From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753389Ab3LGIWh (ORCPT ); Sat, 7 Dec 2013 03:22:37 -0500 Received: from moutng.kundenserver.de ([212.227.126.171]:56117 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751990Ab3LGIWg (ORCPT ); Sat, 7 Dec 2013 03:22:36 -0500 Message-ID: <1386404544.5673.88.camel@marge.simpson.net> Subject: Re: [PATCH] Trace idle entry and exit times From: Mike Galbraith To: Andi Kleen Cc: linux-kernel@vger.kernel.org, x86@kernel.org, Andi Kleen , lenb@kernel.org, tglx@linutronix.de, peterz@infradead.org Date: Sat, 07 Dec 2013 09:22:24 +0100 In-Reply-To: <20131207052029.GC21717@two.firstfloor.org> References: <1386377164-2223-1-git-send-email-andi@firstfloor.org> <1386392252.5673.31.camel@marge.simpson.net> <20131207052029.GC21717@two.firstfloor.org> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.3 Content-Transfer-Encoding: 7bit Mime-Version: 1.0 X-Provags-ID: V02:K0:DinOpTPZ98tbWvIQaKE5oxNyx/N4qJgZiO96kuNC687 kwGzBqfVbXUOaEEoCono43W41fNW7yJAnR6Isyr9QYI48frjHn lEGMzCP2PCk6yLu3wI4lJimj3fu2hpGY1YX828MlbBjKUNrt2w 5uIqW0z/S+5GjjWZ8XMaVuCL68RFy88d3rvL7ZmSATeowi5qEs aNAl8vGIxnXNODvOppwXsLqAuvX3+jqO7TyycVDwNEn8YP1I3i y2jt2QKZVyBQ2u0vEIMdVAoTFV01lY89psA6QMkhiKOSqxb2ld vF7PGzM8u5/Rk+N1ACLReJWpifoKa1lMmSFN7Du9FNgTLmF9jo azDrzwRQR57ZzATDYk+olvJpZoR/HS+Qp/sPuOlWzZo4wrnMBE RLvb8o02/NnKw== Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, 2013-12-07 at 06:20 +0100, Andi Kleen wrote: > On Sat, Dec 07, 2013 at 05:57:32AM +0100, Mike Galbraith wrote: > > On Fri, 2013-12-06 at 16:46 -0800, Andi Kleen wrote: > > > From: Andi Kleen > > > > > > This patch creates 4 new trace points: one at the beginning of idle, > > > one before entering mwait, one after exiting mwait, and one after > > > finishing the idle code. This is useful to trace down any > > > additional latencies caused by the ever-growing idle code. > > > > > > Idle latencies are a common source of performance problems, so it's > > > important to be able to measure them. > > > > You mean stuff like.. oh, say 60 cores doing zero work every 12ms > > munching ~20% cpu? > > Well that one is obvious. I mean more the subtle problems. Yeah, not subtle. SGI could use C2 as power supply stress tester. -Mike