occasional core dump

[Eric Hanchrow]

This bug report will be sent to the Free Software Foundation,
 not to your local site managers!!
Please write in English, because the Emacs maintainers do not have
translators to read other languages for them.

In GNU Emacs 20.7.2 (i386-debian-linux-gnu, X toolkit)
 of Mon Feb 19 2001 on offby1
configured using `configure  i386-debian-linux-gnu --prefix=/usr 
--sharedstatedir=/var/lib --libexecdir=/usr/lib --localstatedir=/var/lib 
--infodir=/usr/share/info --with-pop=yes --with-x=yes --with-x-toolkit=yes'

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug:

Emacs has dumped core occasionally in the last few weeks, after I've
started using it in a new way.  I assume the core dumps are related to
the way I'm using it, but am not sure.

Unfortunately I can't remember exactly what I was doing at the times
it dumped core.  I can describe, however, the "new way" that I've been
using it:

I have a computer at home, on which I run Emacs, but I use it from a
computer at work (running Windows NT 4, Cygwin 1.1.8, and XWinPro 5.3
{from http://labf.com}).  I leave Emacs running at home all the time,
and when I'm at work, I connect to my home machine via SSH, with this
command:

        DISPLAY=10.10.10.35 ssh -X -l offby1 offby1.atm01.sea.blarg.net

This gives me a "command prompt" window on my box at work (that's
similar to an xterm) that is running a shell on my home machine.  Then
I run this shell script on my home machine to get Emacs to create a
new frame on my work machine:

    #!/bin/sh

    xauth -f ~/.Xauthority merge $XAUTHORITY 

    gnudoit "(make-frame-on-display \"$DISPLAY\" '((width . 105) (height . 39) 
(name . \"home\") (font . 
\"-Adobe-Courier-Medium-R-Normal--24-240-75-75-M-150-ISO8859-1\")))"

Thus I have Emacs displaying a new frame onto an X server, with the
communication between Emacs and the X server "tunneled" over SSH.
(ssh is the only way to connect to my home machine from the outside.)

Now, the core dumps *seem* to come when the frame goes away -- perhaps
because I typed "C-x 5 0", or because the connection "drops", as it
often does (by which I mean: typing "ping gnu.org" at work hangs
instead of printing "64 bytes from 199.232.76.164").  However, as I
said, I'm not sure of the precise circumstances that cause the core
dump.  On the other hand, I'm reasonably confident that I can get
Emacs to dump core again, since it's happened a number of times in the
last few weeks.

Here's the backtrace that I got the last time it happened:

  No symbol table info available.
  #48 0x401434a5 in _XimCheckIfLocalProcessing () from 
/usr/X11R6/lib/libX11.so.6
  No symbol table info available.
  #49 0x401434a5 in _XimCheckIfLocalProcessing () from 
/usr/X11R6/lib/libX11.so.6
  No symbol table info available.
  #50 0x40143493 in _XimCheckIfLocalProcessing () from 
/usr/X11R6/lib/libX11.so.6
  No symbol table info available.
  #51 0x40143502 in _XimLocalIMFree () from /usr/X11R6/lib/libX11.so.6
  No symbol table info available.
  #52 0x4014362b in _XimLocalIMFree () from /usr/X11R6/lib/libX11.so.6
  No symbol table info available.
  #53 0x4011b0a8 in XCloseIM () from /usr/X11R6/lib/libX11.so.6
  No symbol table info available.
  #54 0x8086d00 in x_destroy_window (f=0x8c52938) at xterm.c:6185
          f = (struct frame *) 0x8c52938
          dpyinfo = (struct x_display_info *) 0x852cf30
  #55 0x8052a36 in Fdelete_frame (frame=1220880696, force=404708916)
      at frame.c:1200
          force = 16711680
          f = (struct frame *) 0x8c52938
          minibuffer_selected = 0
  #56 0x80ebf3d in Ffuncall (nargs=1, args=0xbffff624) at eval.c:2470
          nargs = -1073744508
          fun = -1073744508
          funcar = 41
          numargs = 404708916
          val = 41
          backtrace = {next = 0xbffff788, function = 0xbffff624, 
    args = 0xbffff628, nargs = 0, evalargs = 0 '\000', debug_on_exit = 0 '\000'}
          internal_args = (int *) 0xbffff584
          i = 41
  #57 0x80e88b9 in Fcall_interactively (function=404809148, 
      record_flag=404708916, keys=1210116888) at callint.c:790
          val = 41
          args = (int *) 0xbffff624
          visargs = (int *) 0xbffff614
          argstrings = (unsigned char **) 0xbffff604
          fun = 1
          funcar = 16711680
          specs = -1073744348
          teml = 1
          enable = 404708916
          speccount = 1
          next_event = 3
          prefix_arg = 404708916
          string = (unsigned char *) 0x8124042 ""
          tem = (unsigned char *) 0x8124042 ""
          varies = (int *) 0xbffff5f4
          i = 1
          j = 1
          count = 0
          foo = 16711680
          prompt1 = 
"\003\000\000\000\002\000\000\000\000\000\000\000\002\000\000\000\002\000\000\000\034÷ÿ¿±\030\017\b¼å
 \0304^\037\030\211\003\000\000\000\000\000\000¼å \0304^\037\030<÷ÿ¿gç\r\b¼å 
\030\003\000\000\000<÷ÿ¿\a\031\017\b¼å \030\224á 
\030\000\000\000\000¬þ^X\224öÿ¿\204öÿ¿"
          tem1 = 0xff0000 <Address 0xff0000 out of bounds>
          arg_from_tty = 0
          gcpro1 = {next = 0x0, var = 0xbffff6a4, nvars = 1}
          gcpro2 = {next = 0xbffff6cc, var = 0xbffff744, nvars = 1}
          gcpro3 = {next = 0xbffff6c0, var = 0xbffff624, nvars = 1}
          gcpro4 = {next = 0xbffff6b4, var = 0xbffff614, nvars = 1}
          key_count = 3
  #58 0x809ec45 in Fcommand_execute (cmd=404809148, record_flag=404708916, 
      keys=404708916, special=404708916) at keyboard.c:7984
          record_flag = 404708916
          final = 135485736
          tem = 41
          prefixarg = 404708916
          backtrace = {next = 0x0, function = 0x81f1378, args = 0xbffff7a4, 
    nargs = 1, evalargs = 0 '\000'}
  #59 0x80976b2 in command_loop_1 () at keyboard.c:1502
          cmd = 404809148
          lose = 404809148
          nonundocount = 0
          keybuf = {24, 53, 48, 134834258, 1489463380, -1073743848, 
-1073743796, 
    134834182, 0, 1, -1073742460, 0, 0, 1702127104, 909385842, 1684171552, 
    1667329312, 1763733362, 1634887028, 1075786817, 0, 1, -1073743888, 
    -1073744028, 0, 0, -1073743540, 135176745, -1073743716, 0}
          i = 404809148
          no_direct = 0
          prev_modiff = 2101
          prev_buffer = (struct buffer *) 0x86a5f08
          was_locked = 0
  #60 0x80ea27d in internal_condition_case (bfun=0x8096af8 <command_loop_1>, 
      handlers=404709420, hfun=0x8096788 <cmd_error>) at eval.c:1150
          val = 41
          c = {tag = 404708916, val = 404708916, next = 0xbffff98c, gcpro = 
0x0, 
    jmp = {{__jmpbuf = {0, 1, -1073742460, -1073743540, -1073743788, 
135176745}, 
        __mask_was_saved = 0, __saved_mask = {__val = {3221223772, 3221223524, 
            135176745, 0, 134525870, 1075685536, 1074425024, 
            0 <repeats 16 times>, 3221223780, 1073781908, 1073806887, 
            1076568096, 7, 1076549792, 3221223772, 1075790477, 135370864}}}}, 
    backlist = 0x0, handlerlist = 0x0, lisp_eval_depth = 0, pdlcount = 1, 
    poll_suppress_count = 1}
          h = {handler = 404709420, var = 404708916, chosen_clause = 404708964, 
    tag = 0xbffff88c, next = 0x0}
  #61 0x80969d8 in command_loop_2 () at keyboard.c:1093
          val = 16711680
  #62 0x80e9e1f in internal_catch (tag=404709396, 
      func=0x80969b4 <command_loop_2>, arg=404708916) at eval.c:925
          tag = 41
          c = {tag = 404709396, val = 404708916, next = 0x0, gcpro = 0x0, jmp = 
{
      {__jmpbuf = {0, 1, -1073742460, -1073743284, -1073743500, 135175692}, 
        __mask_was_saved = 0, __saved_mask = {__val = {3221224028, 3221223812, 
            135175692, 0, 1075978075, 136639176, 136639176, 1023, 1075977933, 
            1076546776, 136639176, 3221223596, 1075971308, 136639176, 0, 
            3221223596, 1075988890, 136639176, 1075957251, 0, 0, 1074247768, 
            135370864, 0, 0, 0, 0, 0, 0, 0, 0, 0}}}}, backlist = 0x0, 
    handlerlist = 0x0, lisp_eval_depth = 0, pdlcount = 1, 
    poll_suppress_count = 1}
  #63 0x8096982 in command_loop () at keyboard.c:1072
  No locals.
  #64 0x80965bd in recursive_edit_1 () at keyboard.c:821
          count = 1
          val = 0
  #65 0x80966b9 in Frecursive_edit () at keyboard.c:869
          count = 0
  #66 0x80957a5 in main (argc=1, argv=0xbffffd84, envp=0xbffffd8c)
      at emacs.c:1361
          argc = 1
          stack_bottom_variable = 0 '\000'
          skip_args = 0
          rlim = {rlim_cur = 8388608, rlim_max = 2147483647}

Before I started using Emacs this way (creating a new frame on my work
machine from an already-running Emacs on my home machine), I would
start a new Emacs when I connected to my home machine from work, and
then kill that Emacs when I was done for the day.  I stopped using
this method, though, because it would have two Emacsen running at
once, which struck me as wasteful and inconvenient (I like being able
to use the same Emacs from both home and work).  Perhaps
significantly, this older method -- have two separate Emacsen instead
of one with multiple frames -- never caused Emacs to dump core.

Please let me know if I can help further.  Thanks.

-- 
The best things in life are non-proprietary.
PGP Fingerprint: 3E7B A3F3 96CA 8958 ACC5  C8BD 6337 0041 C01C 5276