From xemacs-m  Sat May 17 12:10:27 1997
Received: from the-great-machine.mit.edu (brnstndkramden.acf.nyu.edu@THE-GREAT-MACHINE.MIT.EDU [18.238.0.254])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id MAA12645
	for <xemacs-beta@xemacs.org>; Sat, 17 May 1997 12:10:26 -0500 (CDT)
Received: (from jered@localhost)
	by the-great-machine.mit.edu (8.8.5/8.8.5) id NAA20479;
	Sat, 17 May 1997 13:10:19 -0400
To: xemacs-beta@xemacs.org
Subject: Mixed frame support bugs in 20.2
From: Jered J Floyd <jered@mit.edu>
Date: 17 May 1997 13:10:19 -0400
Message-ID: <s56zptuqb2c.fsf@the-great-machine.mit.edu>
Lines: 108
X-Mailer: Gnus v5.4.52/XEmacs 20.2


   One of my users here is very good at tickling bugs in XEmacs'
multiple display support.  I tried to debug this, but I just don't
understand the XEmacs internals.  I appers that d->devmeths->symbol on
the display is a bogus (but not NULL pointer.) Any suggestions?

--Jered
jered@mit.edu 

------- Start of forwarded message -------
Reply-To: gutnik@mtl.mit.edu
To: jered@MIT.EDU
Subject: xemacs bug report stuff.
From: Vadim Gutnik <gutnik@MIT.EDU>
Date: 16 May 1997 18:16:47 -0400
Message-Id: <yjag1vn9i5s.fsf@godot.mit.edu>


I'm running xemacs on Bill.
gutnik@bill-the-cat> xemacs -version
XEmacs 20.2 [Lucid] (sparc-sun-solaris2.4) of Wed May 14 1997 on portnoy

Real live bugs:
	1) xemacs -nw
	   make-frame-on-display :0
	   C-x 5 2 (in the X frame)
	   move one of the frames to another virtual desktop (I run ctwm)
	   C-x 5 0 (in either X frame)
	    ==> Fatal error (11)
			<instructions>
		Lisp backtrace follows:
 
		  # (condition-case ... . error)
		  # (catch top-level ...)
	gdb -> where gives:
		 #0  0xef4a1cac in _kill ()

	The core is in /mit/gutnik/Public/xemacs_coredumps/core.1

Jered> Here's the C backtrace:
#0  0xef4a1cac in _kill ()
#1  0x654f0 in fatal_error_signal (sig=11)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/emacs.c:204
#2  <signal handler called>
#3  0xe3c74 in device_matches_specifier_tag_set_p (device=2482180, 
    tag_set=540150100)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/specifier.c:785
#4  0xe6744 in specifier_instance_from_inst_list (specifier=271095808, 
    matchspec=270692088, domain=271109888, inst_list=540149948, 
    errb=ERROR_ME_WARN, no_quit=540149948, depth=1)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/specifier.c:2274
#5  0xe6b1c in specifier_instance (specifier=271095808, matchspec=270692088, 
    domain=271109888, errb=ERROR_ME_WARN, no_quit=0, no_fallback=0, depth=1)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/specifier.c:2393
#6  0xa3868 in glyph_image_instance (glyph=271095872, domain=271109888, 
    errb=ERROR_ME_WARN, no_quit=0)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/glyphs.c:2358
#7  0x9d1c8 in update_frame_icon (f=0x28cf00)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/frame.c:2799
#8  0xcd3d4 in redisplay_device (d=0x3a2c00)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/redisplay.c:5466
#9  0xcd790 in redisplay_without_hooks ()
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/redisplay.c:5592
#10 0xcd970 in redisplay ()
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/redisplay.c:5659
#11 0x75344 in Fnext_event (event=272177736, prompt=1344659460)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/event-stream.c:2129
#12 0x52d38 in Fcommand_loop_1 ()
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/cmdloop.c:551
#13 0x52b6c in command_loop_1 (dummy=1344659460)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/cmdloop.c:480
#14 0x691d8 in condition_case_1 (handlers=1344659540, 
    bfun=0x52b48 <command_loop_1>, barg=1344659460, hfun=0x52464 <cmd_error>, 
    harg=1344659460)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/eval.c:1657
#15 0x52580 in command_loop_3 ()
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/cmdloop.c:242
#16 0x525a8 in command_loop_2 (dummy=1344659460)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/cmdloop.c:253
#17 0x68e38 in internal_catch (tag=1344736416, func=0x5259c <command_loop_2>, 
    arg=1344659460, threw=0x0)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/eval.c:1334
#18 0x527b4 in initial_command_loop (load_me=337308)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/cmdloop.c:291
#19 0x66658 in xemacs_20_2_ (argc=2, argv=0xeffff94c, envp=0xeffff958)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/emacs.c:1461
#20 0x66dac in main (argc=2, argv=0xeffff94c, envp=0xeffff958)
    at /afs/athena.mit.edu/contrib/xemacs/src/xemacs-20.2/src/emacs.c:1812

[...]

	3) xemacs -unmapped &
	   gnuclient -nw (in a different xterm, to avoid #2)
	   make-frame-on-display :0
	   C-x 5 0 (in the X frame)
	    ==> Fatal error (11)
		same stuff as #1. 

	The core is in /mit/gutnik/Public/xemacs_coredumps/core.3
Jered> C backtrace is the same as in item #1. Same bug, I'd guess.

I'm sure I could generatea couple more segfaults like 1 & 3, but you
get the idea. It sometimes happens when there's an X frame and a text
frame, and you close the X one. Note that if you use xemacs -nw and
close windows in the same workspace, you *don't* get a coredump. 
[...]

------- End of forwarded message -------

