Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964827AbWBMT5w (ORCPT ); Mon, 13 Feb 2006 14:57:52 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S964839AbWBMT5w (ORCPT ); Mon, 13 Feb 2006 14:57:52 -0500 Received: from mx3.mail.elte.hu ([157.181.1.138]:17095 "EHLO mx3.mail.elte.hu") by vger.kernel.org with ESMTP id S964827AbWBMT5v (ORCPT ); Mon, 13 Feb 2006 14:57:51 -0500 Date: Mon, 13 Feb 2006 20:55:50 +0100 From: Ingo Molnar To: Roman Zippel Cc: Thomas Gleixner , Andrew Morton , linux-kernel@vger.kernel.org, John Stultz Subject: Re: [PATCH 01/13] hrtimer: round up relative start time Message-ID: <20060213195550.GB30679@elte.hu> References: <1139827927.4932.17.camel@localhost.localdomain> <20060213130143.GA10771@elte.hu> <20060213144403.GA21317@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.1i X-ELTE-SpamScore: 0.0 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=0.0 required=5.9 tests=AWL autolearn=no SpamAssassin version=3.0.3 0.0 AWL AWL: From: address is in the auto white-list X-ELTE-VirusStatus: clean Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1376 Lines: 37 * Roman Zippel wrote: > Hi, > > On Mon, 13 Feb 2006, Ingo Molnar wrote: > > > In other words: your patch re-introduces half of the bug on low-res > > platforms. Users doing a series of one-shot itimer calls would be > > exposed to the same kind of (incorrect and unnecessary) summing-up > > errors. What's the point? > > I don't fully agree with the interval behaviour either, [...] i.e. you'd want to reintroduce the comulative interval rounding bug that users noticed? Or do you have some other way to change it? I really dont see your point. > [...] but here one could at least say it's correct on average. [...] i'm not sure i understand. Are you implying by this that some current code is not "correct on average"? > Since hrtimer is also used for nanosleep(), which I consider more > important (as e.g. posix timer), this one should at least be correct > and consistent with previous 2.6 releases. [...] for me it's simple: i dont think we should reintroduce the same type of concept that was clearly causing regressions in previous 2.6 releases. Thomas, what do you think? Ingo - 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/