bug#3659: FW: 23.0.95; crash on set-frame-font

[Marshall, Simon]

With a CFLAGS=-g version of yesterday's trunk, the crash happens in a
different place (it aborts semi-cleanly).  This is true of the
2009-05-20 version as well:

risksun2 134> gdb ./emacs
GNU gdb 6.7.1
Copyright (C) 2007 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show
copying"
and "show warranty" for details.
This GDB was configured as "sparc-sun-solaris2.8"...
SIGINT is used by the debugger.
Are you sure you want to change it? (y or n) [answered Y; input not from
terminal]
DISPLAY = dtp020071.misys.global.ad:0.0
TERM = xterm
Breakpoint 1 at 0x1a331c: file emacs.c, line 433.
Breakpoint 2 at 0x1d6a9c: file sysdep.c, line 1132.
(gdb) r -Q
Starting program:
/homedev/marshals/ftp/emacs-23.1-pretests/emacs-2009-07-23-g/src/emacs
-Q
warning: Temporarily disabling breakpoints for unloaded shared library
"/usr/lib/ld.so.1"

Breakpoint 1, abort () at emacs.c:433
433       kill (getpid (), SIGABRT);
(gdb) where
#0  abort () at emacs.c:433
#1  0x00280904 in internal_condition_case (bfun=0x1a913c
<command_loop_1>, handlers=5385361, 
    hfun=0x1a85c4 <cmd_error>) at eval.c:1487
#2  0x001a8c24 in command_loop_2 () at keyboard.c:1359
#3  0x002801e0 in internal_catch (tag=5378409, func=0x1a8bf8
<command_loop_2>, arg=5314561)
    at eval.c:1248
#4  0x001a8b9c in command_loop () at keyboard.c:1338
#5  0x001a7e54 in recursive_edit_1 () at keyboard.c:953
#6  0x001a8114 in Frecursive_edit () at keyboard.c:1015
#7  0x001a5948 in main (argc=2, argv=0xffbee394) at emacs.c:1854
(gdb) up
#1  0x00280904 in internal_condition_case (bfun=0x1a913c
<command_loop_1>, handlers=5385361, 
    hfun=0x1a85c4 <cmd_error>) at eval.c:1487
1487        abort ();
(gdb) list 
1482
1483      /* Since Fsignal will close off all calls to x_catch_errors,
1484         we will get the wrong results if some are not closed now.
*/
1485    #if HAVE_X_WINDOWS
1486      if (x_catching_errors ())
1487        abort ();
1488    #endif
1489
1490      c.tag = Qnil;
1491      c.val = Qnil;
(gdb) 

With a CFLAGS="-g -O2" version I do not get a crash anymore.

So: -g aborts in 2009-05-22 and 2009-07-23 versions
But: -g -O2 crashes in 2009-05-22 and appears ok in 2009-07-23 versions

What to do?  

-----Original Message-----
From: Kenichi Handa [mailto:handa@m17n.org] 
Sent: 24 July 2009 02:08
To: Marshall, Simon
Cc: cyd@stupidchicken.com; monnier@iro.umontreal.ca; rgm@gnu.org;
3659@emacsbugs.donarmstrong.com
Subject: Re: FW: 23.0.95; crash on set-frame-font

In article
<63C2A154B1708946B60726AFDBA00AC00668060C@ukmailemea01.misys.global.ad>,
"Marshall, Simon" <Simon.Marshall@misys.com> writes:

> (gdb) p *charset
> $1 = {
>   id = 1, 
>   hash_index = 3678208, 
>   dimension = 1073758208, 
>   code_space = {1294009101, 8830976, 1, 40, -4271696, 1, -4271232, 0,
0,
> 1483848, 805306368, 
>     15794175, 3737601, 3737601, 1, 4}, 

Ummm, the content of this structure is completely broken, or
"charset" points incorrect memory.

> (gdb) xcharset charset->id
> Attempt to extract a component of a value that is not a structure
> pointer.

> Was the last cmd wrong?  Let me know what else I can do...

It's a bug of old .gdbinit, and is now fixed both in the
trunk and EMACS_23_1_RC branch.  It seems that you are not
using the latest emacs code, but as we already found the
culprit change, we can debug using the latest code.  Please
recompile the latest code of the trunk or branch with
  % make CFLAGS=-g clean all

When you get segfault with that code, please show me the
backtrace again, and the results of these commands.

(gdb) up 2   ;; go up to the function xfont_list_pattern
(gdb) pp registry
(gdb) pp script

---
Kenichi Handa
handa@m17n.org


"Misys" is the trade name for Misys plc (registered in England and Wales). 
Registration Number: 01360027. Registered office: One Kingdom Street, London W2 
6BL, United Kingdom. For a list of Misys group operating companies please go to 
http://www.misys.com/corp/About_Us/misys_operating_companies.html. This email 
and any attachments have been scanned for known viruses using multiple 
scanners. This email message is intended for the named recipient only. It may 
be privileged and/or confidential. If you are not the named recipient of this 
email please notify us immediately and do not copy it or use it for any 
purpose, nor disclose its contents to any other person. This email does not 
constitute the commencement of legal relations between you and Misys plc. 
Please refer to the executed contract between you and the relevant member of 
the Misys group for the identity of the contracting party with which you are 
dealing.