bug#14889: 23.1;Emacs does not exits after log out

bug#14889: 23.1;Emacs does not exits after log out

[Petr Hracek]

Hi developer,

I have a pretty old version emacs-23.1-21.
On the system I run emacs and afterwards logout from the desktop.
But emacs does not exist but becom a child of init process.

Some traces:

[root@localhost ~]# gdb emacs 2984
...
(gdb) bt
#0  pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:162
#1  0x00000032ca00889f in _xcb_conn_wait (c=0x1082570, cond=<value optimized out>, vector=0x0, count=0x0) at xcb_conn.c:274
#2  0x00000032ca00a1ec in xcb_wait_for_reply (c=0x1082570, request=8134, e=0x7fff819bad70) at xcb_in.c:390
#3  0x00000032ca44d5a3 in _XReply (dpy=0x1038490, rep=0x7fff819badd0, extra=0, discard=1) at xcb_io.c:454
#4  0x00000032ca42fa54 in XInternAtom (dpy=0x1038490, name=0x32ca51e750 "XIM_SERVERS", onlyIfExists=<value optimized out>) at IntAtom.c:184
#5  0x00000032ca47197a in _XimFilterPropertyNotify (display=0x1038490, window=<value optimized out>, event=0x7fff819baea0, client_data=<value optimized out>) at imInsClbk.c:97
#6  0x00000032cf85fa31 in _gdk_events_queue (display=0x1089000) at gdkevents-x11.c:2292
#7  0x00000032cf85ffae in gdk_event_dispatch (source=<value optimized out>, callback=<value optimized out>, user_data=<value optimized out>) at gdkevents-x11.c:2366
#8  0x00000032c7c38f0e in g_main_dispatch (context=0xdc6340) at gmain.c:1960
#9  IA__g_main_context_dispatch (context=0xdc6340) at gmain.c:2513
#10 0x00000032c7c3c938 in g_main_context_iterate (context=0xdc6340, block=1, dispatch=1, self=<value optimized out>) at gmain.c:2591
#11 0x00000032c7c3ca3a in IA__g_main_context_iteration (context=0xdc6340, may_block=1) at gmain.c:2654
#12 0x00000032cfd4c1b1 in IA__gtk_main_iteration () at gtkmain.c:1306
#13 0x00000000004a301c in XTread_socket (terminal=0xe44890, expected=<value optimized out>, hold_quit=0x7fff819bb420) at xterm.c:7279
#14 0x00000000004e6a43 in read_avail_input (expected=1) at keyboard.c:7093
#15 0x00000000004e6b5a in handle_async_input () at keyboard.c:7335
#16 0x00000000004dc45d in fatal_error_signal (sig=1) at emacs.c:382
#17 <signal handler called>
#18 0x00000032c68df218 in __poll (fds=0x7fff819bbb80, nfds=1, timeout=-1) at ../sysdeps/unix/sysv/linux/poll.c:83
#19 0x00000032ca00885a in _xcb_conn_wait (c=0x1082570, cond=<value optimized out>, vector=0x0, count=0x0) at xcb_conn.c:306
#20 0x00000032ca00a1ec in xcb_wait_for_reply (c=0x1082570, request=8133, e=0x7fff819bbc90) at xcb_in.c:390
#21 0x00000032ca44d5a3 in _XReply (dpy=0x1038490, rep=0x7fff819bbcf0, extra=0, discard=1) at xcb_io.c:454
#22 0x00000032ca42fa54 in XInternAtom (dpy=0x1038490, name=0x32ca51e750 "XIM_SERVERS", onlyIfExists=<value optimized out>) at IntAtom.c:184
#23 0x00000032ca47197a in _XimFilterPropertyNotify (display=0x1038490, window=<value optimized out>, event=0x7fff819bbdc0, client_data=<value optimized out>) at imInsClbk.c:97
#24 0x00000032cf85fa31 in _gdk_events_queue (display=0x1089000) at gdkevents-x11.c:2292
#25 0x00000032cf85ffae in gdk_event_dispatch (source=<value optimized out>, callback=<value optimized out>, user_data=<value optimized out>) at gdkevents-x11.c:2366
#26 0x00000032c7c38f0e in g_main_dispatch (context=0xdc6340) at gmain.c:1960
#27 IA__g_main_context_dispatch (context=0xdc6340) at gmain.c:2513
#28 0x00000032c7c3c938 in g_main_context_iterate (context=0xdc6340, block=1, dispatch=1, self=<value optimized out>) at gmain.c:2591
#29 0x00000032c7c3ca3a in IA__g_main_context_iteration (context=0xdc6340, may_block=1) at gmain.c:2654
#30 0x00000032cfd4c1b1 in IA__gtk_main_iteration () at gtkmain.c:1306
#31 0x00000000004a301c in XTread_socket (terminal=0xe44890, expected=<value optimized out>, hold_quit=0x7fff819bc340) at xterm.c:7279
#32 0x00000000004e6a43 in read_avail_input (expected=1) at keyboard.c:7093
#33 0x00000000004e6b5a in handle_async_input () at keyboard.c:7335
#34 0x00000000004e6d45 in process_pending_signals () at keyboard.c:7351
#35 0x000000000058f9bd in wait_reading_process_output (time_limit=0, microsecs=0, read_kbd=<value optimized out>, do_display=1, wait_for_cell=11020689, wait_proc=0x0, just_wait_proc=0)
     at process.c:4566
#36 0x00000000004e9e8b in kbd_buffer_get_event (commandflag=1, nmaps=2, maps=0x7fff819bcba0, prev_event=11020689, used_mouse_menu=0x7fff819bccb4, end_time=0x0) at keyboard.c:4099
#37 read_char (commandflag=1, nmaps=2, maps=0x7fff819bcba0, prev_event=11020689, used_mouse_menu=0x7fff819bccb4, end_time=0x0) at keyboard.c:3059
#38 0x00000000004ebf95 in read_key_sequence (keybuf=<value optimized out>, bufsize=30, prompt=11020689, dont_downcase_last=0, can_return_switch_frame=1, fix_current_buffer=1) at keyboard.c:9460
#39 0x00000000004ee30d in command_loop_1 () at keyboard.c:1642
#40 0x000000000054fecf in internal_condition_case (bfun=0x4ee0f0 <command_loop_1>, handlers=11107841, hfun=0x4e5730 <cmd_error>) at eval.c:1512
#41 0x00000000004e4b6e in command_loop_2 () at keyboard.c:1359
#42 0x000000000054fffa in internal_catch (tag=140735367855032, func=0x4e4b50 <command_loop_2>, arg=11020689) at eval.c:1248
---Type <return> to continue, or q <return> to quit---
#43 0x00000000004e5571 in command_loop () at keyboard.c:1338
#44 0x00000000004e596c in recursive_edit_1 () at keyboard.c:953
#45 0x00000000004e5aa7 in Frecursive_edit () at keyboard.c:1015
#46 0x00000000004db2b7 in main (argc=<value optimized out>, argv=<value optimized out>) at emacs.c:1852

It seems emacs receive a signal, and in XTread_socket function, it stop at here:

  7274   while (gtk_events_pending ())
  7275     {
  7276       current_count = count;
  7277       current_hold_quit = hold_quit;
  7278
  7279       gtk_main_iteration (); <---- run here, and never return
  7280
  7281       count = current_count;
  7282       current_count = -1;
  7283       current_hold_quit = 0;
  7284
  7285       if (current_finish == X_EVENT_GOTO_OUT)
  7286         break;
  7287     }


I'm not sure whether it is a emacs problem or a gtk problem.


-- 
Best regards / S pozdravem
Petr Hracek

bug#14889: 23.1;Emacs does not exits after log out

[Paul Eggert]

Can you reproduce the problem with the latest stable Emacs version
(24.3), or better yet with the latest development version?
If so, please send a backtrace.  If not, then I wouldn't
worry about it too much, as this part of Emacs has mutated
ince 23.1 and if you can't reproduce the problem
then it's likely already been fixed.

bug#14889: 23.1;Emacs does not exits after log out

[Petr Hracek]

Hi Paul,

hmm, it seems that I can not reproduced that error with 24.3 version.
I have checked also that error in version 24.2 which is pretty old and 
all is working properly.

More or less I have checked really old versions and it seems that since
version 23.3 problem is not reproducible.

But question is if there is any way how to find out patch?

best regards
Petr

On 07/23/2013 06:51 AM, Paul Eggert wrote:
> Can you reproduce the problem with the latest stable Emacs version
> (24.3), or better yet with the latest development version?
> If so, please send a backtrace.  If not, then I wouldn't
> worry about it too much, as this part of Emacs has mutated
> ince 23.1 and if you can't reproduce the problem
> then it's likely already been fixed.


-- 
Best regards / S pozdravem
Petr Hracek

bug#14889: 23.1;Emacs does not exits after log out

[Paul Eggert]

On 07/23/2013 11:36 AM, Petr Hracek wrote:
> But question is if there is any way how to find out patch? 

I suppose a patch could be written, if someone would take
the time and trouble to figure out exactly what the problem
is and how it was fixed in later versions of Emacs.
It's low priority for me, though.

bug#14889: 23.1;Emacs does not exits after log out

[Glenn Morris]

Paul Eggert wrote:

> I suppose a patch could be written, if someone would take
> the time and trouble to figure out exactly what the problem
> is and how it was fixed in later versions of Emacs.
> It's low priority for me, though.

I think it's extremely low priority for everyone other than maintainer
of RHEL6's Emacs. :)
I wonder if RHEL would be better served by rebasing on a more recent
Emacs (I know that's unlikely to happen).

I'm afraid that nobody from the Emacs side is likely to spend much (or
any) time supporting such an old version. Unless someone happens to
remember the issue, or the fix is easy to find in the ChangeLogs, I
think you are out of luck.