cvs commit: src/sys/netinet tcp_syncache.c

[Andre Oppermann]

Kip Macy wrote:
> Did you talk to the original submitter? Note that FreeBSD's TCP stack
> is for use in servers and is not intending as a validating TCP stack.
> If you would like it to serve as such you would better served by
> tracking down the ANVL tests that FreeBSD fails. Also note that there
> is no MUST in the following sentence:
> 
> 
> "For simplicity and symmetry, we specify that
>       timestamps always be sent and echoed in both directions."
> 
> So it is clearly open to interpretation.

No, it is not.  RFC1323 was written in 1992 before RFCs contained the
boiler plate definition of MUST, SHOULD, MAY and so on.  I, at least
as a non-native English speaker, find the sentence perfectly clear
and without any doubt.  The IETF TCPM working group comes to the
same conclusion.  And I suppose many native English speakers too.
Despite that arguing over whether "always" lacks a "MUST" to make
it really always always and never not you cited the wrong part of
RFC1323 as reason to completely remove the check.  That's what I'm
complaining about.  Everyone in FreeBSD, including you and me, should
at least provide the correct citation and rationale for any code
change irrespective of the eventual merit of the change itself which
is a separate issue.

-- 
Andre