Jan Kratochvil [Wed, 12 Oct 2011 22:04:12 +0000 (22:04 +0000)]
gdb/
Fix empty DWARF expressions DATA vs. SIZE conditionals.
* dwarf2loc.c (dwarf2_find_location_expression): Clear *LOCEXPR_LENGTH.
(dwarf_expr_frame_base_1): Indicate unavailability via zero *LENGTH.
(locexpr_tracepoint_var_ref): Check only zero SIZE, not zero DATA.
(loclist_read_variable, loclist_tracepoint_var_ref): Do not check for
zero DATA.
* dwarf2loc.h (struct dwarf2_locexpr_baton): Comment DATA vs. SIZE
validity.
* dwarf2read.c (struct dwarf_block): Comment DATA validity.
(dwarf2_fetch_die_location_block, dwarf2_symbol_mark_computed): Do not
clear DATA on zero SIZE.
gdb/testsuite/
Fix empty DWARF expressions DATA vs. SIZE conditionals.
* gdb.dwarf2/dw2-op-call.S (arraycallnoloc, arraynoloc): New DIEs.
(loclist): New.
(4): New abbrev.
* gdb.dwarf2/dw2-op-call.exp: Remove variable srcfile and executable.
Use prepare_for_testing, remove clean_restart.
(p arraynoloc, p arraycallnoloc): New tests.
Jan Kratochvil [Wed, 12 Oct 2011 15:55:04 +0000 (15:55 +0000)]
gdb/
Fix compatibility with texinfo versions older than 4.12.
* Makefile.in (MAKEINFO): Set to @MAKEINFO@.
(MAKEINFOFLAGS, MAKEINFO_EXTRA_FLAGS, MAKEINFO_CMD): New.
(MAKEHTMLFLAGS): Use MAKEINFO_CMD.
(FLAGS_TO_PASS): Add MAKEINFOFLAGS and MAKEINFO_EXTRA_FLAGS.
* configure: Regenerate.
* configure.ac (MAKEINFO): Find it, from libiberty/configure.ac.
(MAKEINFOFLAGS): Pre-set it to --split-size=5000000.
(MAKEINFO_EXTRA_FLAGS): New test for -DHAVE_MAKEINFO_CLICK.
gdb/doc/
Fix compatibility with texinfo versions older than 4.12.
* Makefile.in (MAKEINFO): Set to @MAKEINFO@.
(MAKEINFOFLAGS, MAKEINFO_EXTRA_FLAGS, MAKEINFO_CMD): New.
(MAKEHTMLFLAGS): Use MAKEINFO_CMD.
(gdb.info, gdbint.info, stabs.info, annotate.info): Use MAKEINFO_CMD.
* gdb.texinfo (Tail Call Frames): Convert @arrow{} to @click, when possible.
Make the conversion conditional on HAVE_MAKEINFO_CLICK, using variables
CALLSEQ1A, CALLSEQ1B, CALLSEQ2A and CALLSEQ2B.
Gary Benson [Wed, 12 Oct 2011 15:43:49 +0000 (15:43 +0000)]
2011-10-12 Gary Benson <gbenson@redhat.com>
* breakpoint.h (pc_at_non_inline_function): Declare.
* breakpoint.c (is_non_inline_function,
pc_at_non_inline_function): New functions.
* infrun.c (handle_inferior_event): Don't call skip_inline_frames
if the stop is at a location where functions cannot be inlined.
Pedro Alves [Wed, 12 Oct 2011 12:17:29 +0000 (12:17 +0000)]
Fix incomplete entry. Should have been:
2011-10-12 Pedro Alves <pedro@codesourcery.com>
* linux-nat.c (stop_and_resume_callback): Don't re-resume LWPs if
the core wanted them stopped, or if they now have a pending event
to report.
(linux_nat_filter_event): New parameter `new_pending_p'. Pass it
down to stop_and_resume_callback.
(linux_nat_wait_1): Always clear `options' when retrying. Handle
having new pending events after calling linux_nat_filter_event.
Jan Kratochvil [Wed, 12 Oct 2011 09:19:42 +0000 (09:19 +0000)]
gdb/testsuite/
* gdb.cp/namespace-enum.exp (executable): New variable.
(binfile): Use ${objdir}/${subdir}/ prefix.
Use ${executable} for clean_restart.
David S. Miller [Tue, 11 Oct 2011 19:08:59 +0000 (19:08 +0000)]
Fix jit.exp on most 32-bit targets.
* gdbarch.sh: New field 'long_long_align_bit'.
* gdbarch.c, gdbarch.h: Regenerate.
* i386-tdep.c (i386_gdbarch_init): Set long_long_align_bit to 32.
* jit.c (jit_read_code_entry): Use it to determine correct size offset.
Nick Clifton [Tue, 11 Oct 2011 15:56:29 +0000 (15:56 +0000)]
PR binutils/13051
Fix a syntax error bug when compiling rc files with the VERSIONINFO resource
containing more than one language block inside a single StringFileInfo block.
* windint.h (rc_ver_stringtable): New structure definition.
(rc_ver_info): Use it.
* rcparse.y (verstringtable): New variable.
(verstringtables): New type.
(verstringtables:): New rule declaration.
(verblocks:): Use it.
* resrc.c (append_ver_stringtable): New function.
(append_ver_stringfileinfo): Update to use stringtables.
* windres.h (append_ver_stringfileinfo): Update declaration.
(append_ver_stringtable): New declaration.
* resrc.c (write_rc_versioninfo): Update to support multiple blocks.
* resbin.c (bin_to_res_version): Likewise.
(res_to_bin_versioninfo): Likewise.
Pedro Alves [Tue, 11 Oct 2011 14:58:21 +0000 (14:58 +0000)]
2011-10-11 Pedro Alves <pedro@codesourcery.com>
* linux-nat.c (linux_handle_extended_wait): Always dump both the
parent and child's pids as soon as we detect a clone event.
Adjust another debug message.
Jan Kratochvil [Tue, 11 Oct 2011 12:58:08 +0000 (12:58 +0000)]
gdb/
Fix separate debuginfo warning with "remote:" access.
* objfiles.h (struct objfile): New fields crc32 and crc32_p.
* symfile.c (get_file_crc): New function with the code moved from ...
(separate_debug_file_exists): ... this function, specifically variables
buffer and count. New variable verified_as_different, set it. Remove
file_crc initialization. Verify also if both files are not the same
manually, if needed.
Alan Modra [Tue, 11 Oct 2011 06:44:13 +0000 (06:44 +0000)]
PR binutils/13278
* archive.c (bfd_generic_archive_p): Only check first element
when target_defaulted.
(_bfd_construct_extended_name_table): Use ar_maxnamelen.
(_bfd_archive_bsd44_construct_extended_name_table): Likewise.
Doug Evans [Tue, 11 Oct 2011 03:31:59 +0000 (03:31 +0000)]
* linux-thread-db.c (thread_db_new_objfile): Only try to load
libthread_db when we load libpthread or the main symbol file.
(thread_db_inferior_created): New function.
(_initialize_thread_db): Attach inferior_created observer.
* linux-nat.c (linux_child_post_attach): Remove call to
check_for_thread_db.
(linux_child_post_startup_inferior): Ditto.
* objfiles.h (OBJF_MAINLINE): Define.
* symfile.c (symbol_file_add_with_addrs_or_offsets): Pass it to
allocate_objfile when appropriate.
Cary Coutant [Mon, 10 Oct 2011 17:29:52 +0000 (17:29 +0000)]
PR gold/13249
* gold/output.cc (Output_reloc::Output_reloc): Add use_plt_offset flag.
(Output_reloc::symbol_value): Return PLT offset if flag is set.
* gold/output.h (class Output_reloc): Add use_plt_offset flag.
(Output_reloc::type_): Adjust size of bit field.
(Output_reloc::use_plt_offset_): New bit field.
(class Output_data_reloc): Adjust all calls to Output_reloc_type.
(Output_data_reloc::add_local_relative): (RELA only) Add use_plt_offset
flag. Adjust all callers.
* gold/x86_64.cc (Target_x86_64::Scan::local): Check for IFUNC when
creating RELATIVE relocations.
Pedro Alves [Mon, 10 Oct 2011 16:40:02 +0000 (16:40 +0000)]
2011-10-10 Pedro Alves <pedro@codesourcery.com>
gdb/
* linux-nat.c (linux_handle_extended_wait): Don't resume the new
new clone lwp if the core asked it to stop. Don't pass on
unexpected signals to the new clone; leave them pending instead.
Pedro Alves [Mon, 10 Oct 2011 13:23:53 +0000 (13:23 +0000)]
2011-10-10 Pedro Alves <pedro@codesourcery.com>
gdb/
* linux-nat.c (linux_nat_wait_1): Copy the event lwp's
last_resume_kind before clearing it, and use the copy instead to
determine whether to report a SIGSTOP as TARGET_SIGNAL_0. Use
resume_clear_callback in the non-stop path too.
Yao Qi [Mon, 10 Oct 2011 02:50:48 +0000 (02:50 +0000)]
gdb/
* valprint.c (value_check_printable): Add one parameter OPTIONS.
Honor OPTIONS and VAL's type.
(common_val_print, value_print): Update to pass one more parameter.
Mike Frysinger [Mon, 10 Oct 2011 02:41:54 +0000 (02:41 +0000)]
sim: dv-cfi: include stdbool.h
The dv-cfi code uses the bool type but doesn't include the stdbool.h
header. I didn't notice for Blackfin targets as the core Blackfin
header will include stdbool.h, but most targets don't do this.
Mike Frysinger [Mon, 10 Oct 2011 02:41:07 +0000 (02:41 +0000)]
sim: dv-cfi: include common headers only
The "devices.h" header is Blackfin-specific, so trying to build dv-cfi
for any other target fails. Include the specific hw headers this file
needs directly.
Jan Kratochvil [Sun, 9 Oct 2011 19:43:41 +0000 (19:43 +0000)]
gdb/
Display @entry parameter values even for references.
* ada-valprint.c (ada_val_print_1) <TYPE_CODE_REF>: Try also
coerce_ref_if_computed.
* c-valprint.c (c_val_print) <TYPE_CODE_REF>: Likewise.
* dwarf2expr.c (dwarf_block_to_dwarf_reg_deref): New function.
(execute_stack_op) <DW_OP_GNU_entry_value>: Add -1 deref_size to the
existing push_dwarf_reg_entry_value call. Add new detection calling
dwarf_block_to_dwarf_reg_deref. Update the error message.
(ctx_no_push_dwarf_reg_entry_value): New parameter deref_size.
* dwarf2expr.h
(struct dwarf_expr_context_funcs) <push_dwarf_reg_entry_value>: Add new
parameter deref_size, describe it in the comment.
(ctx_no_push_dwarf_reg_entry_value): Add new parameter deref_size.
(dwarf_block_to_dwarf_reg_deref): New declaration.
* dwarf2loc.c (dwarf_entry_parameter_to_value): Add new parameter
deref_size, describe it in the function comment. New variables
data_src and size, fetch the alternative block accoring to DEREF_SIZE.
(dwarf_expr_push_dwarf_reg_entry_value): Add new parameter deref_size,
describe it in the function comment. Fetch the alternative block
accoring to DEREF_SIZE.
(entry_data_value_coerce_ref, entry_data_value_copy_closure)
(entry_data_value_free_closure, entry_data_value_funcs): New.
(value_of_dwarf_reg_entry): New variables checked_type, target_type,
outer_val, target_val, val and addr. Try to fetch and create also
referenced value content.
(pieced_value_funcs): NULL value for coerce_ref.
(needs_dwarf_reg_entry_value): Add new parameter deref_size.
* f-valprint.c (f_val_print) <TYPE_CODE_REF>: Try also
coerce_ref_if_computed.
* opencl-lang.c (opencl_value_funcs): NULL value for coerce_ref.
* p-valprint.c (pascal_val_print) <TYPE_CODE_REF>: Likewise.
* stack.c (read_frame_arg): Compare also dereferenced values.
* value.c (value_computed_funcs): Make the parameter v const, use
value_lval_const for it.
(value_lval_const, coerce_ref_if_computed): New function.
(coerce_ref): New variable retval. Call also coerce_ref_if_computed.
* value.h (struct lval_funcs): New field coerce_ref.
(value_computed_funcs): Make the parameter v const.
(value_lval_const, coerce_ref_if_computed): New declarations.
gdb/testsuite/
Display @entry parameter values even for references.
* gdb.arch/amd64-entry-value.cc (reference, datap, datap_input): New
functions.
(main): New variables regvar, nodatavarp, stackvar1, stackvar2. Call
reference and datap_input.
* gdb.arch/amd64-entry-value.exp (reference, breakhere_reference): New
breakpoints.
(continue to breakpoint: entry_reference: reference)
(entry_reference: bt at entry)
(continue to breakpoint: entry_reference: breakhere_reference)
(entry_reference: bt, entry_reference: ptype regparam)
(entry_reference: p regparam, entry_reference: ptype regparam@entry)
(entry_reference: p regparam@entry, entry_reference: p ®param@entry)
(entry_reference: p regcopy, entry_reference: p nodataparam)
(entry_reference: p nodataparam@entry): New tests.
Jan Kratochvil [Sun, 9 Oct 2011 19:41:17 +0000 (19:41 +0000)]
gdb/
Support @entry in input expressions.
* c-exp.y (ENTRY, unknown_cpp_name): New.
(exp: UNKNOWN_CPP_NAME): Change to `exp: unknown_cpp_name'.
(unknown_cpp_name: UNKNOWN_CPP_NAME, unknown_cpp_name: ENTRY)
(variable: name_not_typename '@' ENTRY, name: ENTRY)
(name_not_typename: ENTRY): New.
(yylex): Recognize ENTRY.
* eval.c (evaluate_subexp_standard): Support also OP_VAR_ENTRY_VALUE.
* expprint.c (print_subexp_standard, dump_subexp_body_standard):
Likewise.
* parse.c (operator_length_standard): Likewise.
* std-operator.def: New operator OP_VAR_ENTRY_VALUE.
gdb/doc/
Support @entry in input expressions.
* gdb.texinfo (Variables): Describe @entry names suffix.
(Print Settings): Add anchor for `set print entry-values'.
gdb/testsuite/
Support @entry in input expressions.
* gdb.arch/amd64-entry-value.exp (entry: p i@entry, entry: p j@entry)
(entry_stack: p s1@entry, entry_stack: p s2@entry)
(entry_stack: p d9@entry, entry_stack: p da@entry, tailcall: p i@entry)
(tailcall: p j@entry): New tests.
* gdb.cp/koenig.cc (A::entry): New function.
(main): Call it.
* gdb.cp/koenig.exp (p entry (c)): New test.
Jan Kratochvil [Sun, 9 Oct 2011 19:36:06 +0000 (19:36 +0000)]
gdb/
Make some lval_funcs methods to default on NULL.
* valops.c (value_fetch_lazy): Check if lval_computed read method is
NULL.
(value_assign): Check if lval_computed write method is NULL.
* value.h (struct lval_funcs): Comment NULL values for read and write
methods.
Jan Kratochvil [Sun, 9 Oct 2011 19:34:19 +0000 (19:34 +0000)]
gdb/
Display @entry parameter values (without references).
* dwarf2expr.c (dwarf_block_to_fb_offset, dwarf_block_to_sp_offset):
New functions.
* dwarf2expr.h (dwarf_block_to_fb_offset, dwarf_block_to_sp_offset):
New declarations.
* dwarf2loc.c (dwarf2_find_location_expression): Support location list
entry record.
(dwarf_entry_parameter_to_value, value_of_dwarf_reg_entry)
(value_of_dwarf_block_entry, locexpr_read_variable_at_entry): New
functions.
(dwarf2_locexpr_funcs): Install locexpr_read_variable_at_entry.
(loclist_read_variable_at_entry): New function.
(dwarf2_loclist_funcs): Install loclist_read_variable_at_entry.
* dwarf2read.c (read_call_site_scope): Support also DW_OP_fbreg in
DW_AT_location, call dwarf_block_to_sp_offset for it.
* frame.h (print_entry_values_no, print_entry_values_only)
(print_entry_values_preferred, print_entry_values_if_needed)
(print_entry_values_both, print_entry_values_compact)
(print_entry_values_default, print_entry_values): New declarations.
(struct frame_arg): New field entry_kind.
(read_frame_arg): New parameter entryargp.
* mi/mi-cmd-stack.c (list_arg_or_local): New gdb_assert for
arg->entry_kind. Optionally print the `@entry' suffix.
(list_args_or_locals): New variable entryarg, initialize it.
Initialize also entry_kind of arg and entryarg. Conditionalize
list_arg_or_local for arg, add list_arg_or_local for entryarg. Call
xfree for entryarg.error.
* stack.c (print_entry_values_no, print_entry_values_only)
(print_entry_values_preferred, print_entry_values_if_needed)
(print_entry_values_both, print_entry_values_compact)
(print_entry_values_default, print_entry_values_choices)
(print_entry_values): New variables.
(print_frame_arg): New gdb_assert for arg->entry_kind. Optionally
print the `@entry' suffix, possibly in combination for
print_entry_values_compact.
(read_frame_arg): New parameter entryargp, new variables entryval,
entryval_error and val_equal. Read in also entryargp, respect
print_entry_values, compare the values using val_equal, fill in also
argp->entry_kind (together with entryargp->entry_kind).
(print_frame_args): New variable entryarg, initialize it.
Conditionalize print_frame_arg for arg, add print_frame_arg for
entryarg. Call xfree for entryarg.error.
(_initialize_stack): Call add_setshow_enum_cmd for `entry-values'.
* symtab.h (struct symbol_computed_ops): New field
read_variable_at_entry.
gdb/doc/
Display @entry parameter values (without references).
* gdb.texinfo (Tail Call Frames): Add anchor. Add self tail call
example.
(Print Settings): New description of set print entry-values and show
print entry-values.
gdb/testsuite/
Display @entry parameter values (without references).
* gdb.arch/amd64-entry-value.cc (locexpr, stacktest, data, data2)
(different, validity, invalid): New functions.
(main): Call them.
* gdb.arch/amd64-entry-value.exp: New breakpoints breakhere_locexpr,
stacktest, breakhere_stacktest, different, breakhere_different,
breakhere_validity and breakhere_invalid.
(entry: bt): Update for @entry.
(entry_locexpr: *, entry_stack: *, entry_equal: *, entry_different: *)
(entry_validity: *, entry_invalid: *): Many new tests.
* gdb.base/break.exp
(run until breakpoint set at small function, optimized file): Accept
also the @entry suffix.
* gdb.mi/Makefile.in (PROGS): Add mi2-amd64-entry-value.
* gdb.mi/mi2-amd64-entry-value.c: New files.
* gdb.mi/mi2-amd64-entry-value.exp: New files.
Jan Kratochvil [Sun, 9 Oct 2011 19:30:25 +0000 (19:30 +0000)]
gdb/
Code reshuffle.
* frame.h (struct frame_arg): New definition.
(read_frame_arg): New declaration.
* mi/mi-cmd-stack.c (list_arg_or_local): New functiom from ...
(list_args_or_locals): ... the code here. New variable arg, call
read_frame_arg and list_arg_or_local with it. Unify the
PRINT_SIMPLE_VALUES and PRINT_ALL_VALUES cases. Call xfree for
arg.error.
* stack.c (print_frame_arg): New functiom from the code of
print_frame_args.
(read_frame_arg): New function.
(print_frame_args): Remove variable val. New variable arg, call
read_frame_arg and print_frame_arg with it. Call xfree for arg.error.
Jan Kratochvil [Sun, 9 Oct 2011 19:23:41 +0000 (19:23 +0000)]
gdb/
Tail call sites reader implementation.
* dwarf2read.c (read_call_site_scope): Recognize DW_AT_GNU_tail_call,
fill in TYPE_TAIL_CALL_LIST.
* gdbtypes.h (struct func_type): New field tail_call_list.
(struct call_site): New field tail_call_next.
(TYPE_TAIL_CALL_LIST): New definition.
Jan Kratochvil [Sun, 9 Oct 2011 19:21:39 +0000 (19:21 +0000)]
gdb/
Implement basic support for DW_TAG_GNU_call_site.
* block.c: Include gdbtypes.h and exceptions.h.
(call_site_for_pc): New function.
* block.h (call_site_for_pc): New declaration.
* defs.h: Include hashtab.h.
(make_cleanup_htab_delete, core_addr_hash, core_addr_eq): New
declarations.
* dwarf2-frame.c (dwarf2_frame_ctx_funcs): Install
ctx_no_push_dwarf_reg_entry_value.
* dwarf2expr.c (read_uleb128, read_sleb128): Support R as NULL.
(dwarf_block_to_dwarf_reg): New function.
(execute_stack_op) <DW_OP_GNU_entry_value>: Implement it.
(ctx_no_push_dwarf_reg_entry_value): New function.
* dwarf2expr.h (struct dwarf_expr_context_funcs): New field
push_dwarf_reg_entry_value.
(ctx_no_push_dwarf_reg_entry_value, dwarf_block_to_dwarf_reg): New
declarations.
* dwarf2loc.c: Include gdbcmd.h.
(dwarf_expr_ctx_funcs): New forward declaration.
(entry_values_debug, show_entry_values_debug, call_site_to_target_addr)
(dwarf_expr_reg_to_entry_parameter)
(dwarf_expr_push_dwarf_reg_entry_value): New.
(dwarf_expr_ctx_funcs): Install dwarf_expr_push_dwarf_reg_entry_value.
(dwarf2_evaluate_loc_desc_full): Handle NO_ENTRY_VALUE_ERROR.
(needs_dwarf_reg_entry_value): New function.
(needs_frame_ctx_funcs): Install it.
(_initialize_dwarf2loc): New function.
* dwarf2loc.h (entry_values_debug): New declaration.
* dwarf2read.c (struct dwarf2_cu): New field call_site_htab.
(read_call_site_scope): New forward declaration.
(process_full_comp_unit): Copy call_site_htab.
(process_die): Support DW_TAG_GNU_call_site.
(read_call_site_scope): New function.
(dwarf2_get_pc_bounds): Support NULL HIGHPC.
(dwarf_tag_name): Support DW_TAG_GNU_call_site.
(cleanup_htab): Delete.
(write_psymtabs_to_index): Use make_cleanup_htab_delete instead of it.
* exceptions.h (enum errors): New NO_ENTRY_VALUE_ERROR.
* gdb-gdb.py (StructMainTypePrettyPrinter): Support
FIELD_LOC_KIND_DWARF_BLOCK.
* gdbtypes.h (enum field_loc_kind): New entry
FIELD_LOC_KIND_DWARF_BLOCK.
(struct main_type): New loc entry dwarf_block.
(struct call_site, FIELD_DWARF_BLOCK, SET_FIELD_DWARF_BLOCK)
(TYPE_FIELD_DWARF_BLOCK): New.
* python/py-type.c: Include dwarf2loc.h.
(check_types_equal): Support FIELD_LOC_KIND_DWARF_BLOCK. New
internal_error call on unknown FIELD_LOC_KIND.
* symtab.h (struct symtab): New field call_site_htab.
* utils.c (do_htab_delete_cleanup, make_cleanup_htab_delete)
(core_addr_hash, core_addr_eq): New functions.
gdb/testsuite/
Implement basic support for DW_TAG_GNU_call_site.
* gdb.arch/Makefile.in (EXECUTABLES): Add amd64-entry-value.
* gdb.arch/amd64-entry-value.cc: New file.
* gdb.arch/amd64-entry-value.exp: New file.
Jan Kratochvil [Sun, 9 Oct 2011 18:46:41 +0000 (18:46 +0000)]
gdb/
Fix DW_OP_GNU_implicit_pointer for DWARF32 v3+ on 64-bit arches.
* dwarf2-frame.c (execute_stack_op): Initialize ctx->ref_addr_size.
* dwarf2expr.c (execute_stack_op) <DW_OP_GNU_implicit_pointer>: Use
ctx->ref_addr_size. Handle its invalid value.
* dwarf2expr.h (struct dwarf_expr_context): New field ref_addr_size.
* dwarf2loc.c (dwarf2_evaluate_loc_desc_full)
(dwarf2_loc_desc_needs_frame): Initialize ctx->ref_addr_size.
* dwarf2loc.h (dwarf2_per_cu_ref_addr_size): New declaration.
* dwarf2read.c (decode_locdesc): Initialize ctx->ref_addr_size.
(dwarf2_per_cu_ref_addr_size): New function.
gdb/testsuite/
Fix DW_OP_GNU_implicit_pointer for DWARF32 v3+ on 64-bit arches.
* gdb.dwarf2/implptr-64bit.S: New file.
* gdb.dwarf2/implptr-64bit.exp: New file.
Jan Kratochvil [Sun, 9 Oct 2011 18:43:32 +0000 (18:43 +0000)]
gdb/
Code cleanup.
* dwarf2read.c (per_cu_header_read_in): New function.
(dwarf2_per_cu_addr_size, dwarf2_per_cu_offset_size): Use it, with new
variables cu_header_local and cu_headerp.
Jan Kratochvil [Sun, 9 Oct 2011 18:37:27 +0000 (18:37 +0000)]
gdb/
Fix initial language detection with -readnow.
* dwarf2read.c (dw2_find_symbol_file): Handle OBJF_READNOW case.
* symfile.h (struct quick_symbol_functions): State find_symbol_file
searches only for global symbols.
gdb/testsuite/
Fix initial language detection with -readnow.
* gdb.cp/readnow-language.cc: New file.
* gdb.cp/readnow-language.exp: New file.
Jan Kratochvil [Sun, 9 Oct 2011 18:33:28 +0000 (18:33 +0000)]
gdb/
Fix printed anonymous struct name.
* dwarf2read.c (fixup_partial_die): Handle for anonymous structs also
DW_TAG_interface_type. Strip for anonymous structs any prefixes.
(anonymous_struct_prefix): New function.
(determine_prefix): New variables retval. Call anonymous_struct_prefix.
(dwarf2_name): Strip for anonymous structs any prefixes.
gdb/testsuite/
Fix printed anonymous struct name.
* gdb.cp/anon-struct.exp (print type of X::t2): New test.
* windows-nat.c: Include wchar.h to avoid compiler warnings.
(clear_win32_environment): New function for Cygwin to clear out
Win32 environment.
(windows_create_inferior): Prepare new environment from in_env
for Cygwin, too.
Ulrich Weigand [Fri, 7 Oct 2011 12:06:48 +0000 (12:06 +0000)]
* inferior.h (disable_randomization): Declare.
* infrun.c (disable_randomization): New global variable.
(show_disable_randomization): New function.
(set_disable_randomization): Likewise.
(_initialize_infrun): Install set/show disable-randomization
commands.
* linux-nat.c (disable_randomization): Remove.
(show_disable_randomization): Likewise.
(set_disable_randomization): Likewise.
(_initialize_linux_nat): No longer install set/show
disable-randomization commands here.
(linux_nat_supports_disable_randomization): New function.
(linux_nat_add_target): Install it.
* remote.c (PACKET_QDisableRandomization): New enum value.
(remote_protocol_packets): Support QDisableRandomization.
(_initialize_remote): Likewise.
(remote_supports_disable_randomization): New function.
(init_remote_ops): Install it.
(extended_remote_supports_disable_randomization): New function.
(init_extended_remote_ops): Install it.
(extended_remote_disable_randomization): New function.
(extended_remote_create_inferior_1): Call it.
* target.h (struct target_ops): Add to_supports_disable_randomization.
(target_supports_disable_randomization): Add prototype.
* target.c (target_supports_disable_randomization): New function.
(find_default_supports_disable_randomization): Likewise.
(init_dummy_target): Install it.
doc/
* gdb.texinfo (Starting your Program): "set disable-randomization"
is no longer Linux-specific.
(Remote Configuration): Document "set remote
disable-randomization-packet".
(General Query Packets): Document "QDisableRandomization" packet
and add it to "qSupported" list.
gdbserver/
* configure.ac: Check support for personality routine.
* configure: Regenerate.
* config.in: Likewise.
* linux-low.c: Include <sys/personality.h>.
Define ADDR_NO_RANDOMIZE if necessary.
(linux_create_inferior): Disable address space randomization when
forking inferior, if requested.
(linux_supports_disable_randomization): New function.
(linux_target_ops): Install it.
* server.h (disable_randomization): Declare.
* server.c (disable_randomization): New global variable.
(handle_general_set): Handle QDisableRandomization.
(handle_query): Likewise for qSupported.
(main): Support --disable-randomization and --no-disable-randomization
command line arguments.
* target.h (struct target_ops): Add supports_disable_randomization.
(target_supports_disable_randomization): New macro.
opcodes/
2011-09-28 Jan Beulich <jbeulich@suse.com>
* ppc-opc.c (insert_nbi, insert_rbx, FRAp, FRBp, FRSp, FRTp, NBI, RAX,
RBX): New.
(insert_bo, insert_boe): Reject bcctr with bit 2 in bo unset.
(powerpc_opcodes): Use RAX for second and RBXC for third operand of
lswx. Use NBI for third operand of lswi. Use FRTp for first operand of
lfdp and lfdpx. Use FRSp for first operand of stfdp and stfdpx, and
mark them as invalid on POWER7. Use FRTp, FRAp, and FRBp repsectively
on DFP quad instructions.