NetBSD Problem Report #723
From gnats Tue Jan 10 18:06:02 1995
Received: from pain.lcs.mit.edu (pain.lcs.mit.edu [128.52.46.239]) by sun-lamp.cs.berkeley.edu (8.6.9/8.6.9) with ESMTP id SAA00685 for <gnats-bugs@sun-lamp.cs.berkeley.edu>; Tue, 10 Jan 1995 18:06:01 -0800
Message-Id: <199501110135.UAA07279@kolvir.blrc.ma.us>
Date: Tue, 10 Jan 1995 20:35:59 -0500
From: John Kohl <jtk@kolvir.blrc.ma.us>
Reply-To: jtk@kolvir.blrc.ma.us
To: gnats-bugs@NetBSD.ORG
Subject: kernel post-mortems on i386 don't work
X-Send-Pr-Version: 3.2
>Number: 723
>Category: port-i386
>Synopsis: kernel post-mortems on i386 don't work
>Confidential: no
>Severity: non-critical
>Priority: medium
>Responsible: gnats-admin
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Tue Jan 10 18:20:06 +0000 1995
>Closed-Date: Sat Aug 23 01:49:31 +0000 1997
>Last-Modified: Sat Aug 23 01:50:07 +0000 1997
>Originator: John Kohl
>Release: 1.0A
>Organization:
NetBSD Kernel Hackers `R` Us
>Environment:
System: NetBSD kolvir 1.0A NetBSD 1.0A (KOLVIR) #39: Wed Dec 21 22:18:44 EST 1994 jtk@kolvir:/u1/NetBSD-current/src/sys/arch/i386/compile/KOLVIR i386
>Description:
'gdb -k' doesn't work properly for debugging i386 crash dumps.
It looks like libkvm is missing support in kvm_i386.c:_kvm_kvatop().
I also noticed that the call to kvm_open() from gdb's
kcorelow.c:kcore_open() doesn't pass in the symbol file name--making it
default to /netbsd.
>How-To-Repeat:
Try to debug a crash dump, you'll get something like:
# gdb -k netbsd.1 vmcore.1
(no debugging symbols found)...
/usr/vice/cache/crash/vmcore.1: No such file or directory.
(kgdb)
>Fix:
I guess it requires some gross hacks in the kvm_i386 support to go page
table walking?
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->closed
State-Changed-By: mrg
State-Changed-When: Sun Jul 6 03:30:11 PDT 1997
State-Changed-Why:
i'm told this is now working...
State-Changed-From-To: closed->open
State-Changed-By: jtk
State-Changed-When: Tue Jul 8 19:57:41 PDT 1997
State-Changed-Why:
reopening (see further details via email) since stack traces for curproc
are still broken
From: John Kohl <jtk@kolvir.arlington-heights.ma.us>
To: mrg@NetBSD.ORG
Cc: gnats-admin@NetBSD.ORG
Subject: Re: port-i386/723
Date: Tue, 8 Jul 1997 22:59:14 -0400 (EDT)
This is still a bug.
I can examine kernel stacks of processes which are not curproc, but gdb
can't find the proper stack traceback for curproc.
% ps -xa -O paddr -M netbsd.2.core
PID PADDR TT STAT TIME COMMAND
0 220bdc ?? DLs 0:00.01 (swapper)
1 7ec400 ?? Ss 0:00.22 /sbin/init
2 7ec200 ?? DL 0:00.09 (pagedaemon)
86 807800 ?? IWs 0:00.00 /sbin/dhclient de0
106 807a00 ?? Rs 0:00.17 syslogd
110 807000 ?? IWs 0:00.04 named
113 807400 ?? IWs 0:00.02 portmap
123 7fea00 ?? IWs 0:00.02 mountd
126 807600 ?? IWs 0:00.02 nfsd: master (nfsd)
128 7ef800 ?? IW 0:00.00 nfsd: server (nfsd)
129 847e00 ?? IW 0:00.01 nfsd: server (nfsd)
130 847c00 ?? IW 0:00.01 nfsd: server (nfsd)
131 847a00 ?? IW 0:00.01 nfsd: server (nfsd)
135 84be00 ?? IW 0:00.01 nfsiod -n 4
136 84bc00 ?? IW 0:00.00 nfsiod -n 4
137 84ba00 ?? IW 0:00.00 nfsiod -n 4
138 84b800 ?? IW 0:00.01 nfsiod -n 4
173 847000 ?? Is 0:00.36 update 30
175 84b600 ?? IWs 0:00.04 cron
188 84b400 ?? S<s 0:00.08 xntpd -p /var/run/xntpd.pid
193 84b200 ?? Ss 0:00.07 dhcpd
199 877e00 ?? IWs 0:00.13 lpd
223 87f800 ?? IWs 0:00.03 sendmail: accepting connections on port 25 (sen
232 877000 ?? IWs 0:00.06 inetd
235 87f000 ?? IWs 0:00.04 /usr/local/sbin/faxq
238 877600 ?? IWs 0:02.36 /usr/local/sbin/sshd
266 807200 ?? IWs 0:00.01 /usr/vice/etc/afsd -nosettime
298 877800 ?? IW 0:00.02 /etc/athena/zhm
307 87f400 ?? IW 0:00.21 /usr/X11R6/bin/xdm -nodaemon ttyv7
309 87fe00 ?? IW 0:00.19 /usr/local/libexec/faxgetty -d tty02
312 7ec000 ?? Ss 0:03.94 /usr/X11R6/bin/X -terminate vt8 -auth /usr/X11R
313 9f3a00 ?? IWs 0:00.14 -:0
324 877a00 ?? IWs 0:00.51 xterm -e /u1/users/jtk/lib/scripts/xinitrc
328 a04400 ?? S 0:00.02 ssh-agent xterm -iconic -xrm *Foreground:cyan -
380 a04200 ?? IWs 0:00.05 /usr/local/libexec/emacs/19.30/i386-unknown-net
329 a04600 p1 IWs+ 0:00.56 -usr/local/bin/tcsh
348 a04800 p1 IW 0:00.02 /bin/csh -f /u1/users/jtk/lib/scripts/xinitrc
349 a04e00 p1 S 0:00.35 vtwm
353 a04c00 p1 R 0:00.22 xconsole
354 a04a00 p1 I 0:03.47 emacs
355 a04000 p1 S 0:00.82 xterm
357 a0fc00 p1 S 0:00.20 xbiff
362 a0f200 p2 Ss 0:00.46 -csh (tcsh)
403 a0f800 p2 S 0:00.25 -csh (csh)
414 a0fe00 p2 R+ 0:00.23 ./reboot-quick
379 a0fa00 p3 IWs+ 0:00.06 /usr/local/libexec/emacs/19.30/i386-unknown-net
308 87fa00 00 IWs+ 0:00.07 /usr/libexec/getty std.9600 tty00
300 9f3800 v0 IWs+ 0:00.06 /usr/libexec/getty console ttyv0
301 9f3600 v1 IWs+ 0:00.06 /usr/libexec/getty console ttyv1
302 9f3400 v2 IWs+ 0:00.06 /usr/libexec/getty console ttyv2
303 9f3200 v3 IWs+ 0:00.06 /usr/libexec/getty console ttyv3
304 9f3000 v4 IWs+ 0:00.07 /usr/libexec/getty console ttyv4
305 9f3c00 v5 IWs+ 0:00.07 /usr/libexec/getty console ttyv5
306 9f3e00 v6 IWs+ 0:00.07 /usr/libexec/getty console ttyv6
pattern:/var/crash % gdb -k netbsd.2*
GDB is free software and you are welcome to 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.
GDB 4.11 (i386-netbsd), Copyright 1993 Free Software Foundation, Inc...
(no debugging symbols found)...
#0 0xfb2e3d60 in end ()
(kgdb) where
#0 0xfb2e3d60 in end ()
#1 0x10 in ?? ()
(kgdb) proc 0xf8a0f800
(kgdb) where
#0 0xf811bda9 in mi_switch ()
#1 0xf811b7cd in bpendtsleep ()
#2 0xf81186de in sys_sigsuspend ()
#3 0xf81af510 in syscall ()
(kgdb) proc 0xf8a0fe00
(kgdb) where
#0 0xfb2e3d60 in end ()
#1 0x10 in ?? ()
(kgdb) quit
State-Changed-From-To: open->closed
State-Changed-By: jtk
State-Changed-When: Fri Aug 22 18:49:31 PDT 1997
State-Changed-Why:
the problem for which this was reopened is distinct from the original report.
please file a separate PR.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.