From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1422694AbcFHINP (ORCPT ); Wed, 8 Jun 2016 04:13:15 -0400 Received: from outbound-smtp03.blacknight.com ([81.17.249.16]:52767 "EHLO outbound-smtp03.blacknight.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1161185AbcFHINJ (ORCPT ); Wed, 8 Jun 2016 04:13:09 -0400 Date: Wed, 8 Jun 2016 09:13:05 +0100 From: Mel Gorman To: Peter Zijlstra Cc: Josh Poimboeuf , Ingo Molnar , linux-kernel@vger.kernel.org, Matt Fleming , Srikar Dronamraju Subject: Re: [PATCH 2/2] sched/debug: fix deadlock when enabling sched events Message-ID: <20160608081305.GG2527@techsingularity.net> References: <50a0f7f2a5f8d69e0440c3430f06be7ccbe59903.1465322027.git.jpoimboe@redhat.com> <20160607195448.GL30154@twins.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20160607195448.GL30154@twins.programming.kicks-ass.net> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Jun 07, 2016 at 09:54:48PM +0200, Peter Zijlstra wrote: > On Tue, Jun 07, 2016 at 02:43:17PM -0500, Josh Poimboeuf wrote: > > 1. Instead of just warning and allowing the tracepoints to be broken, > > I'd argue that it would be better to make them work by forcing > > schedstats enabled and printing a warning about that, which is what's > > Forcing them enabled doesn't make them useful per se; some of these > tracepoint rely on previously recoded state, which now wasn't previously > recorded because back then it was disabled. > This is unfortunately true. It means there is a window when the data is unreliable after the stats are enabled. It is assumed that either someone enabling the stats knows that or is willing to enable them on the command line. -- Mel Gorman SUSE Labs