GNU bug report logs - #69287
30.0.50; Pasting text from KDE clipboard sometimes crashes Emacs

Previous Next

To reply to this bug, email your comments to 69287 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: David Ponce <da_vid <at> orange.fr>
To: bug-gnu-emacs <at> gnu.org
Subject: 30.0.50; Pasting text from KDE clipboard sometimes crashes Emacs
Date: Tue, 20 Feb 2024 15:59:08 +0100

[Message part 1 (text/plain, inline)]

Hello,

While working in Emacs, I often paste text from the 'KDE clipboard popup
menu at mouse position' into the *scratch* buffer, and sometimes this
crashes Emacs, but not systematically.

Since a few days I run Emacs under GDB, and I managed to get the
attached backtrace.

Please, eventually let me know how I can help to get more useful details
when another crash will happen.

Thanks.


In GNU Emacs 30.0.50 (build 1, x86_64-pc-linux-gnu, GTK+ Version
 3.24.41, cairo version 1.18.0) of 2024-02-20
Repository revision: a1cbc4d810bc1b525fa46b23249b414c1ad6b031
Repository branch: master
Windowing system distributor 'The X.Org Foundation', version 11.0.12014000
System Description: Fedora Linux 39 (KDE Plasma)

Configured using:
 'configure --with-x-toolkit=gtk3 --with-cairo-xcb
 --with-native-compilation=no
 PKG_CONFIG_PATH=/usr/local/lib/pkgconfig:/usr/lib/pkgconfig'

Configured features:
ACL CAIRO DBUS FREETYPE GIF GLIB GMP GNUTLS GPM GSETTINGS HARFBUZZ JPEG
JSON LCMS2 LIBOTF LIBSELINUX LIBSYSTEMD LIBXML2 M17N_FLT MODULES NOTIFY
INOTIFY PDUMPER PNG RSVG SECCOMP SOUND SQLITE3 THREADS TIFF
TOOLKIT_SCROLL_BARS TREE_SITTER WEBP X11 XDBE XIM XINPUT2 XPM GTK3 ZLIB

Important settings:
  value of $LC_TIME: fr_FR.utf8
  value of $LANG: fr_FR.UTF-8
  locale-coding-system: utf-8-unix

[emacs-crash-bt.txt (text/plain, attachment)]

Message #8 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: David Ponce <da_vid <at> orange.fr>
To: 69287 <at> debbugs.gnu.org
Subject: 30.0.50; Pasting text from KDE clipboard sometimes crashes Emacs
Date: Tue, 20 Feb 2024 17:22:06 +0100

[Message part 1 (text/plain, inline)]

Another backtrace I just got, which looks different, but still involves GC.

[emacs-crash-bt2.txt (text/plain, attachment)]

Message #11 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: Po Lu <luangruo <at> yahoo.com>
To: David Ponce <da_vid <at> orange.fr>
Cc: 69287 <at> debbugs.gnu.org
Subject: Re: bug#69287: 30.0.50; Pasting text from KDE clipboard sometimes
 crashes Emacs
Date: Wed, 21 Feb 2024 09:56:39 +0800

David Ponce <da_vid <at> orange.fr> writes:

> Hello,
>
> While working in Emacs, I often paste text from the 'KDE clipboard popup
> menu at mouse position' into the *scratch* buffer, and sometimes this
> crashes Emacs, but not systematically.
>
> Since a few days I run Emacs under GDB, and I managed to get the
> attached backtrace.
>
> Please, eventually let me know how I can help to get more useful details
> when another crash will happen.
>
> Thanks.

I suspect invalid Lisp objects are being recorded into one of several
lists that hold temporary data during selection conversion.  Please
build Emacs with --enable-checking and post a backtrace from there.

Message #14 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: David Ponce <da_vid <at> orange.fr>
To: Po Lu <luangruo <at> yahoo.com>
Cc: 69287 <at> debbugs.gnu.org
Subject: Re: bug#69287: 30.0.50; Pasting text from KDE clipboard sometimes
 crashes Emacs
Date: Wed, 21 Feb 2024 10:58:47 +0100

[Message part 1 (text/plain, inline)]

On 21/02/2024 02:56, Po Lu wrote:
> David Ponce <da_vid <at> orange.fr> writes:
> 
>> Hello,
>>
>> While working in Emacs, I often paste text from the 'KDE clipboard popup
>> menu at mouse position' into the *scratch* buffer, and sometimes this
>> crashes Emacs, but not systematically.
>>
>> Since a few days I run Emacs under GDB, and I managed to get the
>> attached backtrace.
>>
>> Please, eventually let me know how I can help to get more useful details
>> when another crash will happen.
>>
>> Thanks.
> 
> I suspect invalid Lisp objects are being recorded into one of several
> lists that hold temporary data during selection conversion.  Please
> build Emacs with --enable-checking and post a backtrace from there.

Please, find attached the requested backtrace got from Emacs rebuilt with
--enable-checking.
Now Emacs consistently crashes after I pasted some Elisp code in a buffer,
as soon as I start to modify the text.

[emacs-crash-bt3.txt (text/plain, attachment)]

Message #17 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: Po Lu <luangruo <at> yahoo.com>
To: David Ponce <da_vid <at> orange.fr>
Cc: 69287 <at> debbugs.gnu.org
Subject: Re: bug#69287: 30.0.50; Pasting text from KDE clipboard sometimes
 crashes Emacs
Date: Wed, 21 Feb 2024 19:34:47 +0800

David Ponce <da_vid <at> orange.fr> writes:

> Please, find attached the requested backtrace got from Emacs rebuilt with
> --enable-checking.
> Now Emacs consistently crashes after I pasted some Elisp code in a buffer,
> as soon as I start to modify the text.

This is an unrelated bug, I think.

Message #20 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Po Lu <luangruo <at> yahoo.com>
Cc: da_vid <at> orange.fr, 69287 <at> debbugs.gnu.org
Subject: Re: bug#69287: 30.0.50;
 Pasting text from KDE clipboard sometimes crashes Emacs
Date: Wed, 21 Feb 2024 14:34:08 +0200

> Cc: 69287 <at> debbugs.gnu.org
> Date: Wed, 21 Feb 2024 09:56:39 +0800
> From:  Po Lu via "Bug reports for GNU Emacs,
>  the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org>
> 
> David Ponce <da_vid <at> orange.fr> writes:
> 
> > Hello,
> >
> > While working in Emacs, I often paste text from the 'KDE clipboard popup
> > menu at mouse position' into the *scratch* buffer, and sometimes this
> > crashes Emacs, but not systematically.
> >
> > Since a few days I run Emacs under GDB, and I managed to get the
> > attached backtrace.
> >
> > Please, eventually let me know how I can help to get more useful details
> > when another crash will happen.
> >
> > Thanks.
> 
> I suspect invalid Lisp objects are being recorded into one of several
> lists that hold temporary data during selection conversion.  Please
> build Emacs with --enable-checking and post a backtrace from there.

But the second backtrace posted by David is unrelated to selections.
It is still in GC, but that GC was invoked from funcall, not from
selection-related code.

So it sounds like some invalid object is indeed created, but not
necessarily during selection conversion.  Unfortunately, debugging GC
crashes is not easy; see etc/DEBUG for some guidance.

Message #23 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: David Ponce <da_vid <at> orange.fr>
Cc: luangruo <at> yahoo.com, 69287 <at> debbugs.gnu.org
Subject: Re: bug#69287: 30.0.50;
 Pasting text from KDE clipboard sometimes crashes Emacs
Date: Wed, 21 Feb 2024 17:31:13 +0200

> Cc: 69287 <at> debbugs.gnu.org
> Date: Wed, 21 Feb 2024 10:58:47 +0100
> From: David Ponce <da_vid <at> orange.fr>
> 
> xdisp.c:21778: Emacs fatal error: assertion failed: w->window_end_valid
> 
> Thread 1 "emacs" hit Breakpoint 1, terminate_due_to_signal (sig=sig <at> entry=6, backtrace_limit=backtrace_limit <at> entry=2147483647)
>     at emacs.c:442
> 442       signal (sig, SIG_DFL);
> (gdb) bt
> #0  terminate_due_to_signal (sig=sig <at> entry=6, backtrace_limit=backtrace_limit <at> entry=2147483647) at emacs.c:442
> #1  0x0000000000436698 in die
>     (msg=msg <at> entry=0x717fce "w->window_end_valid", file=file <at> entry=0x717900 "xdisp.c", line=line <at> entry=21778) at alloc.c:8061
> #2  0x0000000000427006 in find_first_unchanged_at_end_row (delta_bytes=<synthetic pointer>, delta=<synthetic pointer>, w=0x1065950)
>     at xdisp.c:21778
> #3  try_window_id (w=w <at> entry=0x1065950) at xdisp.c:22342
> #4  0x00000000004c2dad in redisplay_window (window=<optimized out>, just_this_one_p=just_this_one_p <at> entry=true) at xdisp.c:20433
> #5  0x00000000004c563e in redisplay_window_1 (window=window <at> entry=XIL(0x1065955)) at xdisp.c:18019

Do you have some optional feature enabled that resizes the mini-window
at random points in time?  Like some optional completion package that
sinerts a lot of stuff into the mini-window and thus causes it to
resize?

IOW, we need a reproducible recipe for debugging this assertion
violation.  If you can trigger this from "emacs -Q", it would be even
better.

However, please note that this is assertion violation has nothing
apparent to do with the previous crashes, which were inside GC.

Message #26 received at 69287 <at> debbugs.gnu.org (full text, mbox):

From: David Ponce <da_vid <at> orange.fr>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: luangruo <at> yahoo.com, 69287 <at> debbugs.gnu.org
Subject: Re: bug#69287: 30.0.50; Pasting text from KDE clipboard sometimes
 crashes Emacs
Date: Thu, 22 Feb 2024 10:14:01 +0100

On 21/02/2024 16:31, Eli Zaretskii wrote:
>> Cc: 69287 <at> debbugs.gnu.org
>> Date: Wed, 21 Feb 2024 10:58:47 +0100
>> From: David Ponce <da_vid <at> orange.fr>
>>
>> xdisp.c:21778: Emacs fatal error: assertion failed: w->window_end_valid
>>
>> Thread 1 "emacs" hit Breakpoint 1, terminate_due_to_signal (sig=sig <at> entry=6, backtrace_limit=backtrace_limit <at> entry=2147483647)
>>      at emacs.c:442
>> 442       signal (sig, SIG_DFL);
>> (gdb) bt
>> #0  terminate_due_to_signal (sig=sig <at> entry=6, backtrace_limit=backtrace_limit <at> entry=2147483647) at emacs.c:442
>> #1  0x0000000000436698 in die
>>      (msg=msg <at> entry=0x717fce "w->window_end_valid", file=file <at> entry=0x717900 "xdisp.c", line=line <at> entry=21778) at alloc.c:8061
>> #2  0x0000000000427006 in find_first_unchanged_at_end_row (delta_bytes=<synthetic pointer>, delta=<synthetic pointer>, w=0x1065950)
>>      at xdisp.c:21778
>> #3  try_window_id (w=w <at> entry=0x1065950) at xdisp.c:22342
>> #4  0x00000000004c2dad in redisplay_window (window=<optimized out>, just_this_one_p=just_this_one_p <at> entry=true) at xdisp.c:20433
>> #5  0x00000000004c563e in redisplay_window_1 (window=window <at> entry=XIL(0x1065955)) at xdisp.c:18019
> 
> Do you have some optional feature enabled that resizes the mini-window
> at random points in time?  Like some optional completion package that
> sinerts a lot of stuff into the mini-window and thus causes it to
> resize?

No.  However I use my own library that provides tabs in the tab-line
(kind of alternative implementation of tab-line.el), and I noticed that
the "assertion failed: w->window_end_valid" only occurs when the tab-line
is used (globally set).  My tab-line extensively uses text properties:
display (images, space), help-echo and keymap, plus some other specific
properties.

> 
> IOW, we need a reproducible recipe for debugging this assertion
> violation.  If you can trigger this from "emacs -Q", it would be even
> better.

I am trying to get a reproducible recipe as simple as possible from
"emacs -Q".  I will post news here, if I manage to get something
interesting.
 
> 
> However, please note that this is assertion violation has nothing
> apparent to do with the previous crashes, which were inside GC.

I agree.  Unfortunately this display issue prevent me to go further
with the previous crash in GC.

Thanks

Previous Next

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