Re: Emacs-w32... still crashing...

[Ken Brown <kbrown@cornell.edu>]

On 5/19/2014 9:20 AM, Zdzislaw Meglicki wrote:
> I got it this time. It crashed on me just a few seconds after I started it,
> while reading mail with rmail. This time I had the right debug info
> installed for this version of emacs and here is the backtrace:
>
> =============================================
> (gdb) backtrace
> #0  0x00007ff9550c9e3b in KERNELBASE!DebugBreak ()
> from /cygdrive/c/WINDOWS/system32/KERNELBASE.dll
> #1  0x000000010061a344 in emacs_abort ()
> at /usr/src/debug/emacs-24.3.90-1/src/w32fns.c:8460
> #2  0x000000010057e7ce in arithcompare (num1=0, num2=4000000,
> comparison=ARITH_GRTR_OR_EQUAL)
> at /usr/src/debug/emacs-24.3.90-1/src/data.c:2319
> #3  0x00000001005df264 in exec_byte_code (bytestr=25771187393,
> vector=4312129309, maxdepth=28, args_template=4305639474, nargs=0,
> args=0x0) at /usr/src/debug/emacs-24.3.90-1/src/bytecode.c:1468
> #4  0x000000010059a5a0 in funcall_lambda (fun=4309134989, nargs=2,
> arg_vector=0x10105df1d <bss_sbrk_buffer+6847133>)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:3049
> #5  0x0000000100599d77 in Ffuncall (nargs=3, args=0x4387d8)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:2864
> #6  0x00000001005dcaa2 in exec_byte_code (bytestr=4303067497,
> vector=4303067533, maxdepth=24, args_template=4305639474, nargs=0,
> args=0x0) at /usr/src/debug/emacs-24.3.90-1/src/bytecode.c:919
> #7  0x000000010059a5a0 in funcall_lambda (fun=4303067373, nargs=4,
> arg_vector=0x1007b998d <pure+1353837>)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:3049
> #8  0x0000000100599d77 in Ffuncall (nargs=5, args=0x438d78)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:2864
> #9  0x00000001005dcaa2 in exec_byte_code (bytestr=4303067977,
> vector=4303068013, maxdepth=28, args_template=4305639474, nargs=0,
> args=0x0) at /usr/src/debug/emacs-24.3.90-1/src/bytecode.c:919
> #10 0x000000010059a5a0 in funcall_lambda (fun=4303067917, nargs=3,
> arg_vector=0x1007b9b6d <pure+1354317>)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:3049
> #11 0x0000000100599d77 in Ffuncall (nargs=4, args=0x4392f8)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:2864
> #12 0x00000001005dcaa2 in exec_byte_code (bytestr=4303070921,
> vector=4303070957, maxdepth=20, args_template=4305639474, nargs=0,
> args=0x0) at /usr/src/debug/emacs-24.3.90-1/src/bytecode.c:919
> #13 0x000000010059a5a0 in funcall_lambda (fun=4303070877, nargs=1,
> arg_vector=0x1007ba6ed <pure+1357261>)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:3049
> #14 0x0000000100599d77 in Ffuncall (nargs=2, args=0x439870)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:2864
> #15 0x00000001005994ca in call1 (fn=4305692866, arg1=4313865909)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:2614
> #16 0x0000000100506af0 in timer_check_2 (timers=25776357158,
> idle_timers=4305639474)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:4510
> #17 0x0000000100506c89 in timer_check ()
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:4577
> #18 0x0000000100504b4b in readable_events (flags=1)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:3444
> #19 0x000000010050afe4 in get_input_pending (flags=1)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:6761
> #20 0x0000000100511142 in detect_input_pending_run_timers (do_display=true)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:9884
> #21 0x00000001005e9939 in wait_reading_process_output (time_limit=30, nsecs=0,
> read_kbd=-1, do_display=true, wait_for_cell=4305639474, wait_proc=0x0,
> just_wait_proc=0) at /usr/src/debug/emacs-24.3.90-1/src/process.c:4699
> #22 0x000000010040df39 in sit_for (timeout=120, reading=true, display_option=1)
> at /usr/src/debug/emacs-24.3.90-1/src/dispnew.c:5805
> #23 0x0000000100503226 in read_char (commandflag=1, map=25776355574,
> prev_event=4305639474, used_mouse_menu=0x43a3ff, end_time=0x0)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:2806
> #24 0x000000010050f622 in read_key_sequence (keybuf=0x43a610, bufsize=30,
> prompt=4305639474, dont_downcase_last=false, can_return_switch_frame=true,
> fix_current_buffer=true, prevent_redisplay=false)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:9079
> #25 0x00000001004ffcf3 in command_loop_1 ()
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:1449
> #26 0x0000000100596707 in internal_condition_case (
> bfun=0x1004ff821 <command_loop_1>, handlers=4305705682,
> hfun=0x1004fef62 <cmd_error>)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:1354
> #27 0x00000001004ff4e5 in command_loop_2 (ignore=4305639474)
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:1174
> #28 0x0000000100595e81 in internal_catch (tag=4305695746,
> func=0x1004ff4b3 <command_loop_2>, arg=4305639474)
> at /usr/src/debug/emacs-24.3.90-1/src/eval.c:1118
> #29 0x00000001004ff474 in command_loop ()
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:1153
> #30 0x00000001004feadf in recursive_edit_1 ()
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:777
> #31 0x00000001004fec83 in Frecursive_edit ()
> at /usr/src/debug/emacs-24.3.90-1/src/keyboard.c:845
> #32 0x00000001004fca49 in main (argc=2, argv=0x43ab30)
> at /usr/src/debug/emacs-24.3.90-1/src/emacs.c:1646
> (gdb)
>
> ================================================

Thanks. These crashes are very mysterious to me.  Two of them have 
recently been reported to the emacs bug tracker, including one by me a 
couple hours ago:

   http://debbugs.gnu.org/cgi/bugreport.cgi?bug=17526

Unfortunately, nothing in the backtrace (yours or mine) is giving me any 
clue.  Maybe someone more knowledgeable, especially about the Cygwin-w32 
build, will have some ideas.  (Daniel?)

For future reference, here are some suggestions:

1. Once you've started gdb, give the command "source 
/usr/share/emacs/24.3.90/etc/.gdbinit".  This defines some gdb commands 
that are useful for debugging emacs.  Among other things, this will 
cause the "backtrace" command to include a lisp backtrace.

2. After getting a backtrace and reporting it somewhere (see item 3 
below), keep the gdb session open, if possible, in case someone wants 
you to try to get more info.

3. If you are willing to help debug these crashes, send your backtraces 
as emacs bug reports instead of sending them to this list.  You can do 
that easily via the command `M-x report-emacs-bug' within emacs, or you 
can just send email to bug-gnu-emacs@gnu.org.

Ken

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple