From owner-freebsd-net Wed Jun 13 17:13:52 2001 Delivered-To: freebsd-net@freebsd.org Received: from imo-m07.mx.aol.com (imo-m07.mx.aol.com [64.12.136.162]) by hub.freebsd.org (Postfix) with ESMTP id 90CA737B401 for ; Wed, 13 Jun 2001 17:13:49 -0700 (PDT) (envelope-from FastPathNow@netscape.net) Received: from FastPathNow@netscape.net by imo-m07.mx.aol.com (mail_out_v30.22.) id x.af.59cb2b (16215); Wed, 13 Jun 2001 20:13:09 -0400 (EDT) Received: from netscape.com (aimmail02.aim.aol.com [205.188.144.194]) by air-in01.mx.aol.com (v78_r3.8) with ESMTP; Wed, 13 Jun 2001 20:13:09 -0400 Date: Wed, 13 Jun 2001 20:13:09 -0400 From: FastPathNow@netscape.net To: luigi@info.iet.unipi.it Cc: jlemon@flugsvamp.com, FastPathNow@netscape.net, net@freebsd.org, bright@rush.net Subject: Re: SACK code patch for FreeBSD 4.3 Mime-Version: 1.0 Message-ID: <7C45F6A6.77A905CF.375A6AF3@netscape.net> References: <200106140001.CAA97906@info.iet.unipi.it> X-Mailer: Franklin Webmailer 1.0 Content-Type: text/plain; charset="ISO-8859-1" Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Thanks to all who responded. Luigi, I will also have a look at the code that you posted. I did have a look at the code from the UCB Daedalus project, which was based on the BSDI 2.0 code. I am currently basing most of my code changes on that work. The changes broadly relate to the relatively new usage of FreeBSD timer-wheel callouts, and some changes due to the way some of the structs are defined and used in FreeBSD. Apart from that I think there might be a couple of small bugs in the code, but I would want to confirm that from the original authors before I commit any changes for the same. Lastly, I also would like to focus on what TCP should do on receiving its 3rd DUPACK (modified fast rexmit and recovery), in the presence of SACK. Recently a draft was posted about this ... http://www.globecom.net/ietf/draft/draft-allman-tcp-sack-02.html . I am planning on spending some time to see, to what extent the old code conforms to/ is different from the recovery algorithm suggested in the draft above. Please let me know if you have other ideas / suggestions Regards -AG --- Luigi Rizzo wrote: > > Looooong ago (1996, sic!) i did some work on this, which you can find at > >     http://www.iet.unipi.it/~luigi/sack.html > > The retransmission logic probably need to be updated, also in relation > to the newer SACK RFC issued i think sometime last year. But the boring > part of the code which handles the list of segments to be SACKed > may be of good use. > > It should not be too hard to port this code to 4.x > >     cheers >     luigi > > > >Hi > > >I am currently working on implementing SACK on FreeBSD 4.3 (STABLE) . At > > >some point in the future I plan to contribute this patch to the FreeBSD > > -----------------------------------+------------------------------------- >   Luigi RIZZO, luigi@iet.unipi.it  . Dip. di Ing. dell'Informazione >   http://www.iet.unipi.it/~luigi/  . Universita` di Pisa >   TEL/FAX: +39-050-568.533/522     . via Diotisalvi 2, 56126 PISA (Italy) >   Mobile   +39-347-0373137 > -----------------------------------+------------------------------------- > __________________________________________________________________ Get your own FREE, personal Netscape Webmail account today at http://webmail.netscape.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message