* sol-thread.c (_initialize_sol_thread): Add FIXME regarding
[deliverable/binutils-gdb.git] / gdb / ChangeLog
index 23ebcebdcb0f1e636b1de671cf351987aa8e4bc9..86ba16ef6fb2502606bdee09031f1fa577fcf0e9 100644 (file)
@@ -1,3 +1,191 @@
+2008-09-12  Doug Evans  <dje@google.com>
+
+       * sol-thread.c (_initialize_sol_thread): Add FIXME regarding
+       order of _initialize_* fns.
+
+       * dwarf2read.c (comp_unit_head): Rename first_die_ptr to
+       first_die_offset.  All uses updated.
+       Delete unused members cu_head_ptr, next.
+       Move members base_known, base_address to ...
+       (dwarf2_cu) ... here.  All uses updated.
+
+2008-09-12  Pedro Alves  <pedro@codesourcery.com>
+
+       * Makefile.in (generated_files): Add $(NAT_GENERATED_FILES).
+       * config/i386/i386gnu.mh (NAT_GENERATED_FILES): New.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * fork-child.c (startup_inferior): Use target_wait and target_resume
+       directly instead of calling wait_for_inferior / resume.
+
+       * infcmd.c (kill_if_already_running): Do not call no_shared_libraries
+       or init_wait_for_inferior.
+       (run_command_1): Call init_wait_for_inferior.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * gdbtypes.h (builtin_type_void_data_ptr, builtin_type_void_func_ptr,
+       builtin_type_CORE_ADDR, builtin_type_char, builtin_type_short,
+       builtin_type_int, builtin_type_long, builtin_type_signed_char,
+       builtin_type_unsigned_char, builtin_type_unsigned_short,
+       builtin_type_unsigned_int, builtin_type_unsigned_long,
+       builtin_type_float, builtin_type_double, builtin_type_long_double,
+       builtin_type_complex, builtin_type_double_complex, builtin_type_string,
+       builtin_type_bool, builtin_type_long_long,
+       builtin_type_unsigned_long_long): Remove macros.
+
+       (builtin_type_f_character, builtin_type_f_integer,
+       builtin_type_f_integer_s2, builtin_type_f_logical,
+       builtin_type_f_logical_s1, builtin_type_f_logical_s2,
+       builtin_type_f_real, builtin_type_f_real_s8, builtin_type_f_real_s16,
+       builtin_type_f_complex_s8, builtin_type_f_complex_s16,
+       builtin_type_f_complex_s32): Likewise.
+
+       (builtin_type_m2_char, builtin_type_m2_int, builtin_type_m2_card,
+       builtin_type_m2_real, builtin_type_m2_bool): Likewise.
+
+       (struct builtin_f_type, builtin_f_type): Move to f-lang.h.
+       (struct builtin_m2_type, builtin_m2_type): Move to m2-lang.h.
+
+       * f-lang.h (struct builtin_f_type, builtin_f_type): Move here.
+       * m2-lang.h (struct builtin_m2_type, builtin_m2_type): Move here.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * solib-svr4.c (LM_ADDR_FROM_LINK_MAP): Use builtin types of
+       target_gdbarch instead of builtin_type_void_data_ptr.
+       (LM_DYNAMIC_FROM_LINK_MAP, LM_NEXT, LM_NAME,
+       IGNORE_FIRST_LINK_MAP_ENTRY, scan_dyntag, elf_locate_base,
+       solib_svr4_r_map, solib_svr4_r_brk, solib_svr4_r_ldsomap,
+       open_symbol_file_object): Likewise.
+       * nto-tdep.c (LM_ADDR): Likewise.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * bsd-uthread.c (bsd_uthread_read_memory_address): New function.
+       (bsd_uthread_fetch_registers, bsd_uthread_store_registers,
+       bsd_uthread_wait, bsd_uthread_find_new_threads): Use it.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * procfs.c (procfs_address_to_host_pointer): Use target_gdbarch
+       and its associated types to perform pointer conversion.
+       (procfs_can_use_hw_breakpoint): Likewise.
+       (procfs_auxv_parse): Remove unused variable.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * auxv.c (default_auxv_parse): Use gdbarch_ptr_bit (target_gdbarch)
+       instead of builtin_type_void_data_ptr.
+       * target.c (default_region_ok_for_hw_watchpoint): Likewise.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * expprint.c (print_subexp_standard): Compare against builtin type
+       associated with exp->gdbarch instead of builtin_type_char.
+
+       * f-valprint.c (f_val_print): Use extract_unsigned_integer to
+       extract values of arbitrary logical type.  Handle arbitrary
+       complex types.
+
+       * printcmd.c (float_type_from_length): New function.
+       (print_scalar_formatted, printf_command): Use it.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * valops.c: Include "objfiles.h" and "symtab.h".
+       (find_function_in_inferior): New argument OBJF_P.  Use it to return
+       objfile where function is defined.  Use per-objfile arch types
+       instead of builtin_type_ to define default return type.
+
+       * linux-fork.c (checkpoint_command): Update calls.  Use per-objfile
+       architecture to define inferior call argument types.
+       * gcore.c (derive_heap_segment): Likewise.
+       * objc-lang.c (value_nsstring): Likewise.
+       * scm-lang.c (scm_lookup_name): Likewise.
+       * scm-valprint.c (scm_inferior_print): Likewise.
+       * valops.c (value_allocate_space_in_inferior): Likewise.
+
+       * eval.c (evaluate_subexp_standard): Update calls.
+       * objc-lang.c (lookup_objc_class, print_object_command): Likewise.
+
+       * linux-fork.c: Include "objfiles.h".
+       * scm-lang.c: Include "objfiles.h".
+       * scm-valprint.c: Include "objfiles.h".
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * gdbarch.sh (name_of_malloc): Remove.
+       * gdbarch.c, gdbarch.h: Re-generate.
+       * valops.c (value_allocate_space_in_inferior): Do not call
+       gdbarch_name_of_malloc.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * valarith.c (value_x_unop): Use builtin_type_int8 as type for
+       UNOP_POSTINCREMENT/UNOP_POSTDECREMENT constant 0 argument.
+       (value_bit_index): Use extract_unsigned_integer
+       instead of unpack_long to read single byte.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * infcall.c (value_arg_coerce): Add GDBARCH parameter.  Use its
+       associates types instead of builtin_type_ macros.
+       (find_function_addr): Leave output VALUE_TYPE NULL if unknown.
+       (call_function_by_hand): Use per-architecture "int" type as
+       fall-back if find_function_addr returns NULL VALUE_TYPE.
+       Update call to value_arg_coerce.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * cp-abi.h (cplus_method_ptr_size): Add TO_TYPE parameter.
+       (cplus_make_method_ptr): Add TYPE parameter.
+       * cp-abi.c (cplus_method_ptr_size): Add TO_TYPE parameter.  Pass it
+       on to current_cp_abi.method_ptr_size callback.
+       (cplus_make_method_ptr): Add TYPE parameter.  Pass it on to
+       current_cp_abi.make_method_ptr callback.
+
+       * gdbtypes.c (lookup_methodptr_type): Pass target type
+       argument to cplus_method_ptr_size.
+       * valops.c (value_cast): Pass type argument to cplus_make_method_ptr.
+       (value_struct_elt_for_reference): Likewise.
+
+       * gnu-v3-abi.c (get_class_arch): New function.
+       (vtable_address_point_offset): Add GDBARCH parameter.  Use it
+       instead of current_gdbarch.  Update all callers.
+       (gnuv3_get_vtable): Likewise.
+       (gnuv3_get_virtual_fn): Likewise.
+       (gnuv3_decode_method_ptr): Likewise.
+       (gnuv3_rtti_type): Call get_class_arch to determine architecture.
+       Use it instead of current_gdbarch.
+       (gnuv3_virtual_fn_field): Likewise.
+       (gnuv3_baseclass_offset): Likewise.
+       (gnuv3_print_method_ptr): Likewise.
+       (gnuv3_method_ptr_to_value): Likewise.
+       (gnuv3_method_ptr_size): Add TYPE parameter.  Use it to determine
+       class architecture.  Use architecture types instead of builtin types.
+       (gnuv3_make_method_ptr): Likewise.
+
+       * cp-valprint.c (cp_print_class_member): Expect pointer type
+       instead of class type.  Use its length when extracting value.
+       * c-valprint.c (c_val_print): Update call to cp_print_class_member.
+
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * stack.c (return_command): Use frame architecture to determine
+       default integer return type.
+
+       * f-valprint.c (f77_get_dynamic_lowerbound): Use frame architecture
+       to determine pointer types.
+       (f77_get_dynamic_upperbound): Likewise.
+
+       * objc-lang.c (OBJC_FETCH_POINTER_ARGUMENT): Remove.
+       (resolve_msgsend): Use architecture of current frame to determine
+       pointer types.  Inline OBJC_FETCH_POINTER_ARGUMENT.
+       (resolve_msgsend_stret, resolve_msgsend_super,
+       resolve_msgsend_super_stret): Likewise.
+
 2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
 
        * alpha-tdep.c (alpha_register_type): Use builtin_type (gdbarch)
        * sh-tdep.c (sh_sh4_build_float_register_type): Likewise.
        * sh64-tdep.c (sh64_build_float_register_type): Likewise.
 
+2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
+
+       * defs.h (struct gdbarch): Add forward declaration.
+       (set_next_address): Add GDBARCH argument.
+       * printcmd.c (set_next_address): Use it to find pointer type.
+       * breakpoint.c (breakpoint_1): Update call.
+       * source.c (line_info): Likewise.
+       * findcmd.c (find_command): Use current_gdbarch to find pointer type.
+
+       * breakpoint.c (set_breakpoint_count): Use platform-neutral
+       types for internal variable values.
+       * infrun.c (handle_inferior_event): Likewise.
+       * source.c (forward_search_command, reverse_search_command): Likewise.
+       * tracepoint.c (set_tracepoint_count, set_traceframe_num,
+       set_tracepoint_num, set_traceframe_context): Likewise.
+
 2008-09-11  Ulrich Weigand  <uweigand@de.ibm.com>
 
        * gdbtypes.h (struct builtin_type): Remove builtin_true_char
This page took 0.025882 seconds and 4 git commands to generate.