More pf work

[YONETANI Tomokazu]

On Thu, Nov 18, 2010 at 12:15:39AM +0100, Jan Lentfer wrote:
> Am 17.11.2010 23:49, schrieb Samuel J. Greear:
> >It has been my experience that panic loops are typically caused by
> >calling a null function pointer. Turning off break to debugger will
> >get you a real trace in this case.
> >
> >It would be great if someone who was more comfortable with kdb and
> >etc. could take a look at my patch on the tracker related to this.
> 
> That backtrace looks strange to me:

It appears that the panic is somewhere inside pf.ko, so you need
the symbols for that module; there's an instruction on how to
obtain them:

  http://www.dragonflybsd.org/docs/developer/KmodDebugging/

Cheers.