Date: Tue, 17 Feb 1998 21:10:57 +0100 (MET) From: Luigi Rizzo <luigi@labinfo.iet.unipi.it> To: eivind@yes.no (Eivind Eklund) Cc: eivind@yes.no, plm@xs4all.nl, freebsd-current@FreeBSD.ORG Subject: Re: cvs commit: src/usr.sbin/ppp command.c datalink.c datalink.h defs.h vars.c vars.h Message-ID: <199802172010.VAA01212@labinfo.iet.unipi.it> In-Reply-To: <19980217222533.63111@follo.net> from "Eivind Eklund" at Feb 17, 98 10:25:14 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > * "preemption" . > > I don't think ppp allows this, but perhaps it could be possible to > > implement preemption of long packets (or transfer them in smaller > > 'cells') so that interactive traffic does not suffer exceedingly ... > I'm not certain how this can be handled. A possibility might be to > fragment the packets for just that connection, ie, in effect lower the > MTU for just part of the use. > > I believe some smart heuristics would be necessary to make this work well. the idea would be that the (non-standard) PPP link can be treated to transfer cells of a fixed max size, each cell tagged with the flow it belongs to. The sender does the fragmentation so the FQ algorithm uses cells instead of packets. The whole things could be (probably) optimized if we can make use of some unused ppp escape sequence to mark a preemption point, rather than always fragment packets. cheers luigi -----------------------------+-------------------------------------- Luigi Rizzo | Dip. di Ingegneria dell'Informazione email: luigi@iet.unipi.it | Universita' di Pisa tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ _____________________________|______________________________________ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802172010.VAA01212>