Problem reported by Keith David Bershatsky in:
https://lists.gnu.org/r/emacs-devel/2019-04/msg00259.html
Solution suggested by Stefan Monnier in:
https://lists.gnu.org/r/emacs-devel/2019-04/msg00282.html
* src/buffer.h (BUFFER_LISP_SIZE): Simplify by using PSEUDOVECSIZE.
(BUFFER_REST_SIZE): Simplify by using VECSIZE and BUFFER_LISP_SIZE.
* src/lisp.h (PSEUDOVECSIZE): Base it on the last Lisp field,
not the first non-Lisp field. All callers changed. Callers
without Lisp fields changed to use ALLOCATE_PLAIN_PSEUDOVECTOR.
(ALLOCATE_PLAIN_PSEUDOVECTOR): New macro.
* src/frame.c (free_monitors) [USE_GTK]: Define in the GTK case as
well.
* src/xfns.c (x-display-monitor-attributes-list) [USE_GTK]: Plug
memory leak.
* src/frame.h (MonitorInfo): Declare name as pointing to const char.
* src/composite.c (fill_gstring_header): Omit first argument HEADER,
since in practice it is always nil. Change caller to match.
Help the compiler by telling it LEN is nonnegative.
Problem found with --enable-gcc-warnings and gcc -O2 -Og.
* src/pdumper.c (PDUMPER_MAX_OBJECT_SIZE): New macro.
(field_relpos): Use PDUMPER_MAX_OBJECT_SIZE, and comment on
why we require that relpos be not too large.
* src/pdumper.c (struct dump_context): New member
old_process_environment.
(Fdump_emacs_portable): Record the original value of
process-environment.
(dump_unwind_cleanup): Restore the original values of
process-environment and post-gc-hook.
Initialize these objects statically, and make them constants.
This is a bit safer and more efficient.
* src/data.c (XBOOLFWD, XKBOARD_OBJFWD, XFIXNUMFWD, XOBJFWD):
* src/lisp.h (XBUFFER_OBJFWD):
Return a pointer-to-const instead of an unrestricted pointer.
(lispfwd): fwdptr is now a pointer-to-const instead of an
unrestricted pointer. All uses changed.
(SET_SYMBOL_FWD): Accept pointer-to-const instead of an
unrestricted pointer.
(DEFVAR_LISP, DEFVAR_LISP_NOPRO, DEFVAR_BOOL, DEFVAR_INT)
(DEFVAR_KBOARD): Initialize static structures statically
instead of dynamically, and make them const.
* src/lread.c (defvar_int, defvar_bool, defvar_lisp_nopro)
(defvar_lisp, defvar_kboard): Accept pointer-to-const instead
of an unrestricted pointer; it’s now the caller’s
responsibility to initialize the pointed-to storage. No need
for a separate address argument any more. All callers
changed.
It's not portable to cast (e.g.) struct Lisp_Objfwd * to union
Lisp_Fwd * and then back again, because the compiler can then assume
that the pointer is aligned for union Lisp_Fwd * when accessing
the struct Lisp_Objfwd * components, and this assumption might
be incorrect becase we don't force that alignment.
* src/lisp.h (lispfwd): New type, replacing ...
(union Lisp_Fwd): ... this type, which was removed.
All uses changed.
(SET_SYMBOL_FWD): 2nd arg is now void *, not lispfwd.
All uses changed (casts no longer needed; they were
not portable anyway).
* lisp/emacs-lisp/bytecomp.el (byte-compile-setq-default): Delete.
(byte-compile-set-default): Inline the part that it used.
* lisp/emacs-lisp/edebug.el (setq-default): Remove the debug spec.
* src/data.c (Fsetq_default): Delete.
(syms_of_data): Don't register.
* src/search.c (looking_at_1, search_buffer_re): Unfreeze the regexp
buffer before signalling a matcher overflow, since the error
processing may require quite some regexp use as well (Bug#34910).
* src/search.c (compile_pattern): Don't give up if the last regexp
cache entry is busy. Instead, use the last (least recently used)
non-busy entry, and only signal a reentrancy error if there is no free
entry at all (Bug#34910).
Note: portions of this change were mistakenly pushed as part
of an unrelated commit a35a1f6a9.
* src/buffer.c (Fget_buffer_create): Set inhibit_buffer_hooks
non-zero for temporary buffers created by coding.c. Don't run
buffer-list-update-hook for such buffers.
(Frename_buffer, Fkill_buffer, record_buffer)
(Fbury_buffer_internal): Don't run hooks for buffers whose
inhibit_buffer_hooks flag is set.
* src/buffer.h (struct buffer): New member
inhibit_buffer_hooks.
* src/pdumper.c (dump_buffer): Dump the new field. Update the
hash value in HASH_buffer_XXX.
* src/coding.c (make_conversion_work_buffer): Function deleted;
code moved to code_conversion_save.
(code_conversion_save): Insert code from
make_conversion_work_buffer, but arrange for unwind-protecting
the current buffer before switching to the work buffer. This
avoids leaving reused_workbuf_in_use set if user presses C-g
during encoding/decoding.
(Vcode_conversion_workbuf_name): Now external variable.
* src/coding.h (Vcode_conversion_reused_workbuf): Declare.
* src/w32term.c (x_draw_image_foreground): Fix detection of
scaled images for sliced images. Scale the original width of
a slice and its coordinates of origin as well.
* src/.gdbinit: Simplify by removing dependency on globals
implementation. This is useful for a future performance
improvement that I have in mind.
* src/alloc.c (enum defined_HAVE_X_WINDOWS, defined_HAVE_X_WINDOWS):
New enum and constant.
(gdb_make_enums_visible) [__GNUC__]: Use it, to make
defined_HAVE_X_WINDOWS visible to GDB.
* src/regex-emacs.c (re_match_2_internal):
No need to allocate one extra trailing search register;
Emacs does not use it. Avoid quadratic behavior on
reallocation.
* src/pdumper.c (dump_discard_mem)
[VM_SUPPORTED == VM_MS_WINDOWS]: Don't pass NULL pointer as
last argument to VirtualProtect. Reported by Martin Rudalics
<rudalics@gmx.at>.
* src/xterm.c (x_cr_draw_image): Add arguments image_width and
image_height and support scaling. All callers changed.
* src/image.c (Fimage_scaling_p): Return t when USE_CAIRO.
(x_set_image_size) [USE_CAIRO]: Record the scaled dimensions
in the image struct.
* src/dispextern.h (HAVE_NATIVE_SCALING): Define when
USE_CAIRO as well.
* etc/NEWS: Update the announcement of native image scaling.
* src/gtkutil.c (xg_get_image_for_pixmap) [USE_CAIRO]: Use cairo image
surface for GtkImage source.
(xg_tool_item_stale_p, update_frame_tool_bar) [USE_CAIRO]: Use cairo
image surface instead of pixmap for data associated with tool bar item.
* src/dispextern.h (struct image) [USE_CAIRO]: Remove member cr_data2.
* src/image.c (set_cairo_image_surface): New function split from
original create_cairo_image_surface. Call cairo_surface_mark_dirty.
(create_cairo_image_surface): Just create image surface and return it.
(x_clear_image): Don't free cr_data2.
(xpm_load, pbm_load, png_load_body, jpeg_load_body, tiff_load)
(gif_load, imagemagick_load_image, svg_load_image) [USE_CAIRO]: Use
new create_cairo_image_surface and cairo_image_surface_get_data
instead of xmalloc. Use set_cairo_image_surface instead of old
create_cairo_image_surface.
(pbm_load) [USE_CAIRO]: Call cairo_surface_destroy for surface instead
of xfree for data.
(gif_load) [USE_CAIRO]: Multiply y-coordinate value by width instead
of subimg_width.
* src/editfns.c: Don’t try to call glibc ngettext;
we’re not ready for that yet.
(Fngettext): Do not restrict integer arguments to fixnums.
Improve doc string a bit.
Problem found on 32-bit sparc, which has stricter alignment
checking than x86-64.
* src/alloc.c (pure_alloc): When TYPE is negative it now specifies
the negation of the required alignment of the result.
(make_pure_bignum): Specify bignum limb alignment.
Also, remove some duplicate comments related to thread.h.
* src/regex-emacs.h (struct re_registers):
* src/regex-emacs.c (SIGN_EXTEND_CHAR): Remove.
(TALLOC, RETALLOC): Remove. All uses replaced by usual
allocators, which check for integer overflow.
(extract_number): Redo without using ‘unsigned’.
(CHARSET_RANGE_TABLE_EXISTS_P): Clearly return a boolean.
(print_fastmap, print_partial_compiled_pattern, CHECK_INFINITE_LOOP)
(regex_compile, analyze_first, bcmp_translate, mutually_exclusive_p)
(re_match_2_internal):
Use bool for booleans.
(print_fastmap, regex_compile, execute_charset):
Prefer int to unsigned where either will do.
(print_double_string): Prefer ptrdiff_t to ssize_t, since the
latter can in theory be narrower than the former. Use fwrite
instead of repeated putchar.
(emacs_re_max_failures, fail_stack_type, compile_stack_type)
(re_wctype_parse, regex_compile, re_search, re_search_2)
(re_match_2, re_match_2_internal, re_compile_pattern):
Prefer ptrdiff_t to size_t where either will do.
(union fail_stack_elt, PUSH_FAILURE_REG, POP_FAILURE_REG_OR_COUNT):
Make the integer an intptr_t, not long.
(GET_BUFFER_SPACE, EXTEND_BUFFER, regex_compile):
Use xpalloc to simplify allocation.
(regex_compile): Check for integer overflow when calculating
register numbers.
* src/regex-emacs.c (re_set_registers, re_match_2_internal):
* src/regex-emacs.h (struct re_registers, struct re_pattern_buffer):
* src/search.c (Freplace_match):
Prefer ptrdiff_t to unsigned where either will do.
* src/regex-emacs.h (struct re_pattern_buffer):
Prefer bool_bf to unsigned where either will do.
