From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753619Ab1ABHXK (ORCPT ); Sun, 2 Jan 2011 02:23:10 -0500 Received: from mail.windriver.com ([147.11.1.11]:35271 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753490Ab1ABHXF (ORCPT ); Sun, 2 Jan 2011 02:23:05 -0500 From: Paul Gortmaker To: stable@kernel.org, linux-kernel@vger.kernel.org Cc: stable-review@kernel.org, Mike Galbraith , Peter Zijlstra , Ingo Molnar , Paul Gortmaker Subject: [34-longterm 093/260] sched: Fix select_idle_sibling() Date: Sun, 2 Jan 2011 02:16:29 -0500 Message-Id: <1293952756-15010-94-git-send-email-paul.gortmaker@windriver.com> X-Mailer: git-send-email 1.7.3.3 In-Reply-To: <1293952756-15010-1-git-send-email-paul.gortmaker@windriver.com> References: <1293952756-15010-1-git-send-email-paul.gortmaker@windriver.com> X-OriginalArrivalTime: 02 Jan 2011 07:21:52.0211 (UTC) FILETIME=[BA20DA30:01CBAA4D] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Mike Galbraith commit 8b911acdf08477c059d1c36c21113ab1696c612b upstream. Don't bother with selection when the current cpu is idle. Recent load balancing changes also make it no longer necessary to check wake_affine() success before returning the selected sibling, so we now always use it. Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra LKML-Reference: <1268301369.6785.36.camel@marge.simson.net> Signed-off-by: Ingo Molnar Signed-off-by: Paul Gortmaker --- kernel/sched_fair.c | 14 ++++++++++---- 1 files changed, 10 insertions(+), 4 deletions(-) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 94993ac..ae4d842 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1454,7 +1454,7 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_ int cpu = smp_processor_id(); int prev_cpu = task_cpu(p); int new_cpu = cpu; - int want_affine = 0; + int want_affine = 0, cpu_idle = !current->pid; int want_sd = 1; int sync = wake_flags & WF_SYNC; @@ -1512,13 +1512,15 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_ * If there's an idle sibling in this domain, make that * the wake_affine target instead of the current cpu. */ - if (tmp->flags & SD_SHARE_PKG_RESOURCES) + if (!cpu_idle && tmp->flags & SD_SHARE_PKG_RESOURCES) target = select_idle_sibling(p, tmp, target); if (target >= 0) { if (tmp->flags & SD_WAKE_AFFINE) { affine_sd = tmp; want_affine = 0; + if (target != cpu) + cpu_idle = 1; } cpu = target; } @@ -1534,6 +1536,7 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_ sd = tmp; } +#ifdef CONFIG_FAIR_GROUP_SCHED if (sched_feat(LB_SHARES_UPDATE)) { /* * Pick the largest domain to update shares over @@ -1550,9 +1553,12 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_ raw_spin_lock(&rq->lock); } } +#endif - if (affine_sd && wake_affine(affine_sd, p, sync)) - return cpu; + if (affine_sd) { + if (cpu_idle || cpu == prev_cpu || wake_affine(affine_sd, p, sync)) + return cpu; + } while (sd) { int load_idx = sd->forkexec_idx; -- 1.7.3.3