Fix DW_AT_lower_bound defaults for DWARF-4+.
* dwarf2read.c (read_subrange_type): Remove initialization of low and
high. New variable low_default_is_valid. Implement DWARF-4+
DW_AT_lower_bound defaults. Print complaint for DW_AT_lower_bound with
no default by the DWARF standard.
* infrun.c (handle_inferior_event): Move the check for return
trampolines ahead of the check for function trampolines.
* mips-tdep.h (MIPS_S2_REGNUM, MIPS_GP_REGNUM): New macros.
* mips-tdep.c (mips_str_mips16_call_stub): New variable.
(mips_str_mips16_ret_stub): Likewise.
(mips_str_call_fp_stub): Likewise.
(mips_str_call_stub): Likewise.
(mips_str_fn_stub): Likewise.
(mips_str_pic): Likewise.
(mips_in_frame_stub): New function.
(mips_unwind_pc): Return the return address rather than the PC
if the PC of an intermediate frame is inside a call thunk.
(mips_is_stub_suffix): New function.
(mips_is_stub_mode): Likewise.
(mips_get_mips16_fn_stub_pc): Likewise.
(mips_skip_mips16_trampoline_code): Update to handle all the
currently generated stub types. Don't recurse into __fn_stub
thunks. Remove heuristics to handle stubs beyond etext/_etext.
Use cooked register accesses.
(mips_in_return_stub): Reintroduce function.
(mips_skip_trampoline_code): Traverse trampolines recursively.
(mips_gdbarch_init): Handle MIPS16 return trampolines.
gdb/testsuite/
* gdb.arch/mips16-thunks-inmain.c: New file.
* gdb.arch/mips16-thunks-main.c: New file.
* gdb.arch/mips16-thunks-sin.c: New file.
* gdb.arch/mips16-thunks-sinfrob.c: New file.
* gdb.arch/mips16-thunks-sinfrob16.c: New file.
* gdb.arch/mips16-thunks-sinmain.c: New file.
* gdb.arch/mips16-thunks-sinmips16.c: New file.
* gdb.arch/mips16-thunks.exp: New file.
* bfd.c (bfd_assert_handler_type): New API type.
(bfd_set_assert_handler, bfd_get_assert_handler): New API functions.
(_bfd_assert_handler): New variable.
(_bfd_default_assert_handler): New function.
(bfd_assert): Call _bfd_assert_handler, not _bfd_error_handler.
* libbfd-in.h (_bfd_assert_handler): Declare.
* libbfd.h, bfd-in2.h: Regenerate.
pointer.
(Stub_addend_reader::operator()): Declare Arm_relocate_functions
as a class, not a struct.
(Target_arm::scan_span_for_cortex_a8_erratum): Likewise.
(Target_arm::apply_cortex_a8_workaround): Likewise.
* gc.h: Declare Reloc_types as a struct, not a class.
* object.h: Declare Symbols_data as a struct.
* reloc.h: Declare Read_relocs_data as a struct.
* target.h: Declare Relocate_info as a struct.
to strip and objcopy.
* objcopy.c (deterministic): New global variable.
(strip_options): Add --enable-deterministic-archives.
(copy_options): Likewise.
(copy_usage): Likewise.
(strip_usage): Likewise.
(copy_archive): When stripping all, don't add archive map; set
deterministic output when requested.
(strip_main): Add -D/--enable-deterministic-archives option.
(copy_main): Likewise.
* procfs.c (procfs_make_note_section): Be sure to NUL-terminate
fname and psargs before trying to concatenate.
* tui/tui-stack.c (tui_get_function_from_frame): NUL-terminate
"name" before applying strchr.
gold/
* sparc.cc (Target_sparc::Relocate::relax_call): New function.
(Target_sparc::Relocate::relocate): Call it for R_SPARC_WDISP30
and R_SPARC_WPLT30.
magic number with symbolic constant.
(dump_incremental_inputs): Update version number.
* incremental.cc (Output_section_incremental_inputs): Update version
number; import symbolic constants from Incremental_inputs_reader.
(Incremental_inputs::create_data_sections): Align relocations
section correctly for 64-bit targets.
(Output_section_incremental_inputs::set_final_data_size): Use symbolic
constants; add padding.
(Output_section_incremental_inputs::write_header): Add assert for
header_size.
(Output_section_incremental_inputs::write_input_files): Add assert
for input_entry_size.
(Output_section_incremental_inputs::write_info_blocks): Add padding;
add assert for object_info_size, input_section_entry_size,
global_sym_entry_size.
* incremental.h (Incremental_inputs_reader): Add symbolic constants
for data structure sizes; use them.
(Incremental_input_entry_reader): Import symbolic constants from
Incremental_inputs_reader; use them.
* ld-arm/arm-elf.exp: Run gc-hidden-1.
* ld-arm/gc-hidden-1.d: New test-file.
* ld-arm/gcdfn.s, ld-arm/hideall.ld, ld-arm/hidfn.s,
ld-arm/main.s: New files.
(PTRACE_ARG4_TYPE): Likewise.
(PTRACE_XFER_TYPE): Likewise.
* linux-arm-low.c (arm_prepare_to_resume): Cast third argument of
ptrace to PTRACE_ARG3_TYPE.
* linux-low.c (PTRACE_ARG3_TYPE): Move macro to linux-low.h.
(PTRACE_ARG4_TYPE): Likewise.
(PTRACE_XFER_TYPE): Likewise.
(linux_detach_one_lwp): Cast fourth argument of
ptrace to long then PTRACE_ARG4_TYPE.
(regsets_fetch_inferior_registers): Cast third argument of
ptrace to long then PTRACE_ARG3_TYPE.
(regsets_store_inferior_registers): Likewise.
Revert this patch to allow breakpoint always-inserted
in record target.
2011-12-05 Pedro Alves <pedro@codesourcery.com>
* breakpoint.c: Include record.h.
(breakpoints_always_inserted_mode): Return false when the record
target is in use.
* breakpoint.c (iterate_over_bp_locations): New.
* breakpoint.h: Declare.
New typedef walk_bp_location_callback.
* record.c (record_open): Call record_init_record_breakpoints.
(record_sync_record_breakpoints): New.
(record_init_record_breakpoints): New.
* NEWS: Mention supporting breakpoint always-inserted mode in
record target.
This appears to be an oversight, since this is done already for
sparc64-*-linux*, and essentially every other linux target.
gdb/
* configure.tgt (sparc-*-linux*): Set build_gdbserver to yes.
gold/
* sparc.cc (class Target_sparc): Add elf_machine_, elf_flags_,
and elf_flags_set_.
(Target_sparc::Target_sparc): Initialize new fields.
(Target_sparc::do_make_elf_object): New function.
(Target_sparc::do_adjust_elf_header): New function.
* elfcpp_swap.h (struct Swap_aligned32): New template.
gold/
* gdb-index.cc (Gdb_index::do_write): Use Swap_aligned32 for writing
CU range table of gdb index.
