From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762943AbZEHJLg (ORCPT ); Fri, 8 May 2009 05:11:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758192AbZEHJLZ (ORCPT ); Fri, 8 May 2009 05:11:25 -0400 Received: from mail46.e.nsc.no ([193.213.115.46]:39362 "EHLO mail46.e.nsc.no" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758274AbZEHJLY (ORCPT ); Fri, 8 May 2009 05:11:24 -0400 Date: Fri, 8 May 2009 11:10:52 +0200 From: Henrik Austad To: GeunSik Lim Cc: finarfin@dreamos.org, linux-kernel@vger.kernel.org Subject: Re: SCHED_EDF infos Message-ID: <20090508091052.GA10429@januz.myftp.org> References: <164c92d827cbee86ba2c5621716309e6@localhost> <200904300939.31754.henrik@austad.us> <49b7c2350905071935kbd2aa22v9d39cf41c537c24b@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <49b7c2350905071935kbd2aa22v9d39cf41c537c24b@mail.gmail.com> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, May 08, 2009 at 11:35:58AM +0900, GeunSik Lim wrote: > Hi Henlik, Hi Lim, First off, to all of you, sorry for my last, very poorly formatted email. I have now tranceded into the realms of mutt. > [..] > I think so. > How can we approach EDF implementation like Pfair as a generic solution > for Multicore in Linux? I am working on an implementation now, and I hope to be able to release a prototype by the end of next week. I think we can continue the discussion then based on that. > > I'm working on SCHED_PFAIR :-) Which is a multicore, ratebased deadline > > driven global scheduling policy where I use the PD^2 rules to solve > > tie-breaks. In theory, it can reach 100% utilization on all cores > > without missing deadlines (but in practice you will only get close to > > 100% as it is not perfect). > > [..] > > I'd suggest getting the scheduler running first, then look at those > > problems later. If you spend all your time trying to learn the > > scheduler and design in every little feature you need, you'll never > > finish in time. > In fact, I also don't have perfect know how to solve PI in Multicore. > [...] > > deadline inversion will be a problem, in fact, whatever you chooose to > > be the 'key' for picking tasks (priority, niceness, deadlines, wind > > direction, ), you can pretty much take that and add a > > -inversion after it. :) No, PI is going to be deadly no matter what you do. Henrik