Re: latest cvs segfault

Re: latest cvs segfault

[Richard Stallman]

    - backtrace:

    #0  set_cursor_from_row (w=0x92e0818, row=0x0, matrix=0x9399ef8, delta=0, 
	delta_bytes=0, dy=0, dvpos=0) at xdisp.c:9439
    #1  0x0806cef8 in try_window_id (w=0x92e0818) at xdisp.c:11773

That appears to be related to a bug fix I made a few days ago.
row_containing_pos must have returned zero.

The change below may fix the problem, but I am not certain it is
correct.  Anyway, can you please run under GDB all the time, with a
breakpoint set in set_cursor_from_row like this?

  b set_cursor_from_row if row==0

If it stops there, please try calling row_containing_pos the same way
it was called by try_window_id, but step through it and see precisely
why it returns zero.  In particular, what line does it return at?
(Show us the code on and around that line, not just the line number!)
What is the value of MATRIX_ROW_BOTTOM_Y (row) for the last row that
it thinks about?  And what is last_y?  Should it have found PT in the
previous row?


*** xdisp.c.~1.794.~	Sun Dec  8 16:06:46 2002
--- xdisp.c	Tue Dec 10 22:22:07 2002
***************
*** 11514,11520 ****
        /* Give up if we have gone too far.  */
        if (end && row >= end)
  	return NULL;
!       if (MATRIX_ROW_BOTTOM_Y (row) >= last_y)
  	return NULL;
  
        /* If it is in this row, return this row.  */
--- 11516,11522 ----
        /* Give up if we have gone too far.  */
        if (end && row >= end)
  	return NULL;
!       if (MATRIX_ROW_BOTTOM_Y (row) > last_y)
  	return NULL;
  
        /* If it is in this row, return this row.  */

Re: latest cvs segfault

[Bill White]

On Wed Dec 11 2002 at 11:45, Richard Stallman <rms@gnu.org> said:

>     - backtrace:
>
>     #0  set_cursor_from_row (w=0x92e0818, row=0x0, matrix=0x9399ef8, delta=0, 
> 	delta_bytes=0, dy=0, dvpos=0) at xdisp.c:9439
>     #1  0x0806cef8 in try_window_id (w=0x92e0818) at xdisp.c:11773
>
> That appears to be related to a bug fix I made a few days ago.
> row_containing_pos must have returned zero.
>
> The change below may fix the problem, but I am not certain it is
> correct.  Anyway, can you please run under GDB all the time, with a
> breakpoint set in set_cursor_from_row like this?

This appears to have fixed the problem.  I haven't had a segfault
since installing this.

bw
-- 
Bill White                                               Office: 5N-X30
Documentation Programmer                      Phone: 217-398-0700 x 234
Wolfram Research                                      Fax: 217-398-0747
http://members.wri.com/billw             Office hours: 4am-noon Mon-Fri

Re: latest cvs segfault

[Richard Stallman]

    > The change below may fix the problem, but I am not certain it is
    > correct.  Anyway, can you please run under GDB all the time, with a
    > breakpoint set in set_cursor_from_row like this?

    This appears to have fixed the problem.  I haven't had a segfault
    since installing this.

It certainly avoids the seg fault.  We don't know if it is correct.

Re: latest cvs segfault

[Robert J. Chassell]

Debian GNU/Linux, `sarge' (`testing'), kernel 2.4.18
GNOME/sawfish

Today's CVS snapshot of Wed, 2002 Dec 18  11:52 UTC,
GNU Emacs 21.3.50.30 (i686-pc-linux-gnu, X toolkit)

    With the second version of Miles' patch that implements the `final
    face realization filtering' function applied to *this mornings*
    version of src/xfaces.c,

started under GDB with my usual .emacs file 

crashed and produced this information:


gdb: Symbol `emacs_ctlx_keymap' has different size in shared object, 
consider re
-linking
GNU gdb 2002-04-01-cvs
Copyright 2002 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-linux"...
Warning: /home/bob/../lwlib: No such f
ile or directory.
DISPLAY = :0.0
TERM = xterm


Program received signal SIGABRT, Aborted.
0x402fd581 in kill () from /lib/libc.so.6
Argument list to give program being debugged when it is started is 
    "-q -l ~bob/.emacs-21 -bg DodgerBlue4 -fg white     -fn '-Misc-Fixed-Medium-
R-Normal--20-200-75-75-C-100-ISO8859-1'     -geometry +160+75     -name "Bob's E
macs"".
(gdb) bt
#0  0x402fd581 in kill () from /lib/libc.so.6
#1  0x080d2e9a in abort () at emacs.c:412
#2  0x0812ae1a in Feval (form=1484565596) at eval.c:1984
#3  0x08128c82 in Fprogn (args=1484565644) at eval.c:424
#4  0x0812bf56 in funcall_lambda (fun=1484565652, nargs=1, 
    arg_vector=0xbffd1380) at eval.c:2921
#5  0x0812bb81 in Ffuncall (nargs=2, args=0xbffd137c) at eval.c:2798
#6  0x0812a250 in internal_condition_case_2 (bfun=0x812b88c <Ffuncall>, 
    nargs=2, args=0xbffd137c, handlers=405462036, 
    hfun=0x805cbf0 <safe_eval_handler>) at eval.c:1435
#7  0x0805cce3 in safe_call (nargs=2, args=0xbffd137c) at xdisp.c:1399
#8  0x0805cd18 in safe_call1 (fn=411272476, arg=1211316344) at xdisp.c:1419
#9  0x080a29f4 in realize_face (cache=0x84fcd18, attrs=0xbffd14f0, c=0, 
    base_face=0x0, former_face_id=-1) at xfaces.c:6716
#10 0x080a3aa5 in face_at_buffer_position (w=0x88bc4d8, pos=6600410, 
    region_beg=0, region_end=0, endptr=0xbffd15f8, limit=6600411, mouse=1)
    at xfaces.c:5648
#11 0x080b2275 in note_mouse_highlight (f=0x84f9ba8, x=624, y=99)
    at xterm.c:7401
#12 0x080b155e in note_mouse_movement (frame=0x84f9ba8, event=0xbffd1b84)
    at xterm.c:6887
#13 0x080b58e5 in XTread_socket (sd=0, bufp=0xbffd2c24, numchars=4096, 
    expected=1) at xterm.c:11147
---Type <return> to continue, or q <return> to quit---
#14 0x080dbc01 in read_avail_input (expected=1) at keyboard.c:6462
#15 0x080dbdc2 in input_available_signal (signo=29) at keyboard.c:6625
#16 0x402fd518 in sigaction () from /lib/libc.so.6
#17 0x08057151 in sit_for (sec=165, usec=0, reading=1, display=1, 
    initial_display=0) at dispnew.c:6247
#18 0x080d7a25 in read_char (commandflag=1, nmaps=4, maps=0xbffff334, 
    prev_event=405461988, used_mouse_menu=0xbffff380) at keyboard.c:2630
#19 0x080de170 in read_key_sequence (keybuf=0xbffff484, bufsize=30, 
    prompt=405461988, dont_downcase_last=0, can_return_switch_frame=1, 
    fix_current_buffer=1) at keyboard.c:8505
#20 0x080d5b18 in command_loop_1 () at keyboard.c:1473
#21 0x0812a059 in internal_condition_case (bfun=0x80d5810 <command_loop_1>, 
    handlers=405558620, hfun=0x80d5424 <cmd_error>) at eval.c:1352
#22 0x080d56e8 in command_loop_2 () at keyboard.c:1274
#23 0x08129bed in internal_catch (tag=405519924, 
    func=0x80d56c4 <command_loop_2>, arg=405461988) at eval.c:1112
#24 0x080d5693 in command_loop () at keyboard.c:1253
#25 0x080d51e0 in recursive_edit_1 () at keyboard.c:969
#26 0x080d5310 in Frecursive_edit () at keyboard.c:1025
#27 0x080d41a3 in main (argc=14, argv=0xbffffa54) at emacs.c:1659
(gdb) c
Continuing.
Fatal error (6).
Program received signal SIGABRT, Aborted.
0x402fd581 in kill () from /lib/libc.so.6

I will continue to run Emacs under GDB; please tell me what else you
would like to see in addition to a backtrace.

-- 
    Robert J. Chassell                         Rattlesnake Enterprises
    http://www.rattlesnake.com                  GnuPG Key ID: 004B4AC8
    http://www.teak.cc                             bob@rattlesnake.com

Re: latest cvs segfault

[Miles Bader]

"Robert J. Chassell" <bob@rattlesnake.com> writes:
>     With the second version of Miles' patch that implements the `final
>     face realization filtering' function applied to *this mornings*
>     version of src/xfaces.c,
> 
> crashed and produced this information:

That backtrace looks pretty much like the one I just got; see my email
with subject `signal handling bogosities' for an explanation.

I think you can avoid the abort by just making sure the filter functions
are byte-compiled, though I can't vouch for this actually being safe...

-Miles
-- 
We live, as we dream -- alone....

Re: latest cvs segfault

[Robert J. Chassell]

Using yesterday's CVS snapshot with your patch to src/xfaces.c

Miles Bader <miles@lsi.nec.co.jp> wrote:

   I think you can avoid the abort by just making sure the filter functions
   are byte-compiled, ....

I byte compiled and installed your face-filter code that you rewrote
with macros.  So far -- about an hour -- everything has gone fine.  I
suffered many aborts yesterday using the uncompiled byte code.

--
    Robert J. Chassell                         Rattlesnake Enterprises
    http://www.rattlesnake.com                  GnuPG Key ID: 004B4AC8
    http://www.teak.cc                             bob@rattlesnake.com