bpf writes on tun device

[Matthew Luckie]

> I'd suggest a name like DLT_PSEUDO might be better -- it may be helpful to
> get review for the change from the NetBSD and OpenBSD guys too, as well as
> the tcpdump.org guys.

I posted a patch to netbsd tech-kern and have had positive feedback:

http://mail-index.netbsd.org/tech-net/2005/06/09/0003.html

I posted to tcpdump-workers and have had no response at this time:

http://marc.theaimsgroup.com/?l=tcpdump-workers&m=111818619022560&w=2

I do, however, feel that this is a safe patch to apply, and one that I'd 
really like to see make it to 6.0.

> Looking at style, it might be better if the driver(s) were changed to
> explicitly use a 32-bit wide int type such as u_int32_t for the address
> family header field in their bpfattach() calls. ICHDRLEN is odd man out,
> but it is #define'd to be the same thing; I would update it there also.

I'll do some house keeping on the patch and then re-send it.  I take it 
a PR is the right place to send the patch?

> A white space pass should probably also be done last thing to be sure.
> I have no idea about ng_sppp.c I'm afraid.

I'll make a note in the PR about ng_sppp.c; but there does not appear to 
be any output function specific to ng_sppp so I think it is safe to 
leave alone.

Thanks

Matthew