tech-net archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: CVS commit: src/sys/net

Mouse <mouse%Rodents-Montreal.ORG@localhost> writes:

>>> Maybe "network interface" is not the right abstraction for bridges
>>> to be instances of, then?  (Though, at least in my experience,
>>> NetBSD's "bridge" interfaces are more like switches than bridges.)
>> A switch is a bridge with more than two ports.

I can buy that definition.

> Then what is my 8-port 10Mb hub?  Switch, bridge, neither?  It has no
> buffering and it always floods all traffic - traffic from port A to a
> machine on port B at the same time as traffic from port C to a machine
> on port D will produce a collision.  This means it is not a switch as
> I've used the term throughout my career (well, throughout after the
> point when devices smart enough to send unicast traffic to only the
> appropriate port appeared).  But it has eight - more than two - ports.

Neither.  It's a hub, which retransmits L1 without decoding L2.

> I also own an AUI multiport transceiver box that's basically the same
> thing, but AUI instead of twisted-pair - it has eight connectors which
> present AUI transceiver interfaces (for connection to hosts) and one
> which presents an AUI host interface (for connection to a transceiver).
> Is it a bridge, a switch, neither, what?  Why?

basically a hub.  Interesting - I never heard of that.

Attachment: signature.asc
Description: PGP signature

Home | Main Index | Thread Index | Old Index