* bug#25883: 25.2; Frequent, but not reproducible, crash on startup
@ 2017-02-26 19:14 John Wiegley
2017-02-26 19:34 ` Eli Zaretskii
0 siblings, 1 reply; 6+ messages in thread
From: John Wiegley @ 2017-02-26 19:14 UTC (permalink / raw)
To: 25883
This only happens when I startup my "alternate" Emacs that I use exclusively
for ERC, and it only happens maybe once or twice a week (I startup that Emacs
once a day). But maybe this stack trace can help give some insight.
If ever this becomes reliably reproducible, I'll comment further. But what
might be causing it to happen?
0 libsystem_kernel.dylib 0x00007fffbfe71dd6 __pthread_kill + 10
1 libsystem_pthread.dylib 0x00007fffbff5d787 pthread_kill + 90
2 libsystem_c.dylib 0x00007fffbfd88497 raise + 26
3 org.gnu.Emacs 0x00000001000b3db8 terminate_due_to_signal + 136
4 org.gnu.Emacs 0x00000001000d0523 emacs_abort + 19
5 org.gnu.Emacs 0x000000010019f5aa ns_term_shutdown + 122
6 org.gnu.Emacs 0x00000001000b3f8f shut_down_emacs + 271
7 org.gnu.Emacs 0x00000001000b3d85 terminate_due_to_signal + 85
8 org.gnu.Emacs 0x00000001000d0523 emacs_abort + 19
9 org.gnu.Emacs 0x000000010009a335 bidi_pop_it + 117
10 org.gnu.Emacs 0x000000010004563b pop_it + 843
11 org.gnu.Emacs 0x000000010001c31f set_iterator_to_next + 63
12 org.gnu.Emacs 0x000000010002afcd display_line + 6461
13 org.gnu.Emacs 0x00000001000294c6 try_window + 422
14 org.gnu.Emacs 0x000000010004f357 redisplay_window + 11607
15 org.gnu.Emacs 0x000000010004c5f6 redisplay_window_0 + 38
16 org.gnu.Emacs 0x000000010012fecc internal_condition_case_1 + 76
17 org.gnu.Emacs 0x000000010004bc90 redisplay_windows + 160
18 org.gnu.Emacs 0x0000000100027921 redisplay_internal + 5297
19 org.gnu.Emacs 0x000000010002877e redisplay_preserve_echo_area + 30
20 org.gnu.Emacs 0x0000000100171430 wait_reading_process_output + 5088
21 org.gnu.Emacs 0x00000001000bcc63 read_char + 7875
22 org.gnu.Emacs 0x00000001000b8d91 read_key_sequence + 1921
23 org.gnu.Emacs 0x00000001000b83bd command_loop_1 + 4861
24 org.gnu.Emacs 0x000000010012fe66 internal_condition_case + 70
25 org.gnu.Emacs 0x00000001000c67e0 command_loop_2 + 48
26 org.gnu.Emacs 0x000000010012f9d6 internal_catch + 54
27 org.gnu.Emacs 0x00000001000b678e command_loop + 158
28 org.gnu.Emacs 0x00000001000b66a1 recursive_edit_1 + 113
29 org.gnu.Emacs 0x00000001000b68ca Frecursive_edit + 218
30 org.gnu.Emacs 0x00000001000b55d2 main + 5602
31 libdyld.dylib 0x00007fffbfd43255 start + 1
--
John Wiegley GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com 60E1 46C4 BD1A 7AC1 4BA2
^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#25883: 25.2; Frequent, but not reproducible, crash on startup
2017-02-26 19:14 bug#25883: 25.2; Frequent, but not reproducible, crash on startup John Wiegley
@ 2017-02-26 19:34 ` Eli Zaretskii
2017-02-26 19:45 ` John Wiegley
0 siblings, 1 reply; 6+ messages in thread
From: Eli Zaretskii @ 2017-02-26 19:34 UTC (permalink / raw)
To: John Wiegley; +Cc: 25883
> From: John Wiegley <jwiegley@gmail.com>
> Date: Sun, 26 Feb 2017 11:14:50 -0800
>
> This only happens when I startup my "alternate" Emacs that I use exclusively
> for ERC, and it only happens maybe once or twice a week (I startup that Emacs
> once a day). But maybe this stack trace can help give some insight.
>
> If ever this becomes reliably reproducible, I'll comment further. But what
> might be causing it to happen?
Some serious messup in the display code. The backtrace basically says
that the display engine became very confused regarding stuff that
comes from buffer and from various display/overlay strings or images.
I haven't seen aborts in bidi_pop_it since I don't remember how long.
I am not familiar enough with ERC support in Emacs -- is it possible
that Emacs tries to display some partial byte stream that includes
incomplete characters, i.e. the byte sequence for some character(s) is
incomplete?
^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#25883: 25.2; Frequent, but not reproducible, crash on startup
2017-02-26 19:34 ` Eli Zaretskii
@ 2017-02-26 19:45 ` John Wiegley
2017-02-26 20:07 ` Eli Zaretskii
0 siblings, 1 reply; 6+ messages in thread
From: John Wiegley @ 2017-02-26 19:45 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: 25883
>>>>> Eli Zaretskii <eliz@gnu.org> writes:
> I am not familiar enough with ERC support in Emacs -- is it possible that
> Emacs tries to display some partial byte stream that includes incomplete
> characters, i.e. the byte sequence for some character(s) is incomplete?
Not only possible, but likely. What happens is this:
1. Emacs starts up
2. ERC begins initializing
3. ERC logins in to ZNC
4. ZNC starts playing black all the messages that came in since I last
exited Emacs.
5. Emacs sometimes survives this, sometimes crashes.
So it's quite possible that a "past message" ZNC is playing back contains a
partial byte sequence.
--
John Wiegley GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com 60E1 46C4 BD1A 7AC1 4BA2
^ permalink raw reply [flat|nested] 6+ messages in thread* bug#25883: 25.2; Frequent, but not reproducible, crash on startup
2017-02-26 19:45 ` John Wiegley
@ 2017-02-26 20:07 ` Eli Zaretskii
2017-02-26 20:37 ` John Wiegley
0 siblings, 1 reply; 6+ messages in thread
From: Eli Zaretskii @ 2017-02-26 20:07 UTC (permalink / raw)
To: John Wiegley; +Cc: 25883
> From: John Wiegley <jwiegley@gmail.com>
> Cc: bug-gnu-emacs@gnu.org
> Date: Sun, 26 Feb 2017 11:45:45 -0800
>
> >>>>> Eli Zaretskii <eliz@gnu.org> writes:
>
> > I am not familiar enough with ERC support in Emacs -- is it possible that
> > Emacs tries to display some partial byte stream that includes incomplete
> > characters, i.e. the byte sequence for some character(s) is incomplete?
>
> Not only possible, but likely. What happens is this:
>
> 1. Emacs starts up
> 2. ERC begins initializing
> 3. ERC logins in to ZNC
> 4. ZNC starts playing black all the messages that came in since I last
> exited Emacs.
> 5. Emacs sometimes survives this, sometimes crashes.
>
> So it's quite possible that a "past message" ZNC is playing back contains a
> partial byte sequence.
It's important to have the text of the buffer at the point of crash.
I could tell you how to do that with GDB, but I don't know how with
other debuggers.
Does ERC display include overlays or display strings or images? Or
maybe (space . PROPS) display properties? Or any other kind of
non-buffer text?
^ permalink raw reply [flat|nested] 6+ messages in thread
* bug#25883: 25.2; Frequent, but not reproducible, crash on startup
2017-02-26 20:07 ` Eli Zaretskii
@ 2017-02-26 20:37 ` John Wiegley
2017-02-26 20:44 ` Eli Zaretskii
0 siblings, 1 reply; 6+ messages in thread
From: John Wiegley @ 2017-02-26 20:37 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: 25883
>>>>> Eli Zaretskii <eliz@gnu.org> writes:
> Does ERC display include overlays or display strings or images? Or maybe
> (space . PROPS) display properties? Or any other kind of non-buffer text?
It does interpret control codes to bold text, etc.
Once I'm in the debugger, how do I obtain the buffer text for you?
--
John Wiegley GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com 60E1 46C4 BD1A 7AC1 4BA2
^ permalink raw reply [flat|nested] 6+ messages in thread* bug#25883: 25.2; Frequent, but not reproducible, crash on startup
2017-02-26 20:37 ` John Wiegley
@ 2017-02-26 20:44 ` Eli Zaretskii
0 siblings, 0 replies; 6+ messages in thread
From: Eli Zaretskii @ 2017-02-26 20:44 UTC (permalink / raw)
To: John Wiegley; +Cc: 25883
> From: John Wiegley <jwiegley@gmail.com>
> Cc: bug-gnu-emacs@gnu.org
> Date: Sun, 26 Feb 2017 12:37:48 -0800
>
> Once I'm in the debugger, how do I obtain the buffer text for you?
At the debugger prompt, type "source etc/emacs-buffer.gdb" and use the
commands defined there. The commentary at the beginning of the file
is a good starting point.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2017-02-26 20:44 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-02-26 19:14 bug#25883: 25.2; Frequent, but not reproducible, crash on startup John Wiegley
2017-02-26 19:34 ` Eli Zaretskii
2017-02-26 19:45 ` John Wiegley
2017-02-26 20:07 ` Eli Zaretskii
2017-02-26 20:37 ` John Wiegley
2017-02-26 20:44 ` Eli Zaretskii
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).