From: Tassilo Horn <thorn@fastmail.fm>
To: Eli Zaretskii <eliz@gnu.org>
Cc: mail@daniel-mendler.de, 48839@debbugs.gnu.org
Subject: bug#48839: 28.0.50; Emacs freezes and takes 100% CPU with C-h v l
Date: Sat, 05 Jun 2021 10:11:28 +0200 [thread overview]
Message-ID: <87wnr8ohk5.fsf@fastmail.fm> (raw)
In-Reply-To: <83bl8kltb8.fsf@gnu.org>
Eli Zaretskii <eliz@gnu.org> writes:
Hi Eli and Daniel,
>> Looking at the marginalia code, I can see no evil, and indeed I can
>> un-marginaliarize the sample to
>>
>> (let ((print-escape-newlines t)
>> (print-escape-control-characters t)
>> (print-escape-multibyte t))
>> (string-width (prin1-to-string load-history)))
>>
>> which freezes my emacs in the same way. Concretely, `prin1-to-string'
>> finishes and returns a 1.3 MB string and then `string-width' will run
>> indefinitely on that.
>
> string-width on the current master has a design bug, which causes it
> to be VERY slow on very long strings, if those strings don't include
> newlines or similar characters. I'm working on fixing that design
> bug, but meanwhile: why does marginalia need to compute the width of
> such very long strings? that sounds like a design bug in marginalia.
> The simplest fix is not to compute string-width of any string whose
> length is greater than, say, 300 characters. Does that help in this
> case?
Yes, with this patch I can make the issue go away with no observable
difference in the things marginalia displays.
--8<---------------cut here---------------start------------->8---
diff -u --label /home/horn/.emacs.d/elpa/marginalia-20210530.158/marginalia.el --label \#\<buffer\ marginalia.el\> /home/horn/.emacs.d/elpa/marginalia-20210530.158/marginalia.el /tmp/buffer-content-yqLXyF
--- /home/horn/.emacs.d/elpa/marginalia-20210530.158/marginalia.el
+++ #<buffer marginalia.el>
@@ -450,8 +450,12 @@
((marginalia--symbol-class sym) :face 'marginalia-type)
((let ((print-escape-newlines t)
(print-escape-control-characters t)
- (print-escape-multibyte t))
- (prin1-to-string (if (boundp sym) (symbol-value sym) 'unbound)))
+ (print-escape-multibyte t)
+ (str-val (prin1-to-string
+ (if (boundp sym)
+ (symbol-value sym)
+ 'unbound))))
+ (substring str-val 0 (min (length str-val) 300)))
:truncate (/ marginalia-truncate-width 3) :face 'marginalia-variable)
((documentation-property sym 'variable-documentation)
:truncate marginalia-truncate-width :face 'marginalia-documentation))))
Diff finished. Sat Jun 5 10:09:24 2021
--8<---------------cut here---------------end--------------->8---
Bye,
Tassilo
next prev parent reply other threads:[~2021-06-05 8:11 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-04 22:01 bug#48839: 28.0.50; Emacs freezes and takes 100% CPU with C-h v l Tassilo Horn
2021-06-05 6:27 ` Eli Zaretskii
2021-06-05 8:11 ` Tassilo Horn [this message]
2021-06-05 8:30 ` Tassilo Horn
2021-06-05 8:39 ` Eli Zaretskii
2021-06-05 11:18 ` Eli Zaretskii
2021-06-05 12:07 ` Tassilo Horn
2021-06-05 12:29 ` Eli Zaretskii
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://www.gnu.org/software/emacs/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87wnr8ohk5.fsf@fastmail.fm \
--to=thorn@fastmail.fm \
--cc=48839@debbugs.gnu.org \
--cc=eliz@gnu.org \
--cc=mail@daniel-mendler.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).