From owner-freebsd-net  Fri Aug  4 12:44:40 2000
Delivered-To: freebsd-net@freebsd.org
Received: from bubba.whistle.com (bubba.whistle.com [207.76.205.7])
	by hub.freebsd.org (Postfix) with ESMTP id 52C5337B887
	for <freebsd-net@FreeBSD.ORG>; Fri,  4 Aug 2000 12:44:38 -0700 (PDT)
	(envelope-from archie@whistle.com)
Received: (from archie@localhost)
	by bubba.whistle.com (8.9.3/8.9.3) id MAA10454;
	Fri, 4 Aug 2000 12:44:04 -0700 (PDT)
	(envelope-from archie)
From: Archie Cobbs <archie@whistle.com>
Message-Id: <200008041944.MAA10454@bubba.whistle.com>
Subject: Re: cvs commit: src/sys/netinet tcp_output.c
In-Reply-To: <200008041815.OAA33020@khavrinen.lcs.mit.edu> from Garrett Wollman
 at "Aug 4, 2000 02:15:32 pm"
To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu>
Date: Fri, 4 Aug 2000 12:44:04 -0700 (PDT)
Cc: freebsd-net@FreeBSD.ORG
X-Mailer: ELM [version 2.4ME+ PL68 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-net@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org

Garrett Wollman writes:
> > I wonder if this has anything to do with why the default
> > interface max transmit queue is IFQ_MAXLEN == 50. Seems
> > like a rather high number to me. I wonder how they came
> > up with this number.
> 
> The default used to be 100.

Yikes. It seems like a more sensible approach would be to:

 (a) Measure queue length by counting bytes instead of packets

 (b) Define the max queue length in terms of time (IFQ_MAXDELAY?)
     instead of bytes, e.g.:

       ifp->if_maxq = ifp->if_baudrate * IFQ_MAXDELAY

"50 packets" can mean wildy different things to a 14.4 modem
vs. a 100baseT Ethernet.

-Archie

___________________________________________________________________________
Archie Cobbs   *   Whistle Communications, Inc.  *   http://www.whistle.com


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message