Date: Fri, 17 Apr 1998 08:03:15 +0200 (MET DST) From: Luigi Rizzo <luigi@labinfo.iet.unipi.it> To: julian@whistle.com (Julian Elischer) Cc: current@FreeBSD.ORG, kjc@csl.sony.co.jp Subject: Re: Networking strategy for -current Message-ID: <199804170603.IAA08337@labinfo.iet.unipi.it> In-Reply-To: <353699BD.ABD322C@whistle.com> from "Julian Elischer" at Apr 16, 98 04:52:10 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> One usage that people have often asked for is the ability to allocate > bandwidth, and to re-visit the queuing strategies we use. right... I have been thinking for a long time to move my dummynet stuff down a level (now it's between TCP and IP) and hook it to the IPFW code so that one can use the same filtering as IPFW and use bw limitation. Haven't had the time to look at ALTQ, obviously if it can do hard BW limiting (and not only WFQ) that would be great (the only advantage of dummynet at this point would be the ability to add programmable delays to simulate long pipes...). > Sony labs have recently announced their ALTQ package in a new release. > I would like to find out from the people involved in networking, what > the general consensus would be to our integrating this into -current. > (under #ifdefs of course) I'd like the idea a lot. > 2/ The CBQ method Sony supply has it's own packet classifier > that is basically a duplicate of the IPFW or ipfilter > classifiers. It might be worth us looking to see if the IPFW > (or ipfilter) code can be used to do this. I would do this by this question can be turned to the author of ALTQ... My understanding is that if he wants a portable package, then the classifier must be within ALTQ -- in case some other OS does not have the classifier. In a sense this would be ok since we could remove our own IPFW classifier and use the ALTQ instead (the issues are efficiency -- i have no idea on this -- and backward compatibility -- but this can be solved at the user level perhaps, in the ipfw control program). > expanding the mbuf pkthdr field to contain a pointer to some meta-data > such as 'traffic class'. This could be set by each protocol a flow ID you mean ? > in any way that makes sense to it. (e.g. by a new IPFW command) > (instead of a special pass). 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?199804170603.IAA08337>