Very Computer

I've hit a tough one this time, folks.  I run FreeBSD on several large
servers at work, but my home system has me baffled.  Here it is:

386-DX25 clone (with i387), 8meg RAM, AMI 1990 BIOS with all BIOS defaults.
Adaptec 1542CF, address 330, irq 11, drq 5.
IBM 400 meg ID#0, IBM 160 meg ID#1, Toshiba CDROM ID#2.
Standard VGA, mse0 mouse, sio0, 1, and 2 (on IRQ 9).
SMC NE2000 clone card at 280, irq 10, d800.

This system has run 2.1.0 for over 6 months without ANY problems.
I just brought it home from work, checked it out, and then started 2.1.5R.

I setup the first drive with 20meg /, 16meg swap, and the rest /usr.
I setup the second drive with 64meg swap, and the rest /var.
The same configuration as before under 2.1.0.

The kernel panics after the "aha0 waiting for SCSI devices to settle" msg:

Fatal trap 12: page fault while in kernel mode
fault virtual address = 0x6
fault code            = supervisor read, page not present
instruction pointer   = 0x8: 0xf051cad8
code segment          = base 0x0, limit 0xfffff, type 0x16
                      = DPL 0, pres 1, def 32 1, gran 1
processor eflags      = interrupt enabled, resume, IOPL = 0
current process       = 0 ()
interrupt mask        = net tty bio
panic: page fault

I have tried this with and without booteasy, and with and without a 10meg
DOS partition to hold DOS and the ethernet card diagnostics programs.

I have low-level formatted both drives with the Adaptec CTRL-A menu, and
all the diags run clean.

This panic first showed up as a different message AFTER the sysinstall
was underway doing:
   newfs -b 8192 -f 1024 /mnt/dev/rsd0s1e, and then I got:
   panic: page fault
   syncing disks...

Now I can't even get to the point of the sysinstall menu.

Any ideas?
Thanks, Tyler Barnett

I don't have a genuine 2.1.5 kernel handy, so i cannot check out the
coredump addresses right now.

--
cheers, J"org

Never trust an operating system you don't have sources for. ;-)