From ortmann@sparc.isl.net  Fri Apr 18 20:38:51 1997
Received: from watcher.isl.net (ppp-83.isl.net [199.3.25.132])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA23456
          for <FreeBSD-gnats-submit@freebsd.org>; Fri, 18 Apr 1997 20:38:48 -0700 (PDT)
Received: (from ortmann@localhost)
	by watcher.isl.net (8.8.5/8.8.5) id KAA00557;
	Fri, 18 Apr 1997 10:44:47 -0500 (CDT)
Message-Id: <199704181544.KAA00557@watcher.isl.net>
Date: Fri, 18 Apr 1997 10:44:47 -0500 (CDT)
From: Daniel Ortmann <ortmann@sparc.isl.net>
Reply-To: ortmann@sparc.isl.net
To: FreeBSD-gnats-submit@freebsd.org
Subject: another kernel panic
X-Send-Pr-Version: 3.2

>Number:         3328
>Category:       kern
>Synopsis:       another kernel panic
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    dyson
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Apr 18 20:40:03 PDT 1997
>Closed-Date:    Tue Apr 14 12:28:24 PDT 1998
>Last-Modified:  Tue Apr 14 12:29:18 PDT 1998
>Originator:     Daniel Ortmann
>Release:        FreeBSD 3.0-CURRENT i386
>Organization:
N/A
>Environment:

EISA, Adaptec 1742 SCSI, Mach32 vidwo, 48MB RAM/80MB swap, 28.8kbps
modem, tcp/ip networking using /usr/sbin/pppd, 11 vty's with getty's
xdm/X on the 12th, several logins including a couple as root, NO
activity on X, one "make install" in /usr/ports/math/, with my config
file's "maxusers" set to 32.  (The system wasn't really loaded very
heavily.)

Here is my current config file:

machine       "i386"
cpu           "I486_CPU"
ident         WATCHER
maxusers      32
options       INET
options       FFS
options       MSDOSFS
options       "CD9660"
options       PROCFS
options       "COMPAT_43"
options       SCSI_DELAY=15
options       BOUNCE_BUFFERS
options       UCONSOLE
options       FAILSAFE
options       USERCONFIG
options       VISUAL_USERCONFIG
options       SYSVSHM
options       SYSVSEM
options       SYSVMSG
options       BROKEN_KEYBOARD_RESET
options       FAILSAFE
options       USER_LDT
options       COMPAT_LINUX
options       IPFIREWALL
options       IPFIREWALL_VERBOSE
options       IPFIREWALL_VERBOSE_LIMIT=10
options       TCPDEBUG
options       SCSI_REPORT_GEOMETRY
config        kernel root on sd0 swap on sd1 dumps on sd1
controller    isa0
controller    eisa0
controller    fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 vector fdintr
disk          fd0 at fdc0 drive 0
disk          fd1 at fdc0 drive 1
controller    ahb0
controller    scbus0
device        sd0
device        st0
device        cd0
device        sc0 at isa? port "IO_KBD" tty irq 1 vector scintr
options       XSERVER
options       FAT_CURSOR
device        npx0 at isa? port "IO_NPX" irq 13 vector npxintr
device        apm0 at isa? disable
options       APM_BROKEN_STATCLOCK
device        sio0 at isa? port "IO_COM1" tty irq 4 vector siointr
device        sio1 at isa? port "IO_COM2" tty irq 3 vector siointr
device        sio2 at isa? port "IO_COM3" tty irq 9 vector siointr
device        sio3 at isa? port "0x368" tty irq 10 vector siointr
device        sio4 at isa? port "0x268" tty irq 12 vector siointr
device        lpt0 at isa? port "IO_LPT1" tty irq 7 vector lptintr
device        lpt1 at isa? port "IO_LPT2" tty irq 5 vector lptintr
pseudo-device loop
pseudo-device ether
pseudo-device log
pseudo-device ppp 1
pseudo-device tun 1
pseudo-device pty 32
pseudo-device gzip
pseudo-device bpfilter 4
pseudo-device speaker
pseudo-device snp 4
options       KTRACE

>Description:

While filling out another problem report about a kernel panic THIS
panic occurred.

I had one background "make install" process occurring in /usr/proc/math/.
The system was lightly loaded, with xdm and X running.

It seems that "make" tends to goof of the process queuing somehow.  I've
seen processes locked up (especially /bin/ps although the procfs was
ok).  I've also experienced periodic blockages of sendmail, and misc
other binaries.

Here's the panic message; I recognize it from previous panics:

panic: page fault

syncing disks...

Fatal trap 12: page fault while in kernel mode
fault virtual address	= 0x44
fault code	= supervisor read, page not present
instruction pointer	= 0x8:0xf010d510
stack pointer	= 0x10:0xf5b68de8
frame pointer	= 0x10:0xf5b68de8
code segment	= base 0x0, limit 0xfffff, type 0x1b
	= DPL 0, pres 1, def32 1, gran 1
processor eflags	= interrupt enabled, resume, IOLL = 0
current process	= 4 (update)
interrupt mask	=
panic: page fault

dumping yadayadayaydayadayadayada

>How-To-Repeat:

See the above descriptions and environment.  Here's what happened this
time:
 - I was running a "make install" as root in /usr/ports/make/
 - pppd died
 - I restarted it
 - it took several seconds so I did <alt> F1 back to vty 0
I was in time to see the panic message listed above.

This started happening in early April on -current.

>Fix:

Unknown.
>Release-Note:
>Audit-Trail:

From: j@uriah.heep.sax.de (J Wunsch)
To: ortmann@sparc.isl.net
Cc: FreeBSD-gnats-submit@freebsd.org
Subject: Re: kern/3328: another kernel panic
Date: Sat, 19 Apr 1997 09:32:05 +0200

 As Daniel Ortmann wrote:
 
 > Here's the panic message; I recognize it from previous panics:
 
 Worthless as it stands...
 
 > panic: page fault
 > 
 > syncing disks...
 > 
 > Fatal trap 12: page fault while in kernel mode
 > fault virtual address	= 0x44
 > fault code	= supervisor read, page not present
 > instruction pointer	= 0x8:0xf010d510
 
 	nm /kernel | sort | more
 
 and lookup the function corresponding to 0xf010d510.
 
 p.s.: I think filing PRs for a quickly moving -current shouldn't be
 your first reaction.  Better report it to the freebsd-current mailing
 list, including the panic information, and only file a PR if no short-
 term solution will be seen.  Also, always report the exact source date
 for -current.
 
 -- 
 cheers, J"org
 
 joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
 Never trust an operating system you don't have sources for. ;-)

From: Daniel Ortmann <ortmann@sparc.isl.net>
To: joerg_wunsch@uriah.heep.sax.de
Cc: ortmann@sparc.isl.net, FreeBSD-gnats-submit@freebsd.org
Subject: Re: kern/3328: another kernel panic
Date: Sat, 19 Apr 1997 00:57:03 -0500 (CDT)

 > As Daniel Ortmann wrote:
 > 
 > > Here's the panic message; I recognize it from previous panics:
 > 
 > Worthless as it stands...
 > 
 > > panic: page fault
 > > 
 > > syncing disks...
 > > 
 > > Fatal trap 12: page fault while in kernel mode
 > > fault virtual address	= 0x44
 > > fault code	= supervisor read, page not present
 > > instruction pointer	= 0x8:0xf010d510
 > 
 > 	nm /kernel | sort | more
 > 
 > and lookup the function corresponding to 0xf010d510.
 
 Point taken.  Here's more information.  Does this help?
 
 f010cf34 T _lkmexists
 f010cf94 t __lkm_syscall
 f010d08c t __lkm_vfs
 f010d1f4 t __lkm_dev
 f010d2f8 t __lkm_exec
 f010d3c4 t _lkm_exec_dummy1
 f010d3cc t _lkm_exec_dummy2
 f010d3d4 t _lkm_exec_dummy3
 f010d3dc t _lkm_exec_dummy4
 f010d3e4 t ___set_execsw_set_sym_lkm_exec_dummy1
 f010d3e8 t ___set_execsw_set_sym_lkm_exec_dummy2
 f010d3ec t ___set_execsw_set_sym_lkm_exec_dummy3
 f010d3f0 t ___set_execsw_set_sym_lkm_exec_dummy4
 f010d3f4 T _lkmdispatch
 f010d47c T _lkm_nullcmd
 f010d484 t _lkm_drvinit
 f010d4b8 t ___set_sysinit_set_sym_lkmdev_sys_init
 f010d4c0 F kern_lock.o
 f010d4c0 T _lockinit
 f010d508 T _lockstatus        <<<=== this appears to be the function.
 f010d634 T _lockmgr
 f010db88 T _lockmgr_printinfo
 f010dbf0 F kern_lockf.o
 f010dbf0 T _lf_advlock
 f010dd8c t _lf_setlock
 f010e0f8 t _lf_clearlock
 
 > p.s.: I think filing PRs for a quickly moving -current shouldn't be
 > your first reaction.  Better report it to the freebsd-current mailing
 > list, including the panic information, and only file a PR if no short-
 > term solution will be seen.  Also, always report the exact source date
 > for -current.
 
 Point again taken; I'll resubscribe to the list, mea culpa.  (I've
 been impressed with the stability of -current and when I began seeing
 this panic persistenly ... that's when I decided to send it.)
 
 The source date had all of the available ctm patches applied and
 was entirely built using make world followed by a kernel rebuild
 completed on Sunday, April 13, 1997 at 2am.
 
 -- 
 Daniel Ortmann       507.288.7732 (h)  ortmann@isl.net
 2414 30 av NW, #D    507.253.6795 (w)  ortmann@vnet.ibm.com
 Rochester, MN 55901  "PERL: The Swiss Army Chainsaw"
State-Changed-From-To: open->feedback 
State-Changed-By: joerg 
State-Changed-When: Sun Aug 24 11:17:08 MEST 1997 
State-Changed-Why:  

Does the problem still happen?  Many things have been done to the 
lock manager lately. 


Responsible-Changed-From-To: freebsd-bugs->dyson 
Responsible-Changed-By: joerg 
Responsible-Changed-When: Sun Aug 24 11:17:08 MEST 1997 
Responsible-Changed-Why:  
John's Mr. Lock Manager these days. 
State-Changed-From-To: feedback->closed 
State-Changed-By: phk 
State-Changed-When: Tue Apr 14 12:28:24 PDT 1998 
State-Changed-Why:  
timed out 
>Unformatted:
