From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752378Ab3AUHpl (ORCPT ); Mon, 21 Jan 2013 02:45:41 -0500 Received: from e23smtp01.au.ibm.com ([202.81.31.143]:35906 "EHLO e23smtp01.au.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752267Ab3AUHpj (ORCPT ); Mon, 21 Jan 2013 02:45:39 -0500 Message-ID: <50FCF212.3010504@linux.vnet.ibm.com> Date: Mon, 21 Jan 2013 15:45:22 +0800 From: Michael Wang User-Agent: Mozilla/5.0 (X11; Linux i686; rv:16.0) Gecko/20121011 Thunderbird/16.0.1 MIME-Version: 1.0 To: Mike Galbraith CC: linux-kernel@vger.kernel.org, mingo@redhat.com, peterz@infradead.org, mingo@kernel.org, a.p.zijlstra@chello.nl Subject: Re: [RFC PATCH 0/2] sched: simplify the select_task_rq_fair() References: <1356588535-23251-1-git-send-email-wangyun@linux.vnet.ibm.com> <50ED384C.1030301@linux.vnet.ibm.com> <1357977704.6796.47.camel@marge.simpson.net> <1357985943.6796.55.camel@marge.simpson.net> <1358155290.5631.19.camel@marge.simpson.net> <50F79256.1010900@linux.vnet.ibm.com> <1358654997.5743.17.camel@marge.simpson.net> <50FCACE3.5000706@linux.vnet.ibm.com> <1358743128.4994.33.camel@marge.simpson.net> <50FCCCF5.30504@linux.vnet.ibm.com> <1358750523.4994.55.camel@marge.simpson.net> <1358752180.4994.65.camel@marge.simpson.net> In-Reply-To: <1358752180.4994.65.camel@marge.simpson.net> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Content-Scanned: Fidelis XPS MAILER x-cbid: 13012107-1618-0000-0000-00000338B108 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/21/2013 03:09 PM, Mike Galbraith wrote: > On Mon, 2013-01-21 at 07:42 +0100, Mike Galbraith wrote: >> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote: > >>> May be we could try change this back to the old way later, after the aim >>> 7 test on my server. >> >> Yeah, something funny is going on. > > Never entering balance path kills the collapse. Asking wake_affine() > wrt the pull as before, but allowing us to continue should no idle cpu > be found, still collapsed. So the source of funny behavior is indeed in > balance_path. Below patch based on the patch set could help to avoid enter balance path if affine_sd could be found, just like the old logical, would you like to take a try and see whether it could help fix the collapse? Regards, Michael Wang --- kernel/sched/fair.c | 14 ++++++++------ 1 files changed, 8 insertions(+), 6 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index d600708..4e95bb0 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3297,6 +3297,8 @@ next: sg = sg->next; } while (sg != sd->groups); } + + return -1; done: return target; } @@ -3349,7 +3351,7 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) * some cases. */ new_cpu = select_idle_sibling(p, prev_cpu); - if (idle_cpu(new_cpu)) + if (new_cpu != -1) goto unlock; /* @@ -3363,15 +3365,15 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) goto balance_path; new_cpu = select_idle_sibling(p, cpu); - if (!idle_cpu(new_cpu)) - goto balance_path; - /* * Invoke wake_affine() finally since it is no doubt a * performance killer. */ - if (wake_affine(sbm->affine_map[prev_cpu], p, sync)) - goto unlock; + if (new_cpu == -1 || + !wake_affine(sbm->affine_map[prev_cpu], p, sync)) + new_cpu = prev_cpu; + + goto unlock; } balance_path: -- 1.7.4.1 > > -Mike > > -- > 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/ >