"Kevin M. Lahey" writes:
> In message <199701171845.TAA14555@vader.runit.sintef.no>, 
> Havard.Eidnes@runit.sintef.no writes:
> >is there any work afoot to bring Path MTU discovery to NetBSD's
> >TCP implementation?
> 
> It seems like a great idea.  I've talked about this with different 
> NetBSD folks, and I've yet to hear of anybody working on it.
> I'm interested in doing it, but have yet to write any code for it,
> so don't let me stop anybody else from doing it.
> 
> It sure seems like path MTU discovery is an important thing to have
> in a modern TCP/IP implementation...

SACK is important, too...

Perry