* varobj.c (varobj_add_child, install_dynamic_child): Wrap into #if
HAVE_PYTHON.
(instantiate_pretty_printer): Move HAVE_PYTHON outside of the function.
fields to allow larger integer sizes.
(read_subrange_type): Increase size of bound values.
Add logic to determine signedness based on base-type size, signedness.
(read_attribute_value): Change format for bad byte size in message.
(read_8_bytes): Increase size of result type.
(dump_die_shallow): Change format for value.
(dwarf2_get_attr_constant_value): Increase size of return type.
Correct comment.
* gdbtypes.c (create_range_type): Change API to increase size of
bounds. struct field -> union field.
Always take signedness from base type.
(check_typedef): Use new API for TYPE_LOW_BOUND, TYPE_HIGH_BOUND.
(recursive_dump_type, copy_type_recursive): Adjust to new
representation of range types.
* gdbtypes.h (fields_or_bounds): New union containing struct field and
new struct range_bounds, used for range types.
(TYPE_RANGE_DATA): New macro to access range_bounds member.
(TYPE_LOW_BOUND, TYPE_HIGH_BOUND): Represent with new TYPE_RANGE_DATA.
(TYPE_LOW_BOUND_UNDEFINED, TYPE_HIGH_BOUND_UNDEFINED): New macros,
taking over the job of TYPE_FIELD_ARTIFICIAL for range bounds.
(SET_TYPE_LOW_BOUND, SET_TYPE_HIGH_BOUND, SET_TYPE_LOW_BOUND_DEFINED)
(SET_TYPE_HIGH_BOUND_DEFINED): New macros.
(TYPE_FIELDS, TYPE_BASECLASS, TYPE_BASECLASS_NAME, TYPE_FIELD)
(TYPE_ARRAY_UPPER_BOUND_IS_UNDEFINED)
(TYPE_ARRAY_LOWER_BOUND_IS_UNDEFINED, TYPE_ARRAY_UPPER_BOUND_VALUE)
(TYPE_ARRAY_LOWER_BOUND_VALUE): Adjust to new representation.
(create_range_type): Adjust API.
* ada-lang.c (ada_modulus): Use new extended bound values.
(discrete_type_low_bound): Rename to...
(ada_discrete_type_low_bound): ... and make external.
(discrete_type_high_bound): Rename to...
(ada_discrete_type_high_bound): ... and make external.
(ada_value_slice_from_ptr, ada_array_bound_from_type)
(ada_evaluate_subexp, to_fixed_range_type):
Use ada_discrete_type_low_bound, ada_discrete_type_high_bound.
* ada-typeprint.c (print_range): Use ada_discrete_type_low_bound,
ada_discrete_type_high_bound. Don't look at field count, which
is no longer meaningful. Print bounds whenever argument is a range
or enumeration.
* ada-lang.h (ada_discrete_type_low_bound,ada_discrete_type_high_bound):
Declare.
* varobj.c (c_describe_child): Adjust to render larger values.
* mdebugread.c (parse_type): Use proper abstractions for range types:
TYPE_RANGE_DATA, SET_TYPE_LOW_BOUND_DEFINED,
SET_TYPE_HIGH_BOUND_DEFINED.
* p-typeprint.c (pascal_type_print_varspec_prefix): Use larger format
for bounds.
Paul Brook <paul@codesourcery.com>
* c-typeprint.c (c_type_print_base): Skip artificial fields.
Use get_vptr_fieldno to skip the vtable pointer.
* dwarf2read.c (dwarf2_add_field): Set FIELD_ARTIFICIAL on artificial
fields.
(dwarf2_add_member_fn): Complain about virtual member functions
without DW_AT_vtable_elem_location and force TYPE_CPLUS_DYNAMIC.
* gdbtypes.c (get_vptr_fieldno): Update comment.
* gdbtypes.h (struct cplus_struct_type): Add is_dynamic.
(TYPE_CPLUS_DYNAMIC): New macro.
* gnu-v3-abi.c (gnuv3_dynamic_class): New.
(gnuv3_get_vtable): Rewrite to use gnuv3_dynamic_class. Move higher.
(gnuv3_rtti_type, gnuv3_get_virtual_fn, gnuv3_baseclass_offset): Use
gnuv3_get_vtable.
* varobj.c (cplus_class_num_children, cplus_describe_child): Skip
artificial fields. Use get_vptr_fieldno to skip the vtable pointer.
* varobj.c (varobj_create): Error if the specified frame was not found
and it is needed.
gdb/testsuite/
* gdb.mi/mi2-var-cmd.exp (create variable with invalid FRAME-ADDR): New.
Replace public function varobj_list by all_root_varobjs iterator.
* mi/mi-cmd-var.c (struct mi_cmd_var_update, mi_cmd_var_update_iter):
New.
(mi_cmd_var_update): Replace the varobj_list call by all_root_varobjs.
Remove the variables rootlist, cr. New variable data.
* varobj.c (rootcount, varobj_list): Remove.
(install_variable, uninstall_variable): Remove the rootcount updates.
(all_root_varobjs): New function.
(varobj_invalidate): Use the all_root_varobjs call. Move the code to...
(varobj_invalidate_iter): ... a new function.
* varobj.h (varobj_list): Remove the prototype.
(all_root_varobjs): New prototype.
smash_to_memberptr_type, lookup_array_range_type, check_stub_method):
Use type architecture instead of current_gdbarch.
* gdbtypes.h (address_space_name_to_int, address_space_int_to_name):
Add GDBARCH paramter.
* gdbtypes.c (address_space_name_to_int, address_space_int_to_name):
Add GDBARCH parameter. Use it instead of current_gdbarch.
* c-typeprint.c (c_type_print_modifier): Update call.
* parse.c (push_type_address_space): Likewise.
* findvar.c (extract_typed_address, store_typed_address): Use type
architecture instead of current_gdbarch.
* value.c (value_as_address, unpack_field_as_long): Use type architecture
instead of current_gdbarch.
* doublest.c (floatformat_from_length): Add GDBARCH argument. Use it
instead of current_gdbarch.
(floatformat_from_type): Pass type architecture.
* infcall.c (find_function_addr): Use type architecture instead
of current_gdbarch.
* valarith.c (value_bitstring_subscript, value_x_binop, value_neg,
value_bit_index): Use type architecture instead of current_gdbarch.
* valops.c (value_cast, value_slice): Likewise.
* value.h (modify_field): Add TYPE argument.
* value.c (modify_field): Add TYPE argument. Use type architecture
instead of current_gdbarch.
(set_internalvar_component): Likewise.
* eval.c (evaluate_struct_tuple): Update call.
* valops.c (value_assign): Likewise.
* ada-lang.c (modify_general_field): Likewise. Add TYPE argument.
(make_array_descriptor): Update calls.
(move_bits): Add BITS_BIG_ENDIAN_P argument. Use it instead of
current_gdbarch.
(ada_value_assign, value_assign_to_component): Update calls.
(decode_packed_array, ada_value_primitive_packed_val, ada_value_assign,
value_assign_to_component): Use type arch instead of current_gdbarch.
* printcmd.c (float_type_from_length): Remove GDBARCH argument,
use type architecture instead.
(print_scalar_formatted, printf_command): Update calls. Use type
architecture instead of current_gdbarch.
* valprint.c (val_print_type_code_int): Use type architecture
instead of current_gdbarch.
* varobj.c (value_get_print_value): Likewise.
* python/python-prettyprint.c (print_string_repr): Add GDBARCH
argument. Use it instead of current_gdbarch.
(apply_val_pretty_printer): Update call.
* ada-valprint.c (ada_val_print_1): Use type architecture instead
of current_gdbarch.
* c-valprint.c (print_function_pointer_address): Add GDBARCH argument.
Use it instead of current_gdbarch.
(c_val_print): Update calls passing type architecture.
* f-valprint.c (f_val_print): Use type architecture instead of
current_gdbarch.
* jv-valprint (java_value_print): Likewise.
* m2-valprint.c (print_function_pointer_address): Add GDBARCH argument.
Use it instead of current_gdbarch.
(print_unpacked_pointer): Update calls passing type architecture.
* scm-valprint.c (scm_scmval_print): Use type architecture instead of
current_gdbarch.
* gnu-v3-abi.c (get_class_arch): Remove.
(gnuv3_rtti_type): Use get_type_arch instead of get_class_arch. Remove
special-case check for Java classes.
(gnuv3_virtual_fn_field, gnuv3_baseclass_offset, gnuv3_print_method_ptr,
gnuv3_method_ptr_size, gnuv3_make_method_ptr, gnuv3_method_ptr_to_value):
Use get_type_arch instead of get_class_arch.
value_bitstring_subscript, value_ptradd): Use LONGEST instead
of value as type of the index argument.
(value_ptrsub): Remove.
* valarith.c (value_subscript, value_subscripted_rvalue,
value_bitstring_subscript, value_ptradd): Use LONGEST instead
of value as type of the index argument.
(value_ptrsub): Remove.
* wrapper.h (gdb_value_subscript): Use LONGEST instead of
value as type of the index argument.
* wrapper.c (gdb_value_subscript): Likewise.
Update calls to gdb_value_subscript, value_subscript,
value_subscripted_rvalue, value_bitstring_subscript and
value_ptradd to use LONGEST instead of value as index
argument type. Use value_ptradd instead of value_ptrsub.
* ada-lang.c (ada_value_subscript, ada_value_ptr_subscript,
ada_tag_name_2): Update.
* ada-tasks.c (read_atcb): Update.
* eval.c (evaluate_subexp_standard): Update.
* valarith.c (value_subscript): Update.
* gnu-v2-abi.c (gnuv2_virtual_fn_field): Update.
* gnu-v3-abi.c (gnuv3_get_virtual_fn, gnuv3_baseclass_offset,
gnuv3_method_ptr_to_value): Update.
* jv-lang.c (evaluate_subexp_java): Update.
* m2-lang.c (evaluate_subexp_modula2): Update.
* python/python-value.c (valpy_getitem, valpy_binop): Update.
* wrapper.c (gdb_value_subscript): Update.
* varobj.c (c_describe_child): Update.
* printcmd.c (print_scalar_formatted): Always truncate
unsigned data types.
* cli-dump.c (struct callback_data): Change type of load_offset
to CORE_ADDR.
(restore_binary_file): Update type casts.
(restore_command): Parse load_offset as address, not long.
* utils.c (string_to_core_addr): Do not sign-extend value.
* varobj.c (find_frame_addr_in_frame_chain): Truncate frame_base
before comparing against requested frame address.
testsuite/ChangeLog:
* gdb.base/dump.exp: Handle SPU like 64-bit platforms.
* varobj.c (varobj_invalidate): Unconditionally free
all_rootvarobj.
* symfile.c (syms_from_objfile): Free local_addr when returning
normally.
* exec.c (exec_file_attach): Do cleanups before returning.
(exec_file_command): Likewise.
* corefile.c (reopen_exec_file): Do cleanups before returning.
* breakpoint.c (insert_breakpoint_locations): Do cleanups before
returning.
(do_vec_free): New function.
(update_global_location_list): Make a cleanup for old_locations.
Do cleanups before returning. Remove unused variable 'e'.
(find_condition_and_thread): Free result of parsing the
expression.
(print_it_typical): Do cleanups before returning.
(breakpoint_re_set_one): Always free sals.sals.
Replace TYPE_ARRAY_{UPPER,LOWER}_BOUND_TYPE by a bit if {un,}defined.
* c-typeprint.c (c_type_print_varspec_suffix), m2-typeprint.c
(m2_array), p-typeprint.c (pascal_type_print_varspec_prefix),
valops.c (value_cast), varobj.c (c_number_of_children): Replace
TYPE_ARRAY_UPPER_BOUND_TYPE compared to BOUND_CANNOT_BE_DETERMINED by
TYPE_ARRAY_UPPER_BOUND_IS_UNDEFINED.
* parse.c (follow_types): Use TYPE_ARRAY_UPPER_BOUND_IS_UNDEFINED.
* f-valprint.c (f77_get_dynamic_upperbound): Replace with ...
(f77_get_upperbound): ... this function handling now only
TYPE_ARRAY_UPPER_BOUND_IS_UNDEFINED.
(f77_get_dynamic_lowerbound): Replace with ...
(f77_get_lowerbound): ... this function handling now only
TYPE_ARRAY_LOWER_BOUND_IS_UNDEFINED.
(f77_get_dynamic_length_of_aggregate, f77_create_arrayprint_offset_tbl):
Update their callers.
* eval.c (evaluate_subexp_standard): Update their callers.
* f-lang.h (f77_get_dynamic_upperbound, f77_get_upperbound)
(f77_get_dynamic_lowerbound, f77_get_lowerbound): Update their
prototypes.
(BOUND_FETCH_OK, BOUND_FETCH_ERROR): Remove.
* f-typeprint.c (f_type_print_varspec_suffix, f_type_print_base): Remove
the lower_bound_was_default variable. Update the
f77_get_dynamic_upperbound, f77_get_upperbound and
TYPE_ARRAY_UPPER_BOUND_TYPE calls.
* gdbtypes.c (print_bound_type): Remove the function.
(recursive_dump_type): Remove its calls printing UPPER_BOUND_TYPE and
LOWER_BOUND_TYPE.
* gdbtypes.h (enum array_bound_type): Remove.
(struct main_type): Remove the fields upper_bound_type and
lower_bound_type. Comment the new overload of the field artificial.
(TYPE_ARRAY_UPPER_BOUND_TYPE): Replace by ...
(TYPE_ARRAY_UPPER_BOUND_IS_UNDEFINED): ... this macro.
(TYPE_ARRAY_LOWER_BOUND_TYPE): Replace by ...
(TYPE_ARRAY_LOWER_BOUND_IS_UNDEFINED): ... this macro.
gdb/testsuite/
* gdb.base/maint.exp (maint print type): Remove printing
UPPER_BOUND_TYPE and LOWER_BOUND_TYPE.
* tracepoint.c (stringify_collection_list): Use xfree.
* remote-fileio.c (remote_fileio_reset): Use xfree.
* mipsread.c (read_alphacoff_dynamic_symtab): Use xfree.
* dfp.c (decimal_from_floating): Use xfree, xstrprintf. Don't use
asprintf.
* cp-support.c (mangled_name_to_comp): Use xfree.
* thread.c (enum thread_state): New.
(thread_state main_thread_running): Delete, in favor of...
(thread_state main_thread_state): ... this. Update throughout.
(clear_thread_inferior_resources): New, split from free_thread.
(free_thread): Call clear_thread_inferior_resources.
(init_thread_list): Set main thread to stopped state.
(add_thread_silent): Take care of PTID reuses.
(delete_thread): If deleting inferior_ptid or a thread with
refcount > 0, mark it as exited, but still keep it in the list.
Only notify of thread exits, if we haven't done so yet.
(iterate_over_threads): Make it safe to delete threads while
iterating over them.
(do_captured_list_thread_ids): Don't account for exited threads.
(thread_alive): Check for the THREAD_EXITED state, and don't set
ptid to -1 on exited threads.
(set_running): Update to account for extra possible states.
(is_thread_state): New.
(is_stopped, is_exited): New.
(is_running): Implement in terms of is_thread_state.
(any_running): Update.
(print_thread_info): Update. Account for exited threads. Don't
warn about missed frame restoring here, its done in the cleanup.
(switch_to_thread): Don't read from a thread that has gone.
(restore_current_thread): In non-stop mode, do a full context
switch.
(restore_selected_frame): Add a frame_level argument. Rewrite.
(struct current_thread_cleanup): Add selected_frame_level and
was_stopped members.
(do_restore_current_thread_cleanup): Check if thread was stopped
and still is, and if the target has registers, stack and memory
before restoring the selected frame. Don't delete the cleanup
argument here.
(restore_current_thread_cleanup_dtor): New.
(make_cleanup_restore_current_thread): Remove all arguments.
Rewrite.
(thread_apply_all_command): Update. Prune threads.
(thread_apply_command): Update.
(thread_command): Account for currently selected exited thread.
(do_captured_thread_select): Check for a running thread. Prune
threads.
(_initialize_thread): Make "info threads", "thread", "thread
apply", and "thread apply all" appliable without a selected thread.
* gdbthread.h (struct thread_info): Replace running_ by state_.
Add refcount.
(is_exited, is_stopped): Declare.
(make_cleanup_restore_current_thread): Remove all arguments.
* infrun.c: Include "event-top.h".
(fetch_inferior_event): In non-stop mode, restore selected thread
and frame after handling the event and running breakpoint
commands. Display GDB prompt if needed.
(normal_stop): In non-stop mode, don't print thread switching
notice.
* cli/cli-decode.c (set_cmd_no_selected_thread_ok)
(get_cmd_no_selected_thread_ok): New.
* cli/cli-decode.h (CMD_NO_SELECTED_THREAD_OK): New.
(set_cmd_no_selected_thread_ok, get_cmd_no_selected_thread_ok):
Declare.
* cli/cli-cmds.c: Set "pwd", "help", "info", "show" as
no-selected-thread ok.
* top.c (execute_command): Check for non no-selected-thread-ok
commands.
* linux-nat.c (struct saved_ptids, threads_to_delete)
(record_dead_thread, prune_lwps): Delete.
(exit_lwp): Unconditionally delete thread.
(linux_nat_resume): Remove prune_lwps call.
* infcmd.c (proceed_thread_callback): Check if !is_stopped instead
of is_running. Adjust to make_cleanup_restore_current_thread
interface change.
* mi/mi-main.c (mi_cmd_execute): Only allow a few commands if the
selected thread has exited.
* inf-loop.c (inferior_event_handler): Don't display the prompt
here.
* varobj.c (c_value_of_root): Update.
* defs.h (make_cleanup_dtor): Declare.
* utils.c (make_cleanup_dtor): New.
* Makefile.in (infrun.o): Depend on $(event_top_h).
* varobj.c (varobj_update): Report changes as vector. Also
return not just a list of varobj, but a list of special structures
that tell what exactly has changed.
* varobj.h (enum varobj_update_error): Rename to
varobj_scope_status.
(struct varobj_update_result_t): New.
(varobj_update): Adjust prototype.
* mi/mi-cmd-var.c: Adjust for changes.
2008-04-09 Marc Khouzam <marc.khouzam@ericsson.com>
* mi/mi-cmd-var.c: Include "mi-getopt.h".
(mi_parse_format): New. Factored out from mi_cmd_var_set_format.
(mi_cmd_var_set_format): Use new mi_parse_format.
(mi_cmd_var_evaluate_expression): Support for -f option to specify
format.
* Makefile.in (mi-cmd-var.o): Update dependencies.
* varobj.h (varobj_get_formatted_value): Declare.
* varobj.c (my_value_of_variable): Added format parameter.
(cplus_value_of_variable): Likewise.
(java_value_of_variable): Likewise.
(c_value_of_variable): Likewise. Evaluate expression based
on format parameter.
(struct language_specific): Add format parameter to function member
*value_of_variable.
(varobj_get_formatted_value): New.
(varobj_get_value): Added format parameter to method call.
gdb/doc/ChangeLog
2008-04-09 Marc Khouzam <marc.khouzam@ericsson.com>
* gdb.texinfo (GDB/MI Variable Objects): Add anchor to
-var-set-format. Add -f option to -var-evaluate-expression.
gdb/testsuite/ChangeLog
2008-04-09 Marc Khouzam <marc.khouzam@ericsson.com>
* gdb.mi/mi2-var-display.exp: Added tests for the new -f
option of -var-evaluate-expression.
* gdb.mi/mi2-var-display.exp: Likewise.
floating, and clarify the meaning.
(varobj_create, varobj_update, new_root_variable): Adjust.
(value_of_root): Don't use type_changed as in variable,
adjust comment.
(c_value_of_root): Adjust.
(varobj_get_thread_id, check_scope): New functions.
(c_value_of_root): Use check_scope. Switch to the
proper thread if necessary.
* varobj.h (varobj_get_thread_id): New extern.
* mi/mi-cmd-var.c (print_varobj): Add thread-id field.
* varobj (adjust_value_for_child_access): Added checking for
returned value from gdb_value_ind.
(c_describe_child): Likewise.
(cplus_describe_child): Fixed a typo.
(java_variable_editable, variable_editable): Delete.
(varobj_editable_p): Replace above functions with one language
independent function. Check for an lvalue.
(varobj_get_attributes, varobj_set_value): Use varobj_editable_p.
(struct language_specific): Delete variable_editable field.
(varobj_get_frozen): New.
(varobj_update): New parameter explicit.
* varobj.c (struct varobj): New fields frozen
and not_fetched.
(varobj_set_frozen, varobj_get_frozen): New.
(install_new_value): Don't fetch values for
frozen variable object, or children thereof. Allow
a frozen variable object to have non-fetched value.
(varobj_update): Allow updating child variables.
Don't traverse frozen children.
(new_variable): Initialize the frozen field.
(c_value_of_variable): Return NULL for frozen
variable without any value yet.
* mi/mi-cmd-var.c (varobj_update_one): New parameter
'explicit'.
(mi_cmd_var_create): Output the 'frozen' field,
as soon as testsuite is adjusted to expect that field.
(mi_cmd_var_set_frozen): New.
(mi_cmd_var_update): Pass the 'explicit' parameter to
varobj_update_one.
* mi/mi-cmds.c (mi_cmds): Register '-var-set-frozen'.
* mi/mi-cmds.h (mi_cmd_var_set_frozen): Declare.
* varobj.h (enum varobj_update_error): New enum.
* varobj.c (struct varobj_root): Add is_valid member.
(varobj_get_type): Check for invalid varobj.
(varobj_get_attributes): Likewise.
(variable_editable):Likewise.
(varobj_update): Likewise. Use varobj_update_error.
(new_root_variable): Set root varobj as valid by default.
(varobj_invalidate): New function.
* symfile.c (clear_symtab_users): Use varobj_invalidate.
* mi/mi-cmd-var.c (varobj_update_one): Change return type to void.
Use varobj_update_error.
for varobjs in C++.
* varobj.c (get_type_deref): Remove.
(adjust_value_for_child_access): New.
(c_number_of_children): Use the above.
(c_describe_child): Likewise.
(enum accessibility): New.
(match_accessibility): New function.
(cplus_describe_child): New function.
(cplus_name_of_child, cplus_value_of_child)
(cplus_type_of_child): Reimplement in terms
of cplus_describe_child.
(cplus_number_of_children): Use
adjust_value_for_child_access.
value changeability for for references.
* varobj.c (get_value_type): New function.
(c_variable_editable): Use get_value_type.
(varobj_value_is_changeable): Likewise.
* varobj.c (struct varobj): Remove the error field.
(varobj_set_value): Don't check var->error.
(install_new_value): Don't set var->error.
(varobj_update): Always pass the new value
of the root via install_new_value.
(create_child): Don't set error field.
(new_variable): Likewise.
(c_value_of_root): Always reevaluate the value
of selected frame varobjs in the selected frame.
Don't call reinit_frame_cache.
(value_get_print_value): Immediately return NULL for missing
values.
* gdb.mi/mi-var-cmd.exp: Expect lpcharacter to update when
lcharacter or linteger change. Correct duplicated test name.
* gdb.mi/mi2-var-cmd.exp: Likewise.
(install_new_value): Compare last printed value with current one
instead of contents.
(new_variable): Initialize var->print_value to NULL.
(free_variable): Free var->print_value.
(value_get_print_value): New function derived from
c_value_of_variable.
(c_value_of_variable): Use value_get_print_value.
* eval.c: Include "ui-out.h" and "exceptions.h".
(evaluate_subexp_standard): Use TRY_CATCH around value_of_variable.
Use value_zero if an error occurs when avoiding side effects.
* varobj.c (c_value_of_root): Initialize new_val.
* gdb.mi/mi-var-cmd.exp: Add tests for unreadable varobjs.
variable objects in C.
* varobj.c (value_struct_element_index): New function.
(c_describe_child): New function.
(c_name_of_child, c_value_of_child)
(c_type_of_child): Rewrite to use c_describe_child.
* varobj.c: Include "vec.h".
(varobj_p): New typedef, declare vector of those.
(struct varobj): Use vector for the 'children' member.
(child_exists): Remove.
(save_child_in_parent): Remove.
(remove_child_from_parent): Remove.
(struct varobj_child): Remove.
(struct vstack): Remove.
(vpush, vpop): Remove.
(varobj_list_children): Adjust to work work vector.
(varobj_update): Likewise. Use vectors for
working stack and result.
(delete_variable_1): Likewise.
* Makefile.in (varobj.o): Update dependencies.
gdb/
* varobj.c (type_changeable): Rename to...
(varobj_value_is_changeable_p): ...this. Adjust all callers.
(is_root_p): New function. Use it everywhere.
* Makefile.in (gnu-v3-abi.o): Delete special rule.
(eval.o, gnu-v3-abi.o, ia64-tdep.o): Update.
* ada-valprint.c (ada_print_scalar): Update for new type codes.
* c-typeprint.c (c_print_type): Update for new type codes.
(c_type_print_varspec_prefix, c_type_print_varspec_suffix)
(c_type_print_base): Likewise.
(c_type_print_args): Rewrite.
* c-valprint.c (c_val_print): Update for new type codes. Remove
support for references to members. Treat methods like functions.
* cp-abi.c (cplus_print_method_ptr, cplus_method_ptr_size)
(cplus_make_method_ptr, cplus_method_ptr_to_value): New.
* cp-abi.h (cplus_print_method_ptr, cplus_method_ptr_size)
(cplus_make_method_ptr, cplus_method_ptr_to_value): New prototypes.
(struct cp_abi_ops): Add corresponding members.
* cp-valprint.c (cp_print_class_method): Delete.
(cp_find_class_member): New function.
(cp_print_class_member): Use it. Simplify support for bogus
member pointers.
* dwarf2read.c (quirk_gcc_member_function_pointer): Use
lookup_methodptr_type.
(read_tag_ptr_to_member_type): Likewise, and lookup_memberptr_type.
* eval.c (evaluate_subexp_standard): Implement EVAL_SKIP for
OP_SCOPE. Update call to value_aggregate_elt. Rewrite member
pointer support.
(evaluate_subexp_for_address): Handle OP_SCOPE explicitly. Handle
references returned by user defined operators.
* f-typeprint.c (f_print_type, f_type_print_varspec_prefix)
(f_type_print_varspec_suffix): Remove support for member pointers.
* gdbtypes.c (lookup_memberptr_type): Renamed from lookup_member_type
and adjusted.
(smash_to_memberptr_type): Likewise, from smash_to_member_type.
(lookup_methodptr_type): New.
(rank_one_type): Adjust for TYPE_CODE_MEMBERPTR.
(recursive_dump_type): Update for new types.
* gdbtypes.h (enum type_code): Replace TYPE_CODE_MEMBER with
TYPE_CODE_MEMBERPTR and TYPE_CODE_METHODPTR.
(lookup_memberptr_type, lookup_methodptr_type)
(smash_to_memberptr_type): New prototypes.
(smash_to_method_type): Formatting fix.
(lookup_member_type, smash_to_member_type): Delete prototypes.
* gnu-v3-abi.c (gnuv3_get_vtable, gnuv3_get_virtual_fn): New.
Do not rely on debug information for the vptr or the method's
enclosing type. Handle function descriptors for IA64.
(gnuv3_virtual_fn_field): Rewrite using the new functions.
(gnuv3_find_method_in, gnuv3_print_method_ptr)
(gnuv3_method_ptr_size, gnuv3_make_method_ptr)
(gnuv3_method_ptr_to_value): New.
(init_gnuv3_ops): Set new members of gnu_v3_abi_ops.
* hpread.c (hpread_type_lookup): Update for new types.
* infcall.c (value_arg_coerce): Likewise.
* m2-typeprint.c (m2_print_type): Remove explicit support
for member pointers.
* m2-valprint.c (m2_val_print): Likewise.
* p-typeprint.c (pascal_type_print_varspec_prefix)
(pascal_type_print_varspec_suffix, pascal_type_print_base): Likewise.
* p-valprint.c (pascal_val_print): Likewise.
(pascal_object_print_class_method, pascal_object_print_class_member):
Delete.
* p-lang.h (pascal_object_print_class_method)
(pascal_object_print_class_member): Delete prototypes.
* stabsread.c (read_type): Update for new types.
* typeprint.c (print_type_scalar): Likewise.
* valops.c (value_struct_elt_for_reference, value_namespace_elt)
(value_maybe_namespace_elt, value_aggregate_elt): Add want_address
argument. Construct a pointer to member if the address of a
function or data member is requested.
(value_cast_pointers): Don't modify the input value.
(value_cast): Adjust pointer to member handling for new types.
Allow null pointer to member constants. Don't modify the input
value.
(value_ind): Remove pointer to member check. Handle function
descriptors for function pointers.
(value_struct_elt, value_find_oload_method_list, check_field):
Remove pointer to member checks.
* value.c (unpack_long): Allow pointers to data members.
(value_from_longest): Allow member pointers.
* value.h (value_aggregate_elt): Add want_address.
* varobj.c (c_variable_editable): Remove check for members.
* gdbarch.sh: Add vtable_function_descriptors and vbit_in_delta.
* ia64-tdep.c (ia64_convert_from_func_ptr_addr): Handle descriptors
in virtual tables.
(ia64_gdbarch_init): Call set_gdbarch_vtable_function_descriptors.
* c-lang.h (cp_print_class_method): Delete prototype.
* arm-tdep.c (arm_gdbarch_init): Call set_gdbarch_vbit_in_delta.
* mips-tdep.c (mips_gdbarch_init): Likewise.
* gdbarch.c, gdbarch.h: Regenerated.
* gdb.cp/classes.exp (test_pointers_to_class_members): Update expected
output. Test the types of members and member pointers.
* gdb.cp/inherit.exp (test_print_mi_member_types): Remove KFAILs for
gdb/2092.
* gdb.cp/member-ptr.exp: Search for a comment instead of a
statement. Enable for GCC. Update expected output for some tests
and add new tests. Remove obsolete GCC KFAILs. Allow GCC's class
layout.
* gdb.cp/member-ptr.cc (Padding, Padding::vspacer, Base, Base::get_x)
(Base::vget_base, Left, Left::vget, Right, Right::vget, Diamond)
(Diamond::vget_base): New.
(main): Add new tests.
* gdb.cp/printmethod.exp: Update expected output for member functions.
* gdb.cp/virtfunc.exp (test_virtual_calls): Add a KFAIL for
print pEe->D::vg().
Fetch varobj values from memory in a single place,
and only fetch the values that are really needed.
* varobj.c (struct varobj): Clarify comment.
(my_value_equal): Remove.
(install_new_value): New function.
(type_of_child): Remove.
(varobj_create): Use install_new_value.
(varobj_set_value): Use value_contents_equal, not
my_value_equal.
(varobj_update): Use install_new_value.
(create_child): Likewise. Inline type_of_child here.
(value_of_child): Don't fetch the value.
(c_value_of_root): Likewise.
(c_value_of_variable): Likewise.
(type_changeable): Improve comments.
