[2949] in linux-net channel archive
Re: More TCP speed fixes
daemon@ATHENA.MIT.EDU (Alan Cox)
Sat May 18 17:19:04 1996
From: alan@lxorguk.ukuu.org.uk (Alan Cox)
To: linux-vger@wab-tis.rabobank.nl
Date: Sat, 18 May 1996 17:46:49 +0100 (BST)
Cc: hjl@innovix.com, schenk@rnode84.cs.toronto.edu, linux-net@vger.rutgers.edu,
linux-kernel@vger.rutgers.edu, alan@cymru.net,
Linus.Torvalds@cs.helsinki.fi
In-Reply-To: <m0uKhRM-0003lPC@sys3.pe1chl.ampr.org> from "Rob Janssen reading Linux mailinglist" at May 18, 96 10:32:44 am
> - make sure the timers tick only between packets.
> requires a lot of coupling to the lower layers (PPP, SLIP).
Too expensive to implement.
> - adjust such "fixed thresholds" depending on the bitrate of the link.
> doesn't always work, because the first visible link may be faster
> than further links in the chain.
The bitrate of a link is a very hard thing to measure and for stuff like
AX.25 its wildly inaccurate anyway.
> This can be done by replacing the "delayed ack" timer with a "drop of
> DCD" event handler in that case. The "drop of DCD" event is of course
> only available on a half-duplex interface.
For a general purpose stack we dont have the instruction time to waste on this
kind of (very effective Im sure for AX.25) hackery.
Alan
