From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1754029Ab0DTIzL (ORCPT <rfc822;w@1wt.eu>);
	Tue, 20 Apr 2010 04:55:11 -0400
Received: from casper.infradead.org ([85.118.1.10]:43821 "EHLO
	casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753871Ab0DTIzJ (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 20 Apr 2010 04:55:09 -0400
Subject: Re: [patch v2 1/2] sched: check for prev_cpu == this_cpu before
 calling wake_affine()
From: Peter Zijlstra <peterz@infradead.org>
To: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Mike Galbraith <efault@gmx.de>, Ingo Molnar <mingo@elte.hu>,
       Arjan van de Ven <arjan@linux.jf.intel.com>,
       "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
       Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
       Yanmin Zhang <yanmin_zhang@linux.jf.intel.com>,
       Gautham R Shenoy <ego@in.ibm.com>
In-Reply-To: <1271753212.1676.400.camel@laptop>
References: <20100308221946.842728363@sbs-t61.sc.intel.com>
	 <1270031119.5003.93.camel@laptop>
	 <1270079265.7835.8.camel@sbs-t61.sc.intel.com>
	 <1271753212.1676.400.camel@laptop>
Content-Type: text/plain; charset="UTF-8"
Date: Tue, 20 Apr 2010 10:55:06 +0200
Message-ID: <1271753706.1676.402.camel@laptop>
Mime-Version: 1.0
X-Mailer: Evolution 2.28.3 
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, 2010-04-20 at 10:46 +0200, Peter Zijlstra wrote:
> So should we be using idle_cpu(i) instead?

something like the below..

---
Index: linux-2.6/kernel/sched_fair.c
===================================================================
--- linux-2.6.orig/kernel/sched_fair.c
+++ linux-2.6/kernel/sched_fair.c
@@ -1375,26 +1375,25 @@ find_idlest_cpu(struct sched_group *grou
 /*
  * Try and locate an idle CPU in the sched_domain.
  */
-static int
-select_idle_sibling(struct task_struct *p, int target)
+static int select_idle_sibling(struct task_struct *p, int target)
 {
 	int cpu = smp_processor_id();
 	int prev_cpu = task_cpu(p);
-	int i;
 	struct sched_domain *sd;
+	int i;
 
 	/*
 	 * If the task is going to be woken-up on this cpu and if it is
 	 * already idle, then it is the right target.
 	 */
-	if (target == cpu && !cpu_rq(cpu)->cfs.nr_running)
+	if (target == cpu && idle_cpu(cpu))
 		return cpu;
 
 	/*
 	 * If the task is going to be woken-up on the cpu where it previously
 	 * ran and if it is currently idle, then it the right target.
 	 */
-	if (target == prev_cpu && !cpu_rq(prev_cpu)->cfs.nr_running)
+	if (target == prev_cpu && idle_cpu(prev_cpu))
 		return prev_cpu;
 
 	/*
@@ -1405,7 +1404,7 @@ select_idle_sibling(struct task_struct *
 			break;
 
 		for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
-			if (!cpu_rq(i)->cfs.nr_running) {
+			if (idle_cpu(i)) {
 				target = i;
 				break;
 			}
@@ -1479,16 +1478,14 @@ select_task_rq_fair(struct rq *rq, struc
 				want_sd = 0;
 		}
 
-		if (want_affine) {
-			/*
-			 * If both cpu and prev_cpu are part of this domain,
-			 * cpu is a valid SD_WAKE_AFFINE target.
-			 */
-			if (cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))
-			    && (tmp->flags & SD_WAKE_AFFINE)) {
-				affine_sd = tmp;
-				want_affine = 0;
-			}
+		/*
+		 * If both cpu and prev_cpu are part of this domain,
+		 * cpu is a valid SD_WAKE_AFFINE target.
+		 */
+		if (want_affine && (tmp->flags & SD_WAKE_AFFINE) &&
+		    cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))) {
+			affine_sd = tmp;
+			want_affine = 0;
 		}
 
 		if (!want_sd && !want_affine)