Re: vlan panic in -CURRENT

From: Brooks Davis <brooks_at_one-eyed-alien.net>
Date: Fri, 6 Feb 2004 10:33:06 -0800
On Thu, Feb 05, 2004 at 04:11:12PM -0600, Kevin Day wrote:
> 
> This crashes -CURRENT (from a couple of days ago) for me:
> 
> ifconfig vlan0 create
> ifconfig vlan0 vlan 2 vlandev fxp0
> ifconfig vlan0 up
> 
> (immediate trap 12/page fault while in kernel mode crash)
> 
> BUT, if I manually bring up fxp0 first, everything is fine.
> 
> ifconfig fxp0 up
> ifconfig vlan0 create
> ifconfig vlan0 vlan 2 vlandev fxp0
> ifconfig vlan0 up
> 
> I don't have a crash dump since I don't have enough space on this 
> drive. but this should be pretty easy to reproduce. If this isn't 
> reproducible by whoever wants to look at this, I'm happy to get a 
> backtrace through some other means. I'm using a stock GENERIC kernel, 
> except I've added bridging and ipfw.
> 
> Sorry if this has already  been reported, but I don't see anything in 
> gnats or the mailing lists about it.

Well, it's not related to simply upping a vlan on a down interface since
I can't reproduce it with fwe0.  I've got a shortage of systems
configured for debugging that have an extra nic in them so I can't seem
to replicate this.  Since it should be possiable to replicate in single
user mode, you might try setting hw.physmem to something small enough
that you can get a crashdump.

-- Brooks

-- 
Any statement of the form "X is the one, true Y" is FALSE.
PGP fingerprint 655D 519C 26A7 82E7 2529  9BF0 5D8E 8BE9 F238 1AD4

Received on Fri Feb 06 2004 - 09:33:14 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:42 UTC