Subject: Re: perhaps time to check our TCP against spec?
To: Jonathan Stone <firstname.lastname@example.org>
From: Jason Thorpe <email@example.com>
Date: 04/06/1998 23:19:49
On Mon, 6 Apr 1998 23:15:54 -0700 (PDT)
Jonathan Stone <jonathan@DSG.Stanford.EDU> wrote:
> Sure. And some people are even of the more extreme opinon that `ack
> every packet' is better. Including most of hte authors in the
...and, in fact, delayed ACK is a "may", not a "must", or even a "should".
> But the in_maxmtu stuff: ``It depends''. If you _do_ assume non-PMTU
> hosts, then there *are* circumstances where the behaviour you suggest
> *is* (relative to the old behaviour) broken.
Before I assume _anything_, I want to know which host you're talking about
when you say "assume non-PMTU hosts". Us, or the peer?
> One example is the most-recent topology i posted. And exactly the
> same issue arises if you have a single-homed Ethernet host talking to
> a mobile host with a wireles MTu that's smaller than the wired MTU.
> I still think the in_maxmtu thing is good for PMTU, but for non-PMTU
> peers, it's broken.
> Did you discuss this to anybody from the mobile community?
It doens't really matter if I did or didn't. In your ether+radio
example, even the "traditional" code would have been bounded by the
MTU of the outgoing interface!
(And, for the record, the MOBILEIP chair attents TCPIMPL meetings, and I
go to the MOBILEIP meetings if I don't have a conflicting session.)
> This isn't current-users, Jason. It's fair to say ``the code is
> broken'' here and refer to relased code.
Fine, and I stated quite early on in the thread that I was talking
about NetBSD-current. I don't run releases, personally. I live in
a development world, so I'm constantly running development code. Cope.
Eventually, all of the recent recent recent changes will make it into
the netbsd-1-3 branch for another patch release. 1.3.1 actually included
several TCP fixes.
> And you still haven't answered the topology issues. Are you just
> constitutionally incapable of acknowledging there are problems?
...probably because I was out getting dinner, or something. "Read the
mail I just sent you."
> That's npot what Kevin says.
Funny, it's what the source-changes archive says, assuming you're talking
about what I think you're talking about (what _are_ you talking about,
Oh, I forgot, you don't read source-changes. "Oh well."
Jason R. Thorpe firstname.lastname@example.org
NASA Ames Research Center Home: +1 408 866 1912
NAS: M/S 258-5 Work: +1 650 604 0935
Moffett Field, CA 94035 Pager: +1 415 428 6939