From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754963Ab1KWMlp (ORCPT ); Wed, 23 Nov 2011 07:41:45 -0500 Received: from dns1.mips.com ([12.201.5.69]:47578 "EHLO dns1.mips.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753693Ab1KWMln (ORCPT ); Wed, 23 Nov 2011 07:41:43 -0500 From: "Zhu, DengCheng" To: Peter Zijlstra CC: "Barzilay, Eyal" , "Fortuna, Zenon" , Paul Mackerras , Ingo Molnar , Arnaldo Carvalho de Melo , "ralf@linux-mips.org" , LKML Subject: RE: [PATCH v2 5/5] perf: Enable applicable siblings when group leader is enable-on-exec Thread-Topic: [PATCH v2 5/5] perf: Enable applicable siblings when group leader is enable-on-exec Thread-Index: AQHMqMcPcUwfPrOJr0SdF29sueMmtJW5Pg2A//+cb6yAAJQ0gP//fV12gACNEACAAWQ9AIAAAGIA//+IhZw= Date: Wed, 23 Nov 2011 12:40:47 +0000 Message-ID: References: <1321932601-21128-1-git-send-email-dczhu@mips.com> ,<1321959115.5148.22.camel@twins> ,<1321969559.14799.7.camel@twins> <1321971798.14799.12.camel@twins> <4ECC6AAC.4010504@mips.com>,<1322048382.14799.41.camel@twins> In-Reply-To: <1322048382.14799.41.camel@twins> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [192.168.36.79] x-ems-proccessed: 6LP3oGfGVdcdb8o1aBnt6w== x-ems-stamp: p6sDGCWvfZHIeHHZ1PSKvg== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from base64 to 8bit by nfs id pANCftQc007911 在 2011年11月23日星期三，Peter Zijlstra 写道： > On Wed, 2011-11-23 at 11:38 +0800, Deng-Cheng Zhu wrote: > >> +     list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { >> +             ret = event_enable_on_exec(event, ctx); >> +             if (ret) >> +                     enabled = 1; >> +     } >> + >>       list_for_each_entry(event, &ctx->pinned_groups, group_entry) { >>               ret = event_enable_on_exec(event, ctx); >>               if (ret) > > This isn't correct either, in this case you should then remove the other > two iterations of pinned/flexible group lists. > > Also your use of list_for_each_entry_rcu() is incorrect, either you then > should also use rcu_read_lock(), or its not needed and not use the _rcu > list primitive at all. > > Now since event_list is modified under ctx->lock, and we hold that lock > no fancy stuff is needed and we can do without. Ah, yes you are right. Thanks. Your following patch looks good except that event_entry should be used for ctx event_list, rather than group_entry. Tomorrow I'll test it and get back to you. Deng-Cheng > --- > Subject: perf: Fix enable_on_exec for sibling events > From: Peter Zijlstra > Date: Tue Nov 22 11:25:43 CET 2011 > > Deng-Cheng Zhu reported that sibling events that were created disabled > with enable_on_exec would never get enabled. Iterate all events instead > of the group lists. > > Reported-by: Deng-Cheng Zhu > Signed-off-by: Peter Zijlstra > Link: http://lkml.kernel.org/n/tip-299rxrpmmle8hp3spyxfo202@git.kernel.org > --- >  kernel/events/core.c |    8 +------- >  1 file changed, 1 insertion(+), 7 deletions(-) > > Index: linux-2.6/kernel/events/core.c > =================================================================== > --- linux-2.6.orig/kernel/events/core.c > +++ linux-2.6/kernel/events/core.c > @@ -2494,13 +2494,7 @@ static void perf_event_enable_on_exec(st >        raw_spin_lock(&ctx->lock); >        task_ctx_sched_out(ctx); > > -       list_for_each_entry(event, &ctx->pinned_groups, group_entry) { > -               ret = event_enable_on_exec(event, ctx); > -               if (ret) > -                       enabled = 1; > -       } > - > -       list_for_each_entry(event, &ctx->flexible_groups, group_entry) { > +       list_for_each_entry(event, &ctx->event_list, group_entry) { >                ret = event_enable_on_exec(event, ctx); >                if (ret) >                        enabled = 1; > > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at  http://vger.kernel.org/majordomo-info.html > Please read the FAQ at  http://www.tux.org/lkml/ >{.n++%ݶw{.n+{G{ayʇڙ,jfhz_(階ݢj"mG?&~iOzv^m ?I