;
|
||
|
||
/* Check for known address space delimiters. */
|
||
if (!strcmp (space_identifier, "code"))
|
||
return TYPE_INSTANCE_FLAG_CODE_SPACE;
|
||
else if (!strcmp (space_identifier, "data"))
|
||
return TYPE_INSTANCE_FLAG_DATA_SPACE;
|
||
else if (gdbarch_address_class_name_to_type_flags_p (gdbarch)
|
||
&& gdbarch_address_class_name_to_type_flags (gdbarch,
|
||
space_identifier,
|
||
&type_flags))
|
||
return type_flags;
|
||
else
|
||
error (_("Unknown address space specifier: \"%s\""), space_identifier);
|
||
}
|
||
|
||
/* Identify address space identifier by integer flag as defined in
|
||
gdbtypes.h -- return the string version of the adress space name. */
|
||
|
||
const char *
|
||
address_space_int_to_name (struct gdbarch *gdbarch, int space_flag)
|
||
{
|
||
if (space_flag & TYPE_INSTANCE_FLAG_CODE_SPACE)
|
||
return "code";
|
||
else if (space_flag & TYPE_INSTANCE_FLAG_DATA_SPACE)
|
||
return "data";
|
||
else if ((space_flag & TYPE_INSTANCE_FLAG_ADDRESS_CLASS_ALL)
|
||
&& gdbarch_address_class_type_flags_to_name_p (gdbarch))
|
||
return gdbarch_address_class_type_flags_to_name (gdbarch, space_flag);
|
||
else
|
||
return NULL;
|
||
}
|
||
|
||
/* Create a new type with instance flags NEW_FLAGS, based on TYPE.
|
||
|
||
If STORAGE is non-NULL, create the new type instance there.
|
||
STORAGE must be in the same obstack as TYPE. */
|
||
|
||
static struct type *
|
||
make_qualified_type (struct type *type, int new_flags,
|
||
struct type *storage)
|
||
{
|
||
struct type *ntype;
|
||
|
||
ntype = type;
|
||
do
|
||
{
|
||
if (TYPE_INSTANCE_FLAGS (ntype) == new_flags)
|
||
return ntype;
|
||
ntype = TYPE_CHAIN (ntype);
|
||
}
|
||
while (ntype != type);
|
||
|
||
/* Create a new type instance. */
|
||
if (storage == NULL)
|
||
ntype = alloc_type_instance (type);
|
||
else
|
||
{
|
||
/* If STORAGE was provided, it had better be in the same objfile
|
||
as TYPE. Otherwise, we can't link it into TYPE's cv chain:
|
||
if one objfile is freed and the other kept, we'd have
|
||
dangling pointers. */
|
||
gdb_assert (TYPE_OBJFILE (type) == TYPE_OBJFILE (storage));
|
||
|
||
ntype = storage;
|
||
TYPE_MAIN_TYPE (ntype) = TYPE_MAIN_TYPE (type);
|
||
TYPE_CHAIN (ntype) = ntype;
|
||
}
|
||
|
||
/* Pointers or references to the original type are not relevant to
|
||
the new type. */
|
||
TYPE_POINTER_TYPE (ntype) = (struct type *) 0;
|
||
TYPE_REFERENCE_TYPE (ntype) = (struct type *) 0;
|
||
|
||
/* Chain the new qualified type to the old type. */
|
||
TYPE_CHAIN (ntype) = TYPE_CHAIN (type);
|
||
TYPE_CHAIN (type) = ntype;
|
||
|
||
/* Now set the instance flags and return the new type. */
|
||
TYPE_INSTANCE_FLAGS (ntype) = new_flags;
|
||
|
||
/* Set length of new type to that of the original type. */
|
||
TYPE_LENGTH (ntype) = TYPE_LENGTH (type);
|
||
|
||
return ntype;
|
||
}
|
||
|
||
/* Make an address-space-delimited variant of a type -- a type that
|
||
is identical to the one supplied except that it has an address
|
||
space attribute attached to it (such as "code" or "data").
|
||
|
||
The space attributes "code" and "data" are for Harvard
|
||
architectures. The address space attributes are for architectures
|
||
which have alternately sized pointers or pointers with alternate
|
||
representations. */
|
||
|
||
struct type *
|
||
make_type_with_address_space (struct type *type, int space_flag)
|
||
{
|
||
int new_flags = ((TYPE_INSTANCE_FLAGS (type)
|
||
& ~(TYPE_INSTANCE_FLAG_CODE_SPACE
|
||
| TYPE_INSTANCE_FLAG_DATA_SPACE
|
||
| TYPE_INSTANCE_FLAG_ADDRESS_CLASS_ALL))
|
||
| space_flag);
|
||
|
||
return make_qualified_type (type, new_flags, NULL);
|
||
}
|
||
|
||
/* Make a "c-v" variant of a type -- a type that is identical to the
|
||
one supplied except that it may have const or volatile attributes
|
||
CNST is a flag for setting the const attribute
|
||
VOLTL is a flag for setting the volatile attribute
|
||
TYPE is the base type whose variant we are creating.
|
||
|
||
If TYPEPTR and *TYPEPTR are non-zero, then *TYPEPTR points to
|
||
storage to hold the new qualified type; *TYPEPTR and TYPE must be
|
||
in the same objfile. Otherwise, allocate fresh memory for the new
|
||
type whereever TYPE lives. If TYPEPTR is non-zero, set it to the
|
||
new type we construct. */
|
||
|
||
struct type *
|
||
make_cv_type (int cnst, int voltl,
|
||
struct type *type,
|
||
struct type **typeptr)
|
||
{
|
||
struct type *ntype; /* New type */
|
||
|
||
int new_flags = (TYPE_INSTANCE_FLAGS (type)
|
||
& ~(TYPE_INSTANCE_FLAG_CONST
|
||
| TYPE_INSTANCE_FLAG_VOLATILE));
|
||
|
||
if (cnst)
|
||
new_flags |= TYPE_INSTANCE_FLAG_CONST;
|
||
|
||
if (voltl)
|
||
new_flags |= TYPE_INSTANCE_FLAG_VOLATILE;
|
||
|
||
if (typeptr && *typeptr != NULL)
|
||
{
|
||
/* TYPE and *TYPEPTR must be in the same objfile. We can't have
|
||
a C-V variant chain that threads across objfiles: if one
|
||
objfile gets freed, then the other has a broken C-V chain.
|
||
|
||
This code used to try to copy over the main type from TYPE to
|
||
*TYPEPTR if they were in different objfiles, but that's
|
||
wrong, too: TYPE may have a field list or member function
|
||
lists, which refer to types of their own, etc. etc. The
|
||
whole shebang would need to be copied over recursively; you
|
||
can't have inter-objfile pointers. The only thing to do is
|
||
to leave stub types as stub types, and look them up afresh by
|
||
name each time you encounter them. */
|
||
gdb_assert (TYPE_OBJFILE (*typeptr) == TYPE_OBJFILE (type));
|
||
}
|
||
|
||
ntype = make_qualified_type (type, new_flags,
|
||
typeptr ? *typeptr : NULL);
|
||
|
||
if (typeptr != NULL)
|
||
*typeptr = ntype;
|
||
|
||
return ntype;
|
||
}
|
||
|
||
/* Make a 'restrict'-qualified version of TYPE. */
|
||
|
||
struct type *
|
||
make_restrict_type (struct type *type)
|
||
{
|
||
return make_qualified_type (type,
|
||
(TYPE_INSTANCE_FLAGS (type)
|
||
| TYPE_INSTANCE_FLAG_RESTRICT),
|
||
NULL);
|
||
}
|
||
|
||
/* Make a type without const, volatile, or restrict. */
|
||
|
||
struct type *
|
||
make_unqualified_type (struct type *type)
|
||
{
|
||
return make_qualified_type (type,
|
||
(TYPE_INSTANCE_FLAGS (type)
|
||
& ~(TYPE_INSTANCE_FLAG_CONST
|
||
| TYPE_INSTANCE_FLAG_VOLATILE
|
||
| TYPE_INSTANCE_FLAG_RESTRICT)),
|
||
NULL);
|
||
}
|
||
|
||
/* Make a '_Atomic'-qualified version of TYPE. */
|
||
|
||
struct type *
|
||
make_atomic_type (struct type *type)
|
||
{
|
||
return make_qualified_type (type,
|
||
(TYPE_INSTANCE_FLAGS (type)
|
||
| TYPE_INSTANCE_FLAG_ATOMIC),
|
||
NULL);
|
||
}
|
||
|
||
/* Replace the contents of ntype with the type *type. This changes the
|
||
contents, rather than the pointer for TYPE_MAIN_TYPE (ntype); thus
|
||
the changes are propogated to all types in the TYPE_CHAIN.
|
||
|
||
In order to build recursive types, it's inevitable that we'll need
|
||
to update types in place --- but this sort of indiscriminate
|
||
smashing is ugly, and needs to be replaced with something more
|
||
controlled. TYPE_MAIN_TYPE is a step in this direction; it's not
|
||
clear if more steps are needed. */
|
||
|
||
void
|
||
replace_type (struct type *ntype, struct type *type)
|
||
{
|
||
struct type *chain;
|
||
|
||
/* These two types had better be in the same objfile. Otherwise,
|
||
the assignment of one type's main type structure to the other
|
||
will produce a type with references to objects (names; field
|
||
lists; etc.) allocated on an objfile other than its own. */
|
||
gdb_assert (TYPE_OBJFILE (ntype) == TYPE_OBJFILE (type));
|
||
|
||
*TYPE_MAIN_TYPE (ntype) = *TYPE_MAIN_TYPE (type);
|
||
|
||
/* The type length is not a part of the main type. Update it for
|
||
each type on the variant chain. */
|
||
chain = ntype;
|
||
do
|
||
{
|
||
/* Assert that this element of the chain has no address-class bits
|
||
set in its flags. Such type variants might have type lengths
|
||
which are supposed to be different from the non-address-class
|
||
variants. This assertion shouldn't ever be triggered because
|
||
symbol readers which do construct address-class variants don't
|
||
call replace_type(). */
|
||
gdb_assert (TYPE_ADDRESS_CLASS_ALL (chain) == 0);
|
||
|
||
TYPE_LENGTH (chain) = TYPE_LENGTH (type);
|
||
chain = TYPE_CHAIN (chain);
|
||
}
|
||
while (ntype != chain);
|
||
|
||
/* Assert that the two types have equivalent instance qualifiers.
|
||
This should be true for at least all of our debug readers. */
|
||
gdb_assert (TYPE_INSTANCE_FLAGS (ntype) == TYPE_INSTANCE_FLAGS (type));
|
||
}
|
||
|
||
/* Implement direct support for MEMBER_TYPE in GNU C++.
|
||
May need to construct such a type if this is the first use.
|
||
The TYPE is the type of the member. The DOMAIN is the type
|
||
of the aggregate that the member belongs to. */
|
||
|
||
struct type *
|
||
lookup_memberptr_type (struct type *type, struct type *domain)
|
||
{
|
||
struct type *mtype;
|
||
|
||
mtype = alloc_type_copy (type);
|
||
smash_to_memberptr_type (mtype, domain, type);
|
||
return mtype;
|
||
}
|
||
|
||
/* Return a pointer-to-method type, for a method of type TO_TYPE. */
|
||
|
||
struct type *
|
||
lookup_methodptr_type (struct type *to_type)
|
||
{
|
||
struct type *mtype;
|
||
|
||
mtype = alloc_type_copy (to_type);
|
||
smash_to_methodptr_type (mtype, to_type);
|
||
return mtype;
|
||
}
|
||
|
||
/* Allocate a stub method whose return type is TYPE. This apparently
|
||
happens for speed of symbol reading, since parsing out the
|
||
arguments to the method is cpu-intensive, the way we are doing it.
|
||
So, we will fill in arguments later. This always returns a fresh
|
||
type. */
|
||
|
||
struct type *
|
||
allocate_stub_method (struct type *type)
|
||
{
|
||
struct type *mtype;
|
||
|
||
mtype = alloc_type_copy (type);
|
||
TYPE_CODE (mtype) = TYPE_CODE_METHOD;
|
||
TYPE_LENGTH (mtype) = 1;
|
||
TYPE_STUB (mtype) = 1;
|
||
TYPE_TARGET_TYPE (mtype) = type;
|
||
/* TYPE_SELF_TYPE (mtype) = unknown yet */
|
||
return mtype;
|
||
}
|
||
|
||
/* Create a range type with a dynamic range from LOW_BOUND to
|
||
HIGH_BOUND, inclusive. See create_range_type for further details. */
|
||
|
||
struct type *
|
||
create_range_type (struct type *result_type, struct type *index_type,
|
||
const struct dynamic_prop *low_bound,
|
||
const struct dynamic_prop *high_bound)
|
||
{
|
||
if (result_type == NULL)
|
||
result_type = alloc_type_copy (index_type);
|
||
TYPE_CODE (result_type) = TYPE_CODE_RANGE;
|
||
TYPE_TARGET_TYPE (result_type) = index_type;
|
||
if (TYPE_STUB (index_type))
|
||
TYPE_TARGET_STUB (result_type) = 1;
|
||
else
|
||
TYPE_LENGTH (result_type) = TYPE_LENGTH (check_typedef (index_type));
|
||
|
||
TYPE_RANGE_DATA (result_type) = (struct range_bounds *)
|
||
TYPE_ZALLOC (result_type, sizeof (struct range_bounds));
|
||
TYPE_RANGE_DATA (result_type)->low = *low_bound;
|
||
TYPE_RANGE_DATA (result_type)->high = *high_bound;
|
||
|
||
if (low_bound->kind == PROP_CONST && low_bound->data.const_val >= 0)
|
||
TYPE_UNSIGNED (result_type) = 1;
|
||
|
||
/* Ada allows the declaration of range types whose upper bound is
|
||
less than the lower bound, so checking the lower bound is not
|
||
enough. Make sure we do not mark a range type whose upper bound
|
||
is negative as unsigned. */
|
||
if (high_bound->kind == PROP_CONST && high_bound->data.const_val < 0)
|
||
TYPE_UNSIGNED (result_type) = 0;
|
||
|
||
return result_type;
|
||
}
|
||
|
||
/* Create a range type using either a blank type supplied in
|
||
RESULT_TYPE, or creating a new type, inheriting the objfile from
|
||
INDEX_TYPE.
|
||
|
||
Indices will be of type INDEX_TYPE, and will range from LOW_BOUND
|
||
to HIGH_BOUND, inclusive.
|
||
|
||
FIXME: Maybe we should check the TYPE_CODE of RESULT_TYPE to make
|
||
sure it is TYPE_CODE_UNDEF before we bash it into a range type? */
|
||
|
||
struct type *
|
||
create_static_range_type (struct type *result_type, struct type *index_type,
|
||
LONGEST low_bound, LONGEST high_bound)
|
||
{
|
||
struct dynamic_prop low, high;
|
||
|
||
low.kind = PROP_CONST;
|
||
low.data.const_val = low_bound;
|
||
|
||
high.kind = PROP_CONST;
|
||
high.data.const_val = high_bound;
|
||
|
||
result_type = create_range_type (result_type, index_type, &low, &high);
|
||
|
||
return result_type;
|
||
}
|
||
|
||
/* Predicate tests whether BOUNDS are static. Returns 1 if all bounds values
|
||
are static, otherwise returns 0. */
|
||
|
||
static int
|
||
has_static_range (const struct range_bounds *bounds)
|
||
{
|
||
return (bounds->low.kind == PROP_CONST
|
||
&& bounds->high.kind == PROP_CONST);
|
||
}
|
||
|
||
|
||
/* Set *LOWP and *HIGHP to the lower and upper bounds of discrete type
|
||
TYPE. Return 1 if type is a range type, 0 if it is discrete (and
|
||
bounds will fit in LONGEST), or -1 otherwise. */
|
||
|
||
int
|
||
get_discrete_bounds (struct type *type, LONGEST *lowp, LONGEST *highp)
|
||
{
|
||
type = check_typedef (type);
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_RANGE:
|
||
*lowp = TYPE_LOW_BOUND (type);
|
||
*highp = TYPE_HIGH_BOUND (type);
|
||
return 1;
|
||
case TYPE_CODE_ENUM:
|
||
if (TYPE_NFIELDS (type) > 0)
|
||
{
|
||
/* The enums may not be sorted by value, so search all
|
||
entries. */
|
||
int i;
|
||
|
||
*lowp = *highp = TYPE_FIELD_ENUMVAL (type, 0);
|
||
for (i = 0; i < TYPE_NFIELDS (type); i++)
|
||
{
|
||
if (TYPE_FIELD_ENUMVAL (type, i) < *lowp)
|
||
*lowp = TYPE_FIELD_ENUMVAL (type, i);
|
||
if (TYPE_FIELD_ENUMVAL (type, i) > *highp)
|
||
*highp = TYPE_FIELD_ENUMVAL (type, i);
|
||
}
|
||
|
||
/* Set unsigned indicator if warranted. */
|
||
if (*lowp >= 0)
|
||
{
|
||
TYPE_UNSIGNED (type) = 1;
|
||
}
|
||
}
|
||
else
|
||
{
|
||
*lowp = 0;
|
||
*highp = -1;
|
||
}
|
||
return 0;
|
||
case TYPE_CODE_BOOL:
|
||
*lowp = 0;
|
||
*highp = 1;
|
||
return 0;
|
||
case TYPE_CODE_INT:
|
||
if (TYPE_LENGTH (type) > sizeof (LONGEST)) /* Too big */
|
||
return -1;
|
||
if (!TYPE_UNSIGNED (type))
|
||
{
|
||
*lowp = -(1 << (TYPE_LENGTH (type) * TARGET_CHAR_BIT - 1));
|
||
*highp = -*lowp - 1;
|
||
return 0;
|
||
}
|
||
/* ... fall through for unsigned ints ... */
|
||
case TYPE_CODE_CHAR:
|
||
*lowp = 0;
|
||
/* This round-about calculation is to avoid shifting by
|
||
TYPE_LENGTH (type) * TARGET_CHAR_BIT, which will not work
|
||
if TYPE_LENGTH (type) == sizeof (LONGEST). */
|
||
*highp = 1 << (TYPE_LENGTH (type) * TARGET_CHAR_BIT - 1);
|
||
*highp = (*highp - 1) | *highp;
|
||
return 0;
|
||
default:
|
||
return -1;
|
||
}
|
||
}
|
||
|
||
/* Assuming TYPE is a simple, non-empty array type, compute its upper
|
||
and lower bound. Save the low bound into LOW_BOUND if not NULL.
|
||
Save the high bound into HIGH_BOUND if not NULL.
|
||
|
||
Return 1 if the operation was successful. Return zero otherwise,
|
||
in which case the values of LOW_BOUND and HIGH_BOUNDS are unmodified.
|
||
|
||
We now simply use get_discrete_bounds call to get the values
|
||
of the low and high bounds.
|
||
get_discrete_bounds can return three values:
|
||
1, meaning that index is a range,
|
||
0, meaning that index is a discrete type,
|
||
or -1 for failure. */
|
||
|
||
int
|
||
get_array_bounds (struct type *type, LONGEST *low_bound, LONGEST *high_bound)
|
||
{
|
||
struct type *index = TYPE_INDEX_TYPE (type);
|
||
LONGEST low = 0;
|
||
LONGEST high = 0;
|
||
int res;
|
||
|
||
if (index == NULL)
|
||
return 0;
|
||
|
||
res = get_discrete_bounds (index, &low, &high);
|
||
if (res == -1)
|
||
return 0;
|
||
|
||
/* Check if the array bounds are undefined. */
|
||
if (res == 1
|
||
&& ((low_bound && TYPE_ARRAY_LOWER_BOUND_IS_UNDEFINED (type))
|
||
|| (high_bound && TYPE_ARRAY_UPPER_BOUND_IS_UNDEFINED (type))))
|
||
return 0;
|
||
|
||
if (low_bound)
|
||
*low_bound = low;
|
||
|
||
if (high_bound)
|
||
*high_bound = high;
|
||
|
||
return 1;
|
||
}
|
||
|
||
/* Assuming that TYPE is a discrete type and VAL is a valid integer
|
||
representation of a value of this type, save the corresponding
|
||
position number in POS.
|
||
|
||
Its differs from VAL only in the case of enumeration types. In
|
||
this case, the position number of the value of the first listed
|
||
enumeration literal is zero; the position number of the value of
|
||
each subsequent enumeration literal is one more than that of its
|
||
predecessor in the list.
|
||
|
||
Return 1 if the operation was successful. Return zero otherwise,
|
||
in which case the value of POS is unmodified.
|
||
*/
|
||
|
||
int
|
||
discrete_position (struct type *type, LONGEST val, LONGEST *pos)
|
||
{
|
||
if (TYPE_CODE (type) == TYPE_CODE_ENUM)
|
||
{
|
||
int i;
|
||
|
||
for (i = 0; i < TYPE_NFIELDS (type); i += 1)
|
||
{
|
||
if (val == TYPE_FIELD_ENUMVAL (type, i))
|
||
{
|
||
*pos = i;
|
||
return 1;
|
||
}
|
||
}
|
||
/* Invalid enumeration value. */
|
||
return 0;
|
||
}
|
||
else
|
||
{
|
||
*pos = val;
|
||
return 1;
|
||
}
|
||
}
|
||
|
||
/* Create an array type using either a blank type supplied in
|
||
RESULT_TYPE, or creating a new type, inheriting the objfile from
|
||
RANGE_TYPE.
|
||
|
||
Elements will be of type ELEMENT_TYPE, the indices will be of type
|
||
RANGE_TYPE.
|
||
|
||
If BIT_STRIDE is not zero, build a packed array type whose element
|
||
size is BIT_STRIDE. Otherwise, ignore this parameter.
|
||
|
||
FIXME: Maybe we should check the TYPE_CODE of RESULT_TYPE to make
|
||
sure it is TYPE_CODE_UNDEF before we bash it into an array
|
||
type? */
|
||
|
||
struct type *
|
||
create_array_type_with_stride (struct type *result_type,
|
||
struct type *element_type,
|
||
struct type *range_type,
|
||
unsigned int bit_stride)
|
||
{
|
||
if (result_type == NULL)
|
||
result_type = alloc_type_copy (range_type);
|
||
|
||
TYPE_CODE (result_type) = TYPE_CODE_ARRAY;
|
||
TYPE_TARGET_TYPE (result_type) = element_type;
|
||
if (has_static_range (TYPE_RANGE_DATA (range_type))
|
||
&& (!type_not_associated (result_type)
|
||
&& !type_not_allocated (result_type)))
|
||
{
|
||
LONGEST low_bound, high_bound;
|
||
|
||
if (get_discrete_bounds (range_type, &low_bound, &high_bound) < 0)
|
||
low_bound = high_bound = 0;
|
||
element_type = check_typedef (element_type);
|
||
/* Be careful when setting the array length. Ada arrays can be
|
||
empty arrays with the high_bound being smaller than the low_bound.
|
||
In such cases, the array length should be zero. */
|
||
if (high_bound < low_bound)
|
||
TYPE_LENGTH (result_type) = 0;
|
||
else if (bit_stride > 0)
|
||
TYPE_LENGTH (result_type) =
|
||
(bit_stride * (high_bound - low_bound + 1) + 7) / 8;
|
||
else
|
||
TYPE_LENGTH (result_type) =
|
||
TYPE_LENGTH (element_type) * (high_bound - low_bound + 1);
|
||
}
|
||
else
|
||
{
|
||
/* This type is dynamic and its length needs to be computed
|
||
on demand. In the meantime, avoid leaving the TYPE_LENGTH
|
||
undefined by setting it to zero. Although we are not expected
|
||
to trust TYPE_LENGTH in this case, setting the size to zero
|
||
allows us to avoid allocating objects of random sizes in case
|
||
we accidently do. */
|
||
TYPE_LENGTH (result_type) = 0;
|
||
}
|
||
|
||
TYPE_NFIELDS (result_type) = 1;
|
||
TYPE_FIELDS (result_type) =
|
||
(struct field *) TYPE_ZALLOC (result_type, sizeof (struct field));
|
||
TYPE_INDEX_TYPE (result_type) = range_type;
|
||
if (bit_stride > 0)
|
||
TYPE_FIELD_BITSIZE (result_type, 0) = bit_stride;
|
||
|
||
/* TYPE_FLAG_TARGET_STUB will take care of zero length arrays. */
|
||
if (TYPE_LENGTH (result_type) == 0)
|
||
TYPE_TARGET_STUB (result_type) = 1;
|
||
|
||
return result_type;
|
||
}
|
||
|
||
/* Same as create_array_type_with_stride but with no bit_stride
|
||
(BIT_STRIDE = 0), thus building an unpacked array. */
|
||
|
||
struct type *
|
||
create_array_type (struct type *result_type,
|
||
struct type *element_type,
|
||
struct type *range_type)
|
||
{
|
||
return create_array_type_with_stride (result_type, element_type,
|
||
range_type, 0);
|
||
}
|
||
|
||
struct type *
|
||
lookup_array_range_type (struct type *element_type,
|
||
LONGEST low_bound, LONGEST high_bound)
|
||
{
|
||
struct gdbarch *gdbarch = get_type_arch (element_type);
|
||
struct type *index_type = builtin_type (gdbarch)->builtin_int;
|
||
struct type *range_type
|
||
= create_static_range_type (NULL, index_type, low_bound, high_bound);
|
||
|
||
return create_array_type (NULL, element_type, range_type);
|
||
}
|
||
|
||
/* Create a string type using either a blank type supplied in
|
||
RESULT_TYPE, or creating a new type. String types are similar
|
||
enough to array of char types that we can use create_array_type to
|
||
build the basic type and then bash it into a string type.
|
||
|
||
For fixed length strings, the range type contains 0 as the lower
|
||
bound and the length of the string minus one as the upper bound.
|
||
|
||
FIXME: Maybe we should check the TYPE_CODE of RESULT_TYPE to make
|
||
sure it is TYPE_CODE_UNDEF before we bash it into a string
|
||
type? */
|
||
|
||
struct type *
|
||
create_string_type (struct type *result_type,
|
||
struct type *string_char_type,
|
||
struct type *range_type)
|
||
{
|
||
result_type = create_array_type (result_type,
|
||
string_char_type,
|
||
range_type);
|
||
TYPE_CODE (result_type) = TYPE_CODE_STRING;
|
||
return result_type;
|
||
}
|
||
|
||
struct type *
|
||
lookup_string_range_type (struct type *string_char_type,
|
||
LONGEST low_bound, LONGEST high_bound)
|
||
{
|
||
struct type *result_type;
|
||
|
||
result_type = lookup_array_range_type (string_char_type,
|
||
low_bound, high_bound);
|
||
TYPE_CODE (result_type) = TYPE_CODE_STRING;
|
||
return result_type;
|
||
}
|
||
|
||
struct type *
|
||
create_set_type (struct type *result_type, struct type *domain_type)
|
||
{
|
||
if (result_type == NULL)
|
||
result_type = alloc_type_copy (domain_type);
|
||
|
||
TYPE_CODE (result_type) = TYPE_CODE_SET;
|
||
TYPE_NFIELDS (result_type) = 1;
|
||
TYPE_FIELDS (result_type)
|
||
= (struct field *) TYPE_ZALLOC (result_type, sizeof (struct field));
|
||
|
||
if (!TYPE_STUB (domain_type))
|
||
{
|
||
LONGEST low_bound, high_bound, bit_length;
|
||
|
||
if (get_discrete_bounds (domain_type, &low_bound, &high_bound) < 0)
|
||
low_bound = high_bound = 0;
|
||
bit_length = high_bound - low_bound + 1;
|
||
TYPE_LENGTH (result_type)
|
||
= (bit_length + TARGET_CHAR_BIT - 1) / TARGET_CHAR_BIT;
|
||
if (low_bound >= 0)
|
||
TYPE_UNSIGNED (result_type) = 1;
|
||
}
|
||
TYPE_FIELD_TYPE (result_type, 0) = domain_type;
|
||
|
||
return result_type;
|
||
}
|
||
|
||
/* Convert ARRAY_TYPE to a vector type. This may modify ARRAY_TYPE
|
||
and any array types nested inside it. */
|
||
|
||
void
|
||
make_vector_type (struct type *array_type)
|
||
{
|
||
struct type *inner_array, *elt_type;
|
||
int flags;
|
||
|
||
/* Find the innermost array type, in case the array is
|
||
multi-dimensional. */
|
||
inner_array = array_type;
|
||
while (TYPE_CODE (TYPE_TARGET_TYPE (inner_array)) == TYPE_CODE_ARRAY)
|
||
inner_array = TYPE_TARGET_TYPE (inner_array);
|
||
|
||
elt_type = TYPE_TARGET_TYPE (inner_array);
|
||
if (TYPE_CODE (elt_type) == TYPE_CODE_INT)
|
||
{
|
||
flags = TYPE_INSTANCE_FLAGS (elt_type) | TYPE_INSTANCE_FLAG_NOTTEXT;
|
||
elt_type = make_qualified_type (elt_type, flags, NULL);
|
||
TYPE_TARGET_TYPE (inner_array) = elt_type;
|
||
}
|
||
|
||
TYPE_VECTOR (array_type) = 1;
|
||
}
|
||
|
||
struct type *
|
||
init_vector_type (struct type *elt_type, int n)
|
||
{
|
||
struct type *array_type;
|
||
|
||
array_type = lookup_array_range_type (elt_type, 0, n - 1);
|
||
make_vector_type (array_type);
|
||
return array_type;
|
||
}
|
||
|
||
/* Internal routine called by TYPE_SELF_TYPE to return the type that TYPE
|
||
belongs to. In c++ this is the class of "this", but TYPE_THIS_TYPE is too
|
||
confusing. "self" is a common enough replacement for "this".
|
||
TYPE must be one of TYPE_CODE_METHODPTR, TYPE_CODE_MEMBERPTR, or
|
||
TYPE_CODE_METHOD. */
|
||
|
||
struct type *
|
||
internal_type_self_type (struct type *type)
|
||
{
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_METHODPTR:
|
||
case TYPE_CODE_MEMBERPTR:
|
||
if (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_NONE)
|
||
return NULL;
|
||
gdb_assert (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_SELF_TYPE);
|
||
return TYPE_MAIN_TYPE (type)->type_specific.self_type;
|
||
case TYPE_CODE_METHOD:
|
||
if (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_NONE)
|
||
return NULL;
|
||
gdb_assert (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_FUNC);
|
||
return TYPE_MAIN_TYPE (type)->type_specific.func_stuff->self_type;
|
||
default:
|
||
gdb_assert_not_reached ("bad type");
|
||
}
|
||
}
|
||
|
||
/* Set the type of the class that TYPE belongs to.
|
||
In c++ this is the class of "this".
|
||
TYPE must be one of TYPE_CODE_METHODPTR, TYPE_CODE_MEMBERPTR, or
|
||
TYPE_CODE_METHOD. */
|
||
|
||
void
|
||
set_type_self_type (struct type *type, struct type *self_type)
|
||
{
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_METHODPTR:
|
||
case TYPE_CODE_MEMBERPTR:
|
||
if (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_NONE)
|
||
TYPE_SPECIFIC_FIELD (type) = TYPE_SPECIFIC_SELF_TYPE;
|
||
gdb_assert (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_SELF_TYPE);
|
||
TYPE_MAIN_TYPE (type)->type_specific.self_type = self_type;
|
||
break;
|
||
case TYPE_CODE_METHOD:
|
||
if (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_NONE)
|
||
INIT_FUNC_SPECIFIC (type);
|
||
gdb_assert (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_FUNC);
|
||
TYPE_MAIN_TYPE (type)->type_specific.func_stuff->self_type = self_type;
|
||
break;
|
||
default:
|
||
gdb_assert_not_reached ("bad type");
|
||
}
|
||
}
|
||
|
||
/* Smash TYPE to be a type of pointers to members of SELF_TYPE with type
|
||
TO_TYPE. A member pointer is a wierd thing -- it amounts to a
|
||
typed offset into a struct, e.g. "an int at offset 8". A MEMBER
|
||
TYPE doesn't include the offset (that's the value of the MEMBER
|
||
itself), but does include the structure type into which it points
|
||
(for some reason).
|
||
|
||
When "smashing" the type, we preserve the objfile that the old type
|
||
pointed to, since we aren't changing where the type is actually
|
||
allocated. */
|
||
|
||
void
|
||
smash_to_memberptr_type (struct type *type, struct type *self_type,
|
||
struct type *to_type)
|
||
{
|
||
smash_type (type);
|
||
TYPE_CODE (type) = TYPE_CODE_MEMBERPTR;
|
||
TYPE_TARGET_TYPE (type) = to_type;
|
||
set_type_self_type (type, self_type);
|
||
/* Assume that a data member pointer is the same size as a normal
|
||
pointer. */
|
||
TYPE_LENGTH (type)
|
||
= gdbarch_ptr_bit (get_type_arch (to_type)) / TARGET_CHAR_BIT;
|
||
}
|
||
|
||
/* Smash TYPE to be a type of pointer to methods type TO_TYPE.
|
||
|
||
When "smashing" the type, we preserve the objfile that the old type
|
||
pointed to, since we aren't changing where the type is actually
|
||
allocated. */
|
||
|
||
void
|
||
smash_to_methodptr_type (struct type *type, struct type *to_type)
|
||
{
|
||
smash_type (type);
|
||
TYPE_CODE (type) = TYPE_CODE_METHODPTR;
|
||
TYPE_TARGET_TYPE (type) = to_type;
|
||
set_type_self_type (type, TYPE_SELF_TYPE (to_type));
|
||
TYPE_LENGTH (type) = cplus_method_ptr_size (to_type);
|
||
}
|
||
|
||
/* Smash TYPE to be a type of method of SELF_TYPE with type TO_TYPE.
|
||
METHOD just means `function that gets an extra "this" argument'.
|
||
|
||
When "smashing" the type, we preserve the objfile that the old type
|
||
pointed to, since we aren't changing where the type is actually
|
||
allocated. */
|
||
|
||
void
|
||
smash_to_method_type (struct type *type, struct type *self_type,
|
||
struct type *to_type, struct field *args,
|
||
int nargs, int varargs)
|
||
{
|
||
smash_type (type);
|
||
TYPE_CODE (type) = TYPE_CODE_METHOD;
|
||
TYPE_TARGET_TYPE (type) = to_type;
|
||
set_type_self_type (type, self_type);
|
||
TYPE_FIELDS (type) = args;
|
||
TYPE_NFIELDS (type) = nargs;
|
||
if (varargs)
|
||
TYPE_VARARGS (type) = 1;
|
||
TYPE_LENGTH (type) = 1; /* In practice, this is never needed. */
|
||
}
|
||
|
||
/* Return a typename for a struct/union/enum type without "struct ",
|
||
"union ", or "enum ". If the type has a NULL name, return NULL. */
|
||
|
||
const char *
|
||
type_name_no_tag (const struct type *type)
|
||
{
|
||
if (TYPE_TAG_NAME (type) != NULL)
|
||
return TYPE_TAG_NAME (type);
|
||
|
||
/* Is there code which expects this to return the name if there is
|
||
no tag name? My guess is that this is mainly used for C++ in
|
||
cases where the two will always be the same. */
|
||
return TYPE_NAME (type);
|
||
}
|
||
|
||
/* A wrapper of type_name_no_tag which calls error if the type is anonymous.
|
||
Since GCC PR debug/47510 DWARF provides associated information to detect the
|
||
anonymous class linkage name from its typedef.
|
||
|
||
Parameter TYPE should not yet have CHECK_TYPEDEF applied, this function will
|
||
apply it itself. */
|
||
|
||
const char *
|
||
type_name_no_tag_or_error (struct type *type)
|
||
{
|
||
struct type *saved_type = type;
|
||
const char *name;
|
||
struct objfile *objfile;
|
||
|
||
type = check_typedef (type);
|
||
|
||
name = type_name_no_tag (type);
|
||
if (name != NULL)
|
||
return name;
|
||
|
||
name = type_name_no_tag (saved_type);
|
||
objfile = TYPE_OBJFILE (saved_type);
|
||
error (_("Invalid anonymous type %s [in module %s], GCC PR debug/47510 bug?"),
|
||
name ? name : "<anonymous>",
|
||
objfile ? objfile_name (objfile) : "<arch>");
|
||
}
|
||
|
||
/* Lookup a typedef or primitive type named NAME, visible in lexical
|
||
block BLOCK. If NOERR is nonzero, return zero if NAME is not
|
||
suitably defined. */
|
||
|
||
struct type *
|
||
lookup_typename (const struct language_defn *language,
|
||
struct gdbarch *gdbarch, const char *name,
|
||
const struct block *block, int noerr)
|
||
{
|
||
struct symbol *sym;
|
||
|
||
sym = lookup_symbol_in_language (name, block, VAR_DOMAIN,
|
||
language->la_language, NULL).symbol;
|
||
if (sym != NULL && SYMBOL_CLASS (sym) == LOC_TYPEDEF)
|
||
return SYMBOL_TYPE (sym);
|
||
|
||
if (noerr)
|
||
return NULL;
|
||
error (_("No type named %s."), name);
|
||
}
|
||
|
||
struct type *
|
||
lookup_unsigned_typename (const struct language_defn *language,
|
||
struct gdbarch *gdbarch, const char *name)
|
||
{
|
||
char *uns = (char *) alloca (strlen (name) + 10);
|
||
|
||
strcpy (uns, "unsigned ");
|
||
strcpy (uns + 9, name);
|
||
return lookup_typename (language, gdbarch, uns, (struct block *) NULL, 0);
|
||
}
|
||
|
||
struct type *
|
||
lookup_signed_typename (const struct language_defn *language,
|
||
struct gdbarch *gdbarch, const char *name)
|
||
{
|
||
struct type *t;
|
||
char *uns = (char *) alloca (strlen (name) + 8);
|
||
|
||
strcpy (uns, "signed ");
|
||
strcpy (uns + 7, name);
|
||
t = lookup_typename (language, gdbarch, uns, (struct block *) NULL, 1);
|
||
/* If we don't find "signed FOO" just try again with plain "FOO". */
|
||
if (t != NULL)
|
||
return t;
|
||
return lookup_typename (language, gdbarch, name, (struct block *) NULL, 0);
|
||
}
|
||
|
||
/* Lookup a structure type named "struct NAME",
|
||
visible in lexical block BLOCK. */
|
||
|
||
struct type *
|
||
lookup_struct (const char *name, const struct block *block)
|
||
{
|
||
struct symbol *sym;
|
||
|
||
sym = lookup_symbol (name, block, STRUCT_DOMAIN, 0).symbol;
|
||
|
||
if (sym == NULL)
|
||
{
|
||
error (_("No struct type named %s."), name);
|
||
}
|
||
if (TYPE_CODE (SYMBOL_TYPE (sym)) != TYPE_CODE_STRUCT)
|
||
{
|
||
error (_("This context has class, union or enum %s, not a struct."),
|
||
name);
|
||
}
|
||
return (SYMBOL_TYPE (sym));
|
||
}
|
||
|
||
/* Lookup a union type named "union NAME",
|
||
visible in lexical block BLOCK. */
|
||
|
||
struct type *
|
||
lookup_union (const char *name, const struct block *block)
|
||
{
|
||
struct symbol *sym;
|
||
struct type *t;
|
||
|
||
sym = lookup_symbol (name, block, STRUCT_DOMAIN, 0).symbol;
|
||
|
||
if (sym == NULL)
|
||
error (_("No union type named %s."), name);
|
||
|
||
t = SYMBOL_TYPE (sym);
|
||
|
||
if (TYPE_CODE (t) == TYPE_CODE_UNION)
|
||
return t;
|
||
|
||
/* If we get here, it's not a union. */
|
||
error (_("This context has class, struct or enum %s, not a union."),
|
||
name);
|
||
}
|
||
|
||
/* Lookup an enum type named "enum NAME",
|
||
visible in lexical block BLOCK. */
|
||
|
||
struct type *
|
||
lookup_enum (const char *name, const struct block *block)
|
||
{
|
||
struct symbol *sym;
|
||
|
||
sym = lookup_symbol (name, block, STRUCT_DOMAIN, 0).symbol;
|
||
if (sym == NULL)
|
||
{
|
||
error (_("No enum type named %s."), name);
|
||
}
|
||
if (TYPE_CODE (SYMBOL_TYPE (sym)) != TYPE_CODE_ENUM)
|
||
{
|
||
error (_("This context has class, struct or union %s, not an enum."),
|
||
name);
|
||
}
|
||
return (SYMBOL_TYPE (sym));
|
||
}
|
||
|
||
/* Lookup a template type named "template NAME<TYPE>",
|
||
visible in lexical block BLOCK. */
|
||
|
||
struct type *
|
||
lookup_template_type (char *name, struct type *type,
|
||
const struct block *block)
|
||
{
|
||
struct symbol *sym;
|
||
char *nam = (char *)
|
||
alloca (strlen (name) + strlen (TYPE_NAME (type)) + 4);
|
||
|
||
strcpy (nam, name);
|
||
strcat (nam, "<");
|
||
strcat (nam, TYPE_NAME (type));
|
||
strcat (nam, " >"); /* FIXME, extra space still introduced in gcc? */
|
||
|
||
sym = lookup_symbol (nam, block, VAR_DOMAIN, 0).symbol;
|
||
|
||
if (sym == NULL)
|
||
{
|
||
error (_("No template type named %s."), name);
|
||
}
|
||
if (TYPE_CODE (SYMBOL_TYPE (sym)) != TYPE_CODE_STRUCT)
|
||
{
|
||
error (_("This context has class, union or enum %s, not a struct."),
|
||
name);
|
||
}
|
||
return (SYMBOL_TYPE (sym));
|
||
}
|
||
|
||
/* Given a type TYPE, lookup the type of the component of type named
|
||
NAME.
|
||
|
||
TYPE can be either a struct or union, or a pointer or reference to
|
||
a struct or union. If it is a pointer or reference, its target
|
||
type is automatically used. Thus '.' and '->' are interchangable,
|
||
as specified for the definitions of the expression element types
|
||
STRUCTOP_STRUCT and STRUCTOP_PTR.
|
||
|
||
If NOERR is nonzero, return zero if NAME is not suitably defined.
|
||
If NAME is the name of a baseclass type, return that type. */
|
||
|
||
struct type *
|
||
lookup_struct_elt_type (struct type *type, const char *name, int noerr)
|
||
{
|
||
int i;
|
||
char *type_name;
|
||
|
||
for (;;)
|
||
{
|
||
type = check_typedef (type);
|
||
if (TYPE_CODE (type) != TYPE_CODE_PTR
|
||
&& TYPE_CODE (type) != TYPE_CODE_REF)
|
||
break;
|
||
type = TYPE_TARGET_TYPE (type);
|
||
}
|
||
|
||
if (TYPE_CODE (type) != TYPE_CODE_STRUCT
|
||
&& TYPE_CODE (type) != TYPE_CODE_UNION)
|
||
{
|
||
type_name = type_to_string (type);
|
||
make_cleanup (xfree, type_name);
|
||
error (_("Type %s is not a structure or union type."), type_name);
|
||
}
|
||
|
||
#if 0
|
||
/* FIXME: This change put in by Michael seems incorrect for the case
|
||
where the structure tag name is the same as the member name.
|
||
I.e. when doing "ptype bell->bar" for "struct foo { int bar; int
|
||
foo; } bell;" Disabled by fnf. */
|
||
{
|
||
char *type_name;
|
||
|
||
type_name = type_name_no_tag (type);
|
||
if (type_name != NULL && strcmp (type_name, name) == 0)
|
||
return type;
|
||
}
|
||
#endif
|
||
|
||
for (i = TYPE_NFIELDS (type) - 1; i >= TYPE_N_BASECLASSES (type); i--)
|
||
{
|
||
const char *t_field_name = TYPE_FIELD_NAME (type, i);
|
||
|
||
if (t_field_name && (strcmp_iw (t_field_name, name) == 0))
|
||
{
|
||
return TYPE_FIELD_TYPE (type, i);
|
||
}
|
||
else if (!t_field_name || *t_field_name == '\0')
|
||
{
|
||
struct type *subtype
|
||
= lookup_struct_elt_type (TYPE_FIELD_TYPE (type, i), name, 1);
|
||
|
||
if (subtype != NULL)
|
||
return subtype;
|
||
}
|
||
}
|
||
|
||
/* OK, it's not in this class. Recursively check the baseclasses. */
|
||
for (i = TYPE_N_BASECLASSES (type) - 1; i >= 0; i--)
|
||
{
|
||
struct type *t;
|
||
|
||
t = lookup_struct_elt_type (TYPE_BASECLASS (type, i), name, 1);
|
||
if (t != NULL)
|
||
{
|
||
return t;
|
||
}
|
||
}
|
||
|
||
if (noerr)
|
||
{
|
||
return NULL;
|
||
}
|
||
|
||
type_name = type_to_string (type);
|
||
make_cleanup (xfree, type_name);
|
||
error (_("Type %s has no component named %s."), type_name, name);
|
||
}
|
||
|
||
/* Store in *MAX the largest number representable by unsigned integer type
|
||
TYPE. */
|
||
|
||
void
|
||
get_unsigned_type_max (struct type *type, ULONGEST *max)
|
||
{
|
||
unsigned int n;
|
||
|
||
type = check_typedef (type);
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_INT && TYPE_UNSIGNED (type));
|
||
gdb_assert (TYPE_LENGTH (type) <= sizeof (ULONGEST));
|
||
|
||
/* Written this way to avoid overflow. */
|
||
n = TYPE_LENGTH (type) * TARGET_CHAR_BIT;
|
||
*max = ((((ULONGEST) 1 << (n - 1)) - 1) << 1) | 1;
|
||
}
|
||
|
||
/* Store in *MIN, *MAX the smallest and largest numbers representable by
|
||
signed integer type TYPE. */
|
||
|
||
void
|
||
get_signed_type_minmax (struct type *type, LONGEST *min, LONGEST *max)
|
||
{
|
||
unsigned int n;
|
||
|
||
type = check_typedef (type);
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_INT && !TYPE_UNSIGNED (type));
|
||
gdb_assert (TYPE_LENGTH (type) <= sizeof (LONGEST));
|
||
|
||
n = TYPE_LENGTH (type) * TARGET_CHAR_BIT;
|
||
*min = -((ULONGEST) 1 << (n - 1));
|
||
*max = ((ULONGEST) 1 << (n - 1)) - 1;
|
||
}
|
||
|
||
/* Internal routine called by TYPE_VPTR_FIELDNO to return the value of
|
||
cplus_stuff.vptr_fieldno.
|
||
|
||
cplus_stuff is initialized to cplus_struct_default which does not
|
||
set vptr_fieldno to -1 for portability reasons (IWBN to use C99
|
||
designated initializers). We cope with that here. */
|
||
|
||
int
|
||
internal_type_vptr_fieldno (struct type *type)
|
||
{
|
||
type = check_typedef (type);
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_STRUCT
|
||
|| TYPE_CODE (type) == TYPE_CODE_UNION);
|
||
if (!HAVE_CPLUS_STRUCT (type))
|
||
return -1;
|
||
return TYPE_RAW_CPLUS_SPECIFIC (type)->vptr_fieldno;
|
||
}
|
||
|
||
/* Set the value of cplus_stuff.vptr_fieldno. */
|
||
|
||
void
|
||
set_type_vptr_fieldno (struct type *type, int fieldno)
|
||
{
|
||
type = check_typedef (type);
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_STRUCT
|
||
|| TYPE_CODE (type) == TYPE_CODE_UNION);
|
||
if (!HAVE_CPLUS_STRUCT (type))
|
||
ALLOCATE_CPLUS_STRUCT_TYPE (type);
|
||
TYPE_RAW_CPLUS_SPECIFIC (type)->vptr_fieldno = fieldno;
|
||
}
|
||
|
||
/* Internal routine called by TYPE_VPTR_BASETYPE to return the value of
|
||
cplus_stuff.vptr_basetype. */
|
||
|
||
struct type *
|
||
internal_type_vptr_basetype (struct type *type)
|
||
{
|
||
type = check_typedef (type);
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_STRUCT
|
||
|| TYPE_CODE (type) == TYPE_CODE_UNION);
|
||
gdb_assert (TYPE_SPECIFIC_FIELD (type) == TYPE_SPECIFIC_CPLUS_STUFF);
|
||
return TYPE_RAW_CPLUS_SPECIFIC (type)->vptr_basetype;
|
||
}
|
||
|
||
/* Set the value of cplus_stuff.vptr_basetype. */
|
||
|
||
void
|
||
set_type_vptr_basetype (struct type *type, struct type *basetype)
|
||
{
|
||
type = check_typedef (type);
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_STRUCT
|
||
|| TYPE_CODE (type) == TYPE_CODE_UNION);
|
||
if (!HAVE_CPLUS_STRUCT (type))
|
||
ALLOCATE_CPLUS_STRUCT_TYPE (type);
|
||
TYPE_RAW_CPLUS_SPECIFIC (type)->vptr_basetype = basetype;
|
||
}
|
||
|
||
/* Lookup the vptr basetype/fieldno values for TYPE.
|
||
If found store vptr_basetype in *BASETYPEP if non-NULL, and return
|
||
vptr_fieldno. Also, if found and basetype is from the same objfile,
|
||
cache the results.
|
||
If not found, return -1 and ignore BASETYPEP.
|
||
Callers should be aware that in some cases (for example,
|
||
the type or one of its baseclasses is a stub type and we are
|
||
debugging a .o file, or the compiler uses DWARF-2 and is not GCC),
|
||
this function will not be able to find the
|
||
virtual function table pointer, and vptr_fieldno will remain -1 and
|
||
vptr_basetype will remain NULL or incomplete. */
|
||
|
||
int
|
||
get_vptr_fieldno (struct type *type, struct type **basetypep)
|
||
{
|
||
type = check_typedef (type);
|
||
|
||
if (TYPE_VPTR_FIELDNO (type) < 0)
|
||
{
|
||
int i;
|
||
|
||
/* We must start at zero in case the first (and only) baseclass
|
||
is virtual (and hence we cannot share the table pointer). */
|
||
for (i = 0; i < TYPE_N_BASECLASSES (type); i++)
|
||
{
|
||
struct type *baseclass = check_typedef (TYPE_BASECLASS (type, i));
|
||
int fieldno;
|
||
struct type *basetype;
|
||
|
||
fieldno = get_vptr_fieldno (baseclass, &basetype);
|
||
if (fieldno >= 0)
|
||
{
|
||
/* If the type comes from a different objfile we can't cache
|
||
it, it may have a different lifetime. PR 2384 */
|
||
if (TYPE_OBJFILE (type) == TYPE_OBJFILE (basetype))
|
||
{
|
||
set_type_vptr_fieldno (type, fieldno);
|
||
set_type_vptr_basetype (type, basetype);
|
||
}
|
||
if (basetypep)
|
||
*basetypep = basetype;
|
||
return fieldno;
|
||
}
|
||
}
|
||
|
||
/* Not found. */
|
||
return -1;
|
||
}
|
||
else
|
||
{
|
||
if (basetypep)
|
||
*basetypep = TYPE_VPTR_BASETYPE (type);
|
||
return TYPE_VPTR_FIELDNO (type);
|
||
}
|
||
}
|
||
|
||
static void
|
||
stub_noname_complaint (void)
|
||
{
|
||
complaint (&symfile_complaints, _("stub type has NULL name"));
|
||
}
|
||
|
||
/* Worker for is_dynamic_type. */
|
||
|
||
static int
|
||
is_dynamic_type_internal (struct type *type, int top_level)
|
||
{
|
||
type = check_typedef (type);
|
||
|
||
/* We only want to recognize references at the outermost level. */
|
||
if (top_level && TYPE_CODE (type) == TYPE_CODE_REF)
|
||
type = check_typedef (TYPE_TARGET_TYPE (type));
|
||
|
||
/* Types that have a dynamic TYPE_DATA_LOCATION are considered
|
||
dynamic, even if the type itself is statically defined.
|
||
From a user's point of view, this may appear counter-intuitive;
|
||
but it makes sense in this context, because the point is to determine
|
||
whether any part of the type needs to be resolved before it can
|
||
be exploited. */
|
||
if (TYPE_DATA_LOCATION (type) != NULL
|
||
&& (TYPE_DATA_LOCATION_KIND (type) == PROP_LOCEXPR
|
||
|| TYPE_DATA_LOCATION_KIND (type) == PROP_LOCLIST))
|
||
return 1;
|
||
|
||
if (TYPE_ASSOCIATED_PROP (type))
|
||
return 1;
|
||
|
||
if (TYPE_ALLOCATED_PROP (type))
|
||
return 1;
|
||
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_RANGE:
|
||
{
|
||
/* A range type is obviously dynamic if it has at least one
|
||
dynamic bound. But also consider the range type to be
|
||
dynamic when its subtype is dynamic, even if the bounds
|
||
of the range type are static. It allows us to assume that
|
||
the subtype of a static range type is also static. */
|
||
return (!has_static_range (TYPE_RANGE_DATA (type))
|
||
|| is_dynamic_type_internal (TYPE_TARGET_TYPE (type), 0));
|
||
}
|
||
|
||
case TYPE_CODE_ARRAY:
|
||
{
|
||
gdb_assert (TYPE_NFIELDS (type) == 1);
|
||
|
||
/* The array is dynamic if either the bounds are dynamic,
|
||
or the elements it contains have a dynamic contents. */
|
||
if (is_dynamic_type_internal (TYPE_INDEX_TYPE (type), 0))
|
||
return 1;
|
||
return is_dynamic_type_internal (TYPE_TARGET_TYPE (type), 0);
|
||
}
|
||
|
||
case TYPE_CODE_STRUCT:
|
||
case TYPE_CODE_UNION:
|
||
{
|
||
int i;
|
||
|
||
for (i = 0; i < TYPE_NFIELDS (type); ++i)
|
||
if (!field_is_static (&TYPE_FIELD (type, i))
|
||
&& is_dynamic_type_internal (TYPE_FIELD_TYPE (type, i), 0))
|
||
return 1;
|
||
}
|
||
break;
|
||
}
|
||
|
||
return 0;
|
||
}
|
||
|
||
/* See gdbtypes.h. */
|
||
|
||
int
|
||
is_dynamic_type (struct type *type)
|
||
{
|
||
return is_dynamic_type_internal (type, 1);
|
||
}
|
||
|
||
static struct type *resolve_dynamic_type_internal
|
||
(struct type *type, struct property_addr_info *addr_stack, int top_level);
|
||
|
||
/* Given a dynamic range type (dyn_range_type) and a stack of
|
||
struct property_addr_info elements, return a static version
|
||
of that type. */
|
||
|
||
static struct type *
|
||
resolve_dynamic_range (struct type *dyn_range_type,
|
||
struct property_addr_info *addr_stack)
|
||
{
|
||
CORE_ADDR value;
|
||
struct type *static_range_type, *static_target_type;
|
||
const struct dynamic_prop *prop;
|
||
struct dynamic_prop low_bound, high_bound;
|
||
|
||
gdb_assert (TYPE_CODE (dyn_range_type) == TYPE_CODE_RANGE);
|
||
|
||
prop = &TYPE_RANGE_DATA (dyn_range_type)->low;
|
||
if (dwarf2_evaluate_property (prop, NULL, addr_stack, &value))
|
||
{
|
||
low_bound.kind = PROP_CONST;
|
||
low_bound.data.const_val = value;
|
||
}
|
||
else
|
||
{
|
||
low_bound.kind = PROP_UNDEFINED;
|
||
low_bound.data.const_val = 0;
|
||
}
|
||
|
||
prop = &TYPE_RANGE_DATA (dyn_range_type)->high;
|
||
if (dwarf2_evaluate_property (prop, NULL, addr_stack, &value))
|
||
{
|
||
high_bound.kind = PROP_CONST;
|
||
high_bound.data.const_val = value;
|
||
|
||
if (TYPE_RANGE_DATA (dyn_range_type)->flag_upper_bound_is_count)
|
||
high_bound.data.const_val
|
||
= low_bound.data.const_val + high_bound.data.const_val - 1;
|
||
}
|
||
else
|
||
{
|
||
high_bound.kind = PROP_UNDEFINED;
|
||
high_bound.data.const_val = 0;
|
||
}
|
||
|
||
static_target_type
|
||
= resolve_dynamic_type_internal (TYPE_TARGET_TYPE (dyn_range_type),
|
||
addr_stack, 0);
|
||
static_range_type = create_range_type (copy_type (dyn_range_type),
|
||
static_target_type,
|
||
&low_bound, &high_bound);
|
||
TYPE_RANGE_DATA (static_range_type)->flag_bound_evaluated = 1;
|
||
return static_range_type;
|
||
}
|
||
|
||
/* Resolves dynamic bound values of an array type TYPE to static ones.
|
||
ADDR_STACK is a stack of struct property_addr_info to be used
|
||
if needed during the dynamic resolution. */
|
||
|
||
static struct type *
|
||
resolve_dynamic_array (struct type *type,
|
||
struct property_addr_info *addr_stack)
|
||
{
|
||
CORE_ADDR value;
|
||
struct type *elt_type;
|
||
struct type *range_type;
|
||
struct type *ary_dim;
|
||
struct dynamic_prop *prop;
|
||
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_ARRAY);
|
||
|
||
type = copy_type (type);
|
||
|
||
elt_type = type;
|
||
range_type = check_typedef (TYPE_INDEX_TYPE (elt_type));
|
||
range_type = resolve_dynamic_range (range_type, addr_stack);
|
||
|
||
/* Resolve allocated/associated here before creating a new array type, which
|
||
will update the length of the array accordingly. */
|
||
prop = TYPE_ALLOCATED_PROP (type);
|
||
if (prop != NULL && dwarf2_evaluate_property (prop, NULL, addr_stack, &value))
|
||
{
|
||
TYPE_DYN_PROP_ADDR (prop) = value;
|
||
TYPE_DYN_PROP_KIND (prop) = PROP_CONST;
|
||
}
|
||
prop = TYPE_ASSOCIATED_PROP (type);
|
||
if (prop != NULL && dwarf2_evaluate_property (prop, NULL, addr_stack, &value))
|
||
{
|
||
TYPE_DYN_PROP_ADDR (prop) = value;
|
||
TYPE_DYN_PROP_KIND (prop) = PROP_CONST;
|
||
}
|
||
|
||
ary_dim = check_typedef (TYPE_TARGET_TYPE (elt_type));
|
||
|
||
if (ary_dim != NULL && TYPE_CODE (ary_dim) == TYPE_CODE_ARRAY)
|
||
elt_type = resolve_dynamic_array (ary_dim, addr_stack);
|
||
else
|
||
elt_type = TYPE_TARGET_TYPE (type);
|
||
|
||
return create_array_type_with_stride (type, elt_type, range_type,
|
||
TYPE_FIELD_BITSIZE (type, 0));
|
||
}
|
||
|
||
/* Resolve dynamic bounds of members of the union TYPE to static
|
||
bounds. ADDR_STACK is a stack of struct property_addr_info
|
||
to be used if needed during the dynamic resolution. */
|
||
|
||
static struct type *
|
||
resolve_dynamic_union (struct type *type,
|
||
struct property_addr_info *addr_stack)
|
||
{
|
||
struct type *resolved_type;
|
||
int i;
|
||
unsigned int max_len = 0;
|
||
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_UNION);
|
||
|
||
resolved_type = copy_type (type);
|
||
TYPE_FIELDS (resolved_type)
|
||
= (struct field *) TYPE_ALLOC (resolved_type,
|
||
TYPE_NFIELDS (resolved_type)
|
||
* sizeof (struct field));
|
||
memcpy (TYPE_FIELDS (resolved_type),
|
||
TYPE_FIELDS (type),
|
||
TYPE_NFIELDS (resolved_type) * sizeof (struct field));
|
||
for (i = 0; i < TYPE_NFIELDS (resolved_type); ++i)
|
||
{
|
||
struct type *t;
|
||
|
||
if (field_is_static (&TYPE_FIELD (type, i)))
|
||
continue;
|
||
|
||
t = resolve_dynamic_type_internal (TYPE_FIELD_TYPE (resolved_type, i),
|
||
addr_stack, 0);
|
||
TYPE_FIELD_TYPE (resolved_type, i) = t;
|
||
if (TYPE_LENGTH (t) > max_len)
|
||
max_len = TYPE_LENGTH (t);
|
||
}
|
||
|
||
TYPE_LENGTH (resolved_type) = max_len;
|
||
return resolved_type;
|
||
}
|
||
|
||
/* Resolve dynamic bounds of members of the struct TYPE to static
|
||
bounds. ADDR_STACK is a stack of struct property_addr_info to
|
||
be used if needed during the dynamic resolution. */
|
||
|
||
static struct type *
|
||
resolve_dynamic_struct (struct type *type,
|
||
struct property_addr_info *addr_stack)
|
||
{
|
||
struct type *resolved_type;
|
||
int i;
|
||
unsigned resolved_type_bit_length = 0;
|
||
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_STRUCT);
|
||
gdb_assert (TYPE_NFIELDS (type) > 0);
|
||
|
||
resolved_type = copy_type (type);
|
||
TYPE_FIELDS (resolved_type)
|
||
= (struct field *) TYPE_ALLOC (resolved_type,
|
||
TYPE_NFIELDS (resolved_type)
|
||
* sizeof (struct field));
|
||
memcpy (TYPE_FIELDS (resolved_type),
|
||
TYPE_FIELDS (type),
|
||
TYPE_NFIELDS (resolved_type) * sizeof (struct field));
|
||
for (i = 0; i < TYPE_NFIELDS (resolved_type); ++i)
|
||
{
|
||
unsigned new_bit_length;
|
||
struct property_addr_info pinfo;
|
||
|
||
if (field_is_static (&TYPE_FIELD (type, i)))
|
||
continue;
|
||
|
||
/* As we know this field is not a static field, the field's
|
||
field_loc_kind should be FIELD_LOC_KIND_BITPOS. Verify
|
||
this is the case, but only trigger a simple error rather
|
||
than an internal error if that fails. While failing
|
||
that verification indicates a bug in our code, the error
|
||
is not severe enough to suggest to the user he stops
|
||
his debugging session because of it. */
|
||
if (TYPE_FIELD_LOC_KIND (type, i) != FIELD_LOC_KIND_BITPOS)
|
||
error (_("Cannot determine struct field location"
|
||
" (invalid location kind)"));
|
||
|
||
pinfo.type = check_typedef (TYPE_FIELD_TYPE (type, i));
|
||
pinfo.valaddr = addr_stack->valaddr;
|
||
pinfo.addr
|
||
= (addr_stack->addr
|
||
+ (TYPE_FIELD_BITPOS (resolved_type, i) / TARGET_CHAR_BIT));
|
||
pinfo.next = addr_stack;
|
||
|
||
TYPE_FIELD_TYPE (resolved_type, i)
|
||
= resolve_dynamic_type_internal (TYPE_FIELD_TYPE (resolved_type, i),
|
||
&pinfo, 0);
|
||
gdb_assert (TYPE_FIELD_LOC_KIND (resolved_type, i)
|
||
== FIELD_LOC_KIND_BITPOS);
|
||
|
||
new_bit_length = TYPE_FIELD_BITPOS (resolved_type, i);
|
||
if (TYPE_FIELD_BITSIZE (resolved_type, i) != 0)
|
||
new_bit_length += TYPE_FIELD_BITSIZE (resolved_type, i);
|
||
else
|
||
new_bit_length += (TYPE_LENGTH (TYPE_FIELD_TYPE (resolved_type, i))
|
||
* TARGET_CHAR_BIT);
|
||
|
||
/* Normally, we would use the position and size of the last field
|
||
to determine the size of the enclosing structure. But GCC seems
|
||
to be encoding the position of some fields incorrectly when
|
||
the struct contains a dynamic field that is not placed last.
|
||
So we compute the struct size based on the field that has
|
||
the highest position + size - probably the best we can do. */
|
||
if (new_bit_length > resolved_type_bit_length)
|
||
resolved_type_bit_length = new_bit_length;
|
||
}
|
||
|
||
/* The length of a type won't change for fortran, but it does for C and Ada.
|
||
For fortran the size of dynamic fields might change over time but not the
|
||
type length of the structure. If we adapt it, we run into problems
|
||
when calculating the element offset for arrays of structs. */
|
||
if (current_language->la_language != language_fortran)
|
||
TYPE_LENGTH (resolved_type)
|
||
= (resolved_type_bit_length + TARGET_CHAR_BIT - 1) / TARGET_CHAR_BIT;
|
||
|
||
/* The Ada language uses this field as a cache for static fixed types: reset
|
||
it as RESOLVED_TYPE must have its own static fixed type. */
|
||
TYPE_TARGET_TYPE (resolved_type) = NULL;
|
||
|
||
return resolved_type;
|
||
}
|
||
|
||
/* Worker for resolved_dynamic_type. */
|
||
|
||
static struct type *
|
||
resolve_dynamic_type_internal (struct type *type,
|
||
struct property_addr_info *addr_stack,
|
||
int top_level)
|
||
{
|
||
struct type *real_type = check_typedef (type);
|
||
struct type *resolved_type = type;
|
||
struct dynamic_prop *prop;
|
||
CORE_ADDR value;
|
||
|
||
if (!is_dynamic_type_internal (real_type, top_level))
|
||
return type;
|
||
|
||
if (TYPE_CODE (type) == TYPE_CODE_TYPEDEF)
|
||
{
|
||
resolved_type = copy_type (type);
|
||
TYPE_TARGET_TYPE (resolved_type)
|
||
= resolve_dynamic_type_internal (TYPE_TARGET_TYPE (type), addr_stack,
|
||
top_level);
|
||
}
|
||
else
|
||
{
|
||
/* Before trying to resolve TYPE, make sure it is not a stub. */
|
||
type = real_type;
|
||
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_REF:
|
||
{
|
||
struct property_addr_info pinfo;
|
||
|
||
pinfo.type = check_typedef (TYPE_TARGET_TYPE (type));
|
||
pinfo.valaddr = NULL;
|
||
if (addr_stack->valaddr != NULL)
|
||
pinfo.addr = extract_typed_address (addr_stack->valaddr, type);
|
||
else
|
||
pinfo.addr = read_memory_typed_address (addr_stack->addr, type);
|
||
pinfo.next = addr_stack;
|
||
|
||
resolved_type = copy_type (type);
|
||
TYPE_TARGET_TYPE (resolved_type)
|
||
= resolve_dynamic_type_internal (TYPE_TARGET_TYPE (type),
|
||
&pinfo, top_level);
|
||
break;
|
||
}
|
||
|
||
case TYPE_CODE_ARRAY:
|
||
resolved_type = resolve_dynamic_array (type, addr_stack);
|
||
break;
|
||
|
||
case TYPE_CODE_RANGE:
|
||
resolved_type = resolve_dynamic_range (type, addr_stack);
|
||
break;
|
||
|
||
case TYPE_CODE_UNION:
|
||
resolved_type = resolve_dynamic_union (type, addr_stack);
|
||
break;
|
||
|
||
case TYPE_CODE_STRUCT:
|
||
resolved_type = resolve_dynamic_struct (type, addr_stack);
|
||
break;
|
||
}
|
||
}
|
||
|
||
/* Resolve data_location attribute. */
|
||
prop = TYPE_DATA_LOCATION (resolved_type);
|
||
if (prop != NULL
|
||
&& dwarf2_evaluate_property (prop, NULL, addr_stack, &value))
|
||
{
|
||
TYPE_DYN_PROP_ADDR (prop) = value;
|
||
TYPE_DYN_PROP_KIND (prop) = PROP_CONST;
|
||
}
|
||
|
||
return resolved_type;
|
||
}
|
||
|
||
/* See gdbtypes.h */
|
||
|
||
struct type *
|
||
resolve_dynamic_type (struct type *type, const gdb_byte *valaddr,
|
||
CORE_ADDR addr)
|
||
{
|
||
struct property_addr_info pinfo
|
||
= {check_typedef (type), valaddr, addr, NULL};
|
||
|
||
return resolve_dynamic_type_internal (type, &pinfo, 1);
|
||
}
|
||
|
||
/* See gdbtypes.h */
|
||
|
||
struct dynamic_prop *
|
||
get_dyn_prop (enum dynamic_prop_node_kind prop_kind, const struct type *type)
|
||
{
|
||
struct dynamic_prop_list *node = TYPE_DYN_PROP_LIST (type);
|
||
|
||
while (node != NULL)
|
||
{
|
||
if (node->prop_kind == prop_kind)
|
||
return &node->prop;
|
||
node = node->next;
|
||
}
|
||
return NULL;
|
||
}
|
||
|
||
/* See gdbtypes.h */
|
||
|
||
void
|
||
add_dyn_prop (enum dynamic_prop_node_kind prop_kind, struct dynamic_prop prop,
|
||
struct type *type, struct objfile *objfile)
|
||
{
|
||
struct dynamic_prop_list *temp;
|
||
|
||
gdb_assert (TYPE_OBJFILE_OWNED (type));
|
||
|
||
temp = XOBNEW (&objfile->objfile_obstack, struct dynamic_prop_list);
|
||
temp->prop_kind = prop_kind;
|
||
temp->prop = prop;
|
||
temp->next = TYPE_DYN_PROP_LIST (type);
|
||
|
||
TYPE_DYN_PROP_LIST (type) = temp;
|
||
}
|
||
|
||
/* Remove dynamic property from TYPE in case it exists. */
|
||
|
||
void
|
||
remove_dyn_prop (enum dynamic_prop_node_kind prop_kind,
|
||
struct type *type)
|
||
{
|
||
struct dynamic_prop_list *prev_node, *curr_node;
|
||
|
||
curr_node = TYPE_DYN_PROP_LIST (type);
|
||
prev_node = NULL;
|
||
|
||
while (NULL != curr_node)
|
||
{
|
||
if (curr_node->prop_kind == prop_kind)
|
||
{
|
||
/* Update the linked list but don't free anything.
|
||
The property was allocated on objstack and it is not known
|
||
if we are on top of it. Nevertheless, everything is released
|
||
when the complete objstack is freed. */
|
||
if (NULL == prev_node)
|
||
TYPE_DYN_PROP_LIST (type) = curr_node->next;
|
||
else
|
||
prev_node->next = curr_node->next;
|
||
|
||
return;
|
||
}
|
||
|
||
prev_node = curr_node;
|
||
curr_node = curr_node->next;
|
||
}
|
||
}
|
||
|
||
/* Find the real type of TYPE. This function returns the real type,
|
||
after removing all layers of typedefs, and completing opaque or stub
|
||
types. Completion changes the TYPE argument, but stripping of
|
||
typedefs does not.
|
||
|
||
Instance flags (e.g. const/volatile) are preserved as typedefs are
|
||
stripped. If necessary a new qualified form of the underlying type
|
||
is created.
|
||
|
||
NOTE: This will return a typedef if TYPE_TARGET_TYPE for the typedef has
|
||
not been computed and we're either in the middle of reading symbols, or
|
||
there was no name for the typedef in the debug info.
|
||
|
||
NOTE: Lookup of opaque types can throw errors for invalid symbol files.
|
||
QUITs in the symbol reading code can also throw.
|
||
Thus this function can throw an exception.
|
||
|
||
If TYPE is a TYPE_CODE_TYPEDEF, its length is updated to the length of
|
||
the target type.
|
||
|
||
If this is a stubbed struct (i.e. declared as struct foo *), see if
|
||
we can find a full definition in some other file. If so, copy this
|
||
definition, so we can use it in future. There used to be a comment
|
||
(but not any code) that if we don't find a full definition, we'd
|
||
set a flag so we don't spend time in the future checking the same
|
||
type. That would be a mistake, though--we might load in more
|
||
symbols which contain a full definition for the type. */
|
||
|
||
struct type *
|
||
check_typedef (struct type *type)
|
||
{
|
||
struct type *orig_type = type;
|
||
/* While we're removing typedefs, we don't want to lose qualifiers.
|
||
E.g., const/volatile. */
|
||
int instance_flags = TYPE_INSTANCE_FLAGS (type);
|
||
|
||
gdb_assert (type);
|
||
|
||
while (TYPE_CODE (type) == TYPE_CODE_TYPEDEF)
|
||
{
|
||
if (!TYPE_TARGET_TYPE (type))
|
||
{
|
||
const char *name;
|
||
struct symbol *sym;
|
||
|
||
/* It is dangerous to call lookup_symbol if we are currently
|
||
reading a symtab. Infinite recursion is one danger. */
|
||
if (currently_reading_symtab)
|
||
return make_qualified_type (type, instance_flags, NULL);
|
||
|
||
name = type_name_no_tag (type);
|
||
/* FIXME: shouldn't we separately check the TYPE_NAME and
|
||
the TYPE_TAG_NAME, and look in STRUCT_DOMAIN and/or
|
||
VAR_DOMAIN as appropriate? (this code was written before
|
||
TYPE_NAME and TYPE_TAG_NAME were separate). */
|
||
if (name == NULL)
|
||
{
|
||
stub_noname_complaint ();
|
||
return make_qualified_type (type, instance_flags, NULL);
|
||
}
|
||
sym = lookup_symbol (name, 0, STRUCT_DOMAIN, 0).symbol;
|
||
if (sym)
|
||
TYPE_TARGET_TYPE (type) = SYMBOL_TYPE (sym);
|
||
else /* TYPE_CODE_UNDEF */
|
||
TYPE_TARGET_TYPE (type) = alloc_type_arch (get_type_arch (type));
|
||
}
|
||
type = TYPE_TARGET_TYPE (type);
|
||
|
||
/* Preserve the instance flags as we traverse down the typedef chain.
|
||
|
||
Handling address spaces/classes is nasty, what do we do if there's a
|
||
conflict?
|
||
E.g., what if an outer typedef marks the type as class_1 and an inner
|
||
typedef marks the type as class_2?
|
||
This is the wrong place to do such error checking. We leave it to
|
||
the code that created the typedef in the first place to flag the
|
||
error. We just pick the outer address space (akin to letting the
|
||
outer cast in a chain of casting win), instead of assuming
|
||
"it can't happen". */
|
||
{
|
||
const int ALL_SPACES = (TYPE_INSTANCE_FLAG_CODE_SPACE
|
||
| TYPE_INSTANCE_FLAG_DATA_SPACE);
|
||
const int ALL_CLASSES = TYPE_INSTANCE_FLAG_ADDRESS_CLASS_ALL;
|
||
int new_instance_flags = TYPE_INSTANCE_FLAGS (type);
|
||
|
||
/* Treat code vs data spaces and address classes separately. */
|
||
if ((instance_flags & ALL_SPACES) != 0)
|
||
new_instance_flags &= ~ALL_SPACES;
|
||
if ((instance_flags & ALL_CLASSES) != 0)
|
||
new_instance_flags &= ~ALL_CLASSES;
|
||
|
||
instance_flags |= new_instance_flags;
|
||
}
|
||
}
|
||
|
||
/* If this is a struct/class/union with no fields, then check
|
||
whether a full definition exists somewhere else. This is for
|
||
systems where a type definition with no fields is issued for such
|
||
types, instead of identifying them as stub types in the first
|
||
place. */
|
||
|
||
if (TYPE_IS_OPAQUE (type)
|
||
&& opaque_type_resolution
|
||
&& !currently_reading_symtab)
|
||
{
|
||
const char *name = type_name_no_tag (type);
|
||
struct type *newtype;
|
||
|
||
if (name == NULL)
|
||
{
|
||
stub_noname_complaint ();
|
||
return make_qualified_type (type, instance_flags, NULL);
|
||
}
|
||
newtype = lookup_transparent_type (name);
|
||
|
||
if (newtype)
|
||
{
|
||
/* If the resolved type and the stub are in the same
|
||
objfile, then replace the stub type with the real deal.
|
||
But if they're in separate objfiles, leave the stub
|
||
alone; we'll just look up the transparent type every time
|
||
we call check_typedef. We can't create pointers between
|
||
types allocated to different objfiles, since they may
|
||
have different lifetimes. Trying to copy NEWTYPE over to
|
||
TYPE's objfile is pointless, too, since you'll have to
|
||
move over any other types NEWTYPE refers to, which could
|
||
be an unbounded amount of stuff. */
|
||
if (TYPE_OBJFILE (newtype) == TYPE_OBJFILE (type))
|
||
type = make_qualified_type (newtype,
|
||
TYPE_INSTANCE_FLAGS (type),
|
||
type);
|
||
else
|
||
type = newtype;
|
||
}
|
||
}
|
||
/* Otherwise, rely on the stub flag being set for opaque/stubbed
|
||
types. */
|
||
else if (TYPE_STUB (type) && !currently_reading_symtab)
|
||
{
|
||
const char *name = type_name_no_tag (type);
|
||
/* FIXME: shouldn't we separately check the TYPE_NAME and the
|
||
TYPE_TAG_NAME, and look in STRUCT_DOMAIN and/or VAR_DOMAIN
|
||
as appropriate? (this code was written before TYPE_NAME and
|
||
TYPE_TAG_NAME were separate). */
|
||
struct symbol *sym;
|
||
|
||
if (name == NULL)
|
||
{
|
||
stub_noname_complaint ();
|
||
return make_qualified_type (type, instance_flags, NULL);
|
||
}
|
||
sym = lookup_symbol (name, 0, STRUCT_DOMAIN, 0).symbol;
|
||
if (sym)
|
||
{
|
||
/* Same as above for opaque types, we can replace the stub
|
||
with the complete type only if they are in the same
|
||
objfile. */
|
||
if (TYPE_OBJFILE (SYMBOL_TYPE(sym)) == TYPE_OBJFILE (type))
|
||
type = make_qualified_type (SYMBOL_TYPE (sym),
|
||
TYPE_INSTANCE_FLAGS (type),
|
||
type);
|
||
else
|
||
type = SYMBOL_TYPE (sym);
|
||
}
|
||
}
|
||
|
||
if (TYPE_TARGET_STUB (type))
|
||
{
|
||
struct type *target_type = check_typedef (TYPE_TARGET_TYPE (type));
|
||
|
||
if (TYPE_STUB (target_type) || TYPE_TARGET_STUB (target_type))
|
||
{
|
||
/* Nothing we can do. */
|
||
}
|
||
else if (TYPE_CODE (type) == TYPE_CODE_RANGE)
|
||
{
|
||
TYPE_LENGTH (type) = TYPE_LENGTH (target_type);
|
||
TYPE_TARGET_STUB (type) = 0;
|
||
}
|
||
}
|
||
|
||
type = make_qualified_type (type, instance_flags, NULL);
|
||
|
||
/* Cache TYPE_LENGTH for future use. */
|
||
TYPE_LENGTH (orig_type) = TYPE_LENGTH (type);
|
||
|
||
return type;
|
||
}
|
||
|
||
/* Parse a type expression in the string [P..P+LENGTH). If an error
|
||
occurs, silently return a void type. */
|
||
|
||
static struct type *
|
||
safe_parse_type (struct gdbarch *gdbarch, char *p, int length)
|
||
{
|
||
struct ui_file *saved_gdb_stderr;
|
||
struct type *type = NULL; /* Initialize to keep gcc happy. */
|
||
|
||
/* Suppress error messages. */
|
||
saved_gdb_stderr = gdb_stderr;
|
||
gdb_stderr = ui_file_new ();
|
||
|
||
/* Call parse_and_eval_type() without fear of longjmp()s. */
|
||
TRY
|
||
{
|
||
type = parse_and_eval_type (p, length);
|
||
}
|
||
CATCH (except, RETURN_MASK_ERROR)
|
||
{
|
||
type = builtin_type (gdbarch)->builtin_void;
|
||
}
|
||
END_CATCH
|
||
|
||
/* Stop suppressing error messages. */
|
||
ui_file_delete (gdb_stderr);
|
||
gdb_stderr = saved_gdb_stderr;
|
||
|
||
return type;
|
||
}
|
||
|
||
/* Ugly hack to convert method stubs into method types.
|
||
|
||
He ain't kiddin'. This demangles the name of the method into a
|
||
string including argument types, parses out each argument type,
|
||
generates a string casting a zero to that type, evaluates the
|
||
string, and stuffs the resulting type into an argtype vector!!!
|
||
Then it knows the type of the whole function (including argument
|
||
types for overloading), which info used to be in the stab's but was
|
||
removed to hack back the space required for them. */
|
||
|
||
static void
|
||
check_stub_method (struct type *type, int method_id, int signature_id)
|
||
{
|
||
struct gdbarch *gdbarch = get_type_arch (type);
|
||
struct fn_field *f;
|
||
char *mangled_name = gdb_mangle_name (type, method_id, signature_id);
|
||
char *demangled_name = gdb_demangle (mangled_name,
|
||
DMGL_PARAMS | DMGL_ANSI);
|
||
char *argtypetext, *p;
|
||
int depth = 0, argcount = 1;
|
||
struct field *argtypes;
|
||
struct type *mtype;
|
||
|
||
/* Make sure we got back a function string that we can use. */
|
||
if (demangled_name)
|
||
p = strchr (demangled_name, '(');
|
||
else
|
||
p = NULL;
|
||
|
||
if (demangled_name == NULL || p == NULL)
|
||
error (_("Internal: Cannot demangle mangled name `%s'."),
|
||
mangled_name);
|
||
|
||
/* Now, read in the parameters that define this type. */
|
||
p += 1;
|
||
argtypetext = p;
|
||
while (*p)
|
||
{
|
||
if (*p == '(' || *p == '<')
|
||
{
|
||
depth += 1;
|
||
}
|
||
else if (*p == ')' || *p == '>')
|
||
{
|
||
depth -= 1;
|
||
}
|
||
else if (*p == ',' && depth == 0)
|
||
{
|
||
argcount += 1;
|
||
}
|
||
|
||
p += 1;
|
||
}
|
||
|
||
/* If we read one argument and it was ``void'', don't count it. */
|
||
if (startswith (argtypetext, "(void)"))
|
||
argcount -= 1;
|
||
|
||
/* We need one extra slot, for the THIS pointer. */
|
||
|
||
argtypes = (struct field *)
|
||
TYPE_ALLOC (type, (argcount + 1) * sizeof (struct field));
|
||
p = argtypetext;
|
||
|
||
/* Add THIS pointer for non-static methods. */
|
||
f = TYPE_FN_FIELDLIST1 (type, method_id);
|
||
if (TYPE_FN_FIELD_STATIC_P (f, signature_id))
|
||
argcount = 0;
|
||
else
|
||
{
|
||
argtypes[0].type = lookup_pointer_type (type);
|
||
argcount = 1;
|
||
}
|
||
|
||
if (*p != ')') /* () means no args, skip while. */
|
||
{
|
||
depth = 0;
|
||
while (*p)
|
||
{
|
||
if (depth <= 0 && (*p == ',' || *p == ')'))
|
||
{
|
||
/* Avoid parsing of ellipsis, they will be handled below.
|
||
Also avoid ``void'' as above. */
|
||
if (strncmp (argtypetext, "...", p - argtypetext) != 0
|
||
&& strncmp (argtypetext, "void", p - argtypetext) != 0)
|
||
{
|
||
argtypes[argcount].type =
|
||
safe_parse_type (gdbarch, argtypetext, p - argtypetext);
|
||
argcount += 1;
|
||
}
|
||
argtypetext = p + 1;
|
||
}
|
||
|
||
if (*p == '(' || *p == '<')
|
||
{
|
||
depth += 1;
|
||
}
|
||
else if (*p == ')' || *p == '>')
|
||
{
|
||
depth -= 1;
|
||
}
|
||
|
||
p += 1;
|
||
}
|
||
}
|
||
|
||
TYPE_FN_FIELD_PHYSNAME (f, signature_id) = mangled_name;
|
||
|
||
/* Now update the old "stub" type into a real type. */
|
||
mtype = TYPE_FN_FIELD_TYPE (f, signature_id);
|
||
/* MTYPE may currently be a function (TYPE_CODE_FUNC).
|
||
We want a method (TYPE_CODE_METHOD). */
|
||
smash_to_method_type (mtype, type, TYPE_TARGET_TYPE (mtype),
|
||
argtypes, argcount, p[-2] == '.');
|
||
TYPE_STUB (mtype) = 0;
|
||
TYPE_FN_FIELD_STUB (f, signature_id) = 0;
|
||
|
||
xfree (demangled_name);
|
||
}
|
||
|
||
/* This is the external interface to check_stub_method, above. This
|
||
function unstubs all of the signatures for TYPE's METHOD_ID method
|
||
name. After calling this function TYPE_FN_FIELD_STUB will be
|
||
cleared for each signature and TYPE_FN_FIELDLIST_NAME will be
|
||
correct.
|
||
|
||
This function unfortunately can not die until stabs do. */
|
||
|
||
void
|
||
check_stub_method_group (struct type *type, int method_id)
|
||
{
|
||
int len = TYPE_FN_FIELDLIST_LENGTH (type, method_id);
|
||
struct fn_field *f = TYPE_FN_FIELDLIST1 (type, method_id);
|
||
int j, found_stub = 0;
|
||
|
||
for (j = 0; j < len; j++)
|
||
if (TYPE_FN_FIELD_STUB (f, j))
|
||
{
|
||
found_stub = 1;
|
||
check_stub_method (type, method_id, j);
|
||
}
|
||
|
||
/* GNU v3 methods with incorrect names were corrected when we read
|
||
in type information, because it was cheaper to do it then. The
|
||
only GNU v2 methods with incorrect method names are operators and
|
||
destructors; destructors were also corrected when we read in type
|
||
information.
|
||
|
||
Therefore the only thing we need to handle here are v2 operator
|
||
names. */
|
||
if (found_stub && !startswith (TYPE_FN_FIELD_PHYSNAME (f, 0), "_Z"))
|
||
{
|
||
int ret;
|
||
char dem_opname[256];
|
||
|
||
ret = cplus_demangle_opname (TYPE_FN_FIELDLIST_NAME (type,
|
||
method_id),
|
||
dem_opname, DMGL_ANSI);
|
||
if (!ret)
|
||
ret = cplus_demangle_opname (TYPE_FN_FIELDLIST_NAME (type,
|
||
method_id),
|
||
dem_opname, 0);
|
||
if (ret)
|
||
TYPE_FN_FIELDLIST_NAME (type, method_id) = xstrdup (dem_opname);
|
||
}
|
||
}
|
||
|
||
/* Ensure it is in .rodata (if available) by workarounding GCC PR 44690. */
|
||
const struct cplus_struct_type cplus_struct_default = { };
|
||
|
||
void
|
||
allocate_cplus_struct_type (struct type *type)
|
||
{
|
||
if (HAVE_CPLUS_STRUCT (type))
|
||
/* Structure was already allocated. Nothing more to do. */
|
||
return;
|
||
|
||
TYPE_SPECIFIC_FIELD (type) = TYPE_SPECIFIC_CPLUS_STUFF;
|
||
TYPE_RAW_CPLUS_SPECIFIC (type) = (struct cplus_struct_type *)
|
||
TYPE_ALLOC (type, sizeof (struct cplus_struct_type));
|
||
*(TYPE_RAW_CPLUS_SPECIFIC (type)) = cplus_struct_default;
|
||
set_type_vptr_fieldno (type, -1);
|
||
}
|
||
|
||
const struct gnat_aux_type gnat_aux_default =
|
||
{ NULL };
|
||
|
||
/* Set the TYPE's type-specific kind to TYPE_SPECIFIC_GNAT_STUFF,
|
||
and allocate the associated gnat-specific data. The gnat-specific
|
||
data is also initialized to gnat_aux_default. */
|
||
|
||
void
|
||
allocate_gnat_aux_type (struct type *type)
|
||
{
|
||
TYPE_SPECIFIC_FIELD (type) = TYPE_SPECIFIC_GNAT_STUFF;
|
||
TYPE_GNAT_SPECIFIC (type) = (struct gnat_aux_type *)
|
||
TYPE_ALLOC (type, sizeof (struct gnat_aux_type));
|
||
*(TYPE_GNAT_SPECIFIC (type)) = gnat_aux_default;
|
||
}
|
||
|
||
/* Helper function to initialize the standard scalar types.
|
||
|
||
If NAME is non-NULL, then it is used to initialize the type name.
|
||
Note that NAME is not copied; it is required to have a lifetime at
|
||
least as long as OBJFILE. */
|
||
|
||
struct type *
|
||
init_type (enum type_code code, int length, int flags,
|
||
const char *name, struct objfile *objfile)
|
||
{
|
||
struct type *type;
|
||
|
||
type = alloc_type (objfile);
|
||
TYPE_CODE (type) = code;
|
||
TYPE_LENGTH (type) = length;
|
||
|
||
gdb_assert (!(flags & (TYPE_FLAG_MIN - 1)));
|
||
if (flags & TYPE_FLAG_UNSIGNED)
|
||
TYPE_UNSIGNED (type) = 1;
|
||
if (flags & TYPE_FLAG_NOSIGN)
|
||
TYPE_NOSIGN (type) = 1;
|
||
if (flags & TYPE_FLAG_STUB)
|
||
TYPE_STUB (type) = 1;
|
||
if (flags & TYPE_FLAG_TARGET_STUB)
|
||
TYPE_TARGET_STUB (type) = 1;
|
||
if (flags & TYPE_FLAG_STATIC)
|
||
TYPE_STATIC (type) = 1;
|
||
if (flags & TYPE_FLAG_PROTOTYPED)
|
||
TYPE_PROTOTYPED (type) = 1;
|
||
if (flags & TYPE_FLAG_INCOMPLETE)
|
||
TYPE_INCOMPLETE (type) = 1;
|
||
if (flags & TYPE_FLAG_VARARGS)
|
||
TYPE_VARARGS (type) = 1;
|
||
if (flags & TYPE_FLAG_VECTOR)
|
||
TYPE_VECTOR (type) = 1;
|
||
if (flags & TYPE_FLAG_STUB_SUPPORTED)
|
||
TYPE_STUB_SUPPORTED (type) = 1;
|
||
if (flags & TYPE_FLAG_FIXED_INSTANCE)
|
||
TYPE_FIXED_INSTANCE (type) = 1;
|
||
if (flags & TYPE_FLAG_GNU_IFUNC)
|
||
TYPE_GNU_IFUNC (type) = 1;
|
||
|
||
TYPE_NAME (type) = name;
|
||
|
||
/* C++ fancies. */
|
||
|
||
if (name && strcmp (name, "char") == 0)
|
||
TYPE_NOSIGN (type) = 1;
|
||
|
||
switch (code)
|
||
{
|
||
case TYPE_CODE_STRUCT:
|
||
case TYPE_CODE_UNION:
|
||
case TYPE_CODE_NAMESPACE:
|
||
INIT_CPLUS_SPECIFIC (type);
|
||
break;
|
||
case TYPE_CODE_FLT:
|
||
TYPE_SPECIFIC_FIELD (type) = TYPE_SPECIFIC_FLOATFORMAT;
|
||
break;
|
||
case TYPE_CODE_FUNC:
|
||
INIT_FUNC_SPECIFIC (type);
|
||
break;
|
||
}
|
||
return type;
|
||
}
|
||
|
||
/* Queries on types. */
|
||
|
||
int
|
||
can_dereference (struct type *t)
|
||
{
|
||
/* FIXME: Should we return true for references as well as
|
||
pointers? */
|
||
t = check_typedef (t);
|
||
return
|
||
(t != NULL
|
||
&& TYPE_CODE (t) == TYPE_CODE_PTR
|
||
&& TYPE_CODE (TYPE_TARGET_TYPE (t)) != TYPE_CODE_VOID);
|
||
}
|
||
|
||
int
|
||
is_integral_type (struct type *t)
|
||
{
|
||
t = check_typedef (t);
|
||
return
|
||
((t != NULL)
|
||
&& ((TYPE_CODE (t) == TYPE_CODE_INT)
|
||
|| (TYPE_CODE (t) == TYPE_CODE_ENUM)
|
||
|| (TYPE_CODE (t) == TYPE_CODE_FLAGS)
|
||
|| (TYPE_CODE (t) == TYPE_CODE_CHAR)
|
||
|| (TYPE_CODE (t) == TYPE_CODE_RANGE)
|
||
|| (TYPE_CODE (t) == TYPE_CODE_BOOL)));
|
||
}
|
||
|
||
/* Return true if TYPE is scalar. */
|
||
|
||
int
|
||
is_scalar_type (struct type *type)
|
||
{
|
||
type = check_typedef (type);
|
||
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_ARRAY:
|
||
case TYPE_CODE_STRUCT:
|
||
case TYPE_CODE_UNION:
|
||
case TYPE_CODE_SET:
|
||
case TYPE_CODE_STRING:
|
||
return 0;
|
||
default:
|
||
return 1;
|
||
}
|
||
}
|
||
|
||
/* Return true if T is scalar, or a composite type which in practice has
|
||
the memory layout of a scalar type. E.g., an array or struct with only
|
||
one scalar element inside it, or a union with only scalar elements. */
|
||
|
||
int
|
||
is_scalar_type_recursive (struct type *t)
|
||
{
|
||
t = check_typedef (t);
|
||
|
||
if (is_scalar_type (t))
|
||
return 1;
|
||
/* Are we dealing with an array or string of known dimensions? */
|
||
else if ((TYPE_CODE (t) == TYPE_CODE_ARRAY
|
||
|| TYPE_CODE (t) == TYPE_CODE_STRING) && TYPE_NFIELDS (t) == 1
|
||
&& TYPE_CODE (TYPE_INDEX_TYPE (t)) == TYPE_CODE_RANGE)
|
||
{
|
||
LONGEST low_bound, high_bound;
|
||
struct type *elt_type = check_typedef (TYPE_TARGET_TYPE (t));
|
||
|
||
get_discrete_bounds (TYPE_INDEX_TYPE (t), &low_bound, &high_bound);
|
||
|
||
return high_bound == low_bound && is_scalar_type_recursive (elt_type);
|
||
}
|
||
/* Are we dealing with a struct with one element? */
|
||
else if (TYPE_CODE (t) == TYPE_CODE_STRUCT && TYPE_NFIELDS (t) == 1)
|
||
return is_scalar_type_recursive (TYPE_FIELD_TYPE (t, 0));
|
||
else if (TYPE_CODE (t) == TYPE_CODE_UNION)
|
||
{
|
||
int i, n = TYPE_NFIELDS (t);
|
||
|
||
/* If all elements of the union are scalar, then the union is scalar. */
|
||
for (i = 0; i < n; i++)
|
||
if (!is_scalar_type_recursive (TYPE_FIELD_TYPE (t, i)))
|
||
return 0;
|
||
|
||
return 1;
|
||
}
|
||
|
||
return 0;
|
||
}
|
||
|
||
/* Return true is T is a class or a union. False otherwise. */
|
||
|
||
int
|
||
class_or_union_p (const struct type *t)
|
||
{
|
||
return (TYPE_CODE (t) == TYPE_CODE_STRUCT
|
||
|| TYPE_CODE (t) == TYPE_CODE_UNION);
|
||
}
|
||
|
||
/* A helper function which returns true if types A and B represent the
|
||
"same" class type. This is true if the types have the same main
|
||
type, or the same name. */
|
||
|
||
int
|
||
class_types_same_p (const struct type *a, const struct type *b)
|
||
{
|
||
return (TYPE_MAIN_TYPE (a) == TYPE_MAIN_TYPE (b)
|
||
|| (TYPE_NAME (a) && TYPE_NAME (b)
|
||
&& !strcmp (TYPE_NAME (a), TYPE_NAME (b))));
|
||
}
|
||
|
||
/* If BASE is an ancestor of DCLASS return the distance between them.
|
||
otherwise return -1;
|
||
eg:
|
||
|
||
class A {};
|
||
class B: public A {};
|
||
class C: public B {};
|
||
class D: C {};
|
||
|
||
distance_to_ancestor (A, A, 0) = 0
|
||
distance_to_ancestor (A, B, 0) = 1
|
||
distance_to_ancestor (A, C, 0) = 2
|
||
distance_to_ancestor (A, D, 0) = 3
|
||
|
||
If PUBLIC is 1 then only public ancestors are considered,
|
||
and the function returns the distance only if BASE is a public ancestor
|
||
of DCLASS.
|
||
Eg:
|
||
|
||
distance_to_ancestor (A, D, 1) = -1. */
|
||
|
||
static int
|
||
distance_to_ancestor (struct type *base, struct type *dclass, int is_public)
|
||
{
|
||
int i;
|
||
int d;
|
||
|
||
base = check_typedef (base);
|
||
dclass = check_typedef (dclass);
|
||
|
||
if (class_types_same_p (base, dclass))
|
||
return 0;
|
||
|
||
for (i = 0; i < TYPE_N_BASECLASSES (dclass); i++)
|
||
{
|
||
if (is_public && ! BASETYPE_VIA_PUBLIC (dclass, i))
|
||
continue;
|
||
|
||
d = distance_to_ancestor (base, TYPE_BASECLASS (dclass, i), is_public);
|
||
if (d >= 0)
|
||
return 1 + d;
|
||
}
|
||
|
||
return -1;
|
||
}
|
||
|
||
/* Check whether BASE is an ancestor or base class or DCLASS
|
||
Return 1 if so, and 0 if not.
|
||
Note: If BASE and DCLASS are of the same type, this function
|
||
will return 1. So for some class A, is_ancestor (A, A) will
|
||
return 1. */
|
||
|
||
int
|
||
is_ancestor (struct type *base, struct type *dclass)
|
||
{
|
||
return distance_to_ancestor (base, dclass, 0) >= 0;
|
||
}
|
||
|
||
/* Like is_ancestor, but only returns true when BASE is a public
|
||
ancestor of DCLASS. */
|
||
|
||
int
|
||
is_public_ancestor (struct type *base, struct type *dclass)
|
||
{
|
||
return distance_to_ancestor (base, dclass, 1) >= 0;
|
||
}
|
||
|
||
/* A helper function for is_unique_ancestor. */
|
||
|
||
static int
|
||
is_unique_ancestor_worker (struct type *base, struct type *dclass,
|
||
int *offset,
|
||
const gdb_byte *valaddr, int embedded_offset,
|
||
CORE_ADDR address, struct value *val)
|
||
{
|
||
int i, count = 0;
|
||
|
||
base = check_typedef (base);
|
||
dclass = check_typedef (dclass);
|
||
|
||
for (i = 0; i < TYPE_N_BASECLASSES (dclass) && count < 2; ++i)
|
||
{
|
||
struct type *iter;
|
||
int this_offset;
|
||
|
||
iter = check_typedef (TYPE_BASECLASS (dclass, i));
|
||
|
||
this_offset = baseclass_offset (dclass, i, valaddr, embedded_offset,
|
||
address, val);
|
||
|
||
if (class_types_same_p (base, iter))
|
||
{
|
||
/* If this is the first subclass, set *OFFSET and set count
|
||
to 1. Otherwise, if this is at the same offset as
|
||
previous instances, do nothing. Otherwise, increment
|
||
count. */
|
||
if (*offset == -1)
|
||
{
|
||
*offset = this_offset;
|
||
count = 1;
|
||
}
|
||
else if (this_offset == *offset)
|
||
{
|
||
/* Nothing. */
|
||
}
|
||
else
|
||
++count;
|
||
}
|
||
else
|
||
count += is_unique_ancestor_worker (base, iter, offset,
|
||
valaddr,
|
||
embedded_offset + this_offset,
|
||
address, val);
|
||
}
|
||
|
||
return count;
|
||
}
|
||
|
||
/* Like is_ancestor, but only returns true if BASE is a unique base
|
||
class of the type of VAL. */
|
||
|
||
int
|
||
is_unique_ancestor (struct type *base, struct value *val)
|
||
{
|
||
int offset = -1;
|
||
|
||
return is_unique_ancestor_worker (base, value_type (val), &offset,
|
||
value_contents_for_printing (val),
|
||
value_embedded_offset (val),
|
||
value_address (val), val) == 1;
|
||
}
|
||
|
||
|
||
/* Overload resolution. */
|
||
|
||
/* Return the sum of the rank of A with the rank of B. */
|
||
|
||
struct rank
|
||
sum_ranks (struct rank a, struct rank b)
|
||
{
|
||
struct rank c;
|
||
c.rank = a.rank + b.rank;
|
||
c.subrank = a.subrank + b.subrank;
|
||
return c;
|
||
}
|
||
|
||
/* Compare rank A and B and return:
|
||
0 if a = b
|
||
1 if a is better than b
|
||
-1 if b is better than a. */
|
||
|
||
int
|
||
compare_ranks (struct rank a, struct rank b)
|
||
{
|
||
if (a.rank == b.rank)
|
||
{
|
||
if (a.subrank == b.subrank)
|
||
return 0;
|
||
if (a.subrank < b.subrank)
|
||
return 1;
|
||
if (a.subrank > b.subrank)
|
||
return -1;
|
||
}
|
||
|
||
if (a.rank < b.rank)
|
||
return 1;
|
||
|
||
/* a.rank > b.rank */
|
||
return -1;
|
||
}
|
||
|
||
/* Functions for overload resolution begin here. */
|
||
|
||
/* Compare two badness vectors A and B and return the result.
|
||
0 => A and B are identical
|
||
1 => A and B are incomparable
|
||
2 => A is better than B
|
||
3 => A is worse than B */
|
||
|
||
int
|
||
compare_badness (struct badness_vector *a, struct badness_vector *b)
|
||
{
|
||
int i;
|
||
int tmp;
|
||
short found_pos = 0; /* any positives in c? */
|
||
short found_neg = 0; /* any negatives in c? */
|
||
|
||
/* differing lengths => incomparable */
|
||
if (a->length != b->length)
|
||
return 1;
|
||
|
||
/* Subtract b from a */
|
||
for (i = 0; i < a->length; i++)
|
||
{
|
||
tmp = compare_ranks (b->rank[i], a->rank[i]);
|
||
if (tmp > 0)
|
||
found_pos = 1;
|
||
else if (tmp < 0)
|
||
found_neg = 1;
|
||
}
|
||
|
||
if (found_pos)
|
||
{
|
||
if (found_neg)
|
||
return 1; /* incomparable */
|
||
else
|
||
return 3; /* A > B */
|
||
}
|
||
else
|
||
/* no positives */
|
||
{
|
||
if (found_neg)
|
||
return 2; /* A < B */
|
||
else
|
||
return 0; /* A == B */
|
||
}
|
||
}
|
||
|
||
/* Rank a function by comparing its parameter types (PARMS, length
|
||
NPARMS), to the types of an argument list (ARGS, length NARGS).
|
||
Return a pointer to a badness vector. This has NARGS + 1
|
||
entries. */
|
||
|
||
struct badness_vector *
|
||
rank_function (struct type **parms, int nparms,
|
||
struct value **args, int nargs)
|
||
{
|
||
int i;
|
||
struct badness_vector *bv = XNEW (struct badness_vector);
|
||
int min_len = nparms < nargs ? nparms : nargs;
|
||
|
||
bv->length = nargs + 1; /* add 1 for the length-match rank. */
|
||
bv->rank = XNEWVEC (struct rank, nargs + 1);
|
||
|
||
/* First compare the lengths of the supplied lists.
|
||
If there is a mismatch, set it to a high value. */
|
||
|
||
/* pai/1997-06-03 FIXME: when we have debug info about default
|
||
arguments and ellipsis parameter lists, we should consider those
|
||
and rank the length-match more finely. */
|
||
|
||
LENGTH_MATCH (bv) = (nargs != nparms)
|
||
? LENGTH_MISMATCH_BADNESS
|
||
: EXACT_MATCH_BADNESS;
|
||
|
||
/* Now rank all the parameters of the candidate function. */
|
||
for (i = 1; i <= min_len; i++)
|
||
bv->rank[i] = rank_one_type (parms[i - 1], value_type (args[i - 1]),
|
||
args[i - 1]);
|
||
|
||
/* If more arguments than parameters, add dummy entries. */
|
||
for (i = min_len + 1; i <= nargs; i++)
|
||
bv->rank[i] = TOO_FEW_PARAMS_BADNESS;
|
||
|
||
return bv;
|
||
}
|
||
|
||
/* Compare the names of two integer types, assuming that any sign
|
||
qualifiers have been checked already. We do it this way because
|
||
there may be an "int" in the name of one of the types. */
|
||
|
||
static int
|
||
integer_types_same_name_p (const char *first, const char *second)
|
||
{
|
||
int first_p, second_p;
|
||
|
||
/* If both are shorts, return 1; if neither is a short, keep
|
||
checking. */
|
||
first_p = (strstr (first, "short") != NULL);
|
||
second_p = (strstr (second, "short") != NULL);
|
||
if (first_p && second_p)
|
||
return 1;
|
||
if (first_p || second_p)
|
||
return 0;
|
||
|
||
/* Likewise for long. */
|
||
first_p = (strstr (first, "long") != NULL);
|
||
second_p = (strstr (second, "long") != NULL);
|
||
if (first_p && second_p)
|
||
return 1;
|
||
if (first_p || second_p)
|
||
return 0;
|
||
|
||
/* Likewise for char. */
|
||
first_p = (strstr (first, "char") != NULL);
|
||
second_p = (strstr (second, "char") != NULL);
|
||
if (first_p && second_p)
|
||
return 1;
|
||
if (first_p || second_p)
|
||
return 0;
|
||
|
||
/* They must both be ints. */
|
||
return 1;
|
||
}
|
||
|
||
/* Compares type A to type B returns 1 if the represent the same type
|
||
0 otherwise. */
|
||
|
||
int
|
||
types_equal (struct type *a, struct type *b)
|
||
{
|
||
/* Identical type pointers. */
|
||
/* However, this still doesn't catch all cases of same type for b
|
||
and a. The reason is that builtin types are different from
|
||
the same ones constructed from the object. */
|
||
if (a == b)
|
||
return 1;
|
||
|
||
/* Resolve typedefs */
|
||
if (TYPE_CODE (a) == TYPE_CODE_TYPEDEF)
|
||
a = check_typedef (a);
|
||
if (TYPE_CODE (b) == TYPE_CODE_TYPEDEF)
|
||
b = check_typedef (b);
|
||
|
||
/* If after resolving typedefs a and b are not of the same type
|
||
code then they are not equal. */
|
||
if (TYPE_CODE (a) != TYPE_CODE (b))
|
||
return 0;
|
||
|
||
/* If a and b are both pointers types or both reference types then
|
||
they are equal of the same type iff the objects they refer to are
|
||
of the same type. */
|
||
if (TYPE_CODE (a) == TYPE_CODE_PTR
|
||
|| TYPE_CODE (a) == TYPE_CODE_REF)
|
||
return types_equal (TYPE_TARGET_TYPE (a),
|
||
TYPE_TARGET_TYPE (b));
|
||
|
||
/* Well, damnit, if the names are exactly the same, I'll say they
|
||
are exactly the same. This happens when we generate method
|
||
stubs. The types won't point to the same address, but they
|
||
really are the same. */
|
||
|
||
if (TYPE_NAME (a) && TYPE_NAME (b)
|
||
&& strcmp (TYPE_NAME (a), TYPE_NAME (b)) == 0)
|
||
return 1;
|
||
|
||
/* Check if identical after resolving typedefs. */
|
||
if (a == b)
|
||
return 1;
|
||
|
||
/* Two function types are equal if their argument and return types
|
||
are equal. */
|
||
if (TYPE_CODE (a) == TYPE_CODE_FUNC)
|
||
{
|
||
int i;
|
||
|
||
if (TYPE_NFIELDS (a) != TYPE_NFIELDS (b))
|
||
return 0;
|
||
|
||
if (!types_equal (TYPE_TARGET_TYPE (a), TYPE_TARGET_TYPE (b)))
|
||
return 0;
|
||
|
||
for (i = 0; i < TYPE_NFIELDS (a); ++i)
|
||
if (!types_equal (TYPE_FIELD_TYPE (a, i), TYPE_FIELD_TYPE (b, i)))
|
||
return 0;
|
||
|
||
return 1;
|
||
}
|
||
|
||
return 0;
|
||
}
|
||
|
||
/* Deep comparison of types. */
|
||
|
||
/* An entry in the type-equality bcache. */
|
||
|
||
typedef struct type_equality_entry
|
||
{
|
||
struct type *type1, *type2;
|
||
} type_equality_entry_d;
|
||
|
||
DEF_VEC_O (type_equality_entry_d);
|
||
|
||
/* A helper function to compare two strings. Returns 1 if they are
|
||
the same, 0 otherwise. Handles NULLs properly. */
|
||
|
||
static int
|
||
compare_maybe_null_strings (const char *s, const char *t)
|
||
{
|
||
if (s == NULL && t != NULL)
|
||
return 0;
|
||
else if (s != NULL && t == NULL)
|
||
return 0;
|
||
else if (s == NULL && t== NULL)
|
||
return 1;
|
||
return strcmp (s, t) == 0;
|
||
}
|
||
|
||
/* A helper function for check_types_worklist that checks two types for
|
||
"deep" equality. Returns non-zero if the types are considered the
|
||
same, zero otherwise. */
|
||
|
||
static int
|
||
check_types_equal (struct type *type1, struct type *type2,
|
||
VEC (type_equality_entry_d) **worklist)
|
||
{
|
||
type1 = check_typedef (type1);
|
||
type2 = check_typedef (type2);
|
||
|
||
if (type1 == type2)
|
||
return 1;
|
||
|
||
if (TYPE_CODE (type1) != TYPE_CODE (type2)
|
||
|| TYPE_LENGTH (type1) != TYPE_LENGTH (type2)
|
||
|| TYPE_UNSIGNED (type1) != TYPE_UNSIGNED (type2)
|
||
|| TYPE_NOSIGN (type1) != TYPE_NOSIGN (type2)
|
||
|| TYPE_VARARGS (type1) != TYPE_VARARGS (type2)
|
||
|| TYPE_VECTOR (type1) != TYPE_VECTOR (type2)
|
||
|| TYPE_NOTTEXT (type1) != TYPE_NOTTEXT (type2)
|
||
|| TYPE_INSTANCE_FLAGS (type1) != TYPE_INSTANCE_FLAGS (type2)
|
||
|| TYPE_NFIELDS (type1) != TYPE_NFIELDS (type2))
|
||
return 0;
|
||
|
||
if (!compare_maybe_null_strings (TYPE_TAG_NAME (type1),
|
||
TYPE_TAG_NAME (type2)))
|
||
return 0;
|
||
if (!compare_maybe_null_strings (TYPE_NAME (type1), TYPE_NAME (type2)))
|
||
return 0;
|
||
|
||
if (TYPE_CODE (type1) == TYPE_CODE_RANGE)
|
||
{
|
||
if (memcmp (TYPE_RANGE_DATA (type1), TYPE_RANGE_DATA (type2),
|
||
sizeof (*TYPE_RANGE_DATA (type1))) != 0)
|
||
return 0;
|
||
}
|
||
else
|
||
{
|
||
int i;
|
||
|
||
for (i = 0; i < TYPE_NFIELDS (type1); ++i)
|
||
{
|
||
const struct field *field1 = &TYPE_FIELD (type1, i);
|
||
const struct field *field2 = &TYPE_FIELD (type2, i);
|
||
struct type_equality_entry entry;
|
||
|
||
if (FIELD_ARTIFICIAL (*field1) != FIELD_ARTIFICIAL (*field2)
|
||
|| FIELD_BITSIZE (*field1) != FIELD_BITSIZE (*field2)
|
||
|| FIELD_LOC_KIND (*field1) != FIELD_LOC_KIND (*field2))
|
||
return 0;
|
||
if (!compare_maybe_null_strings (FIELD_NAME (*field1),
|
||
FIELD_NAME (*field2)))
|
||
return 0;
|
||
switch (FIELD_LOC_KIND (*field1))
|
||
{
|
||
case FIELD_LOC_KIND_BITPOS:
|
||
if (FIELD_BITPOS (*field1) != FIELD_BITPOS (*field2))
|
||
return 0;
|
||
break;
|
||
case FIELD_LOC_KIND_ENUMVAL:
|
||
if (FIELD_ENUMVAL (*field1) != FIELD_ENUMVAL (*field2))
|
||
return 0;
|
||
break;
|
||
case FIELD_LOC_KIND_PHYSADDR:
|
||
if (FIELD_STATIC_PHYSADDR (*field1)
|
||
!= FIELD_STATIC_PHYSADDR (*field2))
|
||
return 0;
|
||
break;
|
||
case FIELD_LOC_KIND_PHYSNAME:
|
||
if (!compare_maybe_null_strings (FIELD_STATIC_PHYSNAME (*field1),
|
||
FIELD_STATIC_PHYSNAME (*field2)))
|
||
return 0;
|
||
break;
|
||
case FIELD_LOC_KIND_DWARF_BLOCK:
|
||
{
|
||
struct dwarf2_locexpr_baton *block1, *block2;
|
||
|
||
block1 = FIELD_DWARF_BLOCK (*field1);
|
||
block2 = FIELD_DWARF_BLOCK (*field2);
|
||
if (block1->per_cu != block2->per_cu
|
||
|| block1->size != block2->size
|
||
|| memcmp (block1->data, block2->data, block1->size) != 0)
|
||
return 0;
|
||
}
|
||
break;
|
||
default:
|
||
internal_error (__FILE__, __LINE__, _("Unsupported field kind "
|
||
"%d by check_types_equal"),
|
||
FIELD_LOC_KIND (*field1));
|
||
}
|
||
|
||
entry.type1 = FIELD_TYPE (*field1);
|
||
entry.type2 = FIELD_TYPE (*field2);
|
||
VEC_safe_push (type_equality_entry_d, *worklist, &entry);
|
||
}
|
||
}
|
||
|
||
if (TYPE_TARGET_TYPE (type1) != NULL)
|
||
{
|
||
struct type_equality_entry entry;
|
||
|
||
if (TYPE_TARGET_TYPE (type2) == NULL)
|
||
return 0;
|
||
|
||
entry.type1 = TYPE_TARGET_TYPE (type1);
|
||
entry.type2 = TYPE_TARGET_TYPE (type2);
|
||
VEC_safe_push (type_equality_entry_d, *worklist, &entry);
|
||
}
|
||
else if (TYPE_TARGET_TYPE (type2) != NULL)
|
||
return 0;
|
||
|
||
return 1;
|
||
}
|
||
|
||
/* Check types on a worklist for equality. Returns zero if any pair
|
||
is not equal, non-zero if they are all considered equal. */
|
||
|
||
static int
|
||
check_types_worklist (VEC (type_equality_entry_d) **worklist,
|
||
struct bcache *cache)
|
||
{
|
||
while (!VEC_empty (type_equality_entry_d, *worklist))
|
||
{
|
||
struct type_equality_entry entry;
|
||
int added;
|
||
|
||
entry = *VEC_last (type_equality_entry_d, *worklist);
|
||
VEC_pop (type_equality_entry_d, *worklist);
|
||
|
||
/* If the type pair has already been visited, we know it is
|
||
ok. */
|
||
bcache_full (&entry, sizeof (entry), cache, &added);
|
||
if (!added)
|
||
continue;
|
||
|
||
if (check_types_equal (entry.type1, entry.type2, worklist) == 0)
|
||
return 0;
|
||
}
|
||
|
||
return 1;
|
||
}
|
||
|
||
/* Return non-zero if types TYPE1 and TYPE2 are equal, as determined by a
|
||
"deep comparison". Otherwise return zero. */
|
||
|
||
int
|
||
types_deeply_equal (struct type *type1, struct type *type2)
|
||
{
|
||
struct gdb_exception except = exception_none;
|
||
int result = 0;
|
||
struct bcache *cache;
|
||
VEC (type_equality_entry_d) *worklist = NULL;
|
||
struct type_equality_entry entry;
|
||
|
||
gdb_assert (type1 != NULL && type2 != NULL);
|
||
|
||
/* Early exit for the simple case. */
|
||
if (type1 == type2)
|
||
return 1;
|
||
|
||
cache = bcache_xmalloc (NULL, NULL);
|
||
|
||
entry.type1 = type1;
|
||
entry.type2 = type2;
|
||
VEC_safe_push (type_equality_entry_d, worklist, &entry);
|
||
|
||
/* check_types_worklist calls several nested helper functions, some
|
||
of which can raise a GDB exception, so we just check and rethrow
|
||
here. If there is a GDB exception, a comparison is not capable
|
||
(or trusted), so exit. */
|
||
TRY
|
||
{
|
||
result = check_types_worklist (&worklist, cache);
|
||
}
|
||
CATCH (ex, RETURN_MASK_ALL)
|
||
{
|
||
except = ex;
|
||
}
|
||
END_CATCH
|
||
|
||
bcache_xfree (cache);
|
||
VEC_free (type_equality_entry_d, worklist);
|
||
|
||
/* Rethrow if there was a problem. */
|
||
if (except.reason < 0)
|
||
throw_exception (except);
|
||
|
||
return result;
|
||
}
|
||
|
||
/* Allocated status of type TYPE. Return zero if type TYPE is allocated.
|
||
Otherwise return one. */
|
||
|
||
int
|
||
type_not_allocated (const struct type *type)
|
||
{
|
||
struct dynamic_prop *prop = TYPE_ALLOCATED_PROP (type);
|
||
|
||
return (prop && TYPE_DYN_PROP_KIND (prop) == PROP_CONST
|
||
&& !TYPE_DYN_PROP_ADDR (prop));
|
||
}
|
||
|
||
/* Associated status of type TYPE. Return zero if type TYPE is associated.
|
||
Otherwise return one. */
|
||
|
||
int
|
||
type_not_associated (const struct type *type)
|
||
{
|
||
struct dynamic_prop *prop = TYPE_ASSOCIATED_PROP (type);
|
||
|
||
return (prop && TYPE_DYN_PROP_KIND (prop) == PROP_CONST
|
||
&& !TYPE_DYN_PROP_ADDR (prop));
|
||
}
|
||
|
||
/* Compare one type (PARM) for compatibility with another (ARG).
|
||
* PARM is intended to be the parameter type of a function; and
|
||
* ARG is the supplied argument's type. This function tests if
|
||
* the latter can be converted to the former.
|
||
* VALUE is the argument's value or NULL if none (or called recursively)
|
||
*
|
||
* Return 0 if they are identical types;
|
||
* Otherwise, return an integer which corresponds to how compatible
|
||
* PARM is to ARG. The higher the return value, the worse the match.
|
||
* Generally the "bad" conversions are all uniformly assigned a 100. */
|
||
|
||
struct rank
|
||
rank_one_type (struct type *parm, struct type *arg, struct value *value)
|
||
{
|
||
struct rank rank = {0,0};
|
||
|
||
if (types_equal (parm, arg))
|
||
return EXACT_MATCH_BADNESS;
|
||
|
||
/* Resolve typedefs */
|
||
if (TYPE_CODE (parm) == TYPE_CODE_TYPEDEF)
|
||
parm = check_typedef (parm);
|
||
if (TYPE_CODE (arg) == TYPE_CODE_TYPEDEF)
|
||
arg = check_typedef (arg);
|
||
|
||
/* See through references, since we can almost make non-references
|
||
references. */
|
||
if (TYPE_CODE (arg) == TYPE_CODE_REF)
|
||
return (sum_ranks (rank_one_type (parm, TYPE_TARGET_TYPE (arg), NULL),
|
||
REFERENCE_CONVERSION_BADNESS));
|
||
if (TYPE_CODE (parm) == TYPE_CODE_REF)
|
||
return (sum_ranks (rank_one_type (TYPE_TARGET_TYPE (parm), arg, NULL),
|
||
REFERENCE_CONVERSION_BADNESS));
|
||
if (overload_debug)
|
||
/* Debugging only. */
|
||
fprintf_filtered (gdb_stderr,
|
||
"------ Arg is %s [%d], parm is %s [%d]\n",
|
||
TYPE_NAME (arg), TYPE_CODE (arg),
|
||
TYPE_NAME (parm), TYPE_CODE (parm));
|
||
|
||
/* x -> y means arg of type x being supplied for parameter of type y. */
|
||
|
||
switch (TYPE_CODE (parm))
|
||
{
|
||
case TYPE_CODE_PTR:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_PTR:
|
||
|
||
/* Allowed pointer conversions are:
|
||
(a) pointer to void-pointer conversion. */
|
||
if (TYPE_CODE (TYPE_TARGET_TYPE (parm)) == TYPE_CODE_VOID)
|
||
return VOID_PTR_CONVERSION_BADNESS;
|
||
|
||
/* (b) pointer to ancestor-pointer conversion. */
|
||
rank.subrank = distance_to_ancestor (TYPE_TARGET_TYPE (parm),
|
||
TYPE_TARGET_TYPE (arg),
|
||
0);
|
||
if (rank.subrank >= 0)
|
||
return sum_ranks (BASE_PTR_CONVERSION_BADNESS, rank);
|
||
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
case TYPE_CODE_ARRAY:
|
||
if (types_equal (TYPE_TARGET_TYPE (parm),
|
||
TYPE_TARGET_TYPE (arg)))
|
||
return EXACT_MATCH_BADNESS;
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
case TYPE_CODE_FUNC:
|
||
return rank_one_type (TYPE_TARGET_TYPE (parm), arg, NULL);
|
||
case TYPE_CODE_INT:
|
||
if (value != NULL && TYPE_CODE (value_type (value)) == TYPE_CODE_INT)
|
||
{
|
||
if (value_as_long (value) == 0)
|
||
{
|
||
/* Null pointer conversion: allow it to be cast to a pointer.
|
||
[4.10.1 of C++ standard draft n3290] */
|
||
return NULL_POINTER_CONVERSION_BADNESS;
|
||
}
|
||
else
|
||
{
|
||
/* If type checking is disabled, allow the conversion. */
|
||
if (!strict_type_checking)
|
||
return NS_INTEGER_POINTER_CONVERSION_BADNESS;
|
||
}
|
||
}
|
||
/* fall through */
|
||
case TYPE_CODE_ENUM:
|
||
case TYPE_CODE_FLAGS:
|
||
case TYPE_CODE_CHAR:
|
||
case TYPE_CODE_RANGE:
|
||
case TYPE_CODE_BOOL:
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
case TYPE_CODE_ARRAY:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_PTR:
|
||
case TYPE_CODE_ARRAY:
|
||
return rank_one_type (TYPE_TARGET_TYPE (parm),
|
||
TYPE_TARGET_TYPE (arg), NULL);
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
case TYPE_CODE_FUNC:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_PTR: /* funcptr -> func */
|
||
return rank_one_type (parm, TYPE_TARGET_TYPE (arg), NULL);
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
case TYPE_CODE_INT:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_INT:
|
||
if (TYPE_LENGTH (arg) == TYPE_LENGTH (parm))
|
||
{
|
||
/* Deal with signed, unsigned, and plain chars and
|
||
signed and unsigned ints. */
|
||
if (TYPE_NOSIGN (parm))
|
||
{
|
||
/* This case only for character types. */
|
||
if (TYPE_NOSIGN (arg))
|
||
return EXACT_MATCH_BADNESS; /* plain char -> plain char */
|
||
else /* signed/unsigned char -> plain char */
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
}
|
||
else if (TYPE_UNSIGNED (parm))
|
||
{
|
||
if (TYPE_UNSIGNED (arg))
|
||
{
|
||
/* unsigned int -> unsigned int, or
|
||
unsigned long -> unsigned long */
|
||
if (integer_types_same_name_p (TYPE_NAME (parm),
|
||
TYPE_NAME (arg)))
|
||
return EXACT_MATCH_BADNESS;
|
||
else if (integer_types_same_name_p (TYPE_NAME (arg),
|
||
"int")
|
||
&& integer_types_same_name_p (TYPE_NAME (parm),
|
||
"long"))
|
||
/* unsigned int -> unsigned long */
|
||
return INTEGER_PROMOTION_BADNESS;
|
||
else
|
||
/* unsigned long -> unsigned int */
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
}
|
||
else
|
||
{
|
||
if (integer_types_same_name_p (TYPE_NAME (arg),
|
||
"long")
|
||
&& integer_types_same_name_p (TYPE_NAME (parm),
|
||
"int"))
|
||
/* signed long -> unsigned int */
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
else
|
||
/* signed int/long -> unsigned int/long */
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
}
|
||
}
|
||
else if (!TYPE_NOSIGN (arg) && !TYPE_UNSIGNED (arg))
|
||
{
|
||
if (integer_types_same_name_p (TYPE_NAME (parm),
|
||
TYPE_NAME (arg)))
|
||
return EXACT_MATCH_BADNESS;
|
||
else if (integer_types_same_name_p (TYPE_NAME (arg),
|
||
"int")
|
||
&& integer_types_same_name_p (TYPE_NAME (parm),
|
||
"long"))
|
||
return INTEGER_PROMOTION_BADNESS;
|
||
else
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
}
|
||
else
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
}
|
||
else if (TYPE_LENGTH (arg) < TYPE_LENGTH (parm))
|
||
return INTEGER_PROMOTION_BADNESS;
|
||
else
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
case TYPE_CODE_ENUM:
|
||
case TYPE_CODE_FLAGS:
|
||
case TYPE_CODE_CHAR:
|
||
case TYPE_CODE_RANGE:
|
||
case TYPE_CODE_BOOL:
|
||
if (TYPE_DECLARED_CLASS (arg))
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
return INTEGER_PROMOTION_BADNESS;
|
||
case TYPE_CODE_FLT:
|
||
return INT_FLOAT_CONVERSION_BADNESS;
|
||
case TYPE_CODE_PTR:
|
||
return NS_POINTER_CONVERSION_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_ENUM:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_INT:
|
||
case TYPE_CODE_CHAR:
|
||
case TYPE_CODE_RANGE:
|
||
case TYPE_CODE_BOOL:
|
||
case TYPE_CODE_ENUM:
|
||
if (TYPE_DECLARED_CLASS (parm) || TYPE_DECLARED_CLASS (arg))
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
case TYPE_CODE_FLT:
|
||
return INT_FLOAT_CONVERSION_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_CHAR:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_RANGE:
|
||
case TYPE_CODE_BOOL:
|
||
case TYPE_CODE_ENUM:
|
||
if (TYPE_DECLARED_CLASS (arg))
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
case TYPE_CODE_FLT:
|
||
return INT_FLOAT_CONVERSION_BADNESS;
|
||
case TYPE_CODE_INT:
|
||
if (TYPE_LENGTH (arg) > TYPE_LENGTH (parm))
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
else if (TYPE_LENGTH (arg) < TYPE_LENGTH (parm))
|
||
return INTEGER_PROMOTION_BADNESS;
|
||
/* >>> !! else fall through !! <<< */
|
||
case TYPE_CODE_CHAR:
|
||
/* Deal with signed, unsigned, and plain chars for C++ and
|
||
with int cases falling through from previous case. */
|
||
if (TYPE_NOSIGN (parm))
|
||
{
|
||
if (TYPE_NOSIGN (arg))
|
||
return EXACT_MATCH_BADNESS;
|
||
else
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
}
|
||
else if (TYPE_UNSIGNED (parm))
|
||
{
|
||
if (TYPE_UNSIGNED (arg))
|
||
return EXACT_MATCH_BADNESS;
|
||
else
|
||
return INTEGER_PROMOTION_BADNESS;
|
||
}
|
||
else if (!TYPE_NOSIGN (arg) && !TYPE_UNSIGNED (arg))
|
||
return EXACT_MATCH_BADNESS;
|
||
else
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_RANGE:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_INT:
|
||
case TYPE_CODE_CHAR:
|
||
case TYPE_CODE_RANGE:
|
||
case TYPE_CODE_BOOL:
|
||
case TYPE_CODE_ENUM:
|
||
return INTEGER_CONVERSION_BADNESS;
|
||
case TYPE_CODE_FLT:
|
||
return INT_FLOAT_CONVERSION_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_BOOL:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
/* n3290 draft, section 4.12.1 (conv.bool):
|
||
|
||
"A prvalue of arithmetic, unscoped enumeration, pointer, or
|
||
pointer to member type can be converted to a prvalue of type
|
||
bool. A zero value, null pointer value, or null member pointer
|
||
value is converted to false; any other value is converted to
|
||
true. A prvalue of type std::nullptr_t can be converted to a
|
||
prvalue of type bool; the resulting value is false." */
|
||
case TYPE_CODE_INT:
|
||
case TYPE_CODE_CHAR:
|
||
case TYPE_CODE_ENUM:
|
||
case TYPE_CODE_FLT:
|
||
case TYPE_CODE_MEMBERPTR:
|
||
case TYPE_CODE_PTR:
|
||
return BOOL_CONVERSION_BADNESS;
|
||
case TYPE_CODE_RANGE:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
case TYPE_CODE_BOOL:
|
||
return EXACT_MATCH_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_FLT:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_FLT:
|
||
if (TYPE_LENGTH (arg) < TYPE_LENGTH (parm))
|
||
return FLOAT_PROMOTION_BADNESS;
|
||
else if (TYPE_LENGTH (arg) == TYPE_LENGTH (parm))
|
||
return EXACT_MATCH_BADNESS;
|
||
else
|
||
return FLOAT_CONVERSION_BADNESS;
|
||
case TYPE_CODE_INT:
|
||
case TYPE_CODE_BOOL:
|
||
case TYPE_CODE_ENUM:
|
||
case TYPE_CODE_RANGE:
|
||
case TYPE_CODE_CHAR:
|
||
return INT_FLOAT_CONVERSION_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_COMPLEX:
|
||
switch (TYPE_CODE (arg))
|
||
{ /* Strictly not needed for C++, but... */
|
||
case TYPE_CODE_FLT:
|
||
return FLOAT_PROMOTION_BADNESS;
|
||
case TYPE_CODE_COMPLEX:
|
||
return EXACT_MATCH_BADNESS;
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_STRUCT:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_STRUCT:
|
||
/* Check for derivation */
|
||
rank.subrank = distance_to_ancestor (parm, arg, 0);
|
||
if (rank.subrank >= 0)
|
||
return sum_ranks (BASE_CONVERSION_BADNESS, rank);
|
||
/* else fall through */
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_UNION:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
case TYPE_CODE_UNION:
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_MEMBERPTR:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_METHOD:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_REF:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
|
||
break;
|
||
case TYPE_CODE_SET:
|
||
switch (TYPE_CODE (arg))
|
||
{
|
||
/* Not in C++ */
|
||
case TYPE_CODE_SET:
|
||
return rank_one_type (TYPE_FIELD_TYPE (parm, 0),
|
||
TYPE_FIELD_TYPE (arg, 0), NULL);
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
}
|
||
break;
|
||
case TYPE_CODE_VOID:
|
||
default:
|
||
return INCOMPATIBLE_TYPE_BADNESS;
|
||
} /* switch (TYPE_CODE (arg)) */
|
||
}
|
||
|
||
/* End of functions for overload resolution. */
|
||
|
||
/* Routines to pretty-print types. */
|
||
|
||
static void
|
||
print_bit_vector (B_TYPE *bits, int nbits)
|
||
{
|
||
int bitno;
|
||
|
||
for (bitno = 0; bitno < nbits; bitno++)
|
||
{
|
||
if ((bitno % 8) == 0)
|
||
{
|
||
puts_filtered (" ");
|
||
}
|
||
if (B_TST (bits, bitno))
|
||
printf_filtered (("1"));
|
||
else
|
||
printf_filtered (("0"));
|
||
}
|
||
}
|
||
|
||
/* Note the first arg should be the "this" pointer, we may not want to
|
||
include it since we may get into a infinitely recursive
|
||
situation. */
|
||
|
||
static void
|
||
print_args (struct field *args, int nargs, int spaces)
|
||
{
|
||
if (args != NULL)
|
||
{
|
||
int i;
|
||
|
||
for (i = 0; i < nargs; i++)
|
||
{
|
||
printfi_filtered (spaces, "[%d] name '%s'\n", i,
|
||
args[i].name != NULL ? args[i].name : "<NULL>");
|
||
recursive_dump_type (args[i].type, spaces + 2);
|
||
}
|
||
}
|
||
}
|
||
|
||
int
|
||
field_is_static (struct field *f)
|
||
{
|
||
/* "static" fields are the fields whose location is not relative
|
||
to the address of the enclosing struct. It would be nice to
|
||
have a dedicated flag that would be set for static fields when
|
||
the type is being created. But in practice, checking the field
|
||
loc_kind should give us an accurate answer. */
|
||
return (FIELD_LOC_KIND (*f) == FIELD_LOC_KIND_PHYSNAME
|
||
|| FIELD_LOC_KIND (*f) == FIELD_LOC_KIND_PHYSADDR);
|
||
}
|
||
|
||
static void
|
||
dump_fn_fieldlists (struct type *type, int spaces)
|
||
{
|
||
int method_idx;
|
||
int overload_idx;
|
||
struct fn_field *f;
|
||
|
||
printfi_filtered (spaces, "fn_fieldlists ");
|
||
gdb_print_host_address (TYPE_FN_FIELDLISTS (type), gdb_stdout);
|
||
printf_filtered ("\n");
|
||
for (method_idx = 0; method_idx < TYPE_NFN_FIELDS (type); method_idx++)
|
||
{
|
||
f = TYPE_FN_FIELDLIST1 (type, method_idx);
|
||
printfi_filtered (spaces + 2, "[%d] name '%s' (",
|
||
method_idx,
|
||
TYPE_FN_FIELDLIST_NAME (type, method_idx));
|
||
gdb_print_host_address (TYPE_FN_FIELDLIST_NAME (type, method_idx),
|
||
gdb_stdout);
|
||
printf_filtered (_(") length %d\n"),
|
||
TYPE_FN_FIELDLIST_LENGTH (type, method_idx));
|
||
for (overload_idx = 0;
|
||
overload_idx < TYPE_FN_FIELDLIST_LENGTH (type, method_idx);
|
||
overload_idx++)
|
||
{
|
||
printfi_filtered (spaces + 4, "[%d] physname '%s' (",
|
||
overload_idx,
|
||
TYPE_FN_FIELD_PHYSNAME (f, overload_idx));
|
||
gdb_print_host_address (TYPE_FN_FIELD_PHYSNAME (f, overload_idx),
|
||
gdb_stdout);
|
||
printf_filtered (")\n");
|
||
printfi_filtered (spaces + 8, "type ");
|
||
gdb_print_host_address (TYPE_FN_FIELD_TYPE (f, overload_idx),
|
||
gdb_stdout);
|
||
printf_filtered ("\n");
|
||
|
||
recursive_dump_type (TYPE_FN_FIELD_TYPE (f, overload_idx),
|
||
spaces + 8 + 2);
|
||
|
||
printfi_filtered (spaces + 8, "args ");
|
||
gdb_print_host_address (TYPE_FN_FIELD_ARGS (f, overload_idx),
|
||
gdb_stdout);
|
||
printf_filtered ("\n");
|
||
print_args (TYPE_FN_FIELD_ARGS (f, overload_idx),
|
||
TYPE_NFIELDS (TYPE_FN_FIELD_TYPE (f, overload_idx)),
|
||
spaces + 8 + 2);
|
||
printfi_filtered (spaces + 8, "fcontext ");
|
||
gdb_print_host_address (TYPE_FN_FIELD_FCONTEXT (f, overload_idx),
|
||
gdb_stdout);
|
||
printf_filtered ("\n");
|
||
|
||
printfi_filtered (spaces + 8, "is_const %d\n",
|
||
TYPE_FN_FIELD_CONST (f, overload_idx));
|
||
printfi_filtered (spaces + 8, "is_volatile %d\n",
|
||
TYPE_FN_FIELD_VOLATILE (f, overload_idx));
|
||
printfi_filtered (spaces + 8, "is_private %d\n",
|
||
TYPE_FN_FIELD_PRIVATE (f, overload_idx));
|
||
printfi_filtered (spaces + 8, "is_protected %d\n",
|
||
TYPE_FN_FIELD_PROTECTED (f, overload_idx));
|
||
printfi_filtered (spaces + 8, "is_stub %d\n",
|
||
TYPE_FN_FIELD_STUB (f, overload_idx));
|
||
printfi_filtered (spaces + 8, "voffset %u\n",
|
||
TYPE_FN_FIELD_VOFFSET (f, overload_idx));
|
||
}
|
||
}
|
||
}
|
||
|
||
static void
|
||
print_cplus_stuff (struct type *type, int spaces)
|
||
{
|
||
printfi_filtered (spaces, "vptr_fieldno %d\n", TYPE_VPTR_FIELDNO (type));
|
||
printfi_filtered (spaces, "vptr_basetype ");
|
||
gdb_print_host_address (TYPE_VPTR_BASETYPE (type), gdb_stdout);
|
||
puts_filtered ("\n");
|
||
if (TYPE_VPTR_BASETYPE (type) != NULL)
|
||
recursive_dump_type (TYPE_VPTR_BASETYPE (type), spaces + 2);
|
||
|
||
printfi_filtered (spaces, "n_baseclasses %d\n",
|
||
TYPE_N_BASECLASSES (type));
|
||
printfi_filtered (spaces, "nfn_fields %d\n",
|
||
TYPE_NFN_FIELDS (type));
|
||
if (TYPE_N_BASECLASSES (type) > 0)
|
||
{
|
||
printfi_filtered (spaces, "virtual_field_bits (%d bits at *",
|
||
TYPE_N_BASECLASSES (type));
|
||
gdb_print_host_address (TYPE_FIELD_VIRTUAL_BITS (type),
|
||
gdb_stdout);
|
||
printf_filtered (")");
|
||
|
||
print_bit_vector (TYPE_FIELD_VIRTUAL_BITS (type),
|
||
TYPE_N_BASECLASSES (type));
|
||
puts_filtered ("\n");
|
||
}
|
||
if (TYPE_NFIELDS (type) > 0)
|
||
{
|
||
if (TYPE_FIELD_PRIVATE_BITS (type) != NULL)
|
||
{
|
||
printfi_filtered (spaces,
|
||
"private_field_bits (%d bits at *",
|
||
TYPE_NFIELDS (type));
|
||
gdb_print_host_address (TYPE_FIELD_PRIVATE_BITS (type),
|
||
gdb_stdout);
|
||
printf_filtered (")");
|
||
print_bit_vector (TYPE_FIELD_PRIVATE_BITS (type),
|
||
TYPE_NFIELDS (type));
|
||
puts_filtered ("\n");
|
||
}
|
||
if (TYPE_FIELD_PROTECTED_BITS (type) != NULL)
|
||
{
|
||
printfi_filtered (spaces,
|
||
"protected_field_bits (%d bits at *",
|
||
TYPE_NFIELDS (type));
|
||
gdb_print_host_address (TYPE_FIELD_PROTECTED_BITS (type),
|
||
gdb_stdout);
|
||
printf_filtered (")");
|
||
print_bit_vector (TYPE_FIELD_PROTECTED_BITS (type),
|
||
TYPE_NFIELDS (type));
|
||
puts_filtered ("\n");
|
||
}
|
||
}
|
||
if (TYPE_NFN_FIELDS (type) > 0)
|
||
{
|
||
dump_fn_fieldlists (type, spaces);
|
||
}
|
||
}
|
||
|
||
/* Print the contents of the TYPE's type_specific union, assuming that
|
||
its type-specific kind is TYPE_SPECIFIC_GNAT_STUFF. */
|
||
|
||
static void
|
||
print_gnat_stuff (struct type *type, int spaces)
|
||
{
|
||
struct type *descriptive_type = TYPE_DESCRIPTIVE_TYPE (type);
|
||
|
||
if (descriptive_type == NULL)
|
||
printfi_filtered (spaces + 2, "no descriptive type\n");
|
||
else
|
||
{
|
||
printfi_filtered (spaces + 2, "descriptive type\n");
|
||
recursive_dump_type (descriptive_type, spaces + 4);
|
||
}
|
||
}
|
||
|
||
static struct obstack dont_print_type_obstack;
|
||
|
||
void
|
||
recursive_dump_type (struct type *type, int spaces)
|
||
{
|
||
int idx;
|
||
|
||
if (spaces == 0)
|
||
obstack_begin (&dont_print_type_obstack, 0);
|
||
|
||
if (TYPE_NFIELDS (type) > 0
|
||
|| (HAVE_CPLUS_STRUCT (type) && TYPE_NFN_FIELDS (type) > 0))
|
||
{
|
||
struct type **first_dont_print
|
||
= (struct type **) obstack_base (&dont_print_type_obstack);
|
||
|
||
int i = (struct type **)
|
||
obstack_next_free (&dont_print_type_obstack) - first_dont_print;
|
||
|
||
while (--i >= 0)
|
||
{
|
||
if (type == first_dont_print[i])
|
||
{
|
||
printfi_filtered (spaces, "type node ");
|
||
gdb_print_host_address (type, gdb_stdout);
|
||
printf_filtered (_(" <same as already seen type>\n"));
|
||
return;
|
||
}
|
||
}
|
||
|
||
obstack_ptr_grow (&dont_print_type_obstack, type);
|
||
}
|
||
|
||
printfi_filtered (spaces, "type node ");
|
||
gdb_print_host_address (type, gdb_stdout);
|
||
printf_filtered ("\n");
|
||
printfi_filtered (spaces, "name '%s' (",
|
||
TYPE_NAME (type) ? TYPE_NAME (type) : "<NULL>");
|
||
gdb_print_host_address (TYPE_NAME (type), gdb_stdout);
|
||
printf_filtered (")\n");
|
||
printfi_filtered (spaces, "tagname '%s' (",
|
||
TYPE_TAG_NAME (type) ? TYPE_TAG_NAME (type) : "<NULL>");
|
||
gdb_print_host_address (TYPE_TAG_NAME (type), gdb_stdout);
|
||
printf_filtered (")\n");
|
||
printfi_filtered (spaces, "code 0x%x ", TYPE_CODE (type));
|
||
switch (TYPE_CODE (type))
|
||
{
|
||
case TYPE_CODE_UNDEF:
|
||
printf_filtered ("(TYPE_CODE_UNDEF)");
|
||
break;
|
||
case TYPE_CODE_PTR:
|
||
printf_filtered ("(TYPE_CODE_PTR)");
|
||
break;
|
||
case TYPE_CODE_ARRAY:
|
||
printf_filtered ("(TYPE_CODE_ARRAY)");
|
||
break;
|
||
case TYPE_CODE_STRUCT:
|
||
printf_filtered ("(TYPE_CODE_STRUCT)");
|
||
break;
|
||
case TYPE_CODE_UNION:
|
||
printf_filtered ("(TYPE_CODE_UNION)");
|
||
break;
|
||
case TYPE_CODE_ENUM:
|
||
printf_filtered ("(TYPE_CODE_ENUM)");
|
||
break;
|
||
case TYPE_CODE_FLAGS:
|
||
printf_filtered ("(TYPE_CODE_FLAGS)");
|
||
break;
|
||
case TYPE_CODE_FUNC:
|
||
printf_filtered ("(TYPE_CODE_FUNC)");
|
||
break;
|
||
case TYPE_CODE_INT:
|
||
printf_filtered ("(TYPE_CODE_INT)");
|
||
break;
|
||
case TYPE_CODE_FLT:
|
||
printf_filtered ("(TYPE_CODE_FLT)");
|
||
break;
|
||
case TYPE_CODE_VOID:
|
||
printf_filtered ("(TYPE_CODE_VOID)");
|
||
break;
|
||
case TYPE_CODE_SET:
|
||
printf_filtered ("(TYPE_CODE_SET)");
|
||
break;
|
||
case TYPE_CODE_RANGE:
|
||
printf_filtered ("(TYPE_CODE_RANGE)");
|
||
break;
|
||
case TYPE_CODE_STRING:
|
||
printf_filtered ("(TYPE_CODE_STRING)");
|
||
break;
|
||
case TYPE_CODE_ERROR:
|
||
printf_filtered ("(TYPE_CODE_ERROR)");
|
||
break;
|
||
case TYPE_CODE_MEMBERPTR:
|
||
printf_filtered ("(TYPE_CODE_MEMBERPTR)");
|
||
break;
|
||
case TYPE_CODE_METHODPTR:
|
||
printf_filtered ("(TYPE_CODE_METHODPTR)");
|
||
break;
|
||
case TYPE_CODE_METHOD:
|
||
printf_filtered ("(TYPE_CODE_METHOD)");
|
||
break;
|
||
case TYPE_CODE_REF:
|
||
printf_filtered ("(TYPE_CODE_REF)");
|
||
break;
|
||
case TYPE_CODE_CHAR:
|
||
printf_filtered ("(TYPE_CODE_CHAR)");
|
||
break;
|
||
case TYPE_CODE_BOOL:
|
||
printf_filtered ("(TYPE_CODE_BOOL)");
|
||
break;
|
||
case TYPE_CODE_COMPLEX:
|
||
printf_filtered ("(TYPE_CODE_COMPLEX)");
|
||
break;
|
||
case TYPE_CODE_TYPEDEF:
|
||
printf_filtered ("(TYPE_CODE_TYPEDEF)");
|
||
break;
|
||
case TYPE_CODE_NAMESPACE:
|
||
printf_filtered ("(TYPE_CODE_NAMESPACE)");
|
||
break;
|
||
default:
|
||
printf_filtered ("(UNKNOWN TYPE CODE)");
|
||
break;
|
||
}
|
||
puts_filtered ("\n");
|
||
printfi_filtered (spaces, "length %d\n", TYPE_LENGTH (type));
|
||
if (TYPE_OBJFILE_OWNED (type))
|
||
{
|
||
printfi_filtered (spaces, "objfile ");
|
||
gdb_print_host_address (TYPE_OWNER (type).objfile, gdb_stdout);
|
||
}
|
||
else
|
||
{
|
||
printfi_filtered (spaces, "gdbarch ");
|
||
gdb_print_host_address (TYPE_OWNER (type).gdbarch, gdb_stdout);
|
||
}
|
||
printf_filtered ("\n");
|
||
printfi_filtered (spaces, "target_type ");
|
||
gdb_print_host_address (TYPE_TARGET_TYPE (type), gdb_stdout);
|
||
printf_filtered ("\n");
|
||
if (TYPE_TARGET_TYPE (type) != NULL)
|
||
{
|
||
recursive_dump_type (TYPE_TARGET_TYPE (type), spaces + 2);
|
||
}
|
||
printfi_filtered (spaces, "pointer_type ");
|
||
gdb_print_host_address (TYPE_POINTER_TYPE (type), gdb_stdout);
|
||
printf_filtered ("\n");
|
||
printfi_filtered (spaces, "reference_type ");
|
||
gdb_print_host_address (TYPE_REFERENCE_TYPE (type), gdb_stdout);
|
||
printf_filtered ("\n");
|
||
printfi_filtered (spaces, "type_chain ");
|
||
gdb_print_host_address (TYPE_CHAIN (type), gdb_stdout);
|
||
printf_filtered ("\n");
|
||
printfi_filtered (spaces, "instance_flags 0x%x",
|
||
TYPE_INSTANCE_FLAGS (type));
|
||
if (TYPE_CONST (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_CONST");
|
||
}
|
||
if (TYPE_VOLATILE (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_VOLATILE");
|
||
}
|
||
if (TYPE_CODE_SPACE (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_CODE_SPACE");
|
||
}
|
||
if (TYPE_DATA_SPACE (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_DATA_SPACE");
|
||
}
|
||
if (TYPE_ADDRESS_CLASS_1 (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_ADDRESS_CLASS_1");
|
||
}
|
||
if (TYPE_ADDRESS_CLASS_2 (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_ADDRESS_CLASS_2");
|
||
}
|
||
if (TYPE_RESTRICT (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_RESTRICT");
|
||
}
|
||
if (TYPE_ATOMIC (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_ATOMIC");
|
||
}
|
||
puts_filtered ("\n");
|
||
|
||
printfi_filtered (spaces, "flags");
|
||
if (TYPE_UNSIGNED (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_UNSIGNED");
|
||
}
|
||
if (TYPE_NOSIGN (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_NOSIGN");
|
||
}
|
||
if (TYPE_STUB (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_STUB");
|
||
}
|
||
if (TYPE_TARGET_STUB (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_TARGET_STUB");
|
||
}
|
||
if (TYPE_STATIC (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_STATIC");
|
||
}
|
||
if (TYPE_PROTOTYPED (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_PROTOTYPED");
|
||
}
|
||
if (TYPE_INCOMPLETE (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_INCOMPLETE");
|
||
}
|
||
if (TYPE_VARARGS (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_VARARGS");
|
||
}
|
||
/* This is used for things like AltiVec registers on ppc. Gcc emits
|
||
an attribute for the array type, which tells whether or not we
|
||
have a vector, instead of a regular array. */
|
||
if (TYPE_VECTOR (type))
|
||
{
|
||
puts_filtered (" TYPE_FLAG_VECTOR");
|
||
}
|
||
if (TYPE_FIXED_INSTANCE (type))
|
||
{
|
||
puts_filtered (" TYPE_FIXED_INSTANCE");
|
||
}
|
||
if (TYPE_STUB_SUPPORTED (type))
|
||
{
|
||
puts_filtered (" TYPE_STUB_SUPPORTED");
|
||
}
|
||
if (TYPE_NOTTEXT (type))
|
||
{
|
||
puts_filtered (" TYPE_NOTTEXT");
|
||
}
|
||
puts_filtered ("\n");
|
||
printfi_filtered (spaces, "nfields %d ", TYPE_NFIELDS (type));
|
||
gdb_print_host_address (TYPE_FIELDS (type), gdb_stdout);
|
||
puts_filtered ("\n");
|
||
for (idx = 0; idx < TYPE_NFIELDS (type); idx++)
|
||
{
|
||
if (TYPE_CODE (type) == TYPE_CODE_ENUM)
|
||
printfi_filtered (spaces + 2,
|
||
"[%d] enumval %s type ",
|
||
idx, plongest (TYPE_FIELD_ENUMVAL (type, idx)));
|
||
else
|
||
printfi_filtered (spaces + 2,
|
||
"[%d] bitpos %s bitsize %d type ",
|
||
idx, plongest (TYPE_FIELD_BITPOS (type, idx)),
|
||
TYPE_FIELD_BITSIZE (type, idx));
|
||
gdb_print_host_address (TYPE_FIELD_TYPE (type, idx), gdb_stdout);
|
||
printf_filtered (" name '%s' (",
|
||
TYPE_FIELD_NAME (type, idx) != NULL
|
||
? TYPE_FIELD_NAME (type, idx)
|
||
: "<NULL>");
|
||
gdb_print_host_address (TYPE_FIELD_NAME (type, idx), gdb_stdout);
|
||
printf_filtered (")\n");
|
||
if (TYPE_FIELD_TYPE (type, idx) != NULL)
|
||
{
|
||
recursive_dump_type (TYPE_FIELD_TYPE (type, idx), spaces + 4);
|
||
}
|
||
}
|
||
if (TYPE_CODE (type) == TYPE_CODE_RANGE)
|
||
{
|
||
printfi_filtered (spaces, "low %s%s high %s%s\n",
|
||
plongest (TYPE_LOW_BOUND (type)),
|
||
TYPE_LOW_BOUND_UNDEFINED (type) ? " (undefined)" : "",
|
||
plongest (TYPE_HIGH_BOUND (type)),
|
||
TYPE_HIGH_BOUND_UNDEFINED (type)
|
||
? " (undefined)" : "");
|
||
}
|
||
|
||
switch (TYPE_SPECIFIC_FIELD (type))
|
||
{
|
||
case TYPE_SPECIFIC_CPLUS_STUFF:
|
||
printfi_filtered (spaces, "cplus_stuff ");
|
||
gdb_print_host_address (TYPE_CPLUS_SPECIFIC (type),
|
||
gdb_stdout);
|
||
puts_filtered ("\n");
|
||
print_cplus_stuff (type, spaces);
|
||
break;
|
||
|
||
case TYPE_SPECIFIC_GNAT_STUFF:
|
||
printfi_filtered (spaces, "gnat_stuff ");
|
||
gdb_print_host_address (TYPE_GNAT_SPECIFIC (type), gdb_stdout);
|
||
puts_filtered ("\n");
|
||
print_gnat_stuff (type, spaces);
|
||
break;
|
||
|
||
case TYPE_SPECIFIC_FLOATFORMAT:
|
||
printfi_filtered (spaces, "floatformat ");
|
||
if (TYPE_FLOATFORMAT (type) == NULL)
|
||
puts_filtered ("(null)");
|
||
else
|
||
{
|
||
puts_filtered ("{ ");
|
||
if (TYPE_FLOATFORMAT (type)[0] == NULL
|
||
|| TYPE_FLOATFORMAT (type)[0]->name == NULL)
|
||
puts_filtered ("(null)");
|
||
else
|
||
puts_filtered (TYPE_FLOATFORMAT (type)[0]->name);
|
||
|
||
puts_filtered (", ");
|
||
if (TYPE_FLOATFORMAT (type)[1] == NULL
|
||
|| TYPE_FLOATFORMAT (type)[1]->name == NULL)
|
||
puts_filtered ("(null)");
|
||
else
|
||
puts_filtered (TYPE_FLOATFORMAT (type)[1]->name);
|
||
|
||
puts_filtered (" }");
|
||
}
|
||
puts_filtered ("\n");
|
||
break;
|
||
|
||
case TYPE_SPECIFIC_FUNC:
|
||
printfi_filtered (spaces, "calling_convention %d\n",
|
||
TYPE_CALLING_CONVENTION (type));
|
||
/* tail_call_list is not printed. */
|
||
break;
|
||
|
||
case TYPE_SPECIFIC_SELF_TYPE:
|
||
printfi_filtered (spaces, "self_type ");
|
||
gdb_print_host_address (TYPE_SELF_TYPE (type), gdb_stdout);
|
||
puts_filtered ("\n");
|
||
break;
|
||
}
|
||
|
||
if (spaces == 0)
|
||
obstack_free (&dont_print_type_obstack, NULL);
|
||
}
|
||
|
||
/* Trivial helpers for the libiberty hash table, for mapping one
|
||
type to another. */
|
||
|
||
struct type_pair
|
||
{
|
||
struct type *old, *newobj;
|
||
};
|
||
|
||
static hashval_t
|
||
type_pair_hash (const void *item)
|
||
{
|
||
const struct type_pair *pair = (const struct type_pair *) item;
|
||
|
||
return htab_hash_pointer (pair->old);
|
||
}
|
||
|
||
static int
|
||
type_pair_eq (const void *item_lhs, const void *item_rhs)
|
||
{
|
||
const struct type_pair *lhs = (const struct type_pair *) item_lhs;
|
||
const struct type_pair *rhs = (const struct type_pair *) item_rhs;
|
||
|
||
return lhs->old == rhs->old;
|
||
}
|
||
|
||
/* Allocate the hash table used by copy_type_recursive to walk
|
||
types without duplicates. We use OBJFILE's obstack, because
|
||
OBJFILE is about to be deleted. */
|
||
|
||
htab_t
|
||
create_copied_types_hash (struct objfile *objfile)
|
||
{
|
||
return htab_create_alloc_ex (1, type_pair_hash, type_pair_eq,
|
||
NULL, &objfile->objfile_obstack,
|
||
hashtab_obstack_allocate,
|
||
dummy_obstack_deallocate);
|
||
}
|
||
|
||
/* Recursively copy (deep copy) a dynamic attribute list of a type. */
|
||
|
||
static struct dynamic_prop_list *
|
||
copy_dynamic_prop_list (struct obstack *objfile_obstack,
|
||
struct dynamic_prop_list *list)
|
||
{
|
||
struct dynamic_prop_list *copy = list;
|
||
struct dynamic_prop_list **node_ptr = ©
|
||
|
||
while (*node_ptr != NULL)
|
||
{
|
||
struct dynamic_prop_list *node_copy;
|
||
|
||
node_copy = ((struct dynamic_prop_list *)
|
||
obstack_copy (objfile_obstack, *node_ptr,
|
||
sizeof (struct dynamic_prop_list)));
|
||
node_copy->prop = (*node_ptr)->prop;
|
||
*node_ptr = node_copy;
|
||
|
||
node_ptr = &node_copy->next;
|
||
}
|
||
|
||
return copy;
|
||
}
|
||
|
||
/* Recursively copy (deep copy) TYPE, if it is associated with
|
||
OBJFILE. Return a new type owned by the gdbarch associated with the type, a
|
||
saved type if we have already visited TYPE (using COPIED_TYPES), or TYPE if
|
||
it is not associated with OBJFILE. */
|
||
|
||
struct type *
|
||
copy_type_recursive (struct objfile *objfile,
|
||
struct type *type,
|
||
htab_t copied_types)
|
||
{
|
||
struct type_pair *stored, pair;
|
||
void **slot;
|
||
struct type *new_type;
|
||
|
||
if (! TYPE_OBJFILE_OWNED (type))
|
||
return type;
|
||
|
||
/* This type shouldn't be pointing to any types in other objfiles;
|
||
if it did, the type might disappear unexpectedly. */
|
||
gdb_assert (TYPE_OBJFILE (type) == objfile);
|
||
|
||
pair.old = type;
|
||
slot = htab_find_slot (copied_types, &pair, INSERT);
|
||
if (*slot != NULL)
|
||
return ((struct type_pair *) *slot)->newobj;
|
||
|
||
new_type = alloc_type_arch (get_type_arch (type));
|
||
|
||
/* We must add the new type to the hash table immediately, in case
|
||
we encounter this type again during a recursive call below. */
|
||
stored = XOBNEW (&objfile->objfile_obstack, struct type_pair);
|
||
stored->old = type;
|
||
stored->newobj = new_type;
|
||
*slot = stored;
|
||
|
||
/* Copy the common fields of types. For the main type, we simply
|
||
copy the entire thing and then update specific fields as needed. */
|
||
*TYPE_MAIN_TYPE (new_type) = *TYPE_MAIN_TYPE (type);
|
||
TYPE_OBJFILE_OWNED (new_type) = 0;
|
||
TYPE_OWNER (new_type).gdbarch = get_type_arch (type);
|
||
|
||
if (TYPE_NAME (type))
|
||
TYPE_NAME (new_type) = xstrdup (TYPE_NAME (type));
|
||
if (TYPE_TAG_NAME (type))
|
||
TYPE_TAG_NAME (new_type) = xstrdup (TYPE_TAG_NAME (type));
|
||
|
||
TYPE_INSTANCE_FLAGS (new_type) = TYPE_INSTANCE_FLAGS (type);
|
||
TYPE_LENGTH (new_type) = TYPE_LENGTH (type);
|
||
|
||
/* Copy the fields. */
|
||
if (TYPE_NFIELDS (type))
|
||
{
|
||
int i, nfields;
|
||
|
||
nfields = TYPE_NFIELDS (type);
|
||
TYPE_FIELDS (new_type) = XCNEWVEC (struct field, nfields);
|
||
for (i = 0; i < nfields; i++)
|
||
{
|
||
TYPE_FIELD_ARTIFICIAL (new_type, i) =
|
||
TYPE_FIELD_ARTIFICIAL (type, i);
|
||
TYPE_FIELD_BITSIZE (new_type, i) = TYPE_FIELD_BITSIZE (type, i);
|
||
if (TYPE_FIELD_TYPE (type, i))
|
||
TYPE_FIELD_TYPE (new_type, i)
|
||
= copy_type_recursive (objfile, TYPE_FIELD_TYPE (type, i),
|
||
copied_types);
|
||
if (TYPE_FIELD_NAME (type, i))
|
||
TYPE_FIELD_NAME (new_type, i) =
|
||
xstrdup (TYPE_FIELD_NAME (type, i));
|
||
switch (TYPE_FIELD_LOC_KIND (type, i))
|
||
{
|
||
case FIELD_LOC_KIND_BITPOS:
|
||
SET_FIELD_BITPOS (TYPE_FIELD (new_type, i),
|
||
TYPE_FIELD_BITPOS (type, i));
|
||
break;
|
||
case FIELD_LOC_KIND_ENUMVAL:
|
||
SET_FIELD_ENUMVAL (TYPE_FIELD (new_type, i),
|
||
TYPE_FIELD_ENUMVAL (type, i));
|
||
break;
|
||
case FIELD_LOC_KIND_PHYSADDR:
|
||
SET_FIELD_PHYSADDR (TYPE_FIELD (new_type, i),
|
||
TYPE_FIELD_STATIC_PHYSADDR (type, i));
|
||
break;
|
||
case FIELD_LOC_KIND_PHYSNAME:
|
||
SET_FIELD_PHYSNAME (TYPE_FIELD (new_type, i),
|
||
xstrdup (TYPE_FIELD_STATIC_PHYSNAME (type,
|
||
i)));
|
||
break;
|
||
default:
|
||
internal_error (__FILE__, __LINE__,
|
||
_("Unexpected type field location kind: %d"),
|
||
TYPE_FIELD_LOC_KIND (type, i));
|
||
}
|
||
}
|
||
}
|
||
|
||
/* For range types, copy the bounds information. */
|
||
if (TYPE_CODE (type) == TYPE_CODE_RANGE)
|
||
{
|
||
TYPE_RANGE_DATA (new_type) = XNEW (struct range_bounds);
|
||
*TYPE_RANGE_DATA (new_type) = *TYPE_RANGE_DATA (type);
|
||
}
|
||
|
||
if (TYPE_DYN_PROP_LIST (type) != NULL)
|
||
TYPE_DYN_PROP_LIST (new_type)
|
||
= copy_dynamic_prop_list (&objfile->objfile_obstack,
|
||
TYPE_DYN_PROP_LIST (type));
|
||
|
||
|
||
/* Copy pointers to other types. */
|
||
if (TYPE_TARGET_TYPE (type))
|
||
TYPE_TARGET_TYPE (new_type) =
|
||
copy_type_recursive (objfile,
|
||
TYPE_TARGET_TYPE (type),
|
||
copied_types);
|
||
|
||
/* Maybe copy the type_specific bits.
|
||
|
||
NOTE drow/2005-12-09: We do not copy the C++-specific bits like
|
||
base classes and methods. There's no fundamental reason why we
|
||
can't, but at the moment it is not needed. */
|
||
|
||
switch (TYPE_SPECIFIC_FIELD (type))
|
||
{
|
||
case TYPE_SPECIFIC_NONE:
|
||
break;
|
||
case TYPE_SPECIFIC_FUNC:
|
||
INIT_FUNC_SPECIFIC (new_type);
|
||
TYPE_CALLING_CONVENTION (new_type) = TYPE_CALLING_CONVENTION (type);
|
||
TYPE_NO_RETURN (new_type) = TYPE_NO_RETURN (type);
|
||
TYPE_TAIL_CALL_LIST (new_type) = NULL;
|
||
break;
|
||
case TYPE_SPECIFIC_FLOATFORMAT:
|
||
TYPE_FLOATFORMAT (new_type) = TYPE_FLOATFORMAT (type);
|
||
break;
|
||
case TYPE_SPECIFIC_CPLUS_STUFF:
|
||
INIT_CPLUS_SPECIFIC (new_type);
|
||
break;
|
||
case TYPE_SPECIFIC_GNAT_STUFF:
|
||
INIT_GNAT_SPECIFIC (new_type);
|
||
break;
|
||
case TYPE_SPECIFIC_SELF_TYPE:
|
||
set_type_self_type (new_type,
|
||
copy_type_recursive (objfile, TYPE_SELF_TYPE (type),
|
||
copied_types));
|
||
break;
|
||
default:
|
||
gdb_assert_not_reached ("bad type_specific_kind");
|
||
}
|
||
|
||
return new_type;
|
||
}
|
||
|
||
/* Make a copy of the given TYPE, except that the pointer & reference
|
||
types are not preserved.
|
||
|
||
This function assumes that the given type has an associated objfile.
|
||
This objfile is used to allocate the new type. */
|
||
|
||
struct type *
|
||
copy_type (const struct type *type)
|
||
{
|
||
struct type *new_type;
|
||
|
||
gdb_assert (TYPE_OBJFILE_OWNED (type));
|
||
|
||
new_type = alloc_type_copy (type);
|
||
TYPE_INSTANCE_FLAGS (new_type) = TYPE_INSTANCE_FLAGS (type);
|
||
TYPE_LENGTH (new_type) = TYPE_LENGTH (type);
|
||
memcpy (TYPE_MAIN_TYPE (new_type), TYPE_MAIN_TYPE (type),
|
||
sizeof (struct main_type));
|
||
if (TYPE_DYN_PROP_LIST (type) != NULL)
|
||
TYPE_DYN_PROP_LIST (new_type)
|
||
= copy_dynamic_prop_list (&TYPE_OBJFILE (type) -> objfile_obstack,
|
||
TYPE_DYN_PROP_LIST (type));
|
||
|
||
return new_type;
|
||
}
|
||
|
||
/* Helper functions to initialize architecture-specific types. */
|
||
|
||
/* Allocate a type structure associated with GDBARCH and set its
|
||
CODE, LENGTH, and NAME fields. */
|
||
|
||
struct type *
|
||
arch_type (struct gdbarch *gdbarch,
|
||
enum type_code code, int length, const char *name)
|
||
{
|
||
struct type *type;
|
||
|
||
type = alloc_type_arch (gdbarch);
|
||
TYPE_CODE (type) = code;
|
||
TYPE_LENGTH (type) = length;
|
||
|
||
if (name)
|
||
TYPE_NAME (type) = gdbarch_obstack_strdup (gdbarch, name);
|
||
|
||
return type;
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_INT type structure associated with GDBARCH.
|
||
BIT is the type size in bits. If UNSIGNED_P is non-zero, set
|
||
the type's TYPE_UNSIGNED flag. NAME is the type name. */
|
||
|
||
struct type *
|
||
arch_integer_type (struct gdbarch *gdbarch,
|
||
int bit, int unsigned_p, const char *name)
|
||
{
|
||
struct type *t;
|
||
|
||
t = arch_type (gdbarch, TYPE_CODE_INT, bit / TARGET_CHAR_BIT, name);
|
||
if (unsigned_p)
|
||
TYPE_UNSIGNED (t) = 1;
|
||
if (name && strcmp (name, "char") == 0)
|
||
TYPE_NOSIGN (t) = 1;
|
||
|
||
return t;
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_CHAR type structure associated with GDBARCH.
|
||
BIT is the type size in bits. If UNSIGNED_P is non-zero, set
|
||
the type's TYPE_UNSIGNED flag. NAME is the type name. */
|
||
|
||
struct type *
|
||
arch_character_type (struct gdbarch *gdbarch,
|
||
int bit, int unsigned_p, const char *name)
|
||
{
|
||
struct type *t;
|
||
|
||
t = arch_type (gdbarch, TYPE_CODE_CHAR, bit / TARGET_CHAR_BIT, name);
|
||
if (unsigned_p)
|
||
TYPE_UNSIGNED (t) = 1;
|
||
|
||
return t;
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_BOOL type structure associated with GDBARCH.
|
||
BIT is the type size in bits. If UNSIGNED_P is non-zero, set
|
||
the type's TYPE_UNSIGNED flag. NAME is the type name. */
|
||
|
||
struct type *
|
||
arch_boolean_type (struct gdbarch *gdbarch,
|
||
int bit, int unsigned_p, const char *name)
|
||
{
|
||
struct type *t;
|
||
|
||
t = arch_type (gdbarch, TYPE_CODE_BOOL, bit / TARGET_CHAR_BIT, name);
|
||
if (unsigned_p)
|
||
TYPE_UNSIGNED (t) = 1;
|
||
|
||
return t;
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_FLT type structure associated with GDBARCH.
|
||
BIT is the type size in bits; if BIT equals -1, the size is
|
||
determined by the floatformat. NAME is the type name. Set the
|
||
TYPE_FLOATFORMAT from FLOATFORMATS. */
|
||
|
||
struct type *
|
||
arch_float_type (struct gdbarch *gdbarch,
|
||
int bit, const char *name,
|
||
const struct floatformat **floatformats)
|
||
{
|
||
struct type *t;
|
||
|
||
if (bit == -1)
|
||
{
|
||
gdb_assert (floatformats != NULL);
|
||
gdb_assert (floatformats[0] != NULL && floatformats[1] != NULL);
|
||
bit = floatformats[0]->totalsize;
|
||
}
|
||
gdb_assert (bit >= 0);
|
||
|
||
t = arch_type (gdbarch, TYPE_CODE_FLT, bit / TARGET_CHAR_BIT, name);
|
||
TYPE_FLOATFORMAT (t) = floatformats;
|
||
|
||
if (floatformats != NULL)
|
||
{
|
||
size_t len = TYPE_LENGTH (t);
|
||
|
||
gdb_assert (len >= floatformat_totalsize_bytes (floatformats[0]));
|
||
gdb_assert (len >= floatformat_totalsize_bytes (floatformats[1]));
|
||
}
|
||
|
||
return t;
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_COMPLEX type structure associated with GDBARCH.
|
||
NAME is the type name. TARGET_TYPE is the component float type. */
|
||
|
||
struct type *
|
||
arch_complex_type (struct gdbarch *gdbarch,
|
||
const char *name, struct type *target_type)
|
||
{
|
||
struct type *t;
|
||
|
||
t = arch_type (gdbarch, TYPE_CODE_COMPLEX,
|
||
2 * TYPE_LENGTH (target_type), name);
|
||
TYPE_TARGET_TYPE (t) = target_type;
|
||
return t;
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_FLAGS type structure associated with GDBARCH.
|
||
NAME is the type name. LENGTH is the size of the flag word in bytes. */
|
||
|
||
struct type *
|
||
arch_flags_type (struct gdbarch *gdbarch, const char *name, int length)
|
||
{
|
||
int max_nfields = length * TARGET_CHAR_BIT;
|
||
struct type *type;
|
||
|
||
type = arch_type (gdbarch, TYPE_CODE_FLAGS, length, name);
|
||
TYPE_UNSIGNED (type) = 1;
|
||
TYPE_NFIELDS (type) = 0;
|
||
/* Pre-allocate enough space assuming every field is one bit. */
|
||
TYPE_FIELDS (type)
|
||
= (struct field *) TYPE_ZALLOC (type, max_nfields * sizeof (struct field));
|
||
|
||
return type;
|
||
}
|
||
|
||
/* Add field to TYPE_CODE_FLAGS type TYPE to indicate the bit at
|
||
position BITPOS is called NAME. Pass NAME as "" for fields that
|
||
should not be printed. */
|
||
|
||
void
|
||
append_flags_type_field (struct type *type, int start_bitpos, int nr_bits,
|
||
struct type *field_type, const char *name)
|
||
{
|
||
int type_bitsize = TYPE_LENGTH (type) * TARGET_CHAR_BIT;
|
||
int field_nr = TYPE_NFIELDS (type);
|
||
|
||
gdb_assert (TYPE_CODE (type) == TYPE_CODE_FLAGS);
|
||
gdb_assert (TYPE_NFIELDS (type) + 1 <= type_bitsize);
|
||
gdb_assert (start_bitpos >= 0 && start_bitpos < type_bitsize);
|
||
gdb_assert (nr_bits >= 1 && nr_bits <= type_bitsize);
|
||
gdb_assert (name != NULL);
|
||
|
||
TYPE_FIELD_NAME (type, field_nr) = xstrdup (name);
|
||
TYPE_FIELD_TYPE (type, field_nr) = field_type;
|
||
SET_FIELD_BITPOS (TYPE_FIELD (type, field_nr), start_bitpos);
|
||
TYPE_FIELD_BITSIZE (type, field_nr) = nr_bits;
|
||
++TYPE_NFIELDS (type);
|
||
}
|
||
|
||
/* Special version of append_flags_type_field to add a flag field.
|
||
Add field to TYPE_CODE_FLAGS type TYPE to indicate the bit at
|
||
position BITPOS is called NAME. */
|
||
|
||
void
|
||
append_flags_type_flag (struct type *type, int bitpos, const char *name)
|
||
{
|
||
struct gdbarch *gdbarch = get_type_arch (type);
|
||
|
||
append_flags_type_field (type, bitpos, 1,
|
||
builtin_type (gdbarch)->builtin_bool,
|
||
name);
|
||
}
|
||
|
||
/* Allocate a TYPE_CODE_STRUCT or TYPE_CODE_UNION type structure (as
|
||
specified by CODE) associated with GDBARCH. NAME is the type name. */
|
||
|
||
struct type *
|
||
arch_composite_type (struct gdbarch *gdbarch, const char *name,
|
||
enum type_code code)
|
||
{
|
||
struct type *t;
|
||
|
||
gdb_assert (code == TYPE_CODE_STRUCT || code == TYPE_CODE_UNION);
|
||
t = arch_type (gdbarch, code, 0, NULL);
|
||
TYPE_TAG_NAME (t) = name;
|
||
INIT_CPLUS_SPECIFIC (t);
|
||
return t;
|
||
}
|
||
|
||
/* Add new field with name NAME and type FIELD to composite type T.
|
||
Do not set the field's position or adjust the type's length;
|
||
the caller should do so. Return the new field. */
|
||
|
||
struct field *
|
||
append_composite_type_field_raw (struct type *t, const char *name,
|
||
struct type *field)
|
||
{
|
||
struct field *f;
|
||
|
||
TYPE_NFIELDS (t) = TYPE_NFIELDS (t) + 1;
|
||
TYPE_FIELDS (t) = XRESIZEVEC (struct field, TYPE_FIELDS (t),
|
||
TYPE_NFIELDS (t));
|
||
f = &(TYPE_FIELDS (t)[TYPE_NFIELDS (t) - 1]);
|
||
memset (f, 0, sizeof f[0]);
|
||
FIELD_TYPE (f[0]) = field;
|
||
FIELD_NAME (f[0]) = name;
|
||
return f;
|
||
}
|
||
|
||
/* Add new field with name NAME and type FIELD to composite type T.
|
||
ALIGNMENT (if non-zero) specifies the minimum field alignment. */
|
||
|
||
void
|
||
append_composite_type_field_aligned (struct type *t, const char *name,
|
||
struct type *field, int alignment)
|
||
{
|
||
struct field *f = append_composite_type_field_raw (t, name, field);
|
||
|
||
if (TYPE_CODE (t) == TYPE_CODE_UNION)
|
||
{
|
||
if (TYPE_LENGTH (t) < TYPE_LENGTH (field))
|
||
TYPE_LENGTH (t) = TYPE_LENGTH (field);
|
||
}
|
||
else if (TYPE_CODE (t) == TYPE_CODE_STRUCT)
|
||
{
|
||
TYPE_LENGTH (t) = TYPE_LENGTH (t) + TYPE_LENGTH (field);
|
||
if (TYPE_NFIELDS (t) > 1)
|
||
{
|
||
SET_FIELD_BITPOS (f[0],
|
||
(FIELD_BITPOS (f[-1])
|
||
+ (TYPE_LENGTH (FIELD_TYPE (f[-1]))
|
||
* TARGET_CHAR_BIT)));
|
||
|
||
if (alignment)
|
||
{
|
||
int left;
|
||
|
||
alignment *= TARGET_CHAR_BIT;
|
||
left = FIELD_BITPOS (f[0]) % alignment;
|
||
|
||
if (left)
|
||
{
|
||
SET_FIELD_BITPOS (f[0], FIELD_BITPOS (f[0]) + (alignment - left));
|
||
TYPE_LENGTH (t) += (alignment - left) / TARGET_CHAR_BIT;
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}
|
||
|
||
/* Add new field with name NAME and type FIELD to composite type T. */
|
||
|
||
void
|
||
append_composite_type_field (struct type *t, const char *name,
|
||
struct type *field)
|
||
{
|
||
append_composite_type_field_aligned (t, name, field, 0);
|
||
}
|
||
|
||
static struct gdbarch_data *gdbtypes_data;
|
||
|
||
const struct builtin_type *
|
||
builtin_type (struct gdbarch *gdbarch)
|
||
{
|
||
return (const struct builtin_type *) gdbarch_data (gdbarch, gdbtypes_data);
|
||
}
|
||
|
||
static void *
|
||
gdbtypes_post_init (struct gdbarch *gdbarch)
|
||
{
|
||
struct builtin_type *builtin_type
|
||
= GDBARCH_OBSTACK_ZALLOC (gdbarch, struct builtin_type);
|
||
|
||
/* Basic types. */
|
||
builtin_type->builtin_void
|
||
= arch_type (gdbarch, TYPE_CODE_VOID, 1, "void");
|
||
builtin_type->builtin_char
|
||
= arch_integer_type (gdbarch, TARGET_CHAR_BIT,
|
||
!gdbarch_char_signed (gdbarch), "char");
|
||
builtin_type->builtin_signed_char
|
||
= arch_integer_type (gdbarch, TARGET_CHAR_BIT,
|
||
0, "signed char");
|
||
builtin_type->builtin_unsigned_char
|
||
= arch_integer_type (gdbarch, TARGET_CHAR_BIT,
|
||
1, "unsigned char");
|
||
builtin_type->builtin_short
|
||
= arch_integer_type (gdbarch, gdbarch_short_bit (gdbarch),
|
||
0, "short");
|
||
builtin_type->builtin_unsigned_short
|
||
= arch_integer_type (gdbarch, gdbarch_short_bit (gdbarch),
|
||
1, "unsigned short");
|
||
builtin_type->builtin_int
|
||
= arch_integer_type (gdbarch, gdbarch_int_bit (gdbarch),
|
||
0, "int");
|
||
builtin_type->builtin_unsigned_int
|
||
= arch_integer_type (gdbarch, gdbarch_int_bit (gdbarch),
|
||
1, "unsigned int");
|
||
builtin_type->builtin_long
|
||
= arch_integer_type (gdbarch, gdbarch_long_bit (gdbarch),
|
||
0, "long");
|
||
builtin_type->builtin_unsigned_long
|
||
= arch_integer_type (gdbarch, gdbarch_long_bit (gdbarch),
|
||
1, "unsigned long");
|
||
builtin_type->builtin_long_long
|
||
= arch_integer_type (gdbarch, gdbarch_long_long_bit (gdbarch),
|
||
0, "long long");
|
||
builtin_type->builtin_unsigned_long_long
|
||
= arch_integer_type (gdbarch, gdbarch_long_long_bit (gdbarch),
|
||
1, "unsigned long long");
|
||
builtin_type->builtin_float
|
||
= arch_float_type (gdbarch, gdbarch_float_bit (gdbarch),
|
||
"float", gdbarch_float_format (gdbarch));
|
||
builtin_type->builtin_double
|
||
= arch_float_type (gdbarch, gdbarch_double_bit (gdbarch),
|
||
"double", gdbarch_double_format (gdbarch));
|
||
builtin_type->builtin_long_double
|
||
= arch_float_type (gdbarch, gdbarch_long_double_bit (gdbarch),
|
||
"long double", gdbarch_long_double_format (gdbarch));
|
||
builtin_type->builtin_complex
|
||
= arch_complex_type (gdbarch, "complex",
|
||
builtin_type->builtin_float);
|
||
builtin_type->builtin_double_complex
|
||
= arch_complex_type (gdbarch, "double complex",
|
||
builtin_type->builtin_double);
|
||
builtin_type->builtin_string
|
||
= arch_type (gdbarch, TYPE_CODE_STRING, 1, "string");
|
||
builtin_type->builtin_bool
|
||
= arch_type (gdbarch, TYPE_CODE_BOOL, 1, "bool");
|
||
|
||
/* The following three are about decimal floating point types, which
|
||
are 32-bits, 64-bits and 128-bits respectively. */
|
||
builtin_type->builtin_decfloat
|
||
= arch_type (gdbarch, TYPE_CODE_DECFLOAT, 32 / 8, "_Decimal32");
|
||
builtin_type->builtin_decdouble
|
||
= arch_type (gdbarch, TYPE_CODE_DECFLOAT, 64 / 8, "_Decimal64");
|
||
builtin_type->builtin_declong
|
||
= arch_type (gdbarch, TYPE_CODE_DECFLOAT, 128 / 8, "_Decimal128");
|
||
|
||
/* "True" character types. */
|
||
builtin_type->builtin_true_char
|
||
= arch_character_type (gdbarch, TARGET_CHAR_BIT, 0, "true character");
|
||
builtin_type->builtin_true_unsigned_char
|
||
= arch_character_type (gdbarch, TARGET_CHAR_BIT, 1, "true character");
|
||
|
||
/* Fixed-size integer types. */
|
||
builtin_type->builtin_int0
|
||
= arch_integer_type (gdbarch, 0, 0, "int0_t");
|
||
builtin_type->builtin_int8
|
||
= arch_integer_type (gdbarch, 8, 0, "int8_t");
|
||
builtin_type->builtin_uint8
|
||
= arch_integer_type (gdbarch, 8, 1, "uint8_t");
|
||
builtin_type->builtin_int16
|
||
= arch_integer_type (gdbarch, 16, 0, "int16_t");
|
||
builtin_type->builtin_uint16
|
||
= arch_integer_type (gdbarch, 16, 1, "uint16_t");
|
||
builtin_type->builtin_int32
|
||
= arch_integer_type (gdbarch, 32, 0, "int32_t");
|
||
builtin_type->builtin_uint32
|
||
= arch_integer_type (gdbarch, 32, 1, "uint32_t");
|
||
builtin_type->builtin_int64
|
||
= arch_integer_type (gdbarch, 64, 0, "int64_t");
|
||
builtin_type->builtin_uint64
|
||
= arch_integer_type (gdbarch, 64, 1, "uint64_t");
|
||
builtin_type->builtin_int128
|
||
= arch_integer_type (gdbarch, 128, 0, "int128_t");
|
||
builtin_type->builtin_uint128
|
||
= arch_integer_type (gdbarch, 128, 1, "uint128_t");
|
||
TYPE_INSTANCE_FLAGS (builtin_type->builtin_int8) |=
|
||
TYPE_INSTANCE_FLAG_NOTTEXT;
|
||
TYPE_INSTANCE_FLAGS (builtin_type->builtin_uint8) |=
|
||
TYPE_INSTANCE_FLAG_NOTTEXT;
|
||
|
||
/* Wide character types. */
|
||
builtin_type->builtin_char16
|
||
= arch_integer_type (gdbarch, 16, 0, "char16_t");
|
||
builtin_type->builtin_char32
|
||
= arch_integer_type (gdbarch, 32, 0, "char32_t");
|
||
|
||
|
||
/* Default data/code pointer types. */
|
||
builtin_type->builtin_data_ptr
|
||
= lookup_pointer_type (builtin_type->builtin_void);
|
||
builtin_type->builtin_func_ptr
|
||
= lookup_pointer_type (lookup_function_type (builtin_type->builtin_void));
|
||
builtin_type->builtin_func_func
|
||
= lookup_function_type (builtin_type->builtin_func_ptr);
|
||
|
||
/* This type represents a GDB internal function. */
|
||
builtin_type->internal_fn
|
||
= arch_type (gdbarch, TYPE_CODE_INTERNAL_FUNCTION, 0,
|
||
"<internal function>");
|
||
|
||
/* This type represents an xmethod. */
|
||
builtin_type->xmethod
|
||
= arch_type (gdbarch, TYPE_CODE_XMETHOD, 0, "<xmethod>");
|
||
|
||
return builtin_type;
|
||
}
|
||
|
||
/* This set of objfile-based types is intended to be used by symbol
|
||
readers as basic types. */
|
||
|
||
static const struct objfile_data *objfile_type_data;
|
||
|
||
const struct objfile_type *
|
||
objfile_type (struct objfile *objfile)
|
||
{
|
||
struct gdbarch *gdbarch;
|
||
struct objfile_type *objfile_type
|
||
= (struct objfile_type *) objfile_data (objfile, objfile_type_data);
|
||
|
||
if (objfile_type)
|
||
return objfile_type;
|
||
|
||
objfile_type = OBSTACK_CALLOC (&objfile->objfile_obstack,
|
||
1, struct objfile_type);
|
||
|
||
/* Use the objfile architecture to determine basic type properties. */
|
||
gdbarch = get_objfile_arch (objfile);
|
||
|
||
/* Basic types. */
|
||
objfile_type->builtin_void
|
||
= init_type (TYPE_CODE_VOID, 1,
|
||
0,
|
||
"void", objfile);
|
||
|
||
objfile_type->builtin_char
|
||
= init_type (TYPE_CODE_INT, TARGET_CHAR_BIT / TARGET_CHAR_BIT,
|
||
(TYPE_FLAG_NOSIGN
|
||
| (gdbarch_char_signed (gdbarch) ? 0 : TYPE_FLAG_UNSIGNED)),
|
||
"char", objfile);
|
||
objfile_type->builtin_signed_char
|
||
= init_type (TYPE_CODE_INT, TARGET_CHAR_BIT / TARGET_CHAR_BIT,
|
||
0,
|
||
"signed char", objfile);
|
||
objfile_type->builtin_unsigned_char
|
||
= init_type (TYPE_CODE_INT, TARGET_CHAR_BIT / TARGET_CHAR_BIT,
|
||
TYPE_FLAG_UNSIGNED,
|
||
"unsigned char", objfile);
|
||
objfile_type->builtin_short
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_short_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "short", objfile);
|
||
objfile_type->builtin_unsigned_short
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_short_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
TYPE_FLAG_UNSIGNED, "unsigned short", objfile);
|
||
objfile_type->builtin_int
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_int_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "int", objfile);
|
||
objfile_type->builtin_unsigned_int
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_int_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
TYPE_FLAG_UNSIGNED, "unsigned int", objfile);
|
||
objfile_type->builtin_long
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_long_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "long", objfile);
|
||
objfile_type->builtin_unsigned_long
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_long_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
TYPE_FLAG_UNSIGNED, "unsigned long", objfile);
|
||
objfile_type->builtin_long_long
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_long_long_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "long long", objfile);
|
||
objfile_type->builtin_unsigned_long_long
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_long_long_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
TYPE_FLAG_UNSIGNED, "unsigned long long", objfile);
|
||
|
||
objfile_type->builtin_float
|
||
= init_type (TYPE_CODE_FLT,
|
||
gdbarch_float_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "float", objfile);
|
||
TYPE_FLOATFORMAT (objfile_type->builtin_float)
|
||
= gdbarch_float_format (gdbarch);
|
||
objfile_type->builtin_double
|
||
= init_type (TYPE_CODE_FLT,
|
||
gdbarch_double_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "double", objfile);
|
||
TYPE_FLOATFORMAT (objfile_type->builtin_double)
|
||
= gdbarch_double_format (gdbarch);
|
||
objfile_type->builtin_long_double
|
||
= init_type (TYPE_CODE_FLT,
|
||
gdbarch_long_double_bit (gdbarch) / TARGET_CHAR_BIT,
|
||
0, "long double", objfile);
|
||
TYPE_FLOATFORMAT (objfile_type->builtin_long_double)
|
||
= gdbarch_long_double_format (gdbarch);
|
||
|
||
/* This type represents a type that was unrecognized in symbol read-in. */
|
||
objfile_type->builtin_error
|
||
= init_type (TYPE_CODE_ERROR, 0, 0, "<unknown type>", objfile);
|
||
|
||
/* The following set of types is used for symbols with no
|
||
debug information. */
|
||
objfile_type->nodebug_text_symbol
|
||
= init_type (TYPE_CODE_FUNC, 1, 0,
|
||
"<text variable, no debug info>", objfile);
|
||
TYPE_TARGET_TYPE (objfile_type->nodebug_text_symbol)
|
||
= objfile_type->builtin_int;
|
||
objfile_type->nodebug_text_gnu_ifunc_symbol
|
||
= init_type (TYPE_CODE_FUNC, 1, TYPE_FLAG_GNU_IFUNC,
|
||
"<text gnu-indirect-function variable, no debug info>",
|
||
objfile);
|
||
TYPE_TARGET_TYPE (objfile_type->nodebug_text_gnu_ifunc_symbol)
|
||
= objfile_type->nodebug_text_symbol;
|
||
objfile_type->nodebug_got_plt_symbol
|
||
= init_type (TYPE_CODE_PTR, gdbarch_addr_bit (gdbarch) / 8, 0,
|
||
"<text from jump slot in .got.plt, no debug info>",
|
||
objfile);
|
||
TYPE_TARGET_TYPE (objfile_type->nodebug_got_plt_symbol)
|
||
= objfile_type->nodebug_text_symbol;
|
||
objfile_type->nodebug_data_symbol
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_int_bit (gdbarch) / HOST_CHAR_BIT, 0,
|
||
"<data variable, no debug info>", objfile);
|
||
objfile_type->nodebug_unknown_symbol
|
||
= init_type (TYPE_CODE_INT, 1, 0,
|
||
"<variable (not text or data), no debug info>", objfile);
|
||
objfile_type->nodebug_tls_symbol
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_int_bit (gdbarch) / HOST_CHAR_BIT, 0,
|
||
"<thread local variable, no debug info>", objfile);
|
||
|
||
/* NOTE: on some targets, addresses and pointers are not necessarily
|
||
the same.
|
||
|
||
The upshot is:
|
||
- gdb's `struct type' always describes the target's
|
||
representation.
|
||
- gdb's `struct value' objects should always hold values in
|
||
target form.
|
||
- gdb's CORE_ADDR values are addresses in the unified virtual
|
||
address space that the assembler and linker work with. Thus,
|
||
since target_read_memory takes a CORE_ADDR as an argument, it
|
||
can access any memory on the target, even if the processor has
|
||
separate code and data address spaces.
|
||
|
||
In this context, objfile_type->builtin_core_addr is a bit odd:
|
||
it's a target type for a value the target will never see. It's
|
||
only used to hold the values of (typeless) linker symbols, which
|
||
are indeed in the unified virtual address space. */
|
||
|
||
objfile_type->builtin_core_addr
|
||
= init_type (TYPE_CODE_INT,
|
||
gdbarch_addr_bit (gdbarch) / 8,
|
||
TYPE_FLAG_UNSIGNED, "__CORE_ADDR", objfile);
|
||
|
||
set_objfile_data (objfile, objfile_type_data, objfile_type);
|
||
return objfile_type;
|
||
}
|
||
|
||
extern initialize_file_ftype _initialize_gdbtypes;
|
||
|
||
void
|
||
_initialize_gdbtypes (void)
|
||
{
|
||
gdbtypes_data = gdbarch_data_register_post_init (gdbtypes_post_init);
|
||
objfile_type_data = register_objfile_data ();
|
||
|
||
add_setshow_zuinteger_cmd ("overload", no_class, &overload_debug,
|
||
_("Set debugging of C++ overloading."),
|
||
_("Show debugging of C++ overloading."),
|
||
_("When enabled, ranking of the "
|
||
"functions is displayed."),
|
||
NULL,
|
||
show_overload_debug,
|
||
&setdebuglist, &showdebuglist);
|
||
|
||
/* Add user knob for controlling resolution of opaque types. */
|
||
add_setshow_boolean_cmd ("opaque-type-resolution", class_support,
|
||
&opaque_type_resolution,
|
||
_("Set resolution of opaque struct/class/union"
|
||
" types (if set before loading symbols)."),
|
||
_("Show resolution of opaque struct/class/union"
|
||
" types (if set before loading symbols)."),
|
||
NULL, NULL,
|
||
show_opaque_type_resolution,
|
||
&setlist, &showlist);
|
||
|
||
/* Add an option to permit non-strict type checking. */
|
||
add_setshow_boolean_cmd ("type", class_support,
|
||
&strict_type_checking,
|
||
_("Set strict type checking."),
|
||
_("Show strict type checking."),
|
||
NULL, NULL,
|
||
show_strict_type_checking,
|
||
&setchecklist, &showchecklist);
|
||
}
|