* gdb.t07/watchpoint.exp: Accept both "Watchpoint" and
"Hardware watchpoint" throughout file.
(test_watchpoint_triggered_in_syscall): New test.
(test_complex_watchpoint): New test.
with the same name:
* objcopy.c (setup_section): Make a new output section even if
one already exists with the given name.
(copy_section): Use isection->output_section rather than trying
to look the output section up by its (possibly non-unique) name.
read. (bfd_error_system_call or bfd_error_file_truncated).
* som.c: Do not blindly set bfd_error_system_call after a
failing bfd_read, bfd_write, or bfd_seek. In a few places
(like som_object_p) override the error status set by bfd_read.
* aix386-core.c, aout-encap,c archive.c, bout.c: Likewise.
* coff-rs6000.c, coffgen.c ecoff.c, elf.c: Likewise.
* elf32-hppa.c, elfcode.h, hp300hpux.c, i386lynx.c: Likewise.
* nlm32-alpha.c, nlm32-i386.c, nlm32-sparc.c: Likewise.
* som.c: Check return values from several bfd_{seek,read,write}
calls that we just assumed were not failing.
(set_demangling_style, set_demangling_command): Use savestring not
strdup. We were not dealing properly with a NULL return from
strdup, and were not declaring strdup (the system header may or
may not have it).
of a function.
(skip_trampoline_code): Rewrite and add support for argument
relocation stubs stubs, import/export stubs, calls through
"_sr4export" and cascaded trampolines.
into. Tests trampoline problems on the PA.
* funcargs.exp (localvars_in_indirect_call): No longer expect
stepping into indirect call to fail on PAs.
(test_stepping_over_trampolines): New test.
branch with an instruction that uses $at, in case the branch is
later expanded.
(macro): If EMBEDDED_PIC, case M_JAL_A may use $at.
(md_pcrel_from): If not OBJ_AOUT, return 4 for an undefined symbol
to make it pcrel_offset.
(tc_gen_reloc): If not OBJ_AOUT, set the reloc addend to
reloc->address; another gruesome hack to get gas reloc handling to
do the right thing.
use a five instruction sequence for funtion calls which are out of
range of the bal instruction.
* libecoff.h (struct ecoff_section_tdata): Define.
(ecoff_section_data): Define.
(ecoff_bfd_relax_section): Don't define.
* ecoff.c (ecoff_final_link_debug_accumulate): Don't read or free
the debugging information if it has already been read.
(ecoff_indirect_link_order): Handle _cooked_size being different
from _raw_size. Don't reread the contents or the relocs if they
have already been read in.
* coff-mips.c (mips_howto_table): Change bitsize of PCREL16 from
18 to 16.
(PCREL16_EXPANSION_ADJUSTMENT): Define.
(mips_relocate_refhi): Take adjust argument.
(mips_relocate_section): Handle reloc offsets stored in section
used_by_bfd field. Call mips_relax_pcrel16 to handle details of
expanding an out of range PCREL16. Keep trace of adjustments
required by expansions. Set s and unset h when converting a reloc
from undefined to section. Change handling of PC relative relocs:
if against a section, they are correct in the object file, if
against an external symbol they are pcrel_offset.
(mips_relax_section): New function.
(mips_relax_pcrel16): New function.
(ecoff_bfd_relax_section): Define.
* coff-alpha.c (ecoff_bfd_relax_section): Define.
* ecofflink.c (bfd_ecoff_debug_accumulate): Handle adjustments
built by mips_relax_section when writing out addresses.
* elf32-mips.c (mips_elf_read_ecoff_info): Clear adjust field.
(REGISTER_NAMES): Add entries for "right-half" of FP registers.
(REGISTER_RAW_SIZE, MAX_REGISTER_RAW_SIZE): Do not treat FP regs
differently. All registers are four bytes.
(REGISTER_BYTES, REGISTER_BYTE): Simplify now that all registers are
the same size.
(REGISTER_VIRTUAL_TYPE): Use builtin_type_float for all FP regs.
* hppa-tdep.c (pa_print_fp_reg): Update to print even numbered FP
registers as both single and double values (fetching 2nd 32bit half
as necessary). Annotate each register printed with its precision.
(T12): New macro.
(emit_insn): New function.
(md_assemble): Call it.
(alpha_force_relocation): Handle BFD_RELOC_26, for call_pal instructions.
(lituse_pending): New variable. Set by anything that generates a LITERAL
reloc, cleared by anything that generates a LITUSE reloc, tested by code that
might want to emit a LITUSE reloc.
(emit_unaligned_io): New function. Currently calls md_assemble, but it should
eventually be converted to generate the insn itself and call emit_insn directly.
(emit_load_unal, emit_store_unal, emit_byte_manip_r, emit_extract_r,
emit_insert_r, emit_mask_r, emit_sign_extend, emit_bis_r): Likewise.
(alpha_ip, case 'I'): Handle with BFD_RELOC_23.
(alpha_ip, label get_macro): Don't emit the final instruction if the opcode is
zero.
(alpha_ip, case 'B', subcase 'd'): New case, for subword and unaligned memory
access macros.
(md_apply_fix): Handle BFD_RELOC_26. Generate an error message if the value
can't be resolved.
real to true for lang_input_file_is_marker_enum. Clear the_bfd.
(lang_add_input_file): Pass true to new_afile for add_to_list.
(lookup_name): Remove force_load argument. Changed all callers.
Pass false to new_afile for add_to_list. Split loading of symbols
out into separate function.
(load_symbols): New function split out of lookup_name. Don't load
the symbols if they are already loaded.
(open_input_bfds): For lang_input_statement_enum call load_symbols
rather than lookup_name.
(lang_process): Pass abs_output_section rather than NULL to
lang_size_sections.
(lang_startup): Set real field of first_file to true.
Ian Lance Taylor
Jeff Law
David MacKenzie
Jim Kingdon
Roland Pesch
Sean Eric Fagan
Ken Raeburn
Stu Grossman
Steve Chamberlain