From mboxrd@z Thu Jan 1 00:00:00 1970 From: Robert =?ISO-8859-1?Q?Sch=F6ne?= Subject: Re: [PATCH 4/4] PERF: fix power:cpu_idle double end events Date: Mon, 01 Nov 2010 09:11:28 +0100 Message-ID: <1288599088.2712.47.camel__40571.3583961644$1288627560$gmane$org@localhost> References: <1288136605-10526-1-git-send-email-trenn@suse.de> <1288136605-10526-5-git-send-email-trenn@suse.de> <201010271742.04545.trenn@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <201010271742.04545.trenn@suse.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-pm-bounces@lists.linux-foundation.org Errors-To: linux-pm-bounces@lists.linux-foundation.org To: Thomas Renninger Cc: jean.pihet@newoldbits.com, linux-trace-users@vger.kernel.org, mingo@elte.hu, linux-omap@vger.kernel.org, arjan@linux.intel.com, linux-pm@lists.linux-foundation.org List-Id: linux-pm@vger.kernel.org Am Mittwoch, den 27.10.2010, 17:42 +0200 schrieb Thomas Renninger: > Robert: I expect you tested this on a machine with no cpuidle > driver registered? You're right, there was no idle driver, but the idle process from process_64.c which called the idle routine. I reported my thoughts on this on 14th of May this year 2010, mostly claiming for a standard on where to report these events. You're also missing the other idle routines from x86/kernel/process.c mwait_idle_with_hints and mwait_idle only throw start events, so they should behave like default_idle. poll_idle on the other hand reports the end event itself. Robert