From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757273Ab0KROMJ (ORCPT ); Thu, 18 Nov 2010 09:12:09 -0500 Received: from hera.kernel.org ([140.211.167.34]:52081 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752644Ab0KROMH (ORCPT ); Thu, 18 Nov 2010 09:12:07 -0500 Date: Thu, 18 Nov 2010 14:11:35 GMT From: tip-bot for Paul Turner Cc: linux-kernel@vger.kernel.org, hpa@zytor.com, mingo@redhat.com, a.p.zijlstra@chello.nl, pjt@google.com, tglx@linutronix.de, mingo@elte.hu Reply-To: mingo@redhat.com, hpa@zytor.com, linux-kernel@vger.kernel.org, a.p.zijlstra@chello.nl, pjt@google.com, tglx@linutronix.de, mingo@elte.hu In-Reply-To: <20101115234937.933428187@google.com> References: <20101115234937.933428187@google.com> To: linux-tip-commits@vger.kernel.org Subject: [tip:sched/core] sched: Fix update_cfs_load() synchronization Message-ID: Git-Commit-ID: e33078baa4d30ad1d0e46d1f62b9e5a63a3e6ee3 X-Mailer: tip-git-log-daemon Robot-ID: Robot-Unsubscribe: Contact to get blacklisted from these emails MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.3 (hera.kernel.org [127.0.0.1]); Thu, 18 Nov 2010 14:11:36 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Commit-ID: e33078baa4d30ad1d0e46d1f62b9e5a63a3e6ee3 Gitweb: http://git.kernel.org/tip/e33078baa4d30ad1d0e46d1f62b9e5a63a3e6ee3 Author: Paul Turner AuthorDate: Mon, 15 Nov 2010 15:47:04 -0800 Committer: Ingo Molnar CommitDate: Thu, 18 Nov 2010 13:27:48 +0100 sched: Fix update_cfs_load() synchronization Using cfs_rq->nr_running is not sufficient to synchronize update_cfs_load with the put path since nr_running accounting occurs at deactivation. It's also not safe to make the removal decision based on load_avg as this fails with both high periods and low shares. Resolve this by clipping history after 4 periods without activity. Note: the above will always occur from update_shares() since in the last-task-sleep-case that task will still be cfs_rq->curr when update_cfs_load is called. Signed-off-by: Paul Turner Signed-off-by: Peter Zijlstra LKML-Reference: <20101115234937.933428187@google.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- kernel/sched_fair.c | 33 +++++++++++++++++++++------------ 2 files changed, 22 insertions(+), 13 deletions(-) diff --git a/kernel/sched.c b/kernel/sched.c index 6268d2d..dadab4d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -355,7 +355,7 @@ struct cfs_rq { u64 load_avg; u64 load_period; - u64 load_stamp; + u64 load_stamp, load_last; unsigned long load_contribution; #endif diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index d52b97a..a543a5b 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -674,10 +674,11 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) } #if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED -static void update_cfs_load(struct cfs_rq *cfs_rq, int lb) +static void update_cfs_load(struct cfs_rq *cfs_rq) { u64 period = sched_avg_period(); u64 now, delta; + unsigned long load = cfs_rq->load.weight; if (!cfs_rq) return; @@ -685,9 +686,19 @@ static void update_cfs_load(struct cfs_rq *cfs_rq, int lb) now = rq_of(cfs_rq)->clock; delta = now - cfs_rq->load_stamp; + /* truncate load history at 4 idle periods */ + if (cfs_rq->load_stamp > cfs_rq->load_last && + now - cfs_rq->load_last > 4 * period) { + cfs_rq->load_period = 0; + cfs_rq->load_avg = 0; + } + cfs_rq->load_stamp = now; cfs_rq->load_period += delta; - cfs_rq->load_avg += delta * cfs_rq->load.weight; + if (load) { + cfs_rq->load_last = now; + cfs_rq->load_avg += delta * load; + } while (cfs_rq->load_period > period) { /* @@ -700,10 +711,8 @@ static void update_cfs_load(struct cfs_rq *cfs_rq, int lb) cfs_rq->load_avg /= 2; } - if (lb && !cfs_rq->nr_running) { - if (cfs_rq->load_avg < (period / 8)) - list_del_leaf_cfs_rq(cfs_rq); - } + if (!cfs_rq->curr && !cfs_rq->nr_running && !cfs_rq->load_avg) + list_del_leaf_cfs_rq(cfs_rq); } static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, @@ -750,7 +759,7 @@ static void update_cfs_shares(struct cfs_rq *cfs_rq, long weight_delta) reweight_entity(cfs_rq_of(se), se, shares); } #else /* CONFIG_FAIR_GROUP_SCHED */ -static inline void update_cfs_load(struct cfs_rq *cfs_rq, int lb) +static inline void update_cfs_load(struct cfs_rq *cfs_rq) { } @@ -880,7 +889,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * Update run-time statistics of the 'current'. */ update_curr(cfs_rq); - update_cfs_load(cfs_rq, 0); + update_cfs_load(cfs_rq); update_cfs_shares(cfs_rq, se->load.weight); account_entity_enqueue(cfs_rq, se); @@ -941,7 +950,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (se != cfs_rq->curr) __dequeue_entity(cfs_rq, se); se->on_rq = 0; - update_cfs_load(cfs_rq, 0); + update_cfs_load(cfs_rq); account_entity_dequeue(cfs_rq, se); update_min_vruntime(cfs_rq); update_cfs_shares(cfs_rq, 0); @@ -1176,7 +1185,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - update_cfs_load(cfs_rq, 0); + update_cfs_load(cfs_rq); update_cfs_shares(cfs_rq, 0); } @@ -1206,7 +1215,7 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - update_cfs_load(cfs_rq, 0); + update_cfs_load(cfs_rq); update_cfs_shares(cfs_rq, 0); } @@ -2023,7 +2032,7 @@ static int tg_shares_up(struct task_group *tg, int cpu) raw_spin_lock_irqsave(&rq->lock, flags); update_rq_clock(rq); - update_cfs_load(cfs_rq, 1); + update_cfs_load(cfs_rq); load_avg = div64_u64(cfs_rq->load_avg, cfs_rq->load_period+1); load_avg -= cfs_rq->load_contribution;