From mboxrd@z Thu Jan  1 00:00:00 1970
From: Peter Zijlstra <peterz@infradead.org>
Subject: Re: [PATCH v2 4/6] KVM-GST: KVM Steal time registration
Date: Tue, 01 Feb 2011 17:17:10 +0100
Message-ID: <1296577030.26581.236.camel@laptop>
References: <1296244340-15173-1-git-send-email-glommer@redhat.com>
	 <1296244340-15173-5-git-send-email-glommer@redhat.com>
	 <4D45649A.4090709@redhat.com>
	 <1296575618.5081.13.camel@mothafucka.localdomain>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Cc: Avi Kivity <avi@redhat.com>, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, aliguori@us.ibm.com,
	Rik van Riel <riel@redhat.com>,
	Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>,
	Venkatesh Pallipadi <venki@google.com>
To: Glauber Costa <glommer@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from bombadil.infradead.org ([18.85.46.34]:46610 "EHLO
	bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751765Ab1BAQQR (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 1 Feb 2011 11:16:17 -0500
Received: from canuck.infradead.org ([134.117.69.58])
	by bombadil.infradead.org with esmtps (Exim 4.72 #1 (Red Hat Linux))
	id 1PkItt-0006Gw-7T
	for kvm@vger.kernel.org; Tue, 01 Feb 2011 16:16:17 +0000
Received: from j77219.upc-j.chello.nl ([24.132.77.219] helo=dyad.programming.kicks-ass.net)
	by canuck.infradead.org with esmtpsa (Exim 4.72 #1 (Red Hat Linux))
	id 1PkIts-0007Aa-NC
	for kvm@vger.kernel.org; Tue, 01 Feb 2011 16:16:16 +0000
In-Reply-To: <1296575618.5081.13.camel@mothafucka.localdomain>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Tue, 2011-02-01 at 13:53 -0200, Glauber Costa wrote:
> 
> And since the granularity of the cpu accounting is too coarse, we end up
> with much more steal time than we should, because things that are less
> than 1 unity of cputime, are often rounded up to 1 unity of cputime.

See, that! is the problem, don't round up like that.

What you can do is: steal_ticks = steal_time_clock() / TICK_NSEC, or
simply keep a steal time delta and every time it overflows
cputime_one_jiffy insert a steal-time tick.

Venki might have created some infrastructure for doing this with the
IRQ_TIME accounting mess, but irqtime_account_process_tick() still gives
me a head-ache.