> Here's an URL: http://www.computex.com.tw/archive-detail_new.asp?index=3044

Some 'interesting' use of English, and a few non-sequiturs
> 
> At least they make it _sound_ like it should work just fine (especially
> with only a few ports in use).

well the max forwarding rates look like the ethernet limits.
What they don't quote is single packet latency (OTOH I've never seen
it quoted)
> 
> I don't know how switches manage buffer space, but I hope they're not
> stupid enough to allocate 32 KB for ports that are not connected at all..

Well it shouldn't actually matter if 'Half-duplex mode for backpressure,
and full-duplex for flow control' (makes more sense if you read it
backwards!) means what I think it does, and if your board hardware
suuports the flow control protocol when running full duplex.
> 
> Besides #N, my server is still running 1.5R and I haven't even rebooted
> it for ages (204 days).  I can't see why it would suddenly start sending
> out data much faster than before (even though Jason has optimized the
> TCP/IP stack recently :-)

Contagious code fixes?
> 
> It sounds more like a new bottleneck in the new NetBSD versions ran by
> the clients..

Ummm.....


	David

-- 
David Laight: david@l8s.co.uk