From owner-freebsd-current Mon Dec 14 15:16:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA28576 for freebsd-current-outgoing; Mon, 14 Dec 1998 15:16:03 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: (from jmb@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA27390; Mon, 14 Dec 1998 15:07:41 -0800 (PST) (envelope-from jmb) Date: Mon, 14 Dec 1998 15:07:41 -0800 (PST) Message-Id: <199812142307.PAA27390@hub.freebsd.org> From: "Jonathan M. Bresler" To: johan@granlund.nu CC: julian@whistle.com, phk@FreeBSD.ORG, lars@akerlings.t.se, current@FreeBSD.ORG, isdn@FreeBSD.ORG In-reply-to: (message from Johan Granlund on Mon, 14 Dec 1998 22:12:29 +0100 (CET)) Subject: Re: if_sppp is BROKEN!!! References: Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > Date: Mon, 14 Dec 1998 22:12:29 +0100 (CET) > From: Johan Granlund > > This is really interesting! > One of the concepts i liked in SYSV (ducking for cover) was streams and > its ability to chain together modules to process a datastream. > If it's coupled with kld to dynamically load/unload modules i think > you have something _very_ good. > > That network thing. Can a module route a package thru different modules > based on contents and state? the idea of streams is wonderful, the realization is costly. each layer added (or module pushed) slows down processing and hurts throughput. ritchie developed streams for serial, if i remember correctly. streams was then applied to networks. there is an RFC about layering being bad for networking and the relative performance of NIT vs BPF prove the case. jmb To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message