Age | Commit message (Collapse) | Author | Files | Lines |
|
* objdump.c (only): Replace with linked list.
(only_size, only_used): Replace with only_list.
(process_section_p): Set seen field on matches sections.
(add_only): New function.
(free_only_list): New function.
(disassemble_section): Check only_list.
(main): Use add_only and free_only_list.
* gas/pe/aligncomm-c.d: Dump all sections.
* ld-sh/refdbg-0-dso.d: Dump all sections.
|
|
* procfs.c (insert_dbx_link_breakpoint): Delete declaration. Move up.
Only define if SYS_syssgi is defined.
(remove_dbx_link_breakpoint): Delete declaration. Move up.
(dbx_link_addr, insert_dbx_link_bpt_in_file)
(insert_dbx_link_bpt_in_region): Move up. Only define if SYS_syssgi
is itself defined.
|
|
|
|
|
|
* NEWS: Mention new feature.
* config/obj-coff.c (obj_coff_section): Accept digits and use
to override default section alignment power if specified.
* doc/as.texinfo (.section directive): Update documentation.
gas/testsuite/ChangeLog:
* gas/pe/section-align-1.s: New test source file.
* gas/pe/section-align-1.d: Likewise control script.
* gas/pe/section-align-2.s: Likewise ...
* gas/pe/section-align-2.d: ... and likewise.
* gas/pe/pe.exp: Invoke new testcases.
|
|
* gdb.base/break-interp.exp: Use [file tail $filename] to omit any
absolute directory pathnames in gdb.sum file.
|
|
(get_windows_debug_event): Issue error when process dies before completely
initializing.
(do_initial_windows_stuff): Set flag to indicate when we are done with the
initial steps of attaching to the child.
|
|
* gdb.cp/namespace-using.exp: Rename to nsusing.exp.
* gdb.cp/namespace-using.cc: Rename to nsusing.cc.
|
|
* vms.h: Remove trailing spaces.
(struct hdr_struct): Remove unused hdr_c_cpr field.
(enum file_format_enum): Add comments.
* vms.c (vms_bfd_print_private_bfd_data): New function that
replaces the macro.
Remove trailing spaces.
* vms-misc.c: Improve comments.
(_bfd_vms_get_object_record): Also handle files without pads.
* vms-hdr.c (_bfd_vms_write_dbg): Fix format string.
* vms-gsd.c (struct flagdescstruct): Make name field const.
(gsyflagdesc): Fix typo.
|
|
(Target_arm::reloc_needs_sym_origin): New method.
(Arm_relocate_functions::calc_grp_kn): New method.
(Arm_relocate_functions::calc_grp_residual): New method.
(Arm_relocate_functions::calc_grp_gn): New method.
(Arm_relocate_functions::arm_grp_alu): New Method.
(Arm_relocate_functions::arm_grp_ldr): New Method.
(Arm_relocate_functions::arm_grp_ldrs): New Method.
(Arm_relocate_functions::arm_grp_ldc): New Method.
(Target_arm::Scan::local): Handle the ARM group relocations.
(Target_arm::Scan::global): Likewise.
(Target_arm::Relocate::relocate): Likewise.
(Target_arm::Relocatable_size_for_reloc::get_size_for_reloc):
Likewise.
|
|
gas/
2010-01-27 H.J. Lu <hongjiu.lu@intel.com>
* config/tc-i386.c (avxscalar): New.
(OPTION_MAVXSCALAR): Likewise.
(build_vex_prefix): Select vector_length for scalar instructions
based on avxscalar.
(md_longopts): Add OPTION_MAVXSCALAR.
(md_parse_option): Handle OPTION_MAVXSCALAR.
(md_show_usage): Add -mavxscalar=.
* doc/c-i386.texi: Document -mavxscalar=.
gas/testsuite/
2010-01-27 H.J. Lu <hongjiu.lu@intel.com>
* gas/i386/avx-scalar-intel.d: New.
* gas/i386/avx-scalar.d: Likewise.
* gas/i386/avx-scalar.s: Likewise.
* gas/i386/x86-64-avx-scalar-intel.d: Likewise.
* gas/i386/x86-64-avx-scalar.d: Likewise.
* gas/i386/x86-64-avx-scalar.s: Likewise.
* gas/i386/i386.exp: Run avx-scalar, avx-scalar-intel,
x86-64-avx-scalar and x86-64-avx-scalar-intel.
opcodes/
2010-01-27 H.J. Lu <hongjiu.lu@intel.com>
* i386-dis.c (XMScalar): New.
(EXdScalar): Likewise.
(EXqScalar): Likewise.
(EXqScalarS): Likewise.
(VexScalar): Likewise.
(EXdVexScalarS): Likewise.
(EXqVexScalarS): Likewise.
(XMVexScalar): Likewise.
(scalar_mode): Likewise.
(d_scalar_mode): Likewise.
(d_scalar_swap_mode): Likewise.
(q_scalar_mode): Likewise.
(q_scalar_swap_mode): Likewise.
(vex_scalar_mode): Likewise.
(vex_len_table): Duplcate entries for VEX_LEN_10_P_1,
VEX_LEN_10_P_3, VEX_LEN_11_P_1, VEX_LEN_11_P_3, VEX_LEN_2A_P_1,
VEX_LEN_2A_P_3, VEX_LEN_2C_P_3, VEX_LEN_2D_P_1, VEX_LEN_2E_P_0,
VEX_LEN_2E_P_2, VEX_LEN_2F_P_2, VEX_LEN_51_P_1, VEX_LEN_51_P_3,
VEX_LEN_52_P_1, VEX_LEN_53_P_1, VEX_LEN_58_P_1, VEX_LEN_58_P_3,
VEX_LEN_59_P_1, VEX_LEN_5A_P_1, VEX_LEN_5A_P_3, VEX_LEN_5C_P_1,
VEX_LEN_5C_P_3, VEX_LEN_5D_P_1, VEX_LEN_5D_P_3, VEX_LEN_5E_P_1,
VEX_LEN_5E_P_3, VEX_LEN_5F_P_1, VEX_LEN_5F_P_3, VEX_LEN_6E_P_2,
VEX_LEN_7E_P_1, VEX_LEN_7E_P_2, VEX_LEN_D6_P_2, VEX_LEN_C2_P_1,
VEX_LEN_C2_P_3, VEX_LEN_3A0A_P_2 and VEX_LEN_3A0B_P_2.
(vex_w_table): Update entries for VEX_W_10_P_1, VEX_W_10_P_3,
VEX_W_11_P_1, VEX_W_11_P_3, VEX_W_2E_P_0, VEX_W_2E_P_2,
VEX_W_2F_P_0, VEX_W_2F_P_2, VEX_W_51_P_1, VEX_W_51_P_3,
VEX_W_52_P_1, VEX_W_53_P_1, VEX_W_58_P_1, VEX_W_58_P_3,
VEX_W_59_P_1, VEX_W_59_P_3, VEX_W_5A_P_1, VEX_W_5A_P_3,
VEX_W_5C_P_1, VEX_W_5C_P_3, VEX_W_5D_P_1, VEX_W_5D_P_3,
VEX_W_5E_P_1, VEX_W_5E_P_3, VEX_W_5F_P_1, VEX_W_5F_P_3,
VEX_W_7E_P_1, VEX_W_D6_P_2 VEX_W_C2_P_1, VEX_W_C2_P_3,
VEX_W_3A0A_P_2 and VEX_W_3A0B_P_2.
(intel_operand_size): Handle d_scalar_mode, d_scalar_swap_mode,
q_scalar_mode, q_scalar_swap_mode.
(OP_XMM): Handle scalar_mode.
(OP_EX): Handle d_scalar_mode, d_scalar_swap_mode, q_scalar_mode
and q_scalar_swap_mode.
(OP_VEX): Handle vex_scalar_mode.
|
|
* elf64-ppc.c (struct ppc_link_hash_table): Add toc_bfd, toc_first_sec.
(ppc64_elf_setup_section_lists): Init them.
(ppc64_elf_next_toc_section): Don't partition multi-toc between .got
and .toc on the same input file.
(ppc64_elf_relocate_section): Correct GOT entry offset.
|
|
|
|
* symtab.h (struct symbol <symtab>): New comment on NULL values.
|
|
|
|
|
|
* printcmd.c (display_uses_solib_p): Remove variable section. Access
objfile via SYMBOL_SYMTAB.
|
|
|
|
* arm.cc (set): Include.
(class Arm_exidx_fixup): Change type of last_input_section_ to const
pointer type.
(Arm_output_section::Text_section_list): New type.
(Arm_output_section::append_text_sections_to_list): New method.
(Arm_output_section::fix_exidx_coverage): Ditto.
(Arm_relobj::Arm_relobj): Initialize exidx_section_map_.
(Arm_relobj::convert_input_section_to_relaxed_section): Use
Relobj::set_section_offset() instead of
Sized_relobj::invalidate_section_offset().
(Arm_relobj::section_needs_reloc_stub_scanning): Add an extra
parameter for section headers. Ignore relocation sections for
unallocated sections and EXIDX sections.
(Target_arm::fix_exidx_coverage): New method.
(Target_arm::output_section_address_less_than): New type.
(Arm_exidx_fixup::add_exidx_cantunwind_as_needed): Use index of the
linked text section instead of the EXIDX section.
(Arm_output_section::create_stub_group): Add an assertion to check
that this is not an EXIDX output section.
(Arm_output_section::append_text_sections_to_list): New method.
(Arm_output_section::fix_exidx_coverage): Ditto.
(Arm_relobj::scan_sections_for_stubs): Adjust call to
Arm_relobj::section_needs_reloc_stub_scanning.
(Target_arm::do_relax): Fix EXIDX output section coverage in the
first pass.
(Target_arm::fix_exidx_coverage): New method.
* object.h (Relobj::set_output_section): New method.
(Sized_relobj::invalidate_section_offset): Remove method.
(Sized_relobj::do_invalidate_section_offset): Remove method.
(Sized_relobj::do_set_section_offset): Handle offset value -1.
|
|
PR exp/7643:
* eval.c (evaluate_subexp_for_address) <UNOP_IND>: Call
coerce_array on result.
gdb/testsuite
PR exp/7643:
* gdb.base/printcmds.exp (test_print_string_constants): Remove
setup_kfail.
|
|
* gdb.cp/nsrecurs.exp: New test.
* gdb.cp/nsrecurs.cc: New test program.
* gdb.cp/nsstress.exp: New test.
* gdb.cp/nsstress.cc: New test program.
* gdb.cp/nsdecl.exp: New test.
* gdb.cp/nsdecl.cc: New test program.
2010-01-26 Sami Wagiaalla <swagiaal@redhat.com>
* cp-namespace.c (cp_lookup_symbol_namespace): Added
search_parent argument.
(cp_add_using): Initialize 'searched' field.
(reset_directive_searched): New function.
* cp-support.h: Add 'searched' field to using_direct struct.
(cp_lookup_symbol_imports): Ditto.
* cp-namespace.c (cp_lookup_symbol_imports): Ditto.
Perform recursive search.
Implement non parent search.
* valops.c (value_maybe_namespace_elt): Updated.
|
|
* gdb.cp/namespace-using.exp: Add test for printing of namespaces
imported into file scope.
Marked test as xfail.
* gdb.cp/namespace-using.cc (marker5): New function.
* gdb.cp/shadow.exp: New test.
* gdb.cp/shadow.cc: New test program.
* gdb.cp/nsimport.exp: New test.
* gdb.cp/nsimport.cc: New test program.
2010-01-26 Sami Wagiaalla <swagiaal@redhat.com>
PR gdb/10929:
* dwarf2read.c (read_lexical_block_scope): Create blocks for
scopes which contain using directives even if they contain no
declarations.
* symtab.c (lookup_symbol_aux): Pass lowest level block to
la_lookup_symbol_nonlocal.
* cp-namespace.c (cp_lookup_symbol_nonlocal): call
cp_lookup_symbol_namespace.
(cp_lookup_symbol_namespace): Perform an import lookup at every
block level.
(cp_lookup_symbol_imports): New function.
(cp_lookup_symbol_in_namespace): New function.
|
|
* targets.c (BFD_JUMP_TABLE_ARCHIVE): Add initializer for write_ar_hdr.
(bfd_target): Add _bfd_write_ar_hdr_fn field.
* archive.c (is_bsd44_extended_name): New macro.
(_bfd_generic_read_ar_hdr_mag): Use it. Add extra_size.
(bfd_slurp_armap): Also check for "__.SYMDEF" as a BSD4.4 extended
name.
(_bfd_archive_bsd44_construct_extended_name_table): New function.
(_bfd_generic_write_ar_hdr): Ditto.
(_bfd_bsd44_write_ar_hdr): Ditto.
(_bfd_write_archive_contents): Call _bfd_write_ar_hdr.
(bsd_write_armap): Adjust firstreal computation.
* libbfd-in.h (struct areltdata): Add extra_size field.
(_bfd_generic_write_ar_hdr): Add prototype.
(_bfd_bsd44_write_ar_hdr): Ditto.
(_bfd_write_ar_hdr): Define.
(_bfd_noarchive_write_ar_hdr): Ditto.
(_bfd_archive_bsd_write_ar_hdr): Ditto.
(_bfd_archive_coff_write_ar_hdr): Ditto.
(_bfd_archive_bsd44_slurp_armap): Ditto.
(_bfd_archive_bsd44_slurp_extended_name_table): Ditto.
(_bfd_archive_bsd44_construct_extended_name_table): New prototype.
(_bfd_archive_bsd44_truncate_arname): Ditto.
(_bfd_archive_bsd44_write_armap): Ditto.
(_bfd_archive_bsd44_read_ar_hdr): Ditto.
(_bfd_archive_bsd44_write_ar_hdr): Ditto.
(_bfd_archive_bsd44_openr_next_archived_file): Ditto.
(_bfd_archive_bsd44_get_elt_at_index): Ditto.
(_bfd_archive_bsd44_generic_stat_arch_elt): Ditto.
(_bfd_archive_bsd44_update_armap_timestamp): Ditto.
* libbfd.h: Regenerate.
* oasys.c (oasys_write_ar_hdr): Define.
* libecoff.h (_bfd_ecoff_write_ar_hdr): Define.
* ieee.c (ieee_write_ar_hdr): Define.
* elf64-mips.c (bfd_elf64_archive_write_ar_hdr): Define.
* coff-rs6000.c (rs6000coff_vec): Adjust for write_ar_hdr field.
(bfd_pmac_xcoff_backend_data): Ditto.
* coff64-rs6000.c (rs6000coff64_vec): Ditto.
(bfd_xcoff_aix5_backend_data): Ditto.
* coff-alpha.c (alpha_ecoff_write_ar_hdr): Define.
* aout-target.h (MY_write_ar_hdr): Define it if not defined.
* aout-tic30.c (MY_write_ar_hdr): Ditto.
* mach-o-target.c (TARGET_NAME): Use _bfd_archive_bsd44 archive.
(bfd_mach_o_mkarchive, bfd_mach_o_read_ar_hdr)
(bfd_mach_o_slurp_armap, bfd_mach_o_slurp_extended_name_table)
(bfd_mach_o_construct_extended_name_table)
(bfd_mach_o_truncate_arname, bfd_mach_o_write_armap)
(bfd_mach_o_get_elt_at_index, bfd_mach_o_generic_stat_arch_elt)
(bfd_mach_o_update_armap_timestamp): Moved to mach-o.c
* mach-o.c (bfd_mach_o_mkarchive, bfd_mach_o_read_ar_hdr)
(bfd_mach_o_slurp_armap, bfd_mach_o_slurp_extended_name_table)
(bfd_mach_o_construct_extended_name_table)
(bfd_mach_o_truncate_arname, bfd_mach_o_write_armap)
(bfd_mach_o_get_elt_at_index, bfd_mach_o_generic_stat_arch_elt)
(bfd_mach_o_update_armap_timestamp): Moved from mach-o-target.c
* bfd-in2.h: Regenerate.
|
|
bfd/
2010-01-26 Alan Modra <amodra@gmail.com>
H.J. Lu <hongjiu.lu@intel.com>
PR ld/11218
* elflink.c (elf_link_output_extsym): Do not ignore undefined
symbols with ref_regular set when gc_sections is active.
ld/testsuite/
2010-01-26 H.J. Lu <hongjiu.lu@intel.com>
PR ld/11218
* ld-gc/dummy.s: New.
* ld-gc/pr11218-1.c: Likewise.
* ld-gc/pr11218-2.c: Likewise.
* ld-gc/pr11218.d: Likewise.
|
|
* Makefile.am (bin2c): Add libintl dependance and library.
* Makefile.in: Regenerate.
|
|
|
|
|
|
PR gdb/11049:
* c-valprint.c (c_val_print): Fix test of extract_unsigned_integer
result.
gdb/testsuite
PR gdb/11049:
* gdb.base/printcmds.exp (test_print_typedef_arrays): Add test
with "set print null-stop on".
|
|
* configure.ac (bootstrap_target_libs): Make inclusion of
target-libgomp conditional on libgomb being in target_configdirs.
* configure: Regenerate.
|
|
* arm.cc (Arm_exidx_merged_section::do_output_offset):
Fix warning due to signed and unsigned comparison on a 32-bit host.
|
|
Use test to check variables, prefix strings with x.
* configure: Regenerate.
* solib-osf.c (osf_current_sos): Initialize tail.
|
|
* windows-nat.c (windows_continue): Use %x to print thread id.
(get_windows_debug_event): Ditto.
|
|
* elf64-ppc.c (ppc64_elf_tls_optimize): Optimize tls sequences
with relocations against undefined weak symbols.
(ppc64_elf_relocate_section): Don't optimize calls to undefined
weak functions if the symbol is dynamic.
(ppc64_elf_relocate_section): Edit tprel tls sequences.
* elf32-ppc.c (ppc_elf_relocate_section): Likewise.
(_bfd_elf_ppc_at_tprel_transform): New function.
* bfd-in.h (_bfd_elf_ppc_at_tprel_transform): Declare.
* bfd-in2.h: Regenerate.
|
|
* cgen-accfp.c (fextsfdf): New arg how. All callers updated.
(ftruncdfsf, floatsisf, flostsidf, ufloatsisf, fixsfsi, fixdfsi,
ufixsfsi): Ditto.
* cgen-fpu.h (CGEN_FPCONV_KIND): New enum.
(struct cgen_fp_ops): Update signatures of floating point conversion
operations.
frv/
* sem.c: Regenerate.
sh64/
* cpu.h: Regenerate.
|
|
* sid-cpu.scm (cgen-desc.h): Ditto.
* enum.scm (enum-builtin!): New function.
* read.scm (reader-install-builtin!): Call it.
* rtl-c.scm (s-convop): Delete, replaced with ...
(s-int-convop, s-float-convop): ... new fns.
(ext, zext, trunc): Update.
(fext, ftrunc, float, ufloat, fix, ufix): Update.
* rtx-funcs.scm (fext, ftrunc, float, ufloat, fix, ufix): New parameter
`how'.
* cpu/mep-fmax.cpu (fcvtsw): Update.
* cpu/sh.cpu (h-fsd, h-fmov): Update.
* doc/rtl.texi (float-convop): Update.
* frv.cpu (floating-point-conversion): Update call to fp conv op.
(floating-point-dual-conversion, ne-floating-point-dual-conversion,
conditional-floating-point-conversion, ne-floating-point-conversion,
float-parallel-mul-add-double-semantics): Ditto.
|
|
(CGEN_INCLUDE_DEPS): Use it.
(CGEN_MAIN_CPU_DEPS): Simplify.
|
|
|
|
|
|
|
|
2010-01-24 H.J. Lu <hongjiu.lu@intel.com>
* i386-dis.c (vex_len_table): Remove trailing { Bad_Opcode }.
|
|
2010-01-24 H.J. Lu <hongjiu.lu@intel.com>
* i386-dis.c (prefix_table): Remove trailing { Bad_Opcode }.
|
|
with { Bad_Opcode }.
2010-01-24 H.J. Lu <hongjiu.lu@intel.com>
* i386-dis.c (Bad_Opcode): New.
(bad_opcode): Likewise.
(dis386): Replace { "(bad)", { XX } } with { Bad_Opcode }.
(dis386_twobyte): Likewise.
(reg_table): Likewise.
(prefix_table): Likewise.
(x86_64_table): Likewise.
(vex_len_table): Likewise.
(vex_w_table): Likewise.
(mod_table): Likewise.
(rm_table): Likewise.
(float_reg): Likewise.
(reg_table): Remove trailing "(bad)" entries.
(prefix_table): Likewise.
(x86_64_table): Likewise.
(vex_len_table): Likewise.
(vex_w_table): Likewise.
(mod_table): Likewise.
(rm_table): Likewise.
(get_valid_dis386): Handle bytemode 0.
|
|
2010-01-24 H.J. Lu <hongjiu.lu@intel.com>
* config/tc-i386.c (build_vex_prefix): Set i.vex.bytes[0] to
0xc4 individually.
|
|
* gdb.arch/i386-bp_permanent.exp (Disassemble function '$function'):
Adjust it for DISASSEMBLY_OMIT_FNAME.
|
|
2010-01-23 H.J. Lu <hongjiu.lu@intel.com>
* i386-opc.h (VEXScalar): New.
* i386-opc.tbl: Replace "Vex" with "Vex=3" on AVX scalar
instructions.
* i386-tbl.h: Regenerated.
|
|
|
|
|
|
2010-01-23 H.J. Lu <hongjiu.lu@intel.com>
* gas/i386/avx.s: Add more tests.
* gas/i386/x86-64-avx.s: Likewise.
* gas/i386/avx-intel.d: Updated.
* gas/i386/avx.d: Likewise.
* gas/i386/x86-64-avx-intel.d: Likewise.
* gas/i386/x86-64-avx.d: Likewise.
|
|
* coff-rs6000.c (xcoff_howto_table): Change size to 0 and bitsize to 1.
(_bfd_xcoff_reloc_type_lookup): Handle BFD_RELOC_NONE.
* coff64-rs6000.c (xcoff64_howto_table): Change size to 0 and
bitsize to 1.
(xcoff64_reloc_type_lookup): Handle BFD_RELOC_NONE.
gas/
* write.h (fix_at_start): Declare.
* write.c (fix_new_internal): Add at_beginning parameter.
Use it instead of REVERSE_SORT_RELOCS. Fix the handling of
seg_fix_tailP for the at_beginning/REVERSE_SORT_RELOCS case.
(fix_new, fix_new_exp): Update accordingly.
(fix_at_start): New function.
* config/tc-ppc.c (md_pseudo_table): Add .ref to the OBJ_XCOFF section.
(ppc_ref): New function, for OBJ_XCOFF.
(md_apply_fix): Handle BFD_RELOC_NONE for OBJ_XCOFF.
* config/te-i386aix.h (REVERSE_SORT_RELOCS): Remove #undef.
gas/testsuite/
* gas/ppc/xcoff-ref-1.s, gas/ppc/xcoff-ref-1.l: New test.
* gas/ppc/aix.exp: Run it.
ld/testsuite/
* ld-powerpc/aix-ref-1-32.od, ld-powerpc/aix-ref-1-64.od,
ld-powerpc/aix-ref-1.s: New tests.
* ld-powerpc/aix52.exp: Run them.
|
|
* configure.ac (bootstrap_target_libs): Include target-libgomp.
* configure: Regenerate.
|