Mercurial > hg > xemacs-beta
view lisp/bytecomp-runtime.el @ 2720:6fa9919a9a0b
[xemacs-hg @ 2005-04-08 23:10:01 by crestani]
ChangeLog addition:
2005-04-01 Marcus Crestani <crestani@xemacs.org>
The new allocator.
New configure flag: `MC_ALLOC':
* configure.ac (XE_COMPLEX_ARG_ENABLE): Add `--enable-mc-alloc' as
a new configure flag.
* configure.in (AC_INIT_PARSE_ARGS): Add `--mc-alloc' as a new
configure flag.
* configure.usage: Add description for `mc-alloc'.
DUMP_IN_EXEC:
* Makefile.in.in: Condition the installation of a separate dump
file on !DUMP_ON_EXEC.
* configure.ac (XE_COMPLEX_ARG_ENABLE): Add
`--enable-dump-in-exec' as a new configure flag.
* configure.ac: DUMP_IN_EXEC is define as default for PDUMP but
not default for MC_ALLOC.
* configure.in (AC_INIT_PARSE_ARGS): Add `--dump-in-exec' as a
new configure flag.
* configure.in: DUMP_IN_EXEC is define as default for PDUMP but
not default for MC_ALLOC.
* configure.usage: Add description for `dump-in-exec'.
lib-src/ChangeLog addition:
2005-04-01 Marcus Crestani <crestani@xemacs.org>
The new allocator.
DUMP_IN_EXEC:
* Makefile.in.in: Only compile insert-data-in-exec if
DUMP_IN_EXEC is defined.
lisp/ChangeLog addition:
2005-04-01 Marcus Crestani <crestani@xemacs.org>
The new allocator.
MEMORY_USAGE_STATS
* diagnose.el: Add new lisp function to pretty print statistics
about the new allocator.
* diagnose.el (show-mc-alloc-memory-usage): New.
modules/ChangeLog addition:
2005-04-01 Marcus Crestani <crestani@xemacs.org>
The new allocator.
Remove Lcrecords:
* postgresql/postgresql.c (allocate_pgconn): Allocate with new
allocator.
* postgresql/postgresql.c (allocate_pgresult): Allocate PGresult
with new allocator.
* postgresql/postgresql.h (struct Lisp_PGconn): Add
lrecord_header.
* postgresql/postgresql.h (struct Lisp_PGresult): Add
lrecord_header.
* ldap/eldap.c (allocate_ldap): Allocate with new allocator.
* ldap/eldap.h (struct Lisp_LDAP): Add lrecord_header.
nt/ChangeLog addition:
2005-04-01 Marcus Crestani <crestani@xemacs.org>
The new allocator.
New configure flag: `MC_ALLOC':
* config.inc.samp: Add new flag `MC_ALLOC'.
* xemacs.mak: Add flag and configuration output for `MC_ALLOC'.
New files:
* xemacs.dsp: Add source files mc-alloc.c and mc-alloc.h.
* xemacs.mak: Add new object file mc-alloc.obj to dependencies.
src/ChangeLog addition:
2005-04-01 Marcus Crestani <crestani@xemacs.org>
The new allocator.
New configure flag: `MC_ALLOC':
* config.h.in: Add new flag `MC_ALLOC'.
New files:
* Makefile.in.in: Add new object file mc-alloc.o.
* depend: Add new files to dependencies.
* mc-alloc.c: New.
* mc-alloc.h: New.
Running the new allocator from XEmacs:
* alloc.c (deadbeef_memory): Moved to mc-alloc.c.
* emacs.c (main_1): Initialize the new allocator and add
syms_of_mc_alloc.
* symsinit.h: Add syms_of_mc_alloc.
New lrecord allocation and free functions:
* alloc.c (alloc_lrecord): New. Allocates an lrecord, includes
type checking and initializing of the lrecord_header.
* alloc.c (noseeum_alloc_lrecord): Same as above, but increments
the NOSEEUM cons counter.
* alloc.c (free_lrecord): New. Calls the finalizer and frees the
lrecord.
* lrecord.h: Add lrecord allocation prototypes and comments.
Remove old lrecord FROB block allocation:
* alloc.c (allocate_lisp_storage): Former function to expand
heap. Not needed anymore, remove.
* alloc.c: Completely remove `Fixed-size type macros'
* alloc.c (release_breathing_space): Remove.
* alloc.c (memory_full): Remove release_breathing_space.
* alloc.c (refill_memory_reserve): Remove.
* alloc.c (TYPE_ALLOC_SIZE): Remove.
* alloc.c (DECLARE_FIXED_TYPE_ALLOC): Remove.
* alloc.c (ALLOCATE_FIXED_TYPE_FROM_BLOCK): Remove.
* alloc.c (ALLOCATE_FIXED_TYPE_1): Remove.
* alloc.c (ALLOCATE_FIXED_TYPE): Remove.
* alloc.c (NOSEEUM_ALLOCATE_FIXED_TYPE): Remove.
* alloc.c (struct Lisp_Free): Remove.
* alloc.c (LRECORD_FREE_P): Remove.
* alloc.c (MARK_LRECORD_AS_FREE): Remove.
* alloc.c (MARK_LRECORD_AS_NOT_FREE): Remove.
* alloc.c (PUT_FIXED_TYPE_ON_FREE_LIST): Remove.
* alloc.c (FREE_FIXED_TYPE): Remove.
* alloc.c (FREE_FIXED_TYPE_WHEN_NOT_IN_GC): Remove.
Allocate old lrecords with new allocator:
* alloc.c: DECLARE_FIXED_TYPE_ALLOC removed for all lrecords
defined in alloc.c.
* alloc.c (Fcons): Allocate with new allocator.
* alloc.c (noseeum_cons): Allocate with new allocator.
* alloc.c (make_float): Allocate with new allocator.
* alloc.c (make_bignum): Allocate with new allocator.
* alloc.c (make_bignum_bg): Allocate with new allocator.
* alloc.c (make_ratio): Allocate with new allocator.
* alloc.c (make_ratio_bg): Allocate with new allocator.
* alloc.c (make_ratio_rt): Allocate with new allocator.
* alloc.c (make_bigfloat): Allocate with new allocator.
* alloc.c (make_bigfloat_bf): Allocate with new allocator.
* alloc.c (make_compiled_function): Allocate with new allocator.
* alloc.c (Fmake_symbol): Allocate with new allocator.
* alloc.c (allocate_extent): Allocate with new allocator.
* alloc.c (allocate_event): Allocate with new allocator.
* alloc.c (make_key_data): Allocate with new allocator.
* alloc.c (make_button_data): Allocate with new allocator.
* alloc.c (make_motion_data): Allocate with new allocator.
* alloc.c (make_process_data): Allocate with new allocator.
* alloc.c (make_timeout_data): Allocate with new allocator.
* alloc.c (make_magic_data): Allocate with new allocator.
* alloc.c (make_magic_eval_data): Allocate with new allocator.
* alloc.c (make_eval_data): Allocate with new allocator.
* alloc.c (make_misc_user_data): Allocate with new allocator.
* alloc.c (Fmake_marker): Allocate with new allocator.
* alloc.c (noseeum_make_marker): Allocate with new allocator.
* alloc.c (make_uninit_string): Allocate with new allocator.
* alloc.c (resize_string): Allocate with new allocator.
* alloc.c (make_string_nocopy): Allocate with new allocator.
Garbage Collection:
* alloc.c (GC_CHECK_NOT_FREE): Remove obsolete assertions.
* alloc.c (SWEEP_FIXED_TYPE_BLOCK): Remove.
* alloc.c (SWEEP_FIXED_TYPE_BLOCK_1): Remove.
* alloc.c (sweep_conses): Remove.
* alloc.c (free_cons): Use new allocator to free.
* alloc.c (sweep_compiled_functions): Remove.
* alloc.c (sweep_floats): Remove.
* alloc.c (sweep_bignums): Remove.
* alloc.c (sweep_ratios): Remove.
* alloc.c (sweep_bigfloats): Remove.
* alloc.c (sweep_symbols): Remove.
* alloc.c (sweep_extents): Remove.
* alloc.c (sweep_events): Remove.
* alloc.c (sweep_key_data): Remove.
* alloc.c (free_key_data): Use new allocator to free.
* alloc.c (sweep_button_data): Remove.
* alloc.c (free_button_data): Use new allocator to free.
* alloc.c (sweep_motion_data): Remove.
* alloc.c (free_motion_data): Use new allocator to free.
* alloc.c (sweep_process_data): Remove.
* alloc.c (free_process_data): Use new allocator to free.
* alloc.c (sweep_timeout_data): Remove.
* alloc.c (free_timeout_data): Use new allocator to free.
* alloc.c (sweep_magic_data): Remove.
* alloc.c (free_magic_data): Use new allocator to free.
* alloc.c (sweep_magic_eval_data): Remove.
* alloc.c (free_magic_eval_data): Use new allocator to free.
* alloc.c (sweep_eval_data): Remove.
* alloc.c (free_eval_data): Use new allocator to free.
* alloc.c (sweep_misc_user_data): Remove.
* alloc.c (free_misc_user_data): Use new allocator to free.
* alloc.c (sweep_markers): Remove.
* alloc.c (free_marker): Use new allocator to free.
* alloc.c (garbage_collect_1): Remove release_breathing_space.
* alloc.c (gc_sweep): Remove all the old lcrecord and lrecord
related stuff. Sweeping now works like this: compact string
chars, finalize, sweep.
* alloc.c (common_init_alloc_early): Remove old lrecord
initializations, remove breathing_space.
* emacs.c (Fdump_emacs): Remove release_breathing_space.
* lisp.h: Remove prototype for release_breathing_space.
* lisp.h: Adjust the special cons mark makros.
Lrecord Finalizer:
* alloc.c: Add finalizer to lrecord definition.
* alloc.c (finalize_string): Add finalizer for string.
* bytecode.c: Add finalizer to lrecord definition.
* bytecode.c (finalize_compiled_function): Add finalizer for
compiled function.
* marker.c: Add finalizer to lrecord definition.
* marker.c (finalize_marker): Add finalizer for marker.
These changes build the interface to mc-alloc:
* lrecord.h (MC_ALLOC_CALL_FINALIZER): Tell mc-alloc how to
finalize lrecords.
* lrecord.h (MC_ALLOC_CALL_FINALIZER_FOR_DISKSAVE): Tell
mc-alloc how to finalize for disksave.
Unify lrecords and lcrecords:
* lisp.h (struct Lisp_String): Adjust string union hack to
new lrecord header.
* lrecord.h: Adjust comments.
* lrecord.h (struct lrecord_header): The new lrecord header
includes type, lisp-readonly, free, and uid.
* lrecord.h (set_lheader_implementation): Adjust to new
lrecord_header.
* lrecord.h (struct lrecord_implementation): The field basic_p
for indication of an old lrecord is not needed anymore, remove.
* lrecord.h (MAKE_LRECORD_IMPLEMENTATION): Remove basic_p.
* lrecord.h (MAKE_EXTERNAL_LRECORD_IMPLEMENTATION): Remove
basic_p.
* lrecord.h (copy_sized_lrecord): Remove distinction between
old lrecords and lcrecords.
* lrecord.h (copy_lrecord): Remove distinction between old
lrecords and lcrecords.
* lrecord.h (zero_sized_lrecord): Remove distinction between
old lrecords and lcrecords.
* lrecord.h (zero_lrecord): Remove distinction between old
lrecords and lcrecords.
Remove lcrecords and lcrecord lists:
* alloc.c (basic_alloc_lcrecord): Not needed anymore, remove.
* alloc.c (very_old_free_lcrecord): Not needed anymore, remove.
* alloc.c (copy_lisp_object): No more distinction between
lrecords and lcrecords.
* alloc.c (all_lcrecords): Not needed anymore, remove.
* alloc.c (make_vector_internal): Allocate as lrecord.
* alloc.c (make_bit_vector_internal): Allocate as lrecord.
* alloc.c: Completely remove `lcrecord lists'.
* alloc.c (free_description): Remove.
* alloc.c (lcrecord_list_description): Remove.
* alloc.c (mark_lcrecord_list): Remove.
* alloc.c (make_lcrecord_list): Remove.
* alloc.c (alloc_managed_lcrecord): Remove.
* alloc.c (free_managed_lcrecord): Remove.
* alloc.c (alloc_automanaged_lcrecord): Remove.
* alloc.c (free_lcrecord): Remove.
* alloc.c (lcrecord_stats): Remove.
* alloc.c (tick_lcrecord_stats): Remove.
* alloc.c (disksave_object_finalization_1): Add call to
mc_finalize_for_disksave. Remove the lcrecord way to visit all
objects.
* alloc.c (kkcc_marking): Remove XD_FLAG_FREE_LISP_OBJECT
* alloc.c (sweep_lcrecords_1): Remove.
* alloc.c (common_init_alloc_early): Remove everything related
to lcrecords, remove old lrecord initializations,
* alloc.c (init_lcrecord_lists): Not needed anymore, remove.
* alloc.c (reinit_alloc_early): Remove everything related to
lcrecords.
* alloc.c (init_alloc_once_early): Remove everything related to
lcrecords.
* buffer.c (allocate_buffer): Allocate as lrecord.
* buffer.c (nuke_all_buffer_slots): Use lrecord functions.
* buffer.c (common_init_complex_vars_of_buffer): Allocate as
lrecord.
* buffer.h (struct buffer): Add lrecord_header.
* casetab.c (allocate_case_table): Allocate as lrecord.
* casetab.h (struct Lisp_Case_Table): Add lrecord_header.
* charset.h (struct Lisp_Charset): Add lrecord_header.
* chartab.c (fill_char_table): Use lrecord functions.
* chartab.c (Fmake_char_table): Allocate as lrecord.
* chartab.c (make_char_table_entry): Allocate as lrecord.
* chartab.c (copy_char_table_entry): Allocate as lrecord.
* chartab.c (Fcopy_char_table): Allocate as lrecord.
* chartab.c (put_char_table): Use lrecord functions.
* chartab.h (struct Lisp_Char_Table_Entry): Add lrecord_header.
* chartab.h (struct Lisp_Char_Table): Add lrecord_header.
* console-impl.h (struct console): Add lrecord_header.
* console-msw-impl.h (struct Lisp_Devmode): Add lrecord_header.
* console-msw-impl.h (struct mswindows_dialog_id): Add
lrecord_header.
* console.c (allocate_console): Allocate as lrecord.
* console.c (nuke_all_console_slots): Use lrecord functions.
* console.c (common_init_complex_vars_of_console): Allocate as
lrecord.
* data.c (make_weak_list): Allocate as lrecord.
* data.c (make_weak_box): Allocate as lrecord.
* data.c (make_ephemeron): Allocate as lrecord.
* database.c (struct Lisp_Database): Add lrecord_header.
* database.c (allocate_database): Allocate as lrecord.
* device-impl.h (struct device): Add lrecord_header.
* device-msw.c (allocate_devmode): Allocate as lrecord.
* device.c (nuke_all_device_slots): Use lrecord functions.
* device.c (allocate_device): Allocate as lrecord.
* dialog-msw.c (handle_question_dialog_box): Allocate as lrecord.
* elhash.c (struct Lisp_Hash_Table): Add lrecord_header.
* elhash.c (make_general_lisp_hash_table): Allocate as lrecord.
* elhash.c (Fcopy_hash_table): Allocate as lrecord.
* event-stream.c: Lcrecord lists Vcommand_builder_free_list and
Vtimeout_free_list are no longer needed. Remove.
* event-stream.c (allocate_command_builder): Allocate as lrecord.
* event-stream.c (free_command_builder): Use lrecord functions.
* event-stream.c (event_stream_generate_wakeup): Allocate as
lrecord.
* event-stream.c (event_stream_resignal_wakeup): Use lrecord
functions.
* event-stream.c (event_stream_disable_wakeup): Use lrecord
functions.
* event-stream.c (reinit_vars_of_event_stream): Lcrecord lists
remove.
* events.h (struct Lisp_Timeout): Add lrecord_header.
* events.h (struct command_builder): Add lrecord_header.
* extents-impl.h (struct extent_auxiliary): Add lrecord_header.
* extents-impl.h (struct extent_info): Add lrecord_header.
* extents.c (allocate_extent_auxiliary): Allocate as lrecord.
* extents.c (allocate_extent_info): Allocate as lrecord.
* extents.c (copy_extent): Allocate as lrecord.
* faces.c (allocate_face): Allocate as lrecord.
* faces.h (struct Lisp_Face): Add lrecord_header.
* file-coding.c (allocate_coding_system): Allocate as lrecord.
* file-coding.c (Fcopy_coding_system): Allocate as lrecord.
* file-coding.h (struct Lisp_Coding_System): Add lrecord_header.
* fns.c (Ffillarray): Allocate as lrecord.
* frame-impl.h (struct frame): Add lrecord_header.
* frame.c (nuke_all_frame_slots): Use lrecord functions.
* frame.c (allocate_frame_core): Allocate as lrecord.
* glyphs.c (allocate_image_instance): Allocate as lrecord.
* glyphs.c (Fcolorize_image_instance): Allocate as lrecord.
* glyphs.c (allocate_glyph): Allocate as lrecord.
* glyphs.h (struct Lisp_Image_Instance): Add lrecord_header.
* glyphs.h (struct Lisp_Glyph): Add lrecord_header.
* gui.c (allocate_gui_item): Allocate as lrecord.
* gui.h (struct Lisp_Gui_Item): Add lrecord_header.
* keymap.c (struct Lisp_Keymap): Add lrecord_header.
* keymap.c (make_keymap): Allocate as lrecord.
* lisp.h (struct Lisp_Vector): Add lrecord_header.
* lisp.h (struct Lisp_Bit_Vector): Add lrecord_header.
* lisp.h (struct weak_box): Add lrecord_header.
* lisp.h (struct ephemeron): Add lrecord_header.
* lisp.h (struct weak_list): Add lrecord_header.
* lrecord.h (struct lcrecord_header): Not used, remove.
* lrecord.h (struct free_lcrecord_header): Not used, remove.
* lrecord.h (struct lcrecord_list): Not needed anymore, remove.
* lrecord.h (lcrecord_list): Not needed anymore, remove.
* lrecord.h: (enum data_description_entry_flags): Remove
XD_FLAG_FREE_LISP_OBJECT.
* lstream.c: Lrecord list Vlstream_free_list remove.
* lstream.c (Lstream_new): Allocate as lrecord.
* lstream.c (Lstream_delete): Use lrecod functions.
* lstream.c (reinit_vars_of_lstream): Vlstream_free_list
initialization remove.
* lstream.h (struct lstream): Add lrecord_header.
* emacs.c (main_1): Remove lstream initialization.
* mule-charset.c (make_charset): Allocate as lrecord.
* objects-impl.h (struct Lisp_Color_Instance): Add
lrecord_header.
* objects-impl.h (struct Lisp_Font_Instance): Add lrecord_header.
* objects.c (Fmake_color_instance): Allocate as lrecord.
* objects.c (Fmake_font_instance): Allocate as lrecord.
* objects.c (reinit_vars_of_objects): Allocate as lrecord.
* opaque.c: Lcreord list Vopaque_ptr_list remove.
* opaque.c (make_opaque): Allocate as lrecord.
* opaque.c (make_opaque_ptr): Allocate as lrecord.
* opaque.c (free_opaque_ptr): Use lrecord functions.
* opaque.c (reinit_opaque_early):
* opaque.c (init_opaque_once_early): Vopaque_ptr_list
initialization remove.
* opaque.h (Lisp_Opaque): Add lrecord_header.
* opaque.h (Lisp_Opaque_Ptr): Add lrecord_header.
* emacs.c (main_1): Remove opaque variable initialization.
* print.c (default_object_printer): Use new lrecord_header.
* print.c (print_internal): Use new lrecord_header.
* print.c (debug_p4): Use new lrecord_header.
* process.c (make_process_internal): Allocate as lrecord.
* procimpl.h (struct Lisp_Process): Add lrecord_header.
* rangetab.c (Fmake_range_table): Allocate as lrecord.
* rangetab.c (Fcopy_range_table): Allocate as lrecord.
* rangetab.h (struct Lisp_Range_Table): Add lrecord_header.
* scrollbar.c (create_scrollbar_instance): Allocate as lrecord.
* scrollbar.h (struct scrollbar_instance): Add lrecord_header.
* specifier.c (make_specifier_internal): Allocate as lrecord.
* specifier.h (struct Lisp_Specifier): Add lrecord_header.
* symbols.c:
* symbols.c (Fmake_variable_buffer_local): Allocate as lrecord.
* symbols.c (Fdontusethis_set_symbol_value_handler): Allocate
as lrecord.
* symbols.c (Fdefvaralias): Allocate as lrecord.
* symeval.h (struct symbol_value_magic): Add lrecord_header.
* toolbar.c (update_toolbar_button): Allocate as lrecord.
* toolbar.h (struct toolbar_button): Add lrecord_header.
* tooltalk.c (struct Lisp_Tooltalk_Message): Add lrecord_header.
* tooltalk.c (make_tooltalk_message): Allocate as lrecord.
* tooltalk.c (struct Lisp_Tooltalk_Pattern): Add lrecord_header.
* tooltalk.c (make_tooltalk_pattern): Allocate as lrecord.
* ui-gtk.c (allocate_ffi_data): Allocate as lrecord.
* ui-gtk.c (allocate_emacs_gtk_object_data): Allocate as lrecord.
* ui-gtk.c (allocate_emacs_gtk_boxed_data): Allocate as lrecord.
* ui-gtk.h (structs): Add lrecord_header.
* window-impl.h (struct window): Add lrecord_header.
* window-impl.h (struct window_mirror): Add lrecord_header.
* window.c (allocate_window): Allocate as lrecord.
* window.c (new_window_mirror): Allocate as lrecord.
* window.c (make_dummy_parent): Allocate as lrecord.
MEMORY_USAGE_STATS
* alloc.c (fixed_type_block_overhead): Not used anymore, remove.
* buffer.c (compute_buffer_usage): Get storage size from new
allocator.
* marker.c (compute_buffer_marker_usage): Get storage size from
new allocator.
* mule-charset.c (compute_charset_usage): Get storage size from
new allocator.
* scrollbar-gtk.c (gtk_compute_scrollbar_instance_usage): Get
storage size from new allocator.
* scrollbar-msw.c (mswindows_compute_scrollbar_instance_usage):
Get storage size from new allocator.
* scrollbar-x.c (x_compute_scrollbar_instance_usage): Get
storage size from new allocator.
* scrollbar.c (compute_scrollbar_instance_usage): Get storage
size from new allocator.
* unicode.c (compute_from_unicode_table_size_1): Get storage
size from new allocator.
* unicode.c (compute_to_unicode_table_size_1): Get storage size
from new allocator.
* window.c (compute_window_mirror_usage): Get storage size from
new allocator.
* window.c (compute_window_usage): Get storage size from new
allocator.
MC_ALLOC_TYPE_STATS:
* alloc.c (alloc_lrecord): Bump lrecord count.
* alloc.c (noseeum_alloc_lrecord): Bump lrecord count.
* alloc.c (struct lrecord_stats): Storage for counts.
* alloc.c (init_lrecord_stats): Zero statistics.
* alloc.c (inc_lrecord_stats): Increase the statistic.
* alloc.c (dec_lrecord_stats): Decrease the statistic.
* alloc.c (gc_plist_hack): Used to print the information.
* alloc.c (Fgarbage_collect): Return the collected information.
* mc-alloc.c (remove_cell): Decrease lrecord count.
* mc-alloc.h: Set flag MC_ALLOC_TYPE_STATS.
* emacs.c (main_1): Init lrecord statistics.
* lrecord.h: Add prototypes for *_lrecord_stats.
Strings:
* alloc.c (Fmake_string): Initialize ascii_begin to zero.
* alloc.c (gc_count_num_short_string_in_use): Remove.
* alloc.c (gc_count_string_total_size): Remove.
* alloc.c (gc_count_short_string_total_size): Remove.
* alloc.c (debug_string_purity): Remove.
* alloc.c (debug_string_purity_print): Remove.
* alloc.c (sweep_strings): Remove.
Remove static C-readonly Lisp objects:
* alloc.c (c_readonly): Not needed anymore, remove.
* alloc.c (GC_CHECK_LHEADER_INVARIANTS): Remove some obsolete
lheader invariants assertions.
* buffer.c (DEFVAR_BUFFER_LOCAL_1): Allocate dynamically.
* console.c (DEFVAR_CONSOLE_LOCAL_1): Allocate dynamically.
* gpmevent.c: Indirection via MC_ALLOC_Freceive_gpm_event.
* gpmevent.c (Fgpm_enable): Allocate dynamically.
* gpmevent.c (syms_of_gpmevent): Allocate dynamically.
* lisp.h (C_READONLY): Not needed anymore, remove.
* lisp.h (DEFUN): Allocate dynamically.
* lrecord.h (C_READONLY_RECORD_HEADER_P): Not needed anymore,
remove.
* lrecord.h (SET_C_READONLY_RECORD_HEADER): Not needed anymore,
remove.
* symbols.c (guts_of_unbound_marker):
* symeval.h (defsubr): Allocate dynamically.
* symeval.h (DEFSUBR_MACRO): Allocate dynamically.
* symeval.h (DEFVAR_ SYMVAL_FWD): Allocate dynamically.
* tests.c (TESTS_DEFSUBR): Allocate dynamically.
Definition of mcpro:
* lisp.h: Add mcpro prototypes.
* alloc.c (common_init_alloc_early): Add initialization for
mcpros.
* alloc.c (mcpro_description_1): New.
* alloc.c (mcpro_description): New.
* alloc.c (mcpros_description_1): New.
* alloc.c (mcpros_description): New.
* alloc.c (mcpro_one_name_description_1): New.
* alloc.c (mcpro_one_name_description): New.
* alloc.c (mcpro_names_description_1): New.
* alloc.c (mcpro_names_description): New.
* alloc.c (mcpros): New.
* alloc.c (mcpro_names): New.
* alloc.c (mcpro_1): New.
* alloc.c (mc_pro): New.
* alloc.c (garbage_collect_1): Add mcpros to root set.
Usage of mcpro:
* alloc.c (make_string_nocopy): Add string to root set.
* symbols.c (init_symbols_once_early): Add Qunbound to root set.
Changes to the Portable Dumper:
* alloc.c (FREE_OR_REALLOC_BEGIN): Since dumped objects can be
freed with the new allocator, remove assertion for !DUMPEDP.
* dumper.c: Adjust comments, increase PDUMP_HASHSIZE.
* dumper.c (pdump_make_hash): Shift address only 2 bytes, to
avoid collisions.
* dumper.c (pdump_objects_unmark): No more mark bits within
the object, remove.
* dumper.c (mc_addr_elt): New. Element data structure for mc
hash table.
* dumper.c (pdump_mc_hash): New hash table: `lookup table'.
* dumper.c (pdump_get_mc_addr): New. Lookup for hash table.
* dumper.c (pdump_get_indirect_mc_addr): New. Lookup for
convertibles.
* dumper.c (pdump_put_mc_addr): New. Putter for hash table.
* dumper.c (pdump_dump_mc_data): New. Writes the table for
relocation at load time to the dump file.
* dumper.c (pdump_scan_lisp_objects_by_alignment): New.
Visits all dumped Lisp objects.
* dumper.c (pdump_scan_non_lisp_objects_by_alignment): New.
Visits all other dumped objects.
* dumper.c (pdump_reloc_one_mc): New. Updates all pointers
of an object by using the hash table pdump_mc_hash.
* dumper.c (pdump_reloc_one): Replaced by pdump_reloc_one_mc.
* dumper.c (pdump): Change the structure of the dump file, add
the mc post dump relocation table to dump file.
* dumper.c (pdump_load_finish): Hand all dumped objects to the
new allocator and use the mc post dump relocation table for
relocating the dumped objects at dump file load time, free not
longer used data structures.
* dumper.c (pdump_load): Free the dump file.
* dumper.h: Remove pdump_objects_unmark.
* lrecord.h (DUMPEDP): Dumped objects can be freed, remove.
DUMP_IN_EXEC:
* Makefile.in.in: Linking for and with dump in executable only if
DUMP_IN_EXEC is defined.
* config.h.in: Add new flag `DUMP_IN_EXEC'
* emacs.c: Condition dump-data.h on DUMP_IN_EXEC.
* emacs.c (main_1): Flag `-si' only works if dump image is
written into executable.
Miscellanious
* lrecord.h (enum lrecord_type): Added numbers to all types,
very handy for debugging.
* xemacs.def.in.in: Add mc-alloc functions to make them visible
to the modules.
author | crestani |
---|---|
date | Fri, 08 Apr 2005 23:11:35 +0000 |
parents | 0e3842cd22e2 |
children | 018e13fdeaeb |
line wrap: on
line source
;;; bytecomp-runtime.el --- byte-compiler support for inlining ;; Copyright (C) 1992, 1997 Free Software Foundation, Inc. ;; Copyright (C) 2002 Ben Wing. ;; Author: Jamie Zawinski <jwz@jwz.org> ;; Author: Hallvard Furuseth <hbf@ulrik.uio.no> ;; Maintainer: XEmacs Development Team ;; Keywords: internal, dumped ;; This file is part of XEmacs. ;; XEmacs is free software; you can redistribute it and/or modify it ;; under the terms of the GNU General Public License as published by ;; the Free Software Foundation; either version 2, or (at your option) ;; any later version. ;; XEmacs is distributed in the hope that it will be useful, but ;; WITHOUT ANY WARRANTY; without even the implied warranty of ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU ;; General Public License for more details. ;; You should have received a copy of the GNU General Public License ;; along with XEmacs; see the file COPYING. If not, write to the ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330, ;; Boston, MA 02111-1307, USA. ;;; Synched up with: FSF 19.30. ;;; Commentary: ;; This file is dumped with XEmacs. ;; The code in this file should always be loaded, because it defines things ;; like "defsubst" which should work interpreted as well. The code in ;; bytecomp.el and byte-optimize.el can be loaded as needed. ;; interface to selectively inlining functions. ;; This only happens when source-code optimization is turned on. ;;; Code: ;; Redefined in byte-optimize.el. ;; This is not documented--it's not clear that we should promote it. (fset 'inline 'progn) (put 'inline 'lisp-indent-hook 0) ;;; Interface to inline functions. ;; FSF comments the next two out, but I see no reason to do so. --ben (defmacro proclaim-inline (&rest fns) "Cause the named functions to be open-coded when called from compiled code. They will only be compiled open-coded when `byte-optimize' is true." (cons 'eval-and-compile (apply 'nconc (mapcar #'(lambda (x) `((or (memq (get ',x 'byte-optimizer) '(nil byte-compile-inline-expand)) (error "%s already has a byte-optimizer, can't make it inline" ',x)) (put ',x 'byte-optimizer 'byte-compile-inline-expand))) fns)))) (defmacro proclaim-notinline (&rest fns) "Cause the named functions to no longer be open-coded." (cons 'eval-and-compile (apply 'nconc (mapcar #'(lambda (x) `((if (eq (get ',x 'byte-optimizer) 'byte-compile-inline-expand) (put ',x 'byte-optimizer nil)))) fns)))) ;; This has a special byte-hunk-handler in bytecomp.el. (defmacro defsubst (name arglist &rest body) "Define an inline function. The syntax is just like that of `defun'." (or (memq (get name 'byte-optimizer) '(nil byte-compile-inline-expand)) (error "`%s' is a primitive" name)) (list 'prog1 (cons 'defun (cons name (cons arglist body))) (list 'proclaim-inline name))) ; Instead of the above line, FSF has this: ; (list 'eval-and-compile ; (list 'put (list 'quote name) ; ''byte-optimizer ''byte-compile-inline-expand)))) (defun make-obsolete (fn new &optional when) "Make the byte-compiler warn that function FN is obsolete. The warning will say that NEW should be used instead. If NEW is a string, that is the `use instead' message. If provided, WHEN should be a string indicating when the function was first made obsolete, for example a date or a release number." (interactive "aMake function obsolete: \nxObsoletion replacement: ") (let ((handler (get fn 'byte-compile))) (if (eq 'byte-compile-obsolete handler) (setcar (get fn 'byte-obsolete-info) new) (put fn 'byte-obsolete-info (cons new handler)) (put fn 'byte-compile 'byte-compile-obsolete))) fn) (defun make-obsolete-variable (var new &optional when) "Make the byte-compiler warn that variable VAR is obsolete, and NEW should be used instead. If NEW is a string, then that is the `use instead' message. If provided, WHEN should be a string indicating when the variable was first made obsolete, for example a date or a release number." (interactive (list (let ((str (completing-read "Make variable obsolete: " obarray 'boundp t))) (if (equal str "") (error "")) (intern str)) (car (read-from-string (read-string "Obsoletion replacement: "))))) (put var 'byte-obsolete-variable new) var) ;; By overwhelming demand, we separate out truly obsolete symbols from ;; those that are present for GNU Emacs compatibility. (defun make-compatible (fn new) "Make the byte-compiler know that function FN is provided for compatibility. The warning will say that NEW should be used instead. If NEW is a string, that is the `use instead' message." (interactive "aMake function compatible: \nxCompatible replacement: ") (let ((handler (get fn 'byte-compile))) (if (eq 'byte-compile-compatible handler) (setcar (get fn 'byte-compatible-info) new) (put fn 'byte-compatible-info (cons new handler)) (put fn 'byte-compile 'byte-compile-compatible))) fn) (defun make-compatible-variable (var new) "Make the byte-compiler know that variable VAR is provided for compatibility, and NEW should be used instead. If NEW is a string, then that is the `use instead' message." (interactive (list (let ((str (completing-read "Make variable compatible: " obarray 'boundp t))) (if (equal str "") (error "")) (intern str)) (car (read-from-string (read-string "Compatible replacement: "))))) (put var 'byte-compatible-variable new) var) (put 'dont-compile 'lisp-indent-hook 0) (defmacro dont-compile (&rest body) "Like `progn', but the body always runs interpreted (not compiled). If you think you need this, you're probably making a mistake somewhere." (list 'eval (list 'quote (if (cdr body) (cons 'progn body) (car body))))) ;;; interface to evaluating things at compile time and/or load time ;;; these macro must come after any uses of them in this file, as their ;;; definition in the file overrides the magic definitions on the ;;; byte-compile-macro-environment. (put 'eval-when-compile 'lisp-indent-hook 0) (defmacro eval-when-compile (&rest body) "Like `progn', but evaluates the body at compile time. The result of the body appears to the compiler as a quoted constant." ;; Not necessary because we have it in b-c-initial-macro-environment ;; (list 'quote (eval (cons 'progn body))) (cons 'progn body)) (put 'eval-and-compile 'lisp-indent-hook 0) (defmacro eval-and-compile (&rest body) "Like `progn', but evaluates the body at compile time and at load time." ;; Remember, it's magic. (cons 'progn body)) ;;; From Emacs 20. (put 'eval-when-feature 'lisp-indent-hook 1) (defmacro eval-when-feature (feature &rest body) "Run the body forms when FEATURE is featurep, be it now or later. Called (eval-when-feature (FEATURE [. FILENAME]) BODYFORMS...). If (featurep 'FEATURE), evals now; otherwise adds an elt to `after-load-alist' (which see), using FEATURE as filename if FILENAME is nil." (let ((file (or (cdr feature) (symbol-name (car feature))))) `(let ((bodythunk #'(lambda () ,@body))) (if (featurep ',(car feature)) (funcall bodythunk) (setq after-load-alist (cons '(,file . (list 'lambda '() bodythunk)) after-load-alist)))))) ;;; Functions to cleanly eliminate warnings about undefined functions ;;; or variables when the code knows what it's doing. These macros DO ;;; NOT rely on any byte-compiler changes, and thus can be copied into ;;; a package and used within it. ;; NOTE: As a result of the above requirement, the macros rely on ;; "tricks" to get the warnings suppressed. A cleaner way, of course, ;; would be to extend the byte compiler to provide a proper interface. ;; #### Should we require an unquoted symbol rather than a quoted one, ;; as we currently do? The quoting gets no generality, as `eval' is ;; called at compile time. But most functions and macros want quoted ;; arguments, and I find it extremely confusing to deal with cases ;; such as `throw' requiring a quoted argument but `block' an unquoted ;; one. (put 'with-boundp 'lisp-indent-function 1) (defmacro with-boundp (variables &rest body) "Evaluate BODY, but do not issue bytecomp warnings about VARIABLES undefined. VARIABLES can be a symbol or a list of symbols and must be quoted. When compiling this file, the warnings `reference to free variable VARIABLE' and `assignment to free variable VARIABLE' will not occur anywhere in BODY, for any of the listed variables. This is a clean way to avoid such warnings. See also `if-boundp', `when-boundp', and `and-boundp' (ways to conditionalize on a variable being bound and avoid warnings), `declare-boundp' (issue a variable call without warnings), and `globally-declare-boundp' (avoid warnings throughout a file about a variable)." (setq variables (eval variables)) (unless (consp variables) (setq variables (list variables))) `(progn (declare (special ,@variables)) ,@body)) (put 'if-boundp 'lisp-indent-function 2) (defmacro if-boundp (variable then &rest else) "Equivalent to (if (boundp VARIABLE) THEN ELSE) but handles bytecomp warnings. VARIABLE should be a quoted symbol. When compiling this file, the warnings `reference to free variable VARIABLE' and `assignment to free variable VARIABLE' will not occur anywhere in the if-statement. This is a clean way to avoid such warnings. See also `with-boundp' and friends." `(with-boundp ,variable (if (boundp ,variable) ,then ,@else))) (put 'when-boundp 'lisp-indent-function 1) (defmacro when-boundp (variable &rest body) "Equivalent to (when (boundp VARIABLE) BODY) but handles bytecomp warnings. VARIABLE should be a quoted symbol. When compiling this file, the warnings `reference to free variable VARIABLE' and `assignment to free variable VARIABLE' will not occur anywhere in the when-statement. This is a clean way to avoid such warnings. See also `with-boundp' and friends." `(with-boundp ,variable (when (boundp ,variable) ,@body))) (put 'and-boundp 'lisp-indent-function 1) (defmacro and-boundp (variable &rest args) "Equivalent to (and (boundp VARIABLE) ARGS) but handles bytecomp warnings. VARIABLE should be a quoted symbol. When compiling this file, the warnings `reference to free variable VARIABLE' and `assignment to free variable VARIABLE' will not occur anywhere in the and-statement. This is a clean way to avoid such warnings. See also `with-boundp' and friends." `(with-boundp ,variable (and (boundp ,variable) ,@args))) (defmacro declare-boundp (variable) "Evaluate VARIABLE without bytecomp warnings about the symbol. Sample usage is (declare-boundp gpm-minor-mode) which is equivalent to (with-boundp 'gpm-minor-mode gpm-minor-mode) See also `with-boundp' and friends." `(with-boundp ',variable ,variable)) (defmacro globally-declare-boundp (variables) "Declare that all free uses of VARIABLES in this file are valid. VARIABLES can be a symbol or a list of symbols and must be quoted. When compiling this file, the warnings `reference to free variable VARIABLE' and `assignment to free variable VARIABLE' will not occur regardless of where references to VARIABLE occur in the file. In general, you should *NOT* use this; use `with-boundp' or its friends to wrap individual uses, as necessary. That way, you're more likely to remember to put in the explicit checks for the variable's existence that are usually necessary. However, `globally-declare-boundp' is better in some circumstances, such as when writing an ELisp package that makes integral use of optionally-compiled-in functionality (typically, an interface onto a system library) and checks for the existence of the functionality at some entry point to the package. See `globally-declare-fboundp' for more information." (setq variables (eval variables)) (if (not (consp variables)) (setq variables (list variables))) `(progn ;; (defvar FOO) has no side effects. ,@(mapcar #'(lambda (sym) `(defvar ,sym)) variables))) (defun byte-compile-with-fboundp (form) (byte-compile-form (cons 'progn (cdr (cdr form)))) ;; Unfortunately, byte-compile-unresolved-functions is used not only ;; for unresolved-function warnings, but also in connection with the ;; following warnings: ;; "defsubst %s was used before it was defined" ;; "%s being defined to take %s%s, but was previously called with %s" ;; By hacking byte-compile-unresolved-functions like this, we ;; effectively disable these warnings. But code should not be using ;; `with-fboundp' with a function defined later on in the same ;; file, so this is not a big deal. (let ((symbols (eval (car (cdr form))))) (unless (consp symbols) (setq symbols (list symbols))) (setq symbols (mapcar #'(lambda (sym) (cons sym nil)) symbols)) (setq byte-compile-unresolved-functions (set-difference byte-compile-unresolved-functions symbols :key #'car)) )) ;; EEEEEEEEVIL hack. We need to create our own byte-compilation ;; method so that the proper variables are bound while compilation ;; takes place (which is when the warnings get noticed and batched ;; up). What we really want to do is make `with-fboundp' a macro ;; that simply `progn's its BODY; but GOD DAMN IT, macros can't have ;; their own byte-compilation methods! So we make `with-fboundp' a ;; macro calling `with-fboundp-1', which is cleverly aliased to ;; progn. This way we can put a byte-compilation method on ;; `with-fboundp-1', and when interpreting, progn will duly skip ;; the first, quoted argument, i.e. the symbol name. (We could make ;; `with-fboundp-1' a regular function, but then we'd have to thunk ;; BODY and eval it at runtime. We could probably just do this using ;; (apply 'progn BODY), but the existing method is more obviously ;; guaranteed to work.) ;; ;; In defense, cl-macs.el does a very similar thing with ;; `cl-block-wrapper'. (put 'with-fboundp-1 'byte-compile 'byte-compile-with-fboundp) (defalias 'with-fboundp-1 'progn) (put 'with-fboundp 'lisp-indent-function 1) (defmacro with-fboundp (functions &rest body) "Evaluate BODY, but do not issue bytecomp warnings about FUNCTIONS undefined. FUNCTIONS can be a symbol or a list of symbols and must be quoted. When compiling this file, the warning `the function FUNCTION is not known to be defined' will not occur anywhere in BODY, for any of the listed functions. This is a clean way to avoid such warnings. See also `if-fboundp', `when-fboundp', and `and-fboundp' (ways to conditionalize on a function being bound and avoid warnings), `declare-fboundp' (issue a function call without warnings), and `globally-declare-fboundp' (avoid warnings throughout a file about a function)." `(with-fboundp-1 ,functions ,@body)) (put 'if-fboundp 'lisp-indent-function 2) (defmacro if-fboundp (function then &rest else) "Equivalent to (if (fboundp FUNCTION) THEN ELSE) but handles bytecomp warnings. FUNCTION should be a quoted symbol. When compiling this file, the warning `the function FUNCTION is not known to be defined' will not occur anywhere in the if-statement. This is a clean way to avoid such warnings. See also `with-fboundp' and friends." `(with-fboundp ,function (if (fboundp ,function) ,then ,@else))) (put 'when-fboundp 'lisp-indent-function 1) (defmacro when-fboundp (function &rest body) "Equivalent to (when (fboundp FUNCTION) BODY) but handles bytecomp warnings. FUNCTION should be a quoted symbol. When compiling this file, the warning `the function FUNCTION is not known to be defined' will not occur anywhere in the when-statement. This is a clean way to avoid such warnings. See also `with-fboundp' and friends." `(with-fboundp ,function (when (fboundp ,function) ,@body))) (put 'and-fboundp 'lisp-indent-function 1) (defmacro and-fboundp (function &rest args) "Equivalent to (and (fboundp FUNCTION) ARGS) but handles bytecomp warnings. FUNCTION should be a quoted symbol. When compiling this file, the warning `the function FUNCTION is not known to be defined' will not occur anywhere in the and-statement. This is a clean way to avoid such warnings. See also `with-fboundp' and friends." `(with-fboundp ,function (and (fboundp ,function) ,@args))) (defmacro declare-fboundp (form) "Execute FORM (a function call) without bytecomp warnings about the call. Sample usage is (declare-fboundp (x-keysym-on-keyboard-sans-modifiers-p 'backspace)) which is equivalent to (with-fboundp 'x-keysym-on-keyboard-sans-modifiers-p (x-keysym-on-keyboard-sans-modifiers-p 'backspace)) See also `with-fboundp' and friends." `(with-fboundp ',(car form) ,form)) (defmacro globally-declare-fboundp (functions) "Declare that all calls to function FUNCTIONS in this file are valid. FUNCTIONS can be a symbol or a list of symbols and must be quoted. When compiling this file, the warning `the function FUNCTION is not known to be defined' will not occur regardless of where calls to FUNCTION occur in the file. In general, you should *NOT* use this; use `with-fboundp' or its friends to wrap individual uses, as necessary. That way, you're more likely to remember to put in the explicit checks for the function's existence that are usually necessary. However, `globally-declare-fboundp' is better in some circumstances, such as when writing an ELisp package that makes integral use of optionally-compiled-in functionality (typically, an interface onto a system library) and checks for the existence of the functionality at some entry point to the package. The file `ldap.el' is a good example: It provides a layer on top of the optional LDAP ELisp primitives, makes calls to them throughout its code, and verifies the presence of LDAP support at load time. Putting calls to `declare-fboundp' throughout the code would be a major annoyance." (when (cl-compiling-file) (setq functions (eval functions)) (if (not (consp functions)) (setq functions (list functions))) ;; Another hack. This works because the autoload environment is ;; currently used ONLY to suppress warnings, and the actual ;; autoload definition is not used. (NOTE: With this definition, ;; we will get spurious "multiple autoloads for %s" warnings if we ;; have an autoload later in the file for any functions in FUNCTIONS. ;; This is not something that code should ever do, though.) (setq byte-compile-autoload-environment (append (mapcar #'(lambda (sym) (cons sym nil)) functions) byte-compile-autoload-environment))) nil) (defun byte-compile-with-byte-compiler-warnings-suppressed (form) (let ((byte-compile-warnings byte-compile-warnings) (types (car (cdr form)))) (unless (consp types) (setq types (list types))) (if (eq byte-compile-warnings t) (setq byte-compile-warnings byte-compile-default-warnings)) (setq byte-compile-warnings (set-difference byte-compile-warnings types)) (byte-compile-form (cons 'progn (cdr (cdr form)))))) ;; Same hack here as with `with-fboundp'. (put 'with-byte-compiler-warnings-suppressed-1 'byte-compile 'byte-compile-with-byte-compiler-warnings-suppressed) (defalias 'with-byte-compiler-warnings-suppressed-1 'progn) (put 'with-byte-compiler-warnings-suppressed 'lisp-indent-function 1) (defmacro with-byte-compiler-warnings-suppressed (type &rest body) "Evaluate BODY, but do not issue bytecomp warnings TYPE. TYPE should be one of `redefine', `callargs', `subr-callargs', `free-vars', `unresolved', `unused-vars', `obsolete', or `pedantic', or a list of one or more of these symbols. (See `byte-compile-warnings'.) TYPE must be quoted. NOTE: You should *NOT* under normal circumstances be using this! There are better ways of avoiding most of these warnings. In particular: -- use (declare (special ...)) if you are making use of dynamically-scoped variables. -- use `with-fboundp' and friends to avoid warnings about undefined functions when you know the function actually exists. -- use `with-boundp' and friends to avoid warnings about undefined variables when you know the variable actually exists. -- use `with-obsolete-variable' or `with-obsolete-function' if you are purposely using such a variable or function." `(with-byte-compiler-warnings-suppressed-1 ,type ,@body)) ;; #### These should be more clever. You could (e.g.) try fletting ;; `byte-compile-obsolete' or temporarily removing the obsolete info ;; from the symbol and putting it back with an unwind-protect. (Or ;; better, modify the byte-compiler to provide a proper solution, and ;; fix these macros to use it if available, or fall back on the way ;; below. Remember, these definitions need to work with an unchanged ;; byte compiler so that they can be copied and used in packages.) (put 'with-obsolete-variable 'lisp-indent-function 1) (defmacro with-obsolete-variable (symbol &rest body) "Evaluate BODY but do not warn about usage of obsolete variable SYMBOL. SYMBOL must be quoted and can be a list of SYMBOLS. See also `with-obsolete-function'." `(with-byte-compiler-warnings-suppressed 'obsolete ,@body)) (put 'with-obsolete-function 'lisp-indent-function 1) (defmacro with-obsolete-function (symbol &rest body) "Evaluate BODY but do not warn about usage of obsolete function SYMBOL. SYMBOL must be quoted and can be a list of SYMBOLS. See also `with-obsolete-variable'." `(with-byte-compiler-warnings-suppressed 'obsolete ,@body)) ;;; Interface to file-local byte-compiler parameters. ;;; Redefined in bytecomp.el. ;;; The great RMS speaketh: ;;; ;;; I nuked this because it's not a good idea for users to think of ;;; using it. These options are a matter of installation preference, ;;; and have nothing to do with particular source files; it's a ;;; mistake to suggest to users that they should associate these with ;;; particular source files. There is hardly any reason to change ;;; these parameters, anyway. --rms. ;;; ;;; But I'll leave this stuff alone. --ben (put 'byte-compiler-options 'lisp-indent-hook 0) (defmacro byte-compiler-options (&rest args) "Set some compilation-parameters for this file. This will affect only the file in which it appears; this does nothing when evaluated, or when loaded from a .el file. Each argument to this macro must be a list of a key and a value. Keys: Values: Corresponding variable: verbose t, nil byte-compile-verbose optimize t, nil, source, byte byte-optimize warnings list of warnings byte-compile-warnings file-format emacs19, emacs20 byte-compile-emacs19-compatibility The value specified with the `warnings' option must be a list, containing some subset of the following flags: free-vars references to variables not in the current lexical scope. unused-vars references to non-global variables bound but not referenced. unresolved calls to unknown functions. callargs lambda calls with args that don't match the definition. subr-callargs calls to subrs with args that don't match the definition. redefine function cell redefined from a macro to a lambda or vice versa, or redefined to take a different number of arguments. obsolete use of an obsolete function or variable. pedantic warn of use of compatible symbols. If the first element if the list is `+' or `-' then the specified elements are added to or removed from the current set of warnings, instead of the entire set of warnings being overwritten. For example, something like this might appear at the top of a source file: (byte-compiler-options (optimize t) (warnings (- callargs)) ; Don't warn about arglist mismatch (warnings (+ unused-vars)) ; Do warn about unused bindings (file-format emacs19))" nil) ;;; bytecomp-runtime.el ends here