From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S272561AbTHPCWs (ORCPT ); Fri, 15 Aug 2003 22:22:48 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S272576AbTHPCWs (ORCPT ); Fri, 15 Aug 2003 22:22:48 -0400 Received: from c210-49-248-224.thoms1.vic.optusnet.com.au ([210.49.248.224]:17098 "EHLO mail.kolivas.org") by vger.kernel.org with ESMTP id S272561AbTHPCWr (ORCPT ); Fri, 15 Aug 2003 22:22:47 -0400 From: Con Kolivas To: Timothy Miller Subject: Re: [PATCH] O12.2int for interactivity Date: Sat, 16 Aug 2003 12:29:03 +1000 User-Agent: KMail/1.5.3 Cc: Timothy Miller , William Lee Irwin III , linux-kernel@vger.kernel.org References: <20030804195058.GA8267@cray.fish.zetnet.co.uk> <200308160235.05105.kernel@kolivas.org> <3F3D23BD.6050608@techsource.com> In-Reply-To: <3F3D23BD.6050608@techsource.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308161229.03334.kernel@kolivas.org> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, 16 Aug 2003 04:17, Timothy Miller wrote: > >All "nice" 0 tasks get the same size timeslice. If their dynamic priority > > is different (the PRI column in top) they still get the same timeslice. > > Why isn't dynamic priority just an extension of static priority? Why do > you modify only the ordering while leaving the timeslice alone? Because master engineer Molnar has determined that's the correct way. > So, tell me if I infer this correctly: If you have a nice 5 and a nice > 7, but the nice 5 is a cpu hog, while the nice 7 is interactive, then > the interactivity scheduler can modify their dynamic priorities so that > the nice 7 is being run before the nice 5. However, despite that, the > nice 7 still gets a shorter timeslice than tha nice 5. > > Have you tried altering this? Yes, not good with fluctuating timeslices all over the place makes for more bounce in the algorithm, and the big problem - the cpu intensive applications get demoted to smaller timeslices and they are the tasks that benefit the most from larger timeslices (for effective cpu cache usage). Con