bug#5946: current HEAD fails "emacs -Q" on OSX command-line

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Randal L. Schwartz]

Repeat by:

  emacs -Q

Aborts with

 Fatal error (10)

then core dumps.

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Eli Zaretskii]

> From: merlyn@stonehenge.com (Randal L. Schwartz)
> Date: Wed, 14 Apr 2010 07:55:38 -0700
> Cc: 
> 
> 
> Repeat by:
> 
>   emacs -Q
> 
> Aborts with
> 
>  Fatal error (10)
> 
> then core dumps.

Can you show the backtrace from the core file?

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Randal L. Schwartz]

>>>>> "Eli" == Eli Zaretskii <eliz@gnu.org> writes:

>> From: merlyn@stonehenge.com (Randal L. Schwartz)
>> Date: Wed, 14 Apr 2010 07:55:38 -0700
>> Cc: 
>> 
>> 
>> Repeat by:
>> 
>> emacs -Q
>> 
>> Aborts with
>> 
>> Fatal error (10)
>> 
>> then core dumps.

Eli> Can you show the backtrace from the core file?

I don't see a core file.  Is there something I have to do on OSX
specially to get that?

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Eli Zaretskii]

> From: merlyn@stonehenge.com (Randal L. Schwartz)
> Cc: 5946@debbugs.gnu.org
> Date: Wed, 14 Apr 2010 10:03:03 -0700
> 
> Eli> Can you show the backtrace from the core file?
> 
> I don't see a core file.  Is there something I have to do on OSX
> specially to get that?

I know nothing about OSX, but if you use a Posix shell, you may need
to issue a ulimit command to set the core file size limitation to a
suitable (non-zero) value.

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Randal L. Schwartz]

>>>>> "Eli" == Eli Zaretskii <eliz@gnu.org> writes:


Eli> I know nothing about OSX, but if you use a Posix shell, you may need
Eli> to issue a ulimit command to set the core file size limitation to a
Eli> suitable (non-zero) value.

Ahh, I forgot they go into ~/Library/Logs/CrashReporter.  File is attached.

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Randal L. Schwartz]

>>>>> "Eli" == Eli Zaretskii <eliz@gnu.org> writes:

Eli> I know nothing about OSX, but if you use a Posix shell, you may need
Eli> to issue a ulimit command to set the core file size limitation to a
Eli> suitable (non-zero) value.

Found it.  Had to create /cores mode 777, *and* unlimit the core size:

#0  0x97967e42 in kill$UNIX2003 ()
#1  0x0009a15a in fatal_error_signal (sig=6) at emacs.c:402
#2  <signal handler called>
#3  0x97967e42 in kill$UNIX2003 ()
#4  0x979da23a in raise ()
#5  0x979e6679 in abort ()
#6  0x00177e15 in ns_term_shutdown (sig=<value temporarily unavailable,
#due to optimizations>) at nsterm.m:3891
#7  0x0009a1af in fatal_error_signal (sig=10) at emacs.c:388
#8  <signal handler called>
#9  ns_set_doc_edited (f=0x14034c8, arg=25165834) at nsfns.m:691
#10 0x000374c2 in prepare_menu_bars () at xdisp.c:9841
#11 0x0003825d in redisplay_internal (preserve_echo_area=<value
#temporarily unavailable, due to optimizations>) at xdisp.c:11698
#12 0x000a7a63 in read_char (commandflag=0, nmaps=0, maps=0x0,
#prev_event=25165882, used_mouse_menu=0x0, end_time=0xbfffe8d8) at
#keyboard.c:2577
#13 0x00134c3b in read_filtered_event (no_switch_frame=0,
#ascii_required=0, error_nonascii=0, input_method=0, seconds=8) at
#lread.c:713
#14 0x00134f77 in Fread_event (prompt=25165834, inherit_input_method=0,
#seconds=0) at lread.c:827
#15 0x00113811 in Ffuncall (nargs=4, args=0xbfffe9b0) at eval.c:3030
#16 0x00150894 in Fbyte_code (bytestr=25726921, vector=8578341,
#maxdepth=32) at bytecode.c:680
#17 0x001131f1 in funcall_lambda (fun=8578693, nargs=0,
#arg_vector=0xbfffeb44) at eval.c:3211
#18 0x0011360b in Ffuncall (nargs=1, args=0xbfffeb40) at eval.c:3081
#19 0x00150894 in Fbyte_code (bytestr=2072217, vector=2072237,
#maxdepth=16) at bytecode.c:680
#20 0x001131f1 in funcall_lambda (fun=2072165, nargs=1,
#arg_vector=0xbfffecc4) at eval.c:3211
#21 0x0011360b in Ffuncall (nargs=2, args=0xbfffecc0) at eval.c:3081
#22 0x00150894 in Fbyte_code (bytestr=2119513, vector=2119533,
#maxdepth=28) at bytecode.c:680
#23 0x001131f1 in funcall_lambda (fun=2119493, nargs=0,
#arg_vector=0xbfffee54) at eval.c:3211
#24 0x0011360b in Ffuncall (nargs=1, args=0xbfffee50) at eval.c:3081
#25 0x00150894 in Fbyte_code (bytestr=2116425, vector=2116445,
#maxdepth=24) at bytecode.c:680
#26 0x001131f1 in funcall_lambda (fun=2116405, nargs=0,
#arg_vector=0xbfffef70) at eval.c:3211
#27 0x001144c9 in apply_lambda (fun=2116405, args=25165834, eval_flag=1)
#at eval.c:3135
#28 0x00112ad2 in Feval (form=5398318) at eval.c:2406
#29 0x0009c29d in top_level_2 () at keyboard.c:1365
#30 0x00110f0d in internal_condition_case (bfun=0x9c280 <top_level_2>,
#handlers=25201354, hfun=0xa2c60 <cmd_error>) at eval.c:1490
#31 0x0009c2f3 in top_level_1 () at keyboard.c:1373
#32 0x00110dec in internal_catch (tag=0, func=0x9c2b0 <top_level_1>,
#arg=25165834) at eval.c:1226
#33 0x0009bfa7 in command_loop () at keyboard.c:1328
#34 0x0009c082 in recursive_edit_1 () at keyboard.c:950
#35 0x0009c201 in Frecursive_edit () at keyboard.c:1012
#36 0x0009ad5a in main (argc=3, argv=0xbffff460) at emacs.c:1784




-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion

bug#5946: current HEAD fails "emacs -Q" on OSX command-line

[Eli Zaretskii]

> From: merlyn@stonehenge.com (Randal L. Schwartz)
> Cc: 5946@debbugs.gnu.org
> Date: Wed, 14 Apr 2010 12:05:44 -0700
> 
> #0  0x97967e42 in kill$UNIX2003 ()
> #1  0x0009a15a in fatal_error_signal (sig=6) at emacs.c:402
> #2  <signal handler called>
> #3  0x97967e42 in kill$UNIX2003 ()
> #4  0x979da23a in raise ()
> #5  0x979e6679 in abort ()
> #6  0x00177e15 in ns_term_shutdown (sig=<value temporarily unavailable,
> #due to optimizations>) at nsterm.m:3891
> #7  0x0009a1af in fatal_error_signal (sig=10) at emacs.c:388
> #8  <signal handler called>
> #9  ns_set_doc_edited (f=0x14034c8, arg=25165834) at nsfns.m:691
> #10 0x000374c2 in prepare_menu_bars () at xdisp.c:9841

Thanks.

This crash looks NS-specific.  Someone who knows how the NS port works
please look into it.