--yLVHuoLXiP9kZBkt
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Mar 02, 2005 at 12:30:10PM -0500, Greg Troxel wrote:
>=20
> Sure.  My real concern is that interface deletion is not safe.  Given
> that these invariants are becoming kind of hairy, I think the only
> safe thing to do (from a software maintenance viewpoint) is to have
> interface destruction always scan all data structures that might have
> ifnbet pointers.

Perhaps I'm reading too much detail into your suggestion. I agree that we=
=20
need to nudge everything that has an ifp, so that it will notice that the=
=20
interface's gone. I think though that the ifdetach routine may not be the=
=20
best place to do it; it may be happening in an interrupt routine. So we do=
=20
enough to get the interface coming down (no new traffic for instnace) then=
=20
start things draining off.

Take care,

Bill

--yLVHuoLXiP9kZBkt
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFCJndQWz+3JHUci9cRAtFkAJ9OxE6Chi6TiHoH5UqIBLPSNA8VrgCgk+G9
WeYpCYkYe4JZxvVGdRfNubc=
=KjOC
-----END PGP SIGNATURE-----

--yLVHuoLXiP9kZBkt--