Date: Wed, 25 Jan 2006 08:22:44 -0800 From: Brooks Davis <brooks@one-eyed-alien.net> To: Andre Oppermann <andre@freebsd.org> Cc: freebsd-current@freebsd.org, Milan Obuch <current@dino.sk> Subject: Re: if_flags usage etc. Message-ID: <20060125162244.GA2848@odin.ac.hmc.edu> In-Reply-To: <43D73395.1020003@freebsd.org> References: <20060124075437.B67285@xorpc.icir.org> <20060124.231504.74682748.sthaug@nethelp.no> <43D72F49.8080304@freebsd.org> <200601250903.01941.current@dino.sk> <43D73395.1020003@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--k1lZvvs/B4yU6o8G Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jan 25, 2006 at 09:15:17AM +0100, Andre Oppermann wrote: > Milan Obuch wrote: > >On Wednesday 25 January 2006 08:56, Andre Oppermann wrote: > > > >>sthaug@nethelp.no wrote: > >> > >>>>We should probably better document the interface "interface". if we a= re > >>>>going to (as Sam suggests) > >>>>do some cleanups we might as well consider what other changes should = be > >>>>put in at the same time. > >>> > >... > > > >>>- One feature sorely missed (which I use a lot in my daily work with > >>>hardware based routers) is the ability to associate a "description" > >>>field with each interface. Note that this should be available both > >>>for physical interfaces (Ethernet etc.) and for logical interfaces > >>>(e.g. vlan). > >> > >>Struct ifnet is the same for all kinds of interfaces, so any change > >>would make it available for everyone. And yes, this looks like a useful > >>addition. > > > >See archives - mailing list freebsd-net, Nov 25. 2005, subject ifconfig= =20 > >description. >=20 > Interesting. Ideally it would not use a static sized buffer for the=20 > description > but a pointer in struct ifnet to a malloc()ed block of memory. This woul= d=20 > allow > for arbitrary sized descriptions. Haven't looked at the difficulties=20 > implementing > the this for the userland/kernel crossing though. See my post to -net a week or two ago suggesting deprecating the ifr_data member in favor of a pointer and a length. I proposed it specificaly to support this feature in a non-broken way. -- Brooks --=20 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 --k1lZvvs/B4yU6o8G Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFD16XUXY6L6fI4GtQRAi/bAJ9PnOH5CSYK3MzkLz6YzMeMpV6HzQCfVdAu 4MwvcDsgfVPDoGslSGC2gBg= =V8aN -----END PGP SIGNATURE----- --k1lZvvs/B4yU6o8G--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060125162244.GA2848>