GNU bug report logs - #76964
31.0.50; completion-preview-mode uses a lot of CPU

Previous Next

To reply to this bug, email your comments to 76964 AT debbugs.gnu.org.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Geza Herman <geza.herman <at> gmail.com>
To: bug-gnu-emacs <at> gnu.org
Subject: 31.0.50; completion-preview-mode uses a lot of CPU
Date: Wed, 12 Mar 2025 00:08:05 +0100

When moving around a buffer with completion-preview-mode enabled, CPU 
usage becomes very high.

Profiling the problem reveals that with completion-preview-mode, 
redisplay_internal takes much more CPU.

The problem is in completion-preview--post-command, which calls 
(completion-preview-active-mode -1). For some reason, even though this 
should be a nop when just moving around a buffer, it forces emacs to do 
some more thorough redisplay. If I change this code to "(when 
completion-preview-active-mode (completion-preview-active-mode -1))", 
the problem goes away.

I think the problem is inside the define-minor-mode macro, because if I 
change completion-preview-active-mode to an empty minor mode definition:

(define-minor-mode completion-preview-active-mode
  "Mode for when the completion preview is shown."
  :interactive nil)

the problem still happens. Note: looking inside the define-minor-mode 
macro, I see a (force-mode-line-update) call. I'm not sure it's related, 
but it has a chance that it is.

In GNU Emacs 31.0.50 (build 1, x86_64-pc-linux-gnu, cairo version
 1.18.2) of 2025-03-11 built on okoska
Repository revision: 57f7d5f44a136c95958030b7920e5eca4b285309
Repository branch: HEAD
Windowing system distributor 'The X.Org Foundation', version 11.0.12101013
System Description: Debian GNU/Linux trixie/sid

Configured using:
 'configure --with-native-compilation --without-compress-install
 --without-gconf --without-gsettings --without-dbus --with-small-ja-dic
 --with-json --with-xinput2 --with-x-toolkit=no --with-tree-sitter
 --with-cairo --with-cairo-xcb --disable-silent-rules
 --disable-gc-mark-trace 'CFLAGS=-mtune=native -march=native -g3 -O3''

Configured features:
ACL CAIRO FREETYPE GIF GLIB GMP GNUTLS HARFBUZZ JPEG LCMS2 LIBOTF
LIBSELINUX LIBSYSTEMD LIBXML2 M17N_FLT MODULES NATIVE_COMP NOTIFY
INOTIFY OLDXMENU PDUMPER PNG RSVG SECCOMP SOUND SQLITE3 THREADS TIFF
TREE_SITTER WEBP X11 XDBE XIM XINERAMA XINPUT2 XPM XRANDR ZLIB

Important settings:
  value of $LC_ALL: C.UTF-8
  value of $LANG: en_US.UTF-8
  value of $XMODIFIERS: @im=none
  locale-coding-system: utf-8-unix

Major mode: Lisp Interaction

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  eldoc-mode: t
  show-paren-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  minibuffer-regexp-mode: t
  line-number-mode: t
  indent-tabs-mode: t
  transient-mark-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr compile comint ansi-osc ansi-color ring comp-run
bytecomp byte-compile comp-common rx emacsbug lisp-mnt message mailcap
yank-media puny dired dired-loaddefs rfc822 mml mml-sec password-cache
epa derived epg rfc6068 epg-config gnus-util text-property-search
time-date subr-x mm-decode mm-bodies mm-encode mail-parse rfc2231
mailabbrev gmm-utils mailheader cl-loaddefs cl-lib sendmail rfc2047
rfc2045 ietf-drums mm-util mail-prsvr mail-utils rmc iso-transl tooltip
cconv eldoc paren electric uniquify ediff-hook vc-hooks lisp-float-type
elisp-mode mwheel term/x-win x-win term/common-win x-dnd touch-screen
tool-bar dnd fontset image regexp-opt fringe tabulated-list replace
newcomment text-mode lisp-mode prog-mode register page tab-bar menu-bar
rfn-eshadow isearch easymenu timer select scroll-bar mouse jit-lock
font-lock syntax font-core term/tty-colors frame minibuffer nadvice seq
simple cl-generic indonesian philippine cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese eucjp-ms
cp51932 hebrew greek romanian slovak czech european ethiopic indian
cyrillic chinese composite emoji-zwj charscript charprop case-table
epa-hook jka-cmpr-hook help abbrev obarray oclosure cl-preloaded button
loaddefs theme-loaddefs faces cus-face macroexp files window
text-properties overlay sha1 md5 base64 format env code-pages mule
custom widget keymap hashtable-print-readable backquote threads inotify
lcms2 dynamic-setting font-render-setting cairo xinput2 x multi-tty
move-toolbar make-network-process tty-child-frames native-compile emacs)

Memory information:
((conses 16 85565 10951) (symbols 48 6794 0) (strings 32 23179 2179)
 (string-bytes 1 688051) (vectors 16 15115)
 (vector-slots 8 174583 11632) (floats 8 35 1) (intervals 56 332 0)
 (buffers 992 12))

Message #8 received at 76964 <at> debbugs.gnu.org (full text, mbox):

From: Eshel Yaron <me <at> eshelyaron.com>
To: Geza Herman <geza.herman <at> gmail.com>
Cc: 76964 <at> debbugs.gnu.org
Subject: Re: bug#76964: 31.0.50; completion-preview-mode uses a lot of CPU
Date: Wed, 12 Mar 2025 07:48:38 +0100

Hi Geza,

Geza Herman <geza.herman <at> gmail.com> writes:

> When moving around a buffer with completion-preview-mode enabled, CPU
> usage becomes very high.
>
> Profiling the problem reveals that with completion-preview-mode,
> redisplay_internal takes much more CPU.
>
> The problem is in completion-preview--post-command, which calls
> (completion-preview-active-mode -1). For some reason, even though this
> should be a nop when just moving around a buffer, it forces emacs to
> do some more thorough redisplay. If I change this code to "(when
> completion-preview-active-mode (completion-preview-active-mode -1))",
> the problem goes away.
>
> I think the problem is inside the define-minor-mode macro, because if
> I change completion-preview-active-mode to an empty minor mode
> definition:
>
> (define-minor-mode completion-preview-active-mode
>   "Mode for when the completion preview is shown."
>   :interactive nil)
>
> the problem still happens. Note: looking inside the define-minor-mode
> macro, I see a (force-mode-line-update) call. I'm not sure it's
> related, but it has a chance that it is.

Thanks for investigating and reporting.  I think that conditioning the
(completion-preview-active-mode -1) in completion-preview--post-command
as you suggest makes sense.  Would you like to provide a simple patch
with that change?


Best,

Eshel

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.