Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752143AbdIAQsX (ORCPT ); Fri, 1 Sep 2017 12:48:23 -0400 Received: from mga09.intel.com ([134.134.136.24]:15986 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751863AbdIAQsV (ORCPT ); Fri, 1 Sep 2017 12:48:21 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.41,459,1498546800"; d="scan'208";a="124747218" From: Andi Kleen To: Milian Wolff Cc: linux-perf-users@vger.kernel.org, Arnaldo Carvalho de Melo , linux-kernel@vger.kernel.org, peterz@infradead.org Subject: Re: broken cycle counts from perf record in frequency mode [Was: Re: deducing CPU clock rate over time from cycle samples] References: <2900948.aeNJFYEL58@agathebauer> <6142297.1ke4zlptR7@milian-kdab2> <20170828172808.GQ2482@two.firstfloor.org> <3838997.nyT3JFTsVA@agathebauer> Date: Fri, 01 Sep 2017 09:48:20 -0700 In-Reply-To: <3838997.nyT3JFTsVA@agathebauer> (Milian Wolff's message of "Fri, 01 Sep 2017 12:34:36 +0200") Message-ID: <87d17al60b.fsf@firstfloor.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.2 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1165 Lines: 25 Milian Wolff writes: > > do you have any input on this issue? I really wonder why noone else is > complaining about the frequency mode being unreliable or right out broken in > many situations. Since it's the default mode, I think this urgently needs to > be investigated - it makes perf unusable for a large group of users who want > to use it but don't know about `-c N` as a workaround... It's likely related due to the frequency algorithm starting with 0. So at the beginning the samples are very fast (like 1 cycle) and likely something breaks down in perf or your frequency calculation for very short samples. Also for inherited events this happens on every fork. If you trace fork events too you'll likely see it correlated. If you use -a and disable inheritance (no-inherit=1) it will also likely be only at the beginning. However I fail to see what it would actually break. The frequency just needs to be roughly accurate over the whole measurement period to get good sampling coverage. But there's nothing in the profile that uses the actual frequency. It's just a means to get samples, not a measurement by itself. -Andi