From owner-cvs-all@FreeBSD.ORG Tue Sep 6 16:39:46 2005 Return-Path: X-Original-To: cvs-all@FreeBSD.org Delivered-To: cvs-all@FreeBSD.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C75A416A41F; Tue, 6 Sep 2005 16:39:46 +0000 (GMT) (envelope-from nate@root.org) Received: from ylpvm12.prodigy.net (ylpvm12-ext.prodigy.net [207.115.57.43]) by mx1.FreeBSD.org (Postfix) with ESMTP id 529D543D45; Tue, 6 Sep 2005 16:39:46 +0000 (GMT) (envelope-from nate@root.org) Received: from ylpvm01.prodigy.net (ylpvm01-int.prodigy.net [207.115.5.207]) by ylpvm12.prodigy.net (8.12.10 outbound/8.12.10) with ESMTP id j86GdemU010722; Tue, 6 Sep 2005 12:39:40 -0400 X-ORBL: [64.171.187.131] Received: from [10.0.5.50] (adsl-64-171-187-131.dsl.snfc21.pacbell.net [64.171.187.131]) by ylpvm01.prodigy.net (8.13.4 dk-milter linux/8.13.4) with ESMTP id j86GdeHp029440; Tue, 6 Sep 2005 12:39:40 -0400 Message-ID: <431DC64E.9010903@root.org> Date: Tue, 06 Sep 2005 09:39:42 -0700 From: Nate Lawson User-Agent: Mozilla Thunderbird 1.0.6 (X11/20050723) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Robert Watson References: <200509051602.j85G2Bpo090258@repoman.freebsd.org> <20050905094341.A23343@xorpc.icir.org> <20050905180050.GB41863@cell.sick.ru> <20050905141451.A27290@xorpc.icir.org> <20050906061828.GQ41863@cell.sick.ru> <20050906012755.B34182@xorpc.icir.org> <20050906112608.N51625@fledge.watson.org> In-Reply-To: <20050906112608.N51625@fledge.watson.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Luigi Rizzo , Gleb Smirnoff , cvs-all@FreeBSD.org, src-committers@FreeBSD.org, cvs-src@FreeBSD.org Subject: Re: cvs commit: src/sys/kern kern_poll.c X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 06 Sep 2005 16:39:47 -0000 Robert Watson wrote: > > On Tue, 6 Sep 2005, Luigi Rizzo wrote: > >> On Tue, Sep 06, 2005 at 10:18:28AM +0400, Gleb Smirnoff wrote: >> >>> Luigi, >> >> ... >> >>> The idlepoll thread is single. >> >> >> ok this is very good. Re. netisr vs idlepoll, perhaps a way could be >> to bump the idlepoll priority very high upon a net soft interrupt, and >> drop it down to its normal value once done with the netisr cycle. so >> we don't have to arbitrate among the two. > > Also, if we gradually move to a polling model that handles polling for > non-network devices, it would result in a rather mixed model. One of > the challenges of moving to a mixed polling model (one that supports > non-network devices) is that network devices have a fairly well > understood currency for work: processing of packets. Other devices may > have less well understood, or at least not easily comparable, workloads... For the case of storage, you actually have a better model since all transactions are initiated from the host (as opposed to packet arrivals). This gives an easy metric for a dynamic polling threshold -- if you have a deep queue of outstanding requests and one completes, you should poll a little more than normal. -- Nate