From owner-freebsd-net@FreeBSD.ORG Wed Nov 17 18:14:00 2004 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DC2B716A4CE for ; Wed, 17 Nov 2004 18:14:00 +0000 (GMT) Received: from comp.chem.msu.su (comp.chem.msu.su [158.250.32.97]) by mx1.FreeBSD.org (Postfix) with ESMTP id 73E2143D39 for ; Wed, 17 Nov 2004 18:13:58 +0000 (GMT) (envelope-from yar@comp.chem.msu.su) Received: from comp.chem.msu.su (localhost [127.0.0.1]) by comp.chem.msu.su (8.12.9p2/8.12.9) with ESMTP id iAHIDqUf050269 for ; Wed, 17 Nov 2004 21:13:52 +0300 (MSK) (envelope-from yar@comp.chem.msu.su) Received: (from yar@localhost) by comp.chem.msu.su (8.12.9p2/8.12.9/Submit) id iAHIDqFT050268 for net@freebsd.org; Wed, 17 Nov 2004 21:13:52 +0300 (MSK) (envelope-from yar) Date: Wed, 17 Nov 2004 21:13:51 +0300 From: Yar Tikhiy To: net@freebsd.org Message-ID: <20041117181351.GA48071@comp.chem.msu.su> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.6i Subject: polling(4) rocks! X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 17 Nov 2004 18:14:01 -0000 Hi there, I can't but remind you that there's polling(4) in FreeBSD :-) Until today, I was convinced for some obscure reason that polling(4) was an experimental feature that might or might not work. Today I tried it on our central router box and got astounding results. The router box is a 1.4GHz Celeron PC with an fxp(4) interface split across a dozen of vlans. There is nothing special about its setup except for ~250 rules loaded into ipfw2. It is running 4.10-RELEASE. Without polling, it was able to switch full 10Mbytes/sec of traffic (~9kpps), but that took from 50 to 70% CPU time spent in interrupts. With polling on, interrupt time never exceeds 5% and it stays as low as 1-2% on average even when traffic is that high. Many thanks to folks who have had a hand in polling(4) development! Just in case: Please be aware that polling(4) won't make KDE run faster unless on a busy router ;-) -- Yar