bug#17753: Cygwin emacs-X11 core dump

[markus . hoenicka]

Hi again,

today I witnessed the second coredump of Cygwin emacs-x11.exe (64
bit). The setup is the same as reported previously in this thread. The
backtraces have some striking similarities:

(gdb) info threads
  Id   Target Id         Frame
  9    Thread 0xd70      0x0004003300001f80 in ?? ()
  8    Thread 0xee8      0x9e84003300001f80 in ?? ()
  7    Thread 0x498      0x0005003300001f80 in ?? ()
  6    Thread 0x85c      0x0005003300001f80 in ?? ()
  5    Thread 0xe34      0x0000003300001f80 in ?? ()
  4    Thread 0xd9c      0x0000003300001f80 in ?? ()
  3    Thread 0x184      0x0000003300001f80 in ?? ()
  2    Thread 0x2d8      0x0000003300001f80 in ?? ()
* 1    Thread 0x2e8      0x0000003300001fa4 in ?? ()

(gdb) thread apply all bt

Thread 9 (Thread 0xd70):
#0  0x0004003300001f80 in ?? ()
#1  0x0000000000000000 in ?? ()

Thread 8 (Thread 0xee8):
#0  0x9e84003300001f80 in ?? ()
#1  0x000007fefd0c10dc in ?? ()
#2  0x0000000000000000 in ?? ()

Thread 7 (Thread 0x498):
#0  0x0005003300001f80 in ?? ()
#1  0x000007fefd0c10dc in ?? ()
#2  0x0000000000000000 in ?? ()

Thread 6 (Thread 0x85c):
#0  0x0005003300001f80 in ?? ()
#1  0x000007fefd0c10dc in ?? ()
#2  0x0000000000000000 in ?? ()

Thread 5 (Thread 0xe34):
#0  0x0000003300001f80 in ?? ()
#1  0x000007fefd0c1203 in ?? ()
#2  0x00000000033aaba8 in ?? ()
#3  0x0000000600042280 in ?? ()
#4  0x0000000000000003 in ?? ()
#5  0x00000001802de988 in ?? ()
#6  0xfffffffffffe7960 in ?? ()
#7  0x00000000033aab10 in ?? ()
#8  0x0000000000000048 in ?? ()
#9  0x0000000000000001 in ?? ()
#10 0x0000000000000000 in ?? ()

Thread 4 (Thread 0xd9c):
#0  0x0000003300001f80 in ?? ()
Cannot access memory at address 0xffffc458
(gdb) thread apply 3 bt

Thread 3 (Thread 0x184):
#0  0x0000003300001f80 in ?? ()
#1  0x000000007705b037 in ?? ()
#2  0x0000000000000002 in ?? ()
#3  0x0000000000000000 in ?? ()
(gdb) thread apply 2 bt

Thread 2 (Thread 0x2d8):
#0  0x0000003300001f80 in ?? ()
#1  0x000007fefd0c1a7a in ?? ()
#2  0x000000000042ce00 in ?? ()
#3  0x0000000100629773 in run_timers ()
    at /usr/src/debug/emacs-24.3.90-1/src/atimer.c:364
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
(gdb) thread apply 1 bt

Thread 1 (Thread 0x2e8):
#0  0x0000003300001fa4 in ?? ()
#1  0x0000000000427870 in ?? ()
#2  0x0000000000427218 in ?? ()
#3  0x00000000004271a0 in ?? ()
#4  0x0000000100a49832 in bss_sbrk_buffer ()
#5  0x0000000000363000 in ?? ()
#6  0x0000000077058f34 in ?? ()
#7  0x0000000000430000 in ?? ()
#8  0x000000010097a5c0 in ?? ()
#9  0xffffffff0000e118 in ?? ()
#10 0x00000000005e6e30 in ?? ()
#11 0x000000060230e8a1 in ?? ()
#12 0xffffffffffffffff in ?? ()
#13 0x0000000000000000 in ?? ()

Thread 1 looks a bit different to me this time, but the remaining
threads look like copies of the first crash.

This time the crash occurred while I was publishing my planner pages
to xhtml.

I'm aware that you won't be able to do much unless I provide a recipe
to reproduce the bug. This looks impossible to me. The second crash
occurred roughly 4 weeks after the first one. I run Cygwin Emacs
roughly 10 h a day, so I experience one crash in like 200 h. Also, as
I have to use Emacs productively at my dayjob, I won't be able to run
emacs-x11 -Q for weeks in order to see if any of my customizations
interfere.

Please let me know if I can pull any additional knowledge from these
coredumps.


regards,
Markus

-- 
Markus Hoenicka
http://www.mhoenicka.de
AQ score 38