Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755444AbXICTzQ (ORCPT ); Mon, 3 Sep 2007 15:55:16 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755329AbXICTyw (ORCPT ); Mon, 3 Sep 2007 15:54:52 -0400 Received: from scrub.xs4all.nl ([194.109.195.176]:1534 "EHLO scrub.xs4all.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755261AbXICTyv (ORCPT ); Mon, 3 Sep 2007 15:54:51 -0400 Date: Mon, 3 Sep 2007 21:55:20 +0200 (CEST) From: Roman Zippel X-X-Sender: roman@scrub.home To: Ingo Molnar cc: linux-kernel@vger.kernel.org, peterz@infradead.org, Mike Galbraith Subject: Re: [ANNOUNCE/RFC] Really Simple Really Fair Scheduler In-Reply-To: <20070903192050.GA29049@elte.hu> Message-ID: References: <20070902120154.GA23769@elte.hu> <20070903185403.GA23479@elte.hu> <20070903192050.GA29049@elte.hu> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1415 Lines: 38 Hi, On Mon, 3 Sep 2007, Ingo Molnar wrote: > My next question then is about this code of yours in the wakeup path: > > +static void > +enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) > +{ > + kclock_t min_time; > + > + verify_queue(cfs_rq, cfs_rq->curr != se, se); > + min_time = get_time_avg(cfs_rq) - se->req_weight_inv; > + if ((kclock_t)(se->time_norm - min_time) < 0) > + se->time_norm = min_time; > > why do you only use the "min_time" if the pre-sleep time_norm is smaller > than the min_time? Here 'min_time' is close to the current average. It's a variation of the sleeper bonus. Let's assume two running tasks which have been running for 95ms and 105ms and a time slice of 10ms, the average is thus 100ms. If the new task has been sleeping for a while it starts at 90ms, if the task had been running lately it doesn't get this bonus again. > Shouldnt here the woken up task be set to the average time, like i did > it in the crude prototype: > > + se->exec_runtime = avg_exec_runtime(cfs_rq); That would be equivalent to simply clearing wait_runtime in CFS. bye, Roman - 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/