Package: emacs;
Reported by: Nicolas Sarlin <nico.sarlin <at> gmail.com>
Date: Wed, 29 Jan 2025 11:16:02 UTC
Severity: normal
Done: Eli Zaretskii <eliz <at> gnu.org>
Bug is archived. No further changes may be made.
Message #14 received at 75922 <at> debbugs.gnu.org (full text, mbox):
From: Nicolas Sarlin <nico.sarlin <at> gmail.com> To: Pip Cet <pipcet <at> protonmail.com> Cc: 75922 <at> debbugs.gnu.org Subject: Re: bug#75922: CPU hogs with pgtk Date: Wed, 29 Jan 2025 16:28:01 +0100
Thanks for your answer!
On Wed, 29 Jan 2025 at 13:41, Pip Cet <pipcet <at> protonmail.com> wrote:
>
> "Nicolas Sarlin" <nico.sarlin <at> gmail.com> writes:
>
> Thanks for the report! Debug suggestions below, but if you cannot or
> would prefer not to any or all of that, that's perfectly okay, too!
>
> > I'm using lsp-mode with corfu on the emacs-30 branch, and I get extremely
> > frequent CPU hogs (CPU runs at 100% for a few seconds) while typing
> > code.
>
> So this happens on the pgtk branch, but behavior on the master branch is
> not noticeably slower than you'd expect?
This happens on the emacs-30 branch with configure=--with-pgtk
> > I report this as an emacs bug because it seems to only occurs when
> > emacs is compiled with pgtk.
>
> Are you running this on a Wayland setup, or on X (ignoring the popup)?
I am running this on Wayland
>
>
> > If I rebuild emacs with "make bootstrap configure=default" the bug
> > disapears.
>
> Entirely?
Yes, I see no slowdown at all without pgtk.
>
>
> > Here is a profiler record:
> > 11152 91% - corfu--post-command
> > 11152 91% - corfu--exhibit
> > 11043 90% - corfu--update
> > 10891 88% - corfu--recompute
> > 10891 88% - corfu--filter-completions
> > 10891 88% - completion-all-completions
> > 10891 88% - completion--nth-completion
> > 10891 88% - seq-some
> > 10891 88% - seq-do
> > 10891 88% - mapc
> > 10891 88% - #<byte-code-function AC1>
> > 10891 88% - #<byte-code-function AD0>
> > 10891 88% - lsp-completion-passthrough-all-completions
> > 10891 88% - #<byte-code-function 4B5>
> > 10891 88% - #<byte-code-function 4DE>
> > 10888 88% - lsp-request-while-no-input
> > 10888 88% - sit-for
> > 82 0% + redisplay_internal (C function)
> > 18 0% + #<byte-code-function 31C>
> > 3 0% + lsp--text-document-position-params
> > 152 1% + redisplay
> > 89 0% + posn-at-point
> > 20 0% + corfu--candidates-popup
> > 674 5% + command-execute
> > 269 2% + redisplay_internal (C function)
> > 74 0% + #<byte-code-function 31C>
> > 59 0% + timer-event-handler
> > 25 0% + ...
> >
> > Please tell me if you need more information,
>
> While a Lisp profile is always a good thing to have, I suspect a C
> profile is more useful in this case. Can you run
>
> perf record -e instructions,cycles ./src/emacs -Q
>
> then find a problematic workflow and wait for the problem to occur
> (ideally, this will happen so often as to dominate the CPU trace), then
> quit emacs in the ordinary fashion. After that
>
> perf report
>
> will produce a basic output C profile. It's both very long and somewhat
> hard to read, but I think posting it to a bug number email should be
> okay. If you do this, please save these files for this specific run
> somewhere:
>
> ./src/emacs
> ./src/emacs.pdmp
> ./perf.data
> any coredumps you might have had
>
> This will allow us to investigate the situation further if the function
> names aren't enough to find the culprits.
Thanks, here is a partial output of perf report, tell me if you need
more (events > 1%):
====
Samples: 1K of event 'cpu_atom/instructions/u', Event count (approx.):
1304155293
10,41% emacs emacs [.]
find_cache_boundary
◆
5,90% emacs emacs [.]
lookup_char_property
▒
4,71% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4c3
▒
3,33% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4d7
▒
3,01% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733a2
▒
2,84% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733d5
▒
2,61% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4df
▒
2,57% emacs emacs [.] find_newline
▒
2,20% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733b8
▒
1,81% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733af
▒
1,75% emacs emacs [.] read_char
▒
1,72% emacs emacs [.]
itree_iter_next_in_subtree
▒
1,65% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733e1
▒
1,62% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4cf
▒
1,59% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733cf
▒
1,56% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733c5
▒
1,56% emacs emacs [.]
re_match_2_internal
▒
1,47% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733aa
▒
1,42% emacs emacs [.] find_interval
▒
1,34% emacs [vdso] [.]
__vdso_clock_gettime
▒
1,21% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733b6
▒
1,19% emacs emacs [.] Fassq
▒
1,18% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733e4
▒
1,14% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733d9
▒
1,10% emacs emacs [.]
gui_produce_glyphs
▒
1,05% emacs emacs [.]
bidi_resolve_explicit
====
Samples: 133K of event 'cpu_core/instructions/u', Event count
(approx.): 186403198903
21,55% emacs [vdso] [.]
__vdso_clock_gettime
20,07% emacs [unknown] [k] 0xffffffff9e200080
11,55% emacs emacs [.] read_char
4,31% emacs emacs [.] restore_getcjmp
2,87% emacs libpixman-1.so.0.44.2 [.] 0x00000000000733a2
2,47% emacs emacs [.]
quit_throw_to_read_char
2,16% emacs emacs [.] some_mouse_moved
1,94% emacs libc.so.6 [.] siglongjmp
1,89% emacs libc.so.6 [.] pthread_sigmask
1,75% emacs emacs [.] unbind_to
1,49% emacs emacs [.] EQ
1,33% emacs libc.so.6 [.] 0x00000000000925c6
1,32% emacs emacs [.] kbd_on_hold_p
1,25% emacs libc.so.6 [.] 0x00000000000925a5
1,19% emacs emacs [.] _longjmp <at> plt
1,18% emacs libc.so.6 [.] 0x0000000000092616
1,11% emacs emacs [.]
record_unwind_protect_ptr
1,10% emacs libpixman-1.so.0.44.2 [.] 0x00000000000733b8
1,08% emacs libc.so.6 [.] 0x000000000003cfaf
1,00% emacs emacs [.] pthread_sigmask <at> plt
====
Samples: 1K of event 'cpu_atom/cycles/u', Event count (approx.): 1083537627
22,42% emacs libpixman-1.so.0.44.2 [.] 0x000000000006f4c3
◆
13,33% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4d7
▒
9,30% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4df
▒
4,77% emacs emacs [.]
find_cache_boundary
▒
3,84% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4cf
▒
2,95% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733a2
▒
2,34% emacs emacs [.]
lookup_char_property
▒
1,51% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4e3
▒
1,28% emacs libpixman-1.so.0.44.2 [.]
0x00000000000733b8
▒
1,23% emacs libpixman-1.so.0.44.2 [.]
0x000000000006f4d3
====
Samples: 140K of event 'cpu_core/cycles/u', Event count (approx.): 79660130854
16,86% emacs [vdso] [.]
__vdso_clock_gettime
13,66% emacs [unknown] [k] 0xffffffff9e200080
12,32% emacs emacs [.] read_char
6,88% emacs emacs [.] unbind_to
6,41% emacs libpixman-1.so.0.44.2 [.] 0x00000000000733a2
5,66% emacs libc.so.6 [.] pthread_sigmask
4,58% emacs emacs [.] restore_getcjmp
3,73% emacs libpixman-1.so.0.44.2 [.] 0x000000000006f4c3
2,64% emacs libpixman-1.so.0.44.2 [.] 0x000000000006f4db
2,22% emacs emacs [.]
quit_throw_to_read_char
1,90% emacs libpixman-1.so.0.44.2 [.] 0x000000000006f4e3
1,24% emacs emacs [.] some_mouse_moved
1,19% emacs libpixman-1.so.0.44.2 [.] 0x000000000006f4cb
1,08% emacs emacs [.]
record_unwind_protect_ptr
1,05% emacs emacs [.] current_timespec
>
> Please also attempt to find the precise version of "corfu", as well as
> any other packages that might seem like they're involved.
I am using corfu 1.7 from melpa-stable, and lsp-mode 20250129.601 from melpa
>
>
> Thanks!
> Pip
>
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.