From: Eli Zaretskii <eliz@gnu.org>
To: Visuwesh <visuweshm@gmail.com>
Cc: luangruo@yahoo.com, 56528@debbugs.gnu.org
Subject: bug#56528: 29.0.50; Emacs lucid segfaults when X dies
Date: Wed, 13 Jul 2022 20:11:41 +0300 [thread overview]
Message-ID: <83v8s1uecy.fsf@gnu.org> (raw)
In-Reply-To: <87mtddgkp9.fsf@gmail.com> (message from Visuwesh on Wed, 13 Jul 2022 19:48:26 +0530)
> From: Visuwesh <visuweshm@gmail.com>
> Cc: luangruo@yahoo.com, 56528@debbugs.gnu.org
> Date: Wed, 13 Jul 2022 19:48:26 +0530
>
> >> > Btw, what do we expect Emacs to do in this scenario? If the X server
> >> > is killed, Emacs cannot display anything, so why an abort is not TRT?
> >> > What practically interesting use case does this emulate?
> >>
> >> Emacs is running as a daemon, I expect it to survive the death of X.
> >
> > "Survive" how? by closing all the frames?
>
> Unfortunately, I do not start the daemon by M-x server-start RET so I do
> not know what the behaviour was here but read below.
Starting a daemon automatically starts the server.
> > IOW, you are saying that when this happens, Emacs should test whether
> > it has a running server, and if it does, unconditionally close all the
> > frames? Is that really useful?
>
> In the past, when I start Emacs daemon by `emacs --daemon' in my
> ~/.xsession file and eventually kill X, Emacs will not die. I can still
> access the Emacs session in the tty or a fresh Xorg session using
> emacsclient. But I seem to recall M-x server-start RET working
> identical to --daemon here.
That's probably just sheer luck. When you kill the X server, any code
in Emacs that tries to display something will crash and burn, because
there's generally no way for us to display anything in that case.
> >> Emacs dying on me because I killed X is an annoyance since sometimes I
> >> need to restart X.
> >
> > Why not close Emacs before that -- this way you get to keep all your
> > edits, instead of relying on error handling to succeed in doing that.
>
> That's not a solution, sorry. Just saving the buffers is not going to
> cut it, I would like to have my shell session, other processes stay
> alive.
Our solution to this is desktop.el. You can customize it to save and
restore more than it does by default. But expecting Emacs to survive
the killing of X is unreasonable.
> Hmm, trying it with --fg-daemon, sometimes Emacs survives, sometimes it
> dies. Backtrace follows,
>
> (gdb) run -Q --fg-daemon
> Starting program: /home/viz/lib/ports/emacs/src/emacs -Q --fg-daemon
> [Switching to thread 1 (process 7339)](running)
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
> [New Thread 0x7ffff15fe640 (LWP 7340)]
> [New Thread 0x7ffff0c6d640 (LWP 7356)]
> [New Thread 0x7fffebfff640 (LWP 7357)]
> [Detaching after vfork from child process 7393]
>
> Thread 1 "emacs" received signal SIGSEGV, Segmentation fault.
> message3_nolog (m=...) at xdisp.c:11770
> 11770 if (FRAME_TERMINAL (f)->frame_up_to_date_hook)
> (gdb) bt
> #0 message3_nolog (m=XIL(0x555556066254)) at xdisp.c:11770
> #1 0x00005555555f3449 in message3 (m=XIL(0x555556066254)) at xdisp.c:11698
> #2 0x0000555555848e28 in Fmessage (nargs=2, args=0x7ffff15ff250) at editfns.c:2881
Here's an excellent example of what I was trying to say: this says
that Emacs tried to show some message, and crashed because that
requires a valid frame with a terminal connection. What do you expect
Emacs to do here?
I think we should close this bug as wontfix. It's unreasonable to
expect a GUI program to stay in the air when its GUI infrastructure is
forcibly killed.
next prev parent reply other threads:[~2022-07-13 17:11 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-13 4:32 bug#56528: 29.0.50; Emacs lucid segfaults when X dies visuweshm
2022-07-13 10:45 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-13 11:05 ` Visuwesh
2022-07-13 12:17 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-13 13:06 ` Visuwesh
2022-07-13 13:13 ` Eli Zaretskii
2022-07-13 13:15 ` Eli Zaretskii
2022-07-13 13:23 ` Visuwesh
2022-07-13 13:49 ` Eli Zaretskii
2022-07-13 14:18 ` Visuwesh
2022-07-13 17:11 ` Eli Zaretskii [this message]
2022-07-13 17:18 ` Eli Zaretskii
2022-07-13 17:29 ` Visuwesh
2022-07-13 17:54 ` Eli Zaretskii
2022-07-13 17:20 ` Visuwesh
2022-07-13 17:50 ` Eli Zaretskii
2022-07-14 1:06 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 5:41 ` Eli Zaretskii
2022-07-14 6:27 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 7:18 ` Eli Zaretskii
2022-07-14 1:04 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 5:39 ` Eli Zaretskii
2022-07-14 6:26 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 7:18 ` Eli Zaretskii
2022-07-14 7:21 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-15 16:17 ` Andrés Ramírez
2022-07-16 3:17 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 0:39 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 5:36 ` Eli Zaretskii
2022-07-14 6:24 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 3:47 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 4:24 ` Visuwesh
2022-07-14 4:52 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-07-14 5:37 ` Visuwesh
2022-07-14 6:25 ` Po Lu via Bug reports for GNU Emacs, the Swiss army knife of text editors
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=83v8s1uecy.fsf@gnu.org \
--to=eliz@gnu.org \
--cc=56528@debbugs.gnu.org \
--cc=luangruo@yahoo.com \
--cc=visuweshm@gmail.com \
/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).