From owner-cvs-all  Wed Mar 12 14:16: 7 2003
Delivered-To: cvs-all@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP
	id 9E5AB37B401; Wed, 12 Mar 2003 14:16:05 -0800 (PST)
Received: from elvis.mu.org (elvis.mu.org [192.203.228.196])
	by mx1.FreeBSD.org (Postfix) with ESMTP
	id 067C843F75; Wed, 12 Mar 2003 14:16:05 -0800 (PST)
	(envelope-from jayanth@elvis.mu.org)
Received: by elvis.mu.org (Postfix, from userid 1092)
	id D8F732ED3F1; Wed, 12 Mar 2003 14:16:04 -0800 (PST)
Date: Wed, 12 Mar 2003 14:16:04 -0800
From: Jayanth Vijayaraghavan <jayanth@elvis.mu.org>
To: Hiten Pandya <hiten@unixdaemons.com>
Cc: Jeffrey Hsu <hsu@FreeBSD.ORG>, src-committers@FreeBSD.ORG,
	cvs-src@FreeBSD.ORG, cvs-all@FreeBSD.ORG
Subject: Re: cvs commit: src/sys/netinet tcp_input.c
Message-ID: <20030312221604.GA997@elvis.mu.org>
References: <200303122027.h2CKRSwR026081@repoman.freebsd.org> <20030312210107.GA56887@unixdaemons.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20030312210107.GA56887@unixdaemons.com>
User-Agent: Mutt/1.4i
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk
List-ID: <cvs-all.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20cvs-all>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20cvs-all>
X-Loop: FreeBSD.ORG

On Wed, Mar 12, 2003 at 04:01:07PM -0500, Hiten Pandya wrote:
> Jeffrey Hsu (Wed, Mar 12, 2003 at 12:27:28PM -0800) wrote:
> > hsu         2003/03/12 12:27:28 PST
> > 
> >   FreeBSD src repository
> > 
> >   Modified files:
> >     sys/netinet          tcp_input.c 
> >   Log:
> >   Implement the Limited Transmit algorithm (RFC 3042).
> 
> Can you elaborate on this?
> 
>   -- Hiten
One of the main uses of the algorithm is when the congestion window is small
by sending new data for every duplicate acknowledgement you end up 
forcing the other end to send another duplicate ack. 
This way you can retransmit the lost segment quickly (fast retransmit).

If you did not have the this algorithm and have a small congestion window you
could end up with only 2 duplicate acknowledgements rather than 3.
This could have the adverse effect of forcing a retransmit timeout rather than
a fast retransmit. 


jayanth

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