>the MAC bridge handles the bdpu timeout values wrong.
>due to ieee 802.1d (9.2.8 page 130) these values are 1/256 of a second.
>problems occur if there are bridges that handle these values correct
>the timeout delays are interpreted 256 times as long as they are, i.e.
>3840 sec forward delay instead of 15.
>the other bridge also may get confused, because the linux bridge always
>gives timouts values of 0.
Yeah, spotted this a long time ago. I worked on the bridging stuff
for a while (Even bought 802.1d for the purpose....) But it seemed
to me that nobody was/is using the bridging functionality anymore.
It's still tagged experimental (Alan Cox told me: "because nobody
is/was maintaining it.") Also, with linux moving to SMP systems the
bridging functionality is going to die out probably. It's pretty hard,
maybe impossible, to get the forwarding database (AVL tree)
locking right without making performance suck. An AVL tree which
needs updating on every packet received will probably not scale
to a lot of processors. So you'd either need a new data structure,
or.... Ideas?
(Please CC me, I'm not on the list)
Greetings,
Lennert
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/