[1514] in linux-net channel archive
Re: double acks?
daemon@ATHENA.MIT.EDU (Rob Janssen reading Linux mailingl)
Sun Dec 10 07:34:36 1995
From: linux@pe1chl.ampr.org (Rob Janssen reading Linux mailinglist)
To: jmorriso@multiactive.com (John Paul Morrison)
Date: Sat, 9 Dec 1995 13:12:50 +0100 (MET)
Cc: linux-vger@wab-tis.rabobank.nl, erb@inss1.etec.uni-karlsruhe.de,
linux-net@vger.rutgers.edu
In-Reply-To: <199512072250.OAA22438@orange.ConcordPacific.Com> from "John Paul Morrison" at Dec 7, 95 02:50:54 pm
Reply-To: linux-vger@wab-tis.rabobank.nl
According to John Paul Morrison:
> [ double ACKs ]
> >
> > Anyway, there was a patch posted on the linux-net list a few weeks ago, which
> > sends the second ACK only when the first one had sent a very small window.
> > This is the same as the KA9Q code does. (it only sends the second ACK
> > when the first one had sent window 0)
>
> Anyone know why it didn't get included into the new kernels?
> It looked like a small patch, but I suppose it could break something.
Alan says it can cause deadlock, but I don't believe that.
(delivery of an ACK is not reliable, so omitting an occasional ACK
can never cause deadlock in a correct TCP implementation. more detail
supplied in a personal message to Alan)
Rob
--
+------------------------------------+--------------------------------------+
| Rob Janssen rob@knoware.nl | BBS: +31-302870036 (2300-0730 local) |
| AMPRnet: rob@pe1chl.ampr.org | AX.25 BBS: PE1CHL@PI8WNO.#UTR.NLD.EU |
+------------------------------------+--------------------------------------+
