NetBSD Problem Report #19897
Received: (qmail 25947 invoked by uid 605); 18 Jan 2003 04:44:51 -0000
Message-Id: <20030118044844.7634.qmail@juice.scl.cwru.edu>
Date: 18 Jan 2003 04:48:44 -0000
From: schmonz@netbsd.org
Sender: gnats-bugs-owner@netbsd.org
To: gnats-bugs@gnats.netbsd.org
Subject: Another powerpc SIGILL
X-Send-Pr-Version: 3.95
>Number: 19897
>Category: port-powerpc
>Synopsis: Another powerpc SIGILL
>Confidential: no
>Severity: serious
>Priority: high
>Responsible: port-powerpc-maintainer
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Sat Jan 18 04:45:00 +0000 2003
>Closed-Date: Sun May 20 06:20:54 +0000 2012
>Last-Modified: Sun May 20 06:20:54 +0000 2012
>Originator: Amitai Schlair
>Release: NetBSD 1.6L
>Organization:
Schmonz Enterprises
>Environment:
NetBSD xtinag3.xtina.schmonz.com 1.6L NetBSD 1.6L (XTINAG3) #32: Sun Jan 12 08:23:27 EST 2003 schmonz@xtinag3.xtina.schmonz.com:/home/schmonz/netbsd/src/sys/arch/macppc/compile/XTINAG3 macppc
>Description:
With a system built from Jan. 11 source (after both of the powerpc
dynamic linking fixes) I got a SIGILL-induced core dump from /bin/sh
while building KDE3 from pkgsrc. I occasionally find that a daemon
or two has died, leaving a core dump in /.
>How-To-Repeat:
$ cd ~/netbsd/pkgsrc/x11/kde3
$ make install
[...]
gmake[2]: Leaving directory `/home/schmonz/netbsd/pkgsrc/x11/kdelibs3/work.xtinag3/kdelibs-3.0.5a/kio'
Making all in kioslave
[1] Illegal instruction (core dumped) (cd ${subdir} &&...
gmake[1]: *** [all-recursive] Error 1
gmake[1]: Leaving directory `/home/schmonz/netbsd/pkgsrc/x11/kdelibs3/work.xtinag3/kdelibs-3.0.5a'
gmake: *** [all] Error 2
*** Error code 2
Stop.
make: stopped in /home/schmonz/netbsd/pkgsrc/x11/kdelibs3
*** Error code 1
Stop.
make: stopped in /home/schmonz/netbsd/pkgsrc/x11/kdelibs3
*** Error code 1
Stop.
make: stopped in /home/schmonz/netbsd/pkgsrc/x11/kdelibs3
*** Error code 1
Stop.
make: stopped in /home/schmonz/netbsd/pkgsrc/x11/kdebase3
*** Error code 1
Stop.
make: stopped in /home/schmonz/netbsd/pkgsrc/x11/kdebase3
*** Signal 4
Stop.
>Fix:
None known, but here's some gdb output as requested by gimpy:
$ gdb sh ../kdelibs3/work.xtinag3/kdelibs-3.0.5a/sh.core
GNU gdb 5.0nb1
Copyright 2000 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 "powerpc--netbsd"...(no debugging symbols found)...
Core was generated by `sh'.
Program terminated with signal 4, Illegal instruction.
Reading symbols from /libexec/ld.elf_so...(no debugging symbols found)...done.
Loaded symbols for /libexec/ld.elf_so
Reading symbols from /lib/libedit.so.2...(no debugging symbols found)...done.
Loaded symbols for /lib/libedit.so.2
Reading symbols from /lib/libtermcap.so.0...(no debugging symbols found)...done.
Loaded symbols for /lib/libtermcap.so.0
Reading symbols from /lib/libc.so.12...(no debugging symbols found)...done.
Loaded symbols for /lib/libc.so.12
#0 0x4183a6c8 in cache_sh () from /libexec/ld.elf_so
(gdb) bt
#0 0x4183a6c8 in cache_sh () from /libexec/ld.elf_so
#1 0x4183450c in _rtld_find_symdef () from /libexec/ld.elf_so
#2 0x41832be0 in _rtld_bind () from /libexec/ld.elf_so
#3 0x41832858 in _rtld_bind_start () from /libexec/ld.elf_so
#4 0x18032b8 in get_alias_text ()
#5 0x1802860 in get_alias_text ()
#6 0x1805964 in get_alias_text ()
#7 0x1804378 in get_alias_text ()
#8 0x18041fc in get_alias_text ()
#9 0x18048a0 in get_alias_text ()
#10 0x1804294 in get_alias_text ()
#11 0x1804230 in get_alias_text ()
#12 0x18042e0 in get_alias_text ()
#13 0x1804604 in get_alias_text ()
#14 0x1804308 in get_alias_text ()
#15 0x18042e0 in get_alias_text ()
#16 0x18041d8 in get_alias_text ()
#17 0x18041d8 in get_alias_text ()
#18 0x180411c in get_alias_text ()
#19 0x180fa30 in main ()
#20 0x1801b8c in _init ()
#21 0x41832828 in _rtld_start () from /libexec/ld.elf_so
(gdb) x/i $pc
0x4183a6c8 <cache_sh+744>: blrl
(gdb) info shared
From To Syms Read Shared Object Library
0x41830000 0x4183ae18 Yes /libexec/ld.elf_so
0x41843000 0x41871a68 Yes /lib/libedit.so.2
0x41872000 0x418848bc Yes /lib/libtermcap.so.0
0x41885000 0x4195b438 Yes /lib/libc.so.12
>Release-Note:
>Audit-Trail:
From: David Laight <david@l8s.co.uk>
To: netbsd-bugs@netbsd.org
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: port-powerpc/19897: Another powerpc SIGILL
Date: Sat, 18 Jan 2003 10:56:00 +0000
> (gdb) bt
> #0 0x4183a6c8 in cache_sh () from /libexec/ld.elf_so
> #1 0x4183450c in _rtld_find_symdef () from /libexec/ld.elf_so
> #2 0x41832be0 in _rtld_bind () from /libexec/ld.elf_so
> #3 0x41832858 in _rtld_bind_start () from /libexec/ld.elf_so
> #4 0x18032b8 in get_alias_text ()
> #5 0x1802860 in get_alias_text ()
> #6 0x1805964 in get_alias_text ()
> #7 0x1804378 in get_alias_text ()
> #8 0x18041fc in get_alias_text ()
> #9 0x18048a0 in get_alias_text ()
> #10 0x1804294 in get_alias_text ()
> #11 0x1804230 in get_alias_text ()
> #12 0x18042e0 in get_alias_text ()
> #13 0x1804604 in get_alias_text ()
> #14 0x1804308 in get_alias_text ()
> #15 0x18042e0 in get_alias_text ()
> #16 0x18041d8 in get_alias_text ()
> #17 0x18041d8 in get_alias_text ()
> #18 0x180411c in get_alias_text ()
> #19 0x180fa30 in main ()
> #20 0x1801b8c in _init ()
> #21 0x41832828 in _rtld_start () from /libexec/ld.elf_so
Something is telling porkies in that back trace.
get_alias_text() is a small routine that is only called during
some (obscure) command line edit actions. It isn't recursive.
I'd do an 'nm -n sh' and sort out the correct symbols for the
return addresses ('objdump -d sh' might help validate them as well).
David
--
David Laight: david@l8s.co.uk
From: David Laight <david@l8s.co.uk>
To: netbsd-bugs@netbsd.org
Cc: gnats-bugs@gnats.netbsd.org
Subject: Re: port-powerpc/19897: Another powerpc SIGILL
Date: Sat, 18 Jan 2003 12:12:34 +0000
> Something is telling porkies in that back trace.
> get_alias_text() is a small routine that is only called during
> some (obscure) command line edit actions. It isn't recursive.
Just realised: The problem with the backtrace is that /bin/sh
has been stripped, so the only symbols available are those in
the dynamic linker symbol table...
If you built /bin/sh from source you ought to have the unstripped
binary in the obj directory.
Try pointing gdb at that file instead.
David
--
David Laight: david@l8s.co.uk
State-Changed-From-To: open->feedback
State-Changed-By: mrg@NetBSD.org
State-Changed-When: Sun, 20 Dec 2009 08:13:06 +0000
State-Changed-Why:
has this occured in recent times?
State-Changed-From-To: feedback->closed
State-Changed-By: matt@NetBSD.org
State-Changed-When: Sun, 20 May 2012 06:20:54 +0000
State-Changed-Why:
Since powerpc moved to secureplt, the PLT is now read-only and doesn't require
cache synchronization.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.