"Perry E. Metzger" <perry@piermont.com> writes:

> I don't know, but SACK is one of the most important developments in
> TCP in years. In conditions where more than a couple of packets in a
> row drop, ordinary TCP dies a horrible death, but TCP with SACK keeps
> on working. SACK is about to become an IETF standard, and it would be
> a Very Very Very Good Thing if we put it into our kernels.

Do both sides of the link have to be SACK aware, or if just one is will that
be better than nothing?

--Michael