This improves performance overall on my benchmark on x86-64,
since the interpreted program-counter resides in a machine
register rather than in RAM.
* etc/DEBUG, src/.gdbinit: Remove xbytecode GDB command, as there
is no longer a byte stack to decode.
* src/bytecode.c (struct byte_stack, byte_stack_list)
(relocate_byte_stack): Remove. All uses removed.
(FETCH): Simplify now that pc is now local (typically, in a
register) and no longer needs to be relocated.
(CHECK_RANGE): Remove. All uses now done inline, in a different way.
(BYTE_CODE_QUIT): Remove; now done by op_relative_branch.
(exec_byte_code): Allocate a copy of the function’s bytecode,
so that there is no problem if GC moves it.
* src/lisp.h (struct handler): Remove byte_stack member.
All uses removed.
* src/thread.c (unmark_threads): Remove. All uses removed.
* src/thread.h (struct thread_state): Remove m_byte_stack_list member.
All uses removed. m_stack_bottom is now the first non-Lisp field.
* src/bytecode.c (BYTE_MAINTAIN_TOP): Remove; no longer needed.
(struct byte_stack) [BYTE_MAINTAIN_TOP]:
Remove unused members ‘top’ and ‘bottom’.
(exec_byte_code): Nest inside { } to avoid GCC warning about
jumping over declaration when compiled with -DBYTE_CODE_SAFE.
See Bug#25120.
* src/lread.c (read_integer): Treat incomplete integer literals as errors.
* test/src/lread-tests.el (lread-empty-int-literal): New unit test for
incomplete integer literals.
* src/charset.c (load_charset_map):
* src/coding.c (decode_coding_object):
* src/frame.c (make_frame):
* src/window.c (Frecenter):
Mark locals with UNINIT to silence false alarms from
-Wmaybe-uninitialized.
* src/lisp.h (SYMBOL_ALIAS, SYMBOL_BLV, SYMBOL_FWD)
(SET_SYMBOL_ALIAS, SET_SYMBOL_BLV, SET_SYMBOL_FWD):
Check and assume that values are nonnull. This pacifies
-Wmaybe-uninitialized in Fmake_variable_buffer_local and
Fmake_local_variable.
* src/alloc.c (mark_object) [GC_CHECK_MARKED_OBJECTS]: Don't abort
for thread objects. They are marked via the all_threads list, and
therefore don't need to be inserted into the red-black tree, so
mem_find will never find them. Reported by Daniel Colascione
<dancol@dancol.org> in
http://lists.gnu.org/archive/html/emacs-devel/2016-12/msg00817.html.
* src/lread.c (load_path_default) [CANNOT_DUMP]:
Use build load-path if we seem to be running uninstalled. (Bug#24974)
I think this became an issue several years ago when we stopped
using EMACSLOADPATH in the Makefiles; however this change should
improve the CANNOT_DUMP uninstalled case in general.
* src/thread.h (struct thread_state): New member not_holding_lock.
(maybe_reacquire_global_lock): Add prototype.
* src/thread.c: Include syssignal.h.
(maybe_reacquire_global_lock): New function.
(really_call_select): Set the not_holding_lock member of the
thread state before releasing the lock, and rest it after
re-acquiring the lock when the select function returns. Block
SIGINT while doing this to make sure we are not interrupted on TTY
frames.
* src/sysdep.c (block_interrupt_signal, restore_signal_mask): New
functions.
* src/syssignal.h (block_interrupt_signal, restore_signal_mask):
Add prototypes.
* src/keyboard.c (read_char) [THREADS_ENABLED]: Call
maybe_reacquire_global_lock. (Bug#25178)
* src/unexw32.c (get_section_info): Make extra_bss_size be the
maximum of extra_bss_size and extra_bss_size_static. This avoids
computing the size of the output file smaller than it actually
needs to be, which then causes copy_executable_and_dump_data to
write beyond the requested size of the file mapping, thus relying
on the OS roundup to page boundary to save us from ourselves. See
http://lists.gnu.org/archive/html/emacs-devel/2016-12/msg00642.html
for the details.
* lib/stdio-impl.h: Revert the workaround fix of not including
errno.h for MinGW.
* src/xgselect.c (xg_select): Don't call Glib functions that use
'context' if we failed to acquire it. This means some other
thread owns the context, in which case both using the context and
calling block_input/unblock_input will step on that thread's toes
and eventually lead to crashes. (Bug#25172)
* src/data.c (Fset_default): Move code into new C level function,
`set_default_internal'.
(set_default_internal): New function, like `Fset_default' but also takes
additional bindflag parameter.
(set_internal): Only call `notify_variable_watchers' if bindflag is not
SET_INTERNAL_THREAD_SWITCH.
* src/eval.c (do_specbind, do_one_unbind): Add bindflag parameter,
passed on to set_internal and set_default_internal. Adjust callers.
(rebind_for_thread_switch, unbind_for_thread_switch): Pass
SET_INTERNAL_THREAD_SWITCH to do_specbind, do_one_unbind instead of
temporarily adjusting symbol's trapped_write field.
* src/eval.c (get_backtrace_starting_at, backtrace_frame_apply)
(Fmapbacktrace, Fbacktrace_frame_internal): New functions.
(get_backtrace_frame, Fbacktrace_debug): Use `get_backtrace_starting_at'.
* lisp/subr.el (backtrace--print-frame): New function.
(backtrace): Reimplement using `backtrace--print-frame' and `mapbacktrace'.
(backtrace-frame): Reimplement using `backtrace-frame--internal'.
* lisp/emacs-lisp/debug.el (debugger-setup-buffer): Pass a base to
`mapbacktrace' instead of searching for "(debug" in the output of
`backtrace'.
* test/lisp/subr-tests.el (subr-test-backtrace-simple-tests)
(subr-test-backtrace-integration-test): New tests.
* doc/lispref/debugging.texi (Internals of Debugger): Document
`mapbacktrace' and missing argument BASE of `backtrace-frame'.
Problem reported by Daniel Colascione in:
http://lists.gnu.org/archive/html/emacs-devel/2016-12/msg00515.html
* src/ftcrfont.c (ftcrfont_driver):
* src/ftfont.c (ftfont_driver):
* src/ftxfont.c (ftxfont_driver):
* src/macfont.m (macfont_driver):
* src/nsfont.m (nsfont_driver):
* src/xfont.c (xfont_driver):
* src/xftfont.c (xftfont_driver):
Use C99 syntax, not the old GNU C syntax.
* src/thread.c (post_acquire_global_lock): Don't raise the pending
signal if the thread's handlers were not yet set up, as that will
cause Emacs to exit with a fatal error. This can happen if a
thread is signaled as soon as make-thread returns, before the new
thread had an opportunity to acquire the global lock, set up the
handlers, and call the thread function.
* test/src/thread-tests.el (thread-signal-early): New test.
* src/thread.c (post_acquire_global_lock): Call
set_buffer_internal_2 instead of tricking set_buffer_internal_1
into resetting the current buffer even if it didn't change. This
avoids bug#25165, caused by failing to record the modified values
of point and mark, because current_buffer was set to NULL. Also,
don't bother re-setting the buffer if there was no thread switch,
as that just wastes cycles.
* src/buffer.c (set_buffer_internal_2): New function, with most of
the body of set_buffer_internal_1, but without the test for B
being identical to the current buffer.
(set_buffer_internal_1): Call set_buffer_internal_2 if B is not
identical to the current buffer.
* src/buffer.h (set_buffer_internal_2): Add prototype.
* test/src/thread-tests.el (thread-sticky-point): New test.
* src/thread.h (struct thread_state): Add members
m_waiting_for_input and m_input_available_clear_time.
(waiting_for_input, input_available_clear_time): New macros.
* src/keyboard.c (waiting_for_input, input_available_clear_time):
Remove; they are now macros that reference the current thread.
(Bug#25171)
* src/w32select.c: Don't include keyboard.h.
* test/src/thread-tests.el (thread-errors): New test.
* src/thread.c (mark_one_thread): Use NILP to compare with
m_saved_last_thing_searched, which is a Lisp object. Reported by
Andreas Politz <politza@hochschule-trier.de>.
* src/thread.h <int select_func>: Make the 5th and 6th arguments
be 'const'.
* src/process.c [WINDOWSNT]:
* src/w32proc.c: Make the 5th and 6th argument to sys_select be
'const'.
* src/eval.c (do_specbind): Don't require a "symbol" that is
actually a cons cell, in order to call set-default, as there are
no longer such bindings. This makes do_specbind work like the
pre-concurrency implementation in specbind for bindings of
forwarded symbols. Use specpdl_kind to access the type of the
binding.
(specpdl_kind): New function.
This is needed if ‘read’ is called soon after startup, before the
Unicode tables have been set up, and it reads a \N escape and
needs to look up a value the Unicode tables, a lookup that in turn
calls read1 recursively. Although this change doesn’t make ‘read’
fully reentrant, it’s good enough to handle this case.
* src/lread.c (read_buffer_size, read_buffer): Remove static vars.
(grow_read_buffer): Revamp to use locals, not statics, and to
record memory allocation un the specpdl. All callers changed.
(read1): Start with a stack-based buffer, and use the heap
only if the stack buffer is too small. Use unbind_to to
free any heap buffer allocated. Use bool for boolean.
Redo symbol loop so that only one call to grow_read_buffer
is needed.
(init_obarray): Remove no-longer-needed initialization.
