From owner-freebsd-isp@FreeBSD.ORG Fri Oct 14 20:48:32 2005 Return-Path: X-Original-To: freebsd-isp@freebsd.org Delivered-To: freebsd-isp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2F7A416A41F for ; Fri, 14 Oct 2005 20:48:32 +0000 (GMT) (envelope-from drew@gothambus.com) Received: from mail.gothambus.com (trixie.gbcx.net [204.89.131.138]) by mx1.FreeBSD.org (Postfix) with SMTP id 9F72143D45 for ; Fri, 14 Oct 2005 20:48:31 +0000 (GMT) (envelope-from drew@gothambus.com) Received: (qmail 33973 invoked by uid 89); 14 Oct 2005 20:48:11 -0000 Message-ID: <20051014204811.33972.qmail@mail.gothambus.com> References: <20051013204631.51300.qmail@mail.gothambus.com> <20051014161100.33377.qmail@web33306.mail.mud.yahoo.com> <20051014181239.GC62233@complx.LF.net> In-Reply-To: <20051014181239.GC62233@complx.LF.net> From: "Drew Linsalata" To: Kurt Jaeger Date: Fri, 14 Oct 2005 16:48:11 -0400 Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8"; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-isp@freebsd.org, Danial Thom Subject: Re: Multiport NICs - VLAN and Polling Support? X-BeenThere: freebsd-isp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Internet Services Providers List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Oct 2005 20:48:32 -0000 Kurt Jaeger writes: > Hi! > >> polling is almost never a performance advantage >> for ethernet, as virtually all modern controllers >> have some sort of interrupt moderation built-in. > > We had the case on fbsd 5.4p7 with SMP where bge drivers produced > much slower throughput if we did not use polling. We measured using > ttcp and the difference was from 2-3Mbyte/sec to 11 Mbyte/sec. What Danial is saying is that polling is whacking your CPU, while manipulating interrupt frequency at the NIC hardware level should make for a "free" (read: no cpu overhead) performance enhancement. If he's correct, you should see the same performance increase if you skip polling and go right to the NIC hardware interrupt tweak (assuming your BGE card supports it), AND you should get it without adding CPU load. This is interesting. I'm going to set up testbed when I have a little time and run both ways to see what happens. Question is, what mechanism do we have available to manipulate interrupt frequency on the NIC itself. The em driver does not appear to offer any options in that area, nor is anything like that configurable via ifconfig. Are we looking at a sysctl variable, or are we tweaking the driver source and rebuilding each time we want to try a new combination of values? Drew Linsalata The Gotham Bus Company, Inc. Dedicated Servers and Colocation Long Island, New York http://www.gothambus.com