Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 23 Dec 1996 21:24:50 +0100 (MET)
From:      Luigi Rizzo <luigi@labinfo.iet.unipi.it>
To:        julian@whistle.com (Julian Elischer)
Cc:        proff@iq.org, danny@panda.hilink.com.au, hackers@FreeBSD.ORG
Subject:   Re: ipretard.c selective tcp/ip queues and throughput limiters
Message-ID:  <199612232024.VAA18083@labinfo.iet.unipi.it>
In-Reply-To: <32BEE215.167EB0E7@whistle.com> from "Julian Elischer" at Dec 23, 96 11:48:18 am

next in thread | previous in thread | raw e-mail | index | archive | help
> Luigi Rizzo wrote:

> > very useful. The only drawback is the overhead that this (and basically
> > everything using the divert socket) has: packets move to user space and
> > back, something you don't want to do on high-speed traffic.
.
> but as Kieth sklower at CSRG told me..
> There's got to be a way to make it possible for essoteric or unusual
> modules to be implimented OUT OF THE KERNEL, or
> they are 
> (1) hard to prototype
> (2) increasing the complexity of what IS in the kernel beyond the 
> point of debuggability :)

which is why I like very much the idea of divert sockets, tun devices
and similar things, but only when used in the proper way (e.g.
prototyping and/or slow speeds).

	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/
_____________________________|______________________________________



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612232024.VAA18083>