In article <rmiodv471h2.fsf@fnord.ir.bbn.com>,
Greg Troxel  <gdt@ir.bbn.com> wrote:
>-=-=-=-=-=-
>
>
>  [802.11 frame writing via bpf]
>
>The plan is to merge this work into NetBSD once it works.  We'll also
>be contributing back changes to make tap(4) be able to do
>point-to-point (e.g. like tun(4)) or 802_11_RADIO, but defaulting to
>the current behavior.  The code will of course be posted for review
>before committing.
>
>The remote interface code will also be available, as a separate
>program with a pkgsrc entry.
>
>My own take on the options is that flags for what encap has been done
>feels kludgy, and that having an explicit raw queue is cleaner.

I also feel that an explicit raw queue is cleaner.

christos