From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751720AbXCLSoY (ORCPT ); Mon, 12 Mar 2007 14:44:24 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751704AbXCLSoY (ORCPT ); Mon, 12 Mar 2007 14:44:24 -0400 Received: from smtp14.wxs.nl ([195.121.247.5]:45648 "EHLO smtp14.wxs.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751458AbXCLSoX (ORCPT ); Mon, 12 Mar 2007 14:44:23 -0400 Date: Mon, 12 Mar 2007 19:47:11 +0100 From: jos poortvliet X-Face: $0>4o"Xx2u2q(Tx!D+6~yPc{ZhEfnQnu:/nthh%Kr%f$aiATk$xjx^X4admsd*)=?utf-8?q?IZz=3A=5FkT=0A=09=7CurITP!=2E?=)L`*)Vw@4\@6>#r;3xSPW`,~C9vb`W/s]}Gq]b!o_/+(lJ:b)=?utf-8?q?T0=26KCLMGvG=7CS=5E=0A=09z=7B=5C=2E7EtehxhFQE=27eYSsir/=7CtQ?= =?utf-8?q?j=23rWQe4o?=>WC>_R To: ck@vds.kolivas.org Cc: Con Kolivas , Al Boldi , linux-kernel@vger.kernel.org Message-id: <200703121947.16008.jos@mijnkamer.nl> MIME-version: 1.0 Content-type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary=nextPart2687731.mml4ruIgb3 Content-transfer-encoding: 7bit User-Agent: KMail/1.9.6 References: <200703042335.26785.a1426z@gawab.com> <200703121714.25034.a1426z@gawab.com> <200703130505.05132.kernel@kolivas.org> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --nextPart2687731.mml4ruIgb3 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Op Monday 12 March 2007, schreef Con Kolivas: > On Tuesday 13 March 2007 01:14, Al Boldi wrote: > > Con Kolivas wrote: > > > > > The higher priority one always get 6-7ms whereas the lower priori= ty > > > > > one runs 6-7ms and then one larger perfectly bound expiration > > > > > amount. Basically exactly as I'd expect. The higher priority task > > > > > gets precisely RR_INTERVAL maximum latency whereas the lower > > > > > priority task gets RR_INTERVAL min and full expiration (according > > > > > to the virtual deadline) as a maximum. That's exactly how I intend > > > > > it to work. Yes I realise that the max latency ends up being long= er > > > > > intermittently on the niced task but that's -in my opinion- > > > > > perfectly fine as a compromise to ensure the nice 0 one always ge= ts > > > > > low latency. > > > > > > > > I think, it should be possible to spread this max expiration latency > > > > across the rotation, should it not? > > > > > > There is a way that I toyed with of creating maps of slots to use for > > > each different priority, but it broke the O(1) nature of the virtual > > > deadline management. Minimising algorithmic complexity seemed more > > > important to maintain than getting slightly better latency spreads for > > > niced tasks. It also appeared to be less cache friendly in design. I > > > could certainly try and implement it but how much importance are we to > > > place on latency of niced tasks? Are you aware of any usage scenario > > > where latency sensitive tasks are ever significantly niced in the real > > > world? > > > > It only takes one negatively nice'd proc to affect X adversely. > > I have an idea. Give me some time to code up my idea. Lack of sleep is > making me very unpleasant. You're excited by RSDL and the positive comments, aren't you? Well, don't=20 forget to sleep, sleeping makes ppl smarter you know ;-) --nextPart2687731.mml4ruIgb3 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQBF9aAw+wgQ1AD35iwRAnYXAJ9O98rS3PKPzZp78xGDo7O9a4f51gCfX3R+ t96LVrpooN5G/kCSd8bl7kk= =Q0y3 -----END PGP SIGNATURE----- --nextPart2687731.mml4ruIgb3--