Re: [tcpinc] TCP's treatment of data in SYN packets

[David Mazieres <[e-mail]>]

Kyle Rose <[e-mail]> writes:

> This is really good information: thanks! I think if tcpinc decides to
> pursue early data, we'll want to learn as much as possible from your
> experience.

So just to scope what we had in mind, "pursue early data" sounds a bit
aggressive.  What I think we had in mind is just saying something about
what receivers MUST do when receiving a SYN segment with data, not
encouraging people to do that.

Since TFO will eventually cause people to address these middlebox
issues, it would be nice to avoid creating a second round of problems
where TCP-ENO implementations also crash or do weird stuff.  And it
would also be nice if TCP-ENO didn't disable this possible optimization.

What I had in mind was something along the lines of saying that the last
TEP suboption in host A's SYN segment is the SYN TEP, and that host B
MUST ignore and MUST NOT ACK any data in SYN segments unless the
negotiated TEP is the SYN TEP and that TEP explicitly defines the use of
data in SYN segments.  Furthermore, SYN data MUST be discarded in the
event that TCP-ENO is disabled, even when such data has already been
ACKed.  TEPs MAY define the use of data in SYN segments for when they
are the SYN TEP, but SHOULD limit such usage to cases in which it is
known that the passive opener and network path both support ENO.

David