Subject: ipv6 ppp / uvm panic
To: None <port-pc532@netbsd.org, tech-net@netbsd.org, tech-kern@netbsd.org>
From: Jon Buller <jon@bullers.net>
List: tech-kern
Date: 09/10/2001 21:58:31
This is with a pc532 running 1.5 GENERIC.

     Building databases...
     Starting syslogd.
     Starting pppd: bullball.
     uvm_fault(0xfc8e0234, 0x0, 0, 1) -> 1
     fatal abort trap in supervisor mode
     trap type=2, pc=0xf80770db, tear=0x30, msr=0xa2
     panic: trap
     syncing disks... 13 13 13 13 13 12 12 12 11 11 11 11 11 10 10 10 10 10 9 9 9 9 9 9 giving up
     
     dumping to dev 0,1 offset 35863
     dump panic: ncr5380_scsi_cmd: poll didn't finish

I'm going to try and get a dump of this, but I suspect it is operator
error on ipv6 ppp configuration.  I wouldn't expect a uvm panic
from simply screwing ip an ifconfig and ppp options though.  If
anyone want's more information, I'll do what I can to help.

Can anyone tell me what the correct way to assign an ipv6 prefix
to a ppp link is?  Based on the panic, I assume it is NOT to make
an ifconfig.ppp0 file with both endpoint addresses and the prefix
length.  I want the pc532 to have a fixed ipv6 address, so I can
give it an AAAA record, but the machine only has serial ports for
connectivity.  Any hints?

Jon