On 10/8/2015 5:36 PM, Rick Jones wrote:
> GRO, TSO, CKO etc have come into being from the fields made fertile by
> the combination of ever increasing link speeds, never increasing de jure
> Ethernet MTUs, no longer increasing processor core frequency, and paying
> customer demand for performance.

Right, and we're just starting to see places where they interfere with
TCP. I've been doing tests on a new TCP option to extend the space
available for options.

We've already seen cases where these extensions aggregate segments that
have different unknown options (which should clearly never happen), or
even when they aggregate segments with unknown options that match (which
I believe should never happen either).

Again, just because nobody's screamed to date doesn't mean these are
known safe.

Joe

_______________________________________________
aqm mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/aqm

Reply via email to