From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755685AbaICIhI (ORCPT ); Wed, 3 Sep 2014 04:37:08 -0400 Received: from mail3.unitn.it ([193.205.206.24]:63255 "EHLO mail3.unitn.it" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755397AbaICIhE (ORCPT ); Wed, 3 Sep 2014 04:37:04 -0400 Message-ID: <5406D32D.9030306@unitn.it> Date: Wed, 03 Sep 2014 10:37:01 +0200 From: Luca Abeni User-Agent: Mozilla/5.0 (X11; Linux i686; rv:31.0) Gecko/20100101 Thunderbird/31.0 MIME-Version: 1.0 To: Henrik Austad CC: Juri Lelli , Peter Zijlstra , rdunlap , Ingo Molnar , Dario Faggioli , Juri Lelli , LKML doc , LKML Subject: Re: [PATCH v3 3/4] Documentation/scheduler/sched-deadline.txt: improve and clarify AC bits References: <1409220029-9002-1-git-send-email-juri.lelli@arm.com> <1409220029-9002-4-git-send-email-juri.lelli@arm.com> <20140902214538.GC22581@sisyphus.home.austad.us> <5406BA02.20806@unitn.it> In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 09/03/2014 09:48 AM, Henrik Austad wrote: > On Wed, Sep 3, 2014 at 8:49 AM, Luca Abeni wrote: > >> Hi, >> >> On 09/02/2014 11:45 PM, Henrik Austad wrote: >> [...] >> >> + On multiprocessor systems with global EDF scheduling (non partitioned >>>> + systems), a sufficient test for schedulability can not be based on the >>>> + utilisations (it can be shown that task sets with utilisations slightly >>>> + larger than 1 can miss deadlines regardless of the number of CPUs M). >>>> + However, as previously stated, enforcing that the total utilisation is >>>> smaller >>>> + than M is enough to guarantee that non real-time tasks are not starved >>>> and >>>> + that the tardiness of real-time tasks has an upper bound. >>>> >>> >>> I'd _really_ appreciate a link to a paper where all of this is presented >>> and proved! >>> >> Well, my original plan was to add the bibliography in the next round of >> patches... >> Is this ok? >> > > Sure, but I don't see why you cannot add this now Well, there are multiple papers to be cited here: the one showing that in a pathological case you can have missed deadlines on M CPUs with utilisation slightly larger than 1 (Dhall's effect), one or more papers presenting a more advanced admission control for global EDF, one or more papers showing the tardiness bound, ... So, I'd like to spend some time doing some research to properly cite the most appropriate references. > you (or Juri?) did update one of the papers in this series, right? In patch 1, I updated the URL for a publically available technical report I wrote in 1998 (the old URL was in a web site I cannot access anymore, so I changed the URL to point to a web site I can control). > As long as an article with all the headache ends up here, I'm happy :) As said, if possible I'd like to do it in the next batch of patches, so that these updates are not delayed... If this is not ok, I'll update the patch adding some references, but this will take some time. Thanks, Luca