Fix a regression by the mmap patch from 2009-06-16.
* symfile.c (reread_symbols): Move sym_finish and clear_objfile_data
calls before deleting the OBJFILE obstack. Extend the comment.
* dwarf2read.c (process_die): Handle import statements
(DW_TAG_imported_declaration, case DW_TAG_imported_module)
(read_import_statement): New.
(read_func_scope): Update using_directives to point to current context
(read_lexical_block_scope): Ditto.
* cp-support.h: Added prototype for cp_add_using.
* cp-namespace.c: Removed local context_stack.
(cp_initialize_namespace): Deleted.
(cp_finalize_namespace): Deleted.
(cp_add_using_directive): Use using_directives instead of using_list.
(cp_add_using): No longer static.
* buildsym.h: Created global using_direct variable.
Created using_direct variable in context_stack.
* buildsym.c (finish_block): Set using directives for the block under
construction.
(start_symtab): Removed call to cp_initialize_namespace().
(end_symtab): Removed call to cp_finalize_namespace().
(push_context): Save and reset using_directives.
* block.c (block_using): Return using directives for given
block instead of static block.
2009-06-23 Sami Wagiaalla <swagiaal@redhat.com>
* gdb.cp/namespace-using.exp: New test.
* gdb.cp/namespace-using.cc: New test.
an array, then use value_ind to dereference it, not ada_value_ind.
(ada_template_to_fixed_record_type_1): Handle the case where a
dynamic field is an aligner type.
(to_fixed_array_type): Add handling for packed array types.
(ada_get_base_type): Do not follow the XVS type if the type is
an aligner type.
(ada_evaluate_subexp): Split a few statements up for clarity
and debuggability.
<OP_FUNCALL>: Add handling of the case of a packed array that
has already been fixed.
* machoread.c (macho_symtab_read): Adjust for bfd changes.
* darwin-nat.h (struct darwin_exception_msg): New type to describe
a mach exception.
(struct private_thread_info): New type to describe the state of the
thread.
(DEF_VEC_I thread_t): Removed, replaced by ...
(DEF_VEC_O darwin_thread_t): ... this new type.
(struct darwin_inferior): Renamed to ...
(struct private_inferior): ... this type. Fields added.
(darwin_not_port): Moved into the private inferior structure.
* darwin-nat.c: Add includes, improve comments.
Rewrite to handle multiple threads and processes.
(darwin_resume_to): New function and protype.
(darwin_resume, darwin_wait_to, darwin_wait): Ditto.
(darwin_kill_inferior): Add ops argument.
(darwin_pid_to_str): New function.
(darwin_thread_alive): Ditto.
(darwin_inf, darwin_not_port): Removed.
(darwin_inf_fake_stop): New variable.
(msgin, msgout, msg_state, exc_msg): Removed.
(mach_check_error): Use warning instead of error.
(darwin_ptrace): Adjust debug level.
(cmp_thread_t): Fix names (typo).
(darwin_check_new_threads): Argument is now an inferior,
adjust for new structures, add no change check, ignore dead ports,
handle first thread case.
(find_inferior_task_it): New function.
(find_inferior_notify_it): Ditto.
(darwin_find_inferior_by_task): Ditto.
(darwin_find_inferior_by_notify): Ditto.
(darwin_find_thread): Ditto.
(darwin_suspend_inferior): Ditto.
(darwin_resume_inferior): Ditto.
(catch_exception_raise_state): Removed.
(catch_exception_raise_state_identity): Removed.
(darwin_suspend_inferior_it): New function.
(darwin_resume_inferior_it): Ditto.
(darwin_dump_message): New function, extracted from darwin_wait.
(darwin_decode_exception_message): New function.
(darwin_encode_reply): New function.
(catch_exception_raise): Removed.
(darwin_send_reply): New function, extracted from darwin_resume.
(darwin_resume_thread): New function, extracted from darwin_resume.
(struct resume_inferior_threads_param): New type.
(darwin_resume_inferior_threads_it): New function.
(darwin_resume_inferior_threads): New function.
(darwin_suspend_inferior_threads): New function.
(darwin_resume): Mostly rewritten to handle multiple threads and
some corner cases.
(darwin_decode_message): New function extracted from darwin_wait.
(cancel_breakpoint): New function.
(darwin_wait): Mostly rewritten. Handle multiple threads.
(darwin_mourn_inferior): Adjust for per process structures.
(darwin_reply_to_all_pending_messages): New function.
(darwin_stop_inferior): Adjust for per inferior structures.
(darwin_attach_pid): Ditto.
(darwin_init_thread_list): Ditto.
(darwin_attach): Ditto.
(darwin_detach): Ditto.
(darwin_files_info): Now empty.
(darwin_pid_to_str): Adjust returns string to match one expected by
the testsuite.
(darwin_read_write_inferior): Rename err variable to match other uses.
Adjust debug message. Handle submaps.
(darwin_xfer_memory): Adjust for per inferior structures.
(set_enable_mach_exceptions): Ditto.
(darwin_pid_to_exec_file): New function.
(darwin_get_ada_task_ptid): Ditto.
(darwin_supports_multi_process): Ditto.
(_initialize_darwin_inferior): Remove useless assertion, adjust for
per inferior structures. Add new target operations.
Pedro Alves <pedro@codesourcery.com>
* target.h (struct target_ops) <supports_multi_process>: New
callback.
(target_supports_multi_process): New.
* server.c (handle_query): Even if GDB reports support, only
enable multi-process if the target also supports it. Report
multi-process support only if the target backend supports it.
* linux-low.c (linux_supports_multi_process): New function.
(linux_target_ops): Install it as target_supports_multi_process
callback.
* 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.
* gdbtypes.h (create_string_type): Add character type argument.
* dwarf2read.c (read_tag_string_type): Pass character type to
create_string_type.
* value.h (value_string): Add character type argument.
* valops.c (value_string): Add character type argument. Pass it to
create_string_type. Do not allocate space in inferior.
* valarith.c (value_concat): Pass character type to value_string.
* value.h (value_typed_string): Rename to ...
(value_cstring): ... this.
* valops.c (value_typed_string): Rename to ...
(value_cstring): ... this.
* c-lang.c (evaluate_subexp_c): Update.
* python/python-value.c (builtin_type_pychar): New define.
(convert_value_from_python): Call value_cstring instead
of value_from_string.
* value.c (value_from_string): Remove.
* value.h (value_from_string): Remove.
* eval.c (evaluate_subexp_standard): Pass character type to
value_string. Pass expression architecture to value_nsstring
and lookup_child_selector.
* objc-lang.h (lookup_objc_class): Add GDBARCH parameter.
(lookup_child_selector): Likewise.
(value_nsstring): Likewise.
* objc-lang.c (lookup_objc_class): Add GDBARCH parameter.
Pass character type to value_string..
(lookup_child_selector): Likewise.
(value_nsstring): Add GDBARCH parameter, use it instead of
objfile architecture. Pass architecture to lookup_objc_class
and lookup_child_selector. Pass character type to value_string.
(end_msglist): Pass architecture to lookup_objc_class.
* objc-exp.y: Pass architecture to lookup_objc_class.
(lookup_typename): Add LANGUAGE and GDBARCH parameters.
(lookup_unsigned_typename): Likewise.
(lookup_signed_typename): Likewise.
* gdbtypes.c (lookup_typename): Add LANGUAGE and GDBARCH parameters.
Use them instead of current_language and current_gdbarch.
(lookup_unsigned_typename): Add LANGUAGE and GDBARCH parameters.
Pass them to lookup_typename.
(lookup_signed_typename): Likewise.
* c-exp.y: Pass parse_language and parse_gdbarch to
lookup_unsigned_typename and lookup_signed_typename.
* objc-exp.y: Likewise.
* m2-exp.y: Pass parse_language and parse_gdbarch to lookup_typename.
* c-lang.c (evaluate_subexp_c): Pass expression language and
gdbarch to lookup_typename.
* printcmd.c (printf_command): Pass current language and
gdbarch to lookup_typename.
* python/python-type.c (typy_lookup_typename): Likewise.
Include "language.h".
* mi/mi-main.c (mi_cmd_data_list_register_names): Use selected
frame architecture instead of current_gdbarch.
(mi_cmd_data_list_changed_registers): Likewise.
(mi_cmd_data_list_register_values): Likewise. Pass selected
frame to get_register.
(get_register): Add FRAME argument, use it as frame to retrieve
registers from (instead of selected frame). Use frame architecture
instead of current_gdbarch.
(mi_cmd_data_write_register_values): Use current regcache
architecture instead of current_gdbarch.
testsuite/ChangeLog:
* gdb.mi/gdb680.exp: Update test for error message.
of current_gdbarch.
(TARGET_LONG_BYTES): Remove.
(read_target_long_array): Add SIZE argument. Use it instead of
TARGET_LONG_BYTES.
(simple_read_overlay_table): Retrieve long size from objfile
architecture; pass it to read_target_long_array.
(simple_read_overlay_region_table): Likewise.
(simple_overlay_update_1): Likewise.
(displaced_step_clear_cleanup): Move to before displaced_step_prepare.
(displaced_step_prepare): Use displaced_step_clear_cleanup instead
of cleanup_displaced_step_closure.
(handle_inferior_event, keep_going): Use thread architecture instead
of current_gdbarch.
* infrun.c (handle_inferior_event): Use current frame architecture
or thread architecture instead of current_gdbarch. Pass to
handle_step_into_function and handle_step_into_function_backward.
(handle_step_into_function): Add GDBARCH parameter. Use it instead
of current_gdbarch.
(handle_step_into_function_backward): Likewise.
(insert_step_resume_breakpoint_at_frame): Use frame architecture
instead of current_gdbarch.
(insert_step_resume_breakpoint_at_caller): Likewise.
testsuite/ChangeLog:
* gdb.base/annota1.exp: Allow multiple occurrences of the
frames-invalid annotation.
* gdb.cp/annota2.exp: Likewise.
* avr-tdep.c (struct gdbarch_tdep): Replace the unused field with
call_length field.
(avr_register_name): Add const to register_names.
(avr_scan_arg_moves): Move inside avr_scan_prologue.
(avr_scan_prologue): Add pc_end argument.
Only read prologue bytes that can be read.
Limit the scan to the known prologue length.
Makes pattern variables static and const.
Fix indentation.
(avr_skip_prologue): Pass func_end argument to avr_scan_prologue.
Fix indentation.
(avr_breakpoint_from_pc): Constify avr_break_insn.
(avr_extract_return_value): Fix function comment.
(avr_frame_unwind_cache): Fix GNU style violations.
Pass current_pc argument to avr_scan_prologue to stop prologue
analysis to the current pc. This fixes the bug with the 'next'
command.
Correctly set the SP register of the previous frame (use call_length).
(avr_frame_prev_register): Fix indentation.
Correctly read PC from the stack on avr6 architectures.
(avr_push_dummy_call): Fix indentation.
(avr_gdbarch_init): Set call_length according to the architecture.
* infcall.c (show_unwind_on_terminating_exception_p): New
function.
(call_function_by_hand): Create breakpoint and clean-up call for
std::terminate.breakpoint. Add unwind_on_terminating_exception_p
gate. Pop frame on breakpoint hit.
(_initialize_infcall): Add add_setshow_boolean_cmd for
unwind-on-terminating-exception.
testsuite/
2009-06-15 Phil Muldoon <pmuldoon@redhat.com>
* gdb.cp/gdb2495.cc: New file.
* gdb.cp/gdb2495.exp: New file.
doc/
2009-06-15 Phil Muldoon <pmuldoon@redhat.com>
* doc/gdb.texinfo (Calling): Document
set-unwind-on-terminating-exception usage.
(i386nto_sigcontext_addr): Change the way context address is calculated.
(i386nto_init_abi): Remove usage of deprecated sc_reg_offset and
sc_sp_offset and set sc_reg_offset and sc_num_regs.
* i386-nto-tdep.c (i386nto_regset_id): Add case for SSE register set.
(i386nto_register_area): Correctly calculate offsets and sizes for
all supported registers.
* infrun.c (sched_multi): New global.
(resume): If sched_multi is set, resume only threads of the
current inferior.
(prepare_to_proceed): Don't switch over to wait_ptid if we're
resuming a different inferior, and sched_multi is off.
(show_schedule_multiple): New.
(_initialize_infrun): Register new "set schedule-multiple" command.
* inferior.h (sched_multi): Declare.
* NEWS: Mention new "schedule-multiple" setting.
gdb/doc/
* gdb.texinfo (All-Stop): Document new 'set schedule-multiple'
command.
(update_current_target): Don't inherit to_sections or
to_sections_end.
(target_get_section_table): New.
(target_section_by_addr): Fetch the section table from the passed
in target.
(memory_xfer_partial): Handle unmapped overlay sections before
anything else. Get the overlay mapped address here. Adjust to
use section_table_xfer_memory_partial.
(get_target_memory): Request a TARGET_OBJECT_RAW_MEMORY object
instead of TARGET_OBJECT_MEMORY.
(target_resize_to_sections): Delete.
(remove_target_sections): Adjust to remove target sections from
`current_target_sections', and use resize_section_table.
* target.h (struct target_ops) <to_sections, to_sections_end>:
Remove fields.
<to_get_section_table>: New method.
(xfer_memory, print_section_info): Delete declarations.
(struct target_section_table): New type.
(target_get_section_table): Declare.
(target_resize_to_sections): Delete declaration.
(remove_target_sections): Delete declaration.
* bfd-target.c (target_bfd_xfer_partial): Get the section table
from to_data.
(target_bfd_get_section_table): New.
(target_bfd_xclose): Adjust.
(target_bfd_reopen): Store the section table in the to_data field.
* corelow.c (core_data): New.
(core_close): Adjust to release core_data and its sections.
(core_open): Allocate core_data, and build its target sections
table.
(deprecated_core_resize_section_table): New.
(core_files_info): Pass core_data to print_section_info.
(core_xfer_partial): Adjust to use
section_table_xfer_memory_partial for TARGET_OBJECT_MEMORY xfers.
(init_core_ops): Do not install a deprecated_xfer_memory callback
anymore.
* solib.c (update_solib_list): Add the shared library sections
to the current target sections table.
* exec.c (current_target_sections_1): New global.
(current_target_sections): New global.
(exec_close_1): New function, refactored from exec_close. Remove
the exec_bfd's sections from the current target sections table.
Adjust to not use to_sections.
(exec_close): Remove all target sections. Call exec_close_1.
(exec_file_clear): Use exec_close_1 instead of unpushing the
target.
(exec_file_attach): Likewise. Adjust to not use to_sections. Add
exec_bfd's sections to the current target sections table. Don't
push the exec_ops target here.
(resize_section_table): New.
(add_target_sections): New.
(remove_target_sections): Moved here.
(section_table_xfer_memory): Adjust to implement the xfer_partial
interface, and rename to...
(section_table_xfer_memory_partial): ... this, replacing the
current function of that same name.
(exec_get_section_table): New.
(exec_xfer_partial): New.
(xfer_memory): Delete.
(print_section_info): Replace the target_ops parameter by a
target_section_table parameter.
(exec_files_info, set_section_command, exec_set_section_address):
Adjust to use the current sections table.
(init_exec_ops): Do not register a deprecated_xfer_memory
callback. Register to_xfer_partial and to_get_section_table
callbacks.
* infrun.c (handle_inferior_event): Update comments around
solib_add.
* rs6000-nat.c (xcoff_relocate_core): Adjust to use
deprecated_core_resize_section_table.
* exec.h (resize_section_table): Declare.
(section_table_xfer_memory_partial): Add const char * argument.
(remove_target_sections): Declare here.
(add_target_sections): Declare.
(print_section_info): Declare here.
* gdbcore.h (deprecated_core_resize_section_table): Declare.
(scm_lookup_name): Add GDBARCH parameter. Use it instead of
objfile architecture. Use per-architecture builtin_scm_type instead
of builtin_type_scm.
(evaluate_exp): Pass expression architecture to scm_lookup_name.
(scm_get_field): Add SIZE parameter. Use it instead of
TYPE_LENGTH (builtin_type_scm).
(build_scm_types): New function.
(scm_type_data): New variable.
(builtin_scm_type): New function.
(_initialize_scheme_language): Do not initialize builtin_type_scm.
Register build_scm_types as post-init function for scm_type_data.
* scm-exp.c (scm_lreadr): Use per-architecture builtin_scm_type
instead of builtin_type_scm.
* scm-lang.h (SCM_CAR, SCM_CDR): Pass SCM_SIZE to scm_get_field.
(scm_get_field): Add SIZE paramter.
(scm_scmval_print): Remove prototype.
(builtin_type_scm): Remove.
(struct builtin_scm_type): Add structure definition.
(builtin_scm_type): Add prototype.
* scm-valprint.c (scm_inferior_print): Add TYPE parameter. Use it
instead of builtin_core_addr type; do not use objfile architecture.
(scm_scmlist_print): Add TYPE parameter. Use it to define SCM_SIZE.
Pass type to scm_scmval_print.
(scm_ipruk): Add TYPE parameter. Use it to define SCM_SIZE.
(scm_scmval_print): Make static. Add TYPE parameter. Use it to
define SCM_SIZE. Pass type to scm_ipruk and scm_scmlist_print.
(scm_val_print): Pass type to scm_inferior_print and scm_scmval_print.
* dfp.c (promote_decimal): Remove.
(decimal_binop): Convert LEN_RESULT to input parameter.
Remove call to decimal_binop.
(decimal_compare): Remove call to decimal_binop.
* valarith.c (value_binop): Pass desired result type length
to decimal_binop.
(check_macro): Return 1 if undefined.
If first test fails, check if macro debug information is available,
and report unsupported test if no macro information is found.
If scheduler-locking is enabled, we're not going to be singlestepping
any other previously stopped thread.
* gdb.threads/hand-call-in-threads.exp: New file.
* gdb.threads/hand-call-in-threads.c: New file.
2009-05-28 Pedro Alves <pedro@codesourcery.com>
* infrun.c (handle_inferior_event): When thread hoping, switch
inferior_ptid to the event thread before removing breakpoints from
the target. If not stopping, also try to revert back to a thread
that was doing a "next". Check if that thread still exists before
resuming.
(currently_stepping_thread): Delete and merge with ...
(currently_stepping): ... this.
(currently_stepping_callback): Rename to ...
(currently_stepping_or_nexting_callback): ... this, and also
return true if the thread was stepping over a call (has a
step-resume breakpoint).
gdb/testsuite/
2009-05-28 Pedro Alves <pedro@codesourcery.com>
* gdb.threads/threxit-hop-specific.c: New.
* gdb.threads/threxit-hop-specific.exp: New.
* gdb.threads/thread-execl.c: New.
* gdb.threads/thread-execl.exp: New.
Global renaming of find_thread_pid to find_thread_ptid.
* gdbthread.h (find_thread_ptid): Renamed from find_thread_pid.
* thread.c (find_thread_ptid): Renamed from find_thread_pid.
All callers updated.
gdbserver:
Global renaming of find_thread_pid to find_thread_ptid.
* server.h (find_thread_ptid): Renamed from find_thread_pid.
* inferiors.c (find_thread_ptid): Renamed from find_thread_pid.
All callers updated.
* gdbthread.h (struct thread_info): New `pending_follow' field.
* thread.c (new_thread): New function.
(add_thread_silent): Use it.
* breakpoint.c (internal_breakpoint_number): New global, moved
from inside...
(create_internal_breakpoint): ... this.
(clone_momentary_breakpoint): New.
* breakpoint.h (clone_momentary_breakpoint): Declare.
* infrun.c (nullify_last_target_wait_ptid): Move declaration
higher.
(pending_follow): Delete.
(follow_fork): Handle pending follow fork event here. Moved the
preserving of thread stepping state here.
(resume): Don't handle pending follow fork events here. Only
install the inferior's terminal modes if we're about to resume it.
(proceed): Handle possible pending follow fork events here.
(init_wait_for_inferior): No need to clear pending_follow anymore,
it's gone.
(handle_inferior_event): Adjust to per-thread `pending_follow'.
Call `follow_fork' to handle following the fork. If the
follow-fork is cancelled, stop stepping.
* linux-nat.c (linux_child_follow_fork): Adjust to per-thread
`pending_follow' events. Remove code that handled preserving the
thread stepping state.
* inf-ptrace.c (inf_ptrace_follow_fork): Ditto.
* inf-ttrace.c (inf_ttrace_follow_fork): Ditto.
gdb/testsuite/
* gdb.threads/fork-thread-pending.c: New.
* gdb.threads/fork-thread-pending.exp: New.
