Re: [pfSense Support] Boot WRAP with etherboot (PXE)

2006-09-26 Thread Rolf Sommerhalder 

On 9/26/06, Scott Ullrich <[EMAIL PROTECTED]> wrote:

Etherbooting + WRAPS do not work.  There have been many discussions of
this in the past on the forum and in the mailing list.


After spending some time on this earlier this year, I finally got
OpenBSD's pxeboot to successfully net-boot OpenBSD on my WRAPs. The
problem was with the Etherboot binary which is included within the
BIOS v1.08 and v1.10 from PCengines. I have sent a corrected and
documented BIOS flash image back to PCengines. However, I am not sure
if a corrected Etherboot binary made it into v1.11 of the BIOS which
is available on pcengines.ch.

Unfortunately, I was not successful at booting FreeBSD over the
networking using PXE. Another WRAP user reported to me that my
proposed BIOS update helped in advancing the boot process of FreeBSD a
bit further, but there seems to be yet another problem which prevents
FreeBSD from net-booting successfully.

Regards,
Rolf

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[pfSense Support] Re: trap 12: page fault while in kernel mode

2006-09-16 Thread Rolf Sommerhalder 

Following up on this thread from earlier this year, I installed the
recent debug kernel
  http://www.pfsense.com/~sullrich/debugging_kernels/1.0-RC2/kernel.gz
on my Nokia IP120 and then ran some ddb commands after the kernel
trapped while booting.

Can anyone please provide some guidance how to further debug this
problem of FreeBSD 6.1 not booting on Nokia Geode plattforms?

Thanks,
Rolf


GDB: no debug ports present
KDB: debugger backends: ddb
KDB: current backend: ddb
Copyright (c) 1992-2006 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
   The Regents of the University of California. All rights reserved.
FreeBSD 6.1-RELEASE-p3 #0: Tue Aug 22 16:59:36 UTC 2006
   [EMAIL PROTECTED]:/usr/obj/usr/src/sys/pfSense_wrap.6.DEBUG
Timecounter "i8254" frequency 1193182 Hz quality 0
CPU: Geode(TM) Integrated Processor by National Semi (266.68-MHz 586-class CPU)
 Origin = "Geode by NSC"  Id = 0x540  Stepping = 0
 Features=0x808131
real memory  = 134217728 (128 MB)
avail memory = 121913344 (116 MB)
wlan: mac acl policy registered
ath_hal: 0.9.16.16 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413, RF5413)
cpu0 on motherboard
pcib0:  pcibus 0 on motherboard
pir0:  on motherboard
pci0:  on pcib0


Fatal trap 12: page fault while in kernel mode
fault virtual address   = 0xeb11f
fault code  = supervisor read, page not present
instruction pointer = 0x20:0xc00eb621
stack pointer   = 0x28:0xc0c209e0
frame pointer   = 0x28:0xc0c209e0
code segment= base 0x0, limit 0xf, type 0x1b
   = DPL 0, pres 1, def32 1, gran 1
processor eflags= interrupt enabled, resume, IOPL = 0
current process = 0 (swapper)
[thread pid 0 tid 0 ]
Stopped at  0xc00eb621: cmpb%cs:0x1(%esi),%bl


db> p
c00eb621


db> show allpcpu
Current CPU: 0

cpuid= 0
curthread= 0xc0942c40: pid 0 "swapper"
curpcb   = 0xc0c20d90
fpcurthread  = none
idlethread   = 0xc14d6780: pid 11 "idle"
APIC ID  = 0
currentldt   = 0x50


db> ps
 pid   proc uid  ppid  pgrp  flag   stat  wmesgwchan  cmd
  19 c156ea3c0 0 0 204 [IWAIT] swi6: +
  18 c14da20c0 0 0 204 [IWAIT] swi6: task queue
   8 c14da4180 0 0 204 [RUNQ] kqueue taskq
  17 c14da6240 0 0 204 [IWAIT] swi2: cambio
  16 c14da8300 0 0 204 [IWAIT] swi5: +
   7 c14daa3c0 0 0 204 [RUNQ] thread taskq
  15 c14dac480 0 0 204 [RUNQ] yarrow
   6 c14f60000 0 0 204 [RUNQ] crypto returns
   5 c14f620c0 0 0 204 [RUNQ] crypto
   4 c14f64180 0 0 204 [RUNQ] g_down
   3 c14d50000 0 0 204 [RUNQ] g_up
   2 c14d520c0 0 0 204 [RUNQ] g_event
  14 c14d54180 0 0 204 [IWAIT] swi3: vm
  13 c14d56240 0 0 20c [IWAIT] swi4: clock
  12 c14d58300 0 0 204 [IWAIT] swi1: net
  11 c14d5a3c0 0 0 20c [Can run] idle
   1 c14d5c480 0 0 200 [INACTIVE] swapper
  10 c14da0000 0 0 204 [RUNQ] ktrace
   0 c0942a200 0 0 200 [CPU 0] swapper



db> set $lines 0
db> allt

Tracing command swi6: + pid 19 tid 100026 td 0xc156f300
fork_trampoline() at fork_trampoline

Tracing command swi6: task queue pid 18 tid 18 td 0xc14d6300
fork_trampoline() at fork_trampoline

Tracing command kqueue taskq pid 8 tid 19 td 0xc14d6180
fork_trampoline() at fork_trampoline

Tracing command swi2: cambio pid 17 tid 100010 td 0xc14d6000
fork_trampoline() at fork_trampoline

Tracing command swi5: + pid 16 tid 100011 td 0xc14db300
fork_trampoline() at fork_trampoline

Tracing command thread taskq pid 7 tid 100012 td 0xc14db180
fork_trampoline() at fork_trampoline

Tracing command yarrow pid 15 tid 100013 td 0xc14db000
fork_trampoline() at fork_trampoline

Tracing command crypto returns pid 6 tid 100014 td 0xc14d7d80
fork_trampoline() at fork_trampoline

Tracing command crypto pid 5 tid 100015 td 0xc14d7c00
fork_trampoline() at fork_trampoline

Tracing command g_down pid 4 tid 100016 td 0xc14d7a80
fork_trampoline() at fork_trampoline

Tracing command g_up pid 3 tid 10 td 0xc14d7000
fork_trampoline() at fork_trampoline

Tracing command g_event pid 2 tid 11 td 0xc14d6d80
fork_trampoline() at fork_trampoline

Tracing command swi3: vm pid 14 tid 12 td 0xc14d6c00
fork_trampoline() at fork_trampoline

Tracing command swi4: clock pid 13 tid 13 td 0xc14d6a80
fork_trampoline() at fork_trampoline

Tracing command swi1: net pid 12 tid 14 td 0xc14d6900
fork_trampoline() at fork_trampoline

Tracing command idle pid 11 tid 15 td 0xc14d6780
fork_trampoline() at fork_trampoline

Tracing command swapper pid 1 tid 16 td 0xc14d6600
fork_trampoline() at fork_trampoline

Tracing command ktrace pid 10 tid 17 td 0xc14d6480
fork_trampoline() at fork_trampoline

Tracing command swapper pid 0 tid 0 td 0xc0942c40
kernbase

Re: [pfSense Support] Re: trap 12: page fault while in kernel mode

2006-06-05 Thread Rolf Sommerhalder 

From the FreeBSD Problem Report filed by Molle

http://www.freebsd.org/cgi/query-pr.cgi?pr=i386/98215
I speculated that eventually probing the Broadcom crypto co-processor
leads to the trap while probing the PCI bus.

Therefore, I disabled the device "ubsec" in
/home/pfsense/tools/builder_scripts/conf/pfSense.6 and rebuilt a
kernel with
sh /home/pfsense/tools/builder_scripts/build_iso.sh

This results in an updated .iso which installs successfully on the
Nokia harddisk in another i386 host. That host also boots successfully
from that Nokia harddisk. However, the Nokia platform still traps at
the same location after transferring that harddisk back.

Trying now to rebuild another kernel with debug support, and to save a
core dump on harddisk for post-mortem debugging with gdb.

Rolf

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [pfSense Support] Re: trap 12: page fault while in kernel mode

2006-06-04 Thread Rolf Sommerhalder 

After stumbling across the very same problem when trying BETA4 on a
Nokia IP120, I am currently installing the development environment on
another i386 platform in order to attempt to build a patched
kernel/.iso that will boot on the Nokia hopefully without panicing, or
provide some some debug output (enable debug symbols, core dump, ddb,
etc.).

I will let the list know about the outcome once the update & build
completes in a few hours hopefully - I had to install the build
environment on an old P-II after I failed to get cvsup working
correctly from within a virtual machine under VMware from behind a firewall...

Regards,
Rolf

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]