From owner-freebsd-hackers Thu Jun 22 18:29:07 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id SAA10971 for hackers-outgoing; Thu, 22 Jun 1995 18:29:07 -0700 Received: from haven.uniserve.com (haven.uniserve.com [198.53.215.121]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id SAA10956 for ; Thu, 22 Jun 1995 18:28:56 -0700 Received: by haven.uniserve.com id <30747>; Thu, 22 Jun 1995 18:29:41 +0100 Date: Thu, 22 Jun 1995 18:29:36 -0700 (PDT) From: Tom Samplonius To: "Rodney W. Grimes" cc: jkh@freebsd.org, evanc@synapse.net, hackers@freefall.cdrom.com Subject: Re: FreeBSD as a router In-Reply-To: <199506230108.SAA09137@gndrsh.aac.dev.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: hackers-owner@freebsd.org Precedence: bulk On Thu, 22 Jun 1995, Rodney W. Grimes wrote: > The bottleneck is that you have to wait for full frame reception > before you get an interrupt to tell you to go look at the header > to decide what to do with the packet. > > In dedicated router hardware they use the trick of interrupting > the CPU after N bytes have been recieved (N is programmable) so > they can actually decide what to do with the packet before it is > even completly received. A bit of extra delay shouldn't affect throughput, especially with TCP large window support, or does it? Can't PC ethernet cards also be handled in the same matter? Or are there none available with that feature? From looking in Netblazer ST around here, it appears to have a SMC ethernet card in it..... Tom