Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755232Ab0HDFCV (ORCPT ); Wed, 4 Aug 2010 01:02:21 -0400 Received: from fafnir.cs.unc.edu ([152.2.129.90]:50765 "EHLO fafnir.cs.unc.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753066Ab0HDFCU convert rfc822-to-8bit (ORCPT ); Wed, 4 Aug 2010 01:02:20 -0400 X-Greylist: delayed 1111 seconds by postgrey-1.27 at vger.kernel.org; Wed, 04 Aug 2010 01:02:20 EDT References: <1278682707.6083.227.camel@Palantir> <1278685133.1900.201.camel@laptop> <51F8E441-58D7-45E1-B7A0-7A717EDF08B5@email.unc.edu> <1278693304.1900.266.camel@laptop> <1278752489.4390.97.camel@Palantir> <1280828775.1923.447.camel@laptop> Message-Id: From: Bjoern Brandenburg To: Peter Zijlstra In-Reply-To: <1280828775.1923.447.camel@laptop> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8BIT X-Mailer: iPad Mail (7B405) Mime-Version: 1.0 (iPad Mail 7B405) Subject: Re: periods and deadlines in SCHED_DEADLINE Date: Wed, 4 Aug 2010 01:02:41 -0400 Cc: Bjoern Brandenburg , Raistlin , linux-kernel , Song Yuan , Dmitry Adamushko , Thomas Gleixner , Nicola Manica , Luca Abeni , Claudio Scordino , Harald Gustafsson , "bastoni@cs.unc.edu" , Giuseppe Lipari Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1784 Lines: 23 On Aug 3, 2010, at 5:46 AM, Peter Zijlstra wrote: > On Sun, 2010-07-11 at 08:42 +0200, Bjoern Brandenburg wrote: >> If you want to do G-EDF with limited and different budgets on each CPU >> (e.g., G-EDF tasks may only run for 100 out of 1000 ms on CPU 0, but >> for 400 out of 1000 ms on CPU 1), then you are entering the domain of >> restricted-supply scheduling, which is significantly more complicated >> to analyze (see [1,2]). > > Would making the thing homogenious by assuming the worst for all cpus > make the analysis easier? That is, in the above example, only allow the > G-EDF scheduler to run for 100 out of 1000 ms on both cpus. It would, but that severely limits your non-G-EDF tasks. What if you want to provision a P-EDF task with a period of ten milliseconds? If you allow a G-EDF slice of guaranteed 100 ms, then your P-EDF task might miss a deadline if it arrives at the wrong time. If you let it preempt the G-EDF slice, then you get "out of sync" and the analysis can become tricky. If you scale down the G-EDF time slice length to less than ten ms, then overheads increase needlessly on all processors. EDF-HSB addresses specifically this problem (for the case where G-EDF is only used for soft tasks): B. Brandenburg and J. Anderson, “Integrating Hard/Soft Real-Time Tasks and Best-Effort Jobs on Multiprocessors”, Proceedings of the 19th Euromicro Conference on Real-Time Systems (ECRTS 2007), pp. 61-70. IEEE, July 2007. http://www.cs.unc.edu/~anderson/papers/ecrts07b.pdf - Björn-- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/