From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755930AbZETKcn (ORCPT ); Wed, 20 May 2009 06:32:43 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757259AbZETKcc (ORCPT ); Wed, 20 May 2009 06:32:32 -0400 Received: from www.tglx.de ([62.245.132.106]:50960 "EHLO www.tglx.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757265AbZETKcb (ORCPT ); Wed, 20 May 2009 06:32:31 -0400 Date: Wed, 20 May 2009 12:31:36 +0200 (CEST) From: Thomas Gleixner To: Anton Blanchard cc: rostedt@goodmis.org, mingo@elte.hu, tzanussi@gmail.com, jbaron@redhat.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH] tracing/events: Add timer and high res timer tracepoints In-Reply-To: <20090520101334.GA23442@kryten> Message-ID: References: <20090520101334.GA23442@kryten> User-Agent: Alpine 2.00 (LFD 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 20 May 2009, Anton Blanchard wrote: > > Add tracepoints for timer and high res timer execution. We add entry and > exit tracepoints so we can calculate timer latency. > > Example ftrace output: > > -0 [000] 264.040506: hrtimer_entry: func=.tick_sched_timer+0x0/0xf0 > -0 [000] 264.040508: hrtimer_exit: func=.tick_sched_timer+0x0/0xf0 restart=HRTIMER_RESTART > -0 [000] 264.040530: timer_entry: func=.e1000_watchdog+0x0/0x8fc > -0 [000] 264.040728: timer_exit: func=.e1000_watchdog+0x0/0x8fc > > Here we can see e1000_watchdog is taking 0.2ms - it might make sense to > move this into a workqueue or kernel thread. Cute. That should help us to pinpoint more such timer abusers :) > Signed-off-by: Anton Blanchard Acked-by: Thomas Gleixner