* etc/e/eterm-color.ti: Added termcaps for entering and leaving an
alternative sub-buffer.
* lisp/term.el (term-reset-size): Resize correctly in an alternative
sub-buffer is in use.
(term-clear-full-screen-programs): New user option.
(term-handle-ansi-escape):
(term-termcap-format): Handle termcaps for entering and leaving an
alternative sub-buffer.
(term-switch-to-alternate-sub-buffer): New (used to be commented out)
function to enter or leave an alternative sub-buffer.
bug#53485
Rebind [header-line mouse-1] from mouse-select-window to
Info-mouse-follow-link. It selects the window anyway,
but also allows repetitively clicking on the Next button
on the Info header line (bug#53170).
When MULTIPLE is nil and the arg DEFAULT is a list, keep its elements
in the "future history" of the minibuffer retrieved by `M-n M-n ...'.
(bug#53255)
* src/keyboard.c (read_char): Don't touch the echo area if
Vecho_keystrokes is nil.
* src/lread.c (read_filtered_event): Bind echo_keystrokes to nil
to avoid outputting prefixes we're not going to act on anyway
(bug#19718).
* etc/NEWS: Document the new user option.
* lisp/minibuffer.el (completions-sort): Add new user option.
(minibuffer-completion-help): Implement it for the Completions
buffer. (Bug#53362)
* lisp/emacs-lisp/ert.el (ert-test):
(ert-deftest): Store the file name.
(ert-test-location): New function.
(ert-run-tests-batch): Use it.
* lisp/emacs-lisp/find-func.el (find-ert-deftest-regexp): New
variable.
(find-function-regexp-alist): Add ert-deftest (bug#22471).
Based on code by Phillip Lord <phillip.lord@russet.org.uk>.
* doc/emacs/help.texi (Help Mode): Document it.
* lisp/help-mode.el (help-mode-map): Add an `I' keystroke.
(help-goto-lispref-info): New command.
* lisp/info-look.el (:mode): Add an entry for the lispref manual.
Since `maybe_quit` is called on many critical paths, inline a slightly
cheaper condition (that ignores Vinhibit_quit).
* src/eval.c (maybe_quit): Rename to `probably_quit`.
* src/lisp.h (maybe_quit): New simplified inline function.
The common case is just to increment `specpdl_ptr`; do that in-line,
but move the uncommon reallocation to a separate subroutine.
* src/eval.c (grow_specpdl): Now inline, most code moved...
(grow_specpdl_allocation): ...here.
* src/eval.c (funcall_general, Ffuncall): Delegate the actual work in
Ffuncall to funcall_general which does exactly this.
This slows down some less used function call paths by a small amount
but the code duplication was just silly.
* src/bytecode.c (exec_byte_code): Inline aref and aset for
vectors and records, since this is important for code that makes heavy
use of arrays and/or objects.
It was implemented but never generated, originally intended for
TCO in the pre-lexbind era (which was semantically dubious anyway).
Removing it speeds up the interpreter because there is no longer any
need for the outermost `count` variable unless checking is enabled.
* lisp/emacs-lisp/bytecomp.el:
* lisp/emacs-lisp/comp.el (comp-limplify-lap-inst):
* src/bytecode.c (BYTE_CODES, exec_byte_code):
Remove definition and implementation of unbind-all, freeing up the opcode
for other purposes.
* src/bytecode.c (exec_byte_code): Perform bytecode unwinding error
check only when building with debugging (NDEBUG not defined, checking
enabled, or BYTE_CODE_SAFE enabled). This improves speed in several
ways.
Since we pass no arguments to a non-lexbind bytecode function, we can
specify its arity as 0 instead of nil and save a test and branch.
* src/bytecode.c (Fbyte_code, exec_byte_code):
* src/eval.c (fetch_and_exec_byte_code, funcall_lambda):
* src/lisp.h:
Change the args_template parameter type to ptrdiff_t, since it is now
always a small integer, in exec_byte_code and
fetch_and_exec_byte_code, all callers adjusted.
Avoid making a copy (in the interpreter C stack frame) of the bytecode
string by making sure it won't be moved by the GC. This is done by
reallocating it to the heap normally only used for large strings,
which isn't compacted.
This requires that we retain an explicit reference to the bytecode
string object (`bytestr`) lest it be GCed away should all other
references vanish during execution. We allocate an extra stack slot
for that, as we already do for the constant vector object.
* src/alloc.c (allocate_string_data): Add `immovable` argument.
(resize_string_data, make_clear_multibyte_string): Use it.
(pin_string): New.
* src/pdumper.c (dump_string): Fix incorrect comment.
Update hash for Lisp_String (only comments changed, not contents).
* src/lread.c (read1):
* src/alloc.c (Fmake_byte_code, purecopy):
* src/bytecode.c (Fbyte_code): Pin bytecode on object creation.
(exec_byte_code): Don't copy bytecode. Retain `bytestr` explicitly.
* src/lisp.h (Lisp_String): Explain special size_byte values.
(string_immovable_p): New.
The function call overhead is nontrivial in comparison to the actual
code which makes this worthwhile.
* src/bytecode.c (exec_byte_code):
Inline code from Fsetcar and Fsetcdr.
Since numeric operations are mostly done on fixnums, this gives a
speed-up for common code.
* src/bytecode.c (exec_byte_code): Inline fixnum comparisons and
operations with fixnum results: =, >, <, <=, >=, -, +, -, *, /, %, max
and min.
Inline parts of the code for function calls to speed up the common
case of calling lexbound byte-code. By eliminating intermediate
functions, this also reduces C stack usage a little.
* src/bytecode.c (exec_byte_code): Inline parts of Ffuncall,
funcall_lambda and fetch_and_exec_byte_code in the Bcall opcode
handler.
* src/eval.c (backtrace_debug_on_exit): Inline and move to lisp.h.
(do_debug_on_call): Make global so that it can be called from
bytecode.c.
(funcall_general): New function, essentially the meat of Ffuncall.
* src/lisp.h (backtrace_debug_on_exit): Moved here from eval.c.
* src/xdisp.c (redisplay_tab_bar, redisplay_tool_bar): If we do
not display a tool or tab bar in these functions, still set
f->tool/tab_bar_redisplayed and f->tool/tab_bar_resized. This
avoids that when one of these are initially turned off, a later
activation resizes the frame despite of the fact that
'frame-inhibit-implied-resize' should prevent it (Bug#52986).
