From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756060Ab3LCXop (ORCPT ); Tue, 3 Dec 2013 18:44:45 -0500 Received: from mga02.intel.com ([134.134.136.20]:42342 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756041Ab3LCXom (ORCPT ); Tue, 3 Dec 2013 18:44:42 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.93,820,1378882800"; d="scan'208";a="418884260" From: Andi Kleen To: Stephane Eranian Cc: Jiri Olsa , Ingo Molnar , LKML , Peter Zijlstra , "mingo\@elte.hu" , David Ahern , Arnaldo Carvalho de Melo , Namhyung Kim Subject: Re: [BUG] perf stat: explicit grouping yields unexpected results References: <20131115063457.GB12442@gmail.com> <20131115103405.GA18264@gmail.com> <20131117034134.GA19762@tassilo.jf.intel.com> <20131129133340.GA25751@krava.brq.redhat.com> <20131202152329.GD22695@tassilo.jf.intel.com> Date: Tue, 03 Dec 2013 15:44:38 -0800 In-Reply-To: (Stephane Eranian's message of "Tue, 3 Dec 2013 03:52:13 +0100") Message-ID: <87r49tcxeh.fsf@tassilo.jf.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (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 Stephane Eranian writes: >> > I don't understand why this need to be so complicated. > Maybe just change the error code in case of group > overcommitment? That way, the tool could distinguish > and report the appropriate error message. The existing errno is an ABI. Also that would handle this case, but there are a bazillion other failure cases in perf. And we can't invent new errnos for all of them. A string error interface would cover it all. Usually when something strange like this happens in perf I have to apply this patch https://lkml.org/lkml/2013/3/8/432 and use ftrace to trace it down. -Andi -- ak@linux.intel.com -- Speaking for myself only