From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI, SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3CD5AC2BBD1 for ; Thu, 17 Sep 2020 10:49:43 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id CFD8521655 for ; Thu, 17 Sep 2020 10:49:42 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=linutronix.de header.i=@linutronix.de header.b="Oz+gjsdf"; dkim=permerror (0-bit key) header.d=linutronix.de header.i=@linutronix.de header.b="tVOqmwi8" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726709AbgIQKtb (ORCPT ); Thu, 17 Sep 2020 06:49:31 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52816 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726657AbgIQKsv (ORCPT ); Thu, 17 Sep 2020 06:48:51 -0400 Received: from galois.linutronix.de (Galois.linutronix.de [IPv6:2a0a:51c0:0:12e:550::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 22F19C061788 for ; Thu, 17 Sep 2020 03:48:50 -0700 (PDT) Message-Id: <20200917101624.096155774@linutronix.de> DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020; t=1600339722; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: references:references; bh=F4LX5M8oToHpRgZAS6ecPSJjSmNxKQ0e3pPx4ZO3+sE=; b=Oz+gjsdfV2rrbArmDXZ5kJ80ksY56I9MhJPxAzE3l/7X7jaHe8tFBTtjrvWbKUXXN3jkXE GxL6Mluqjtyh92//He6wkEoDxZ2jkPBiZYArKQp97M5FkV5fUEZe+APteJnT2WKIOxkxXI bwkVgplEjwQNV6tLtJb0ch4KnNfUE45CMOzBe/wEb+bz5OcJY73LLM9hR/Zed0YYJ0udXT 9BFNQyAASjYeDdfYBrme8x2csGYHIkDJL8ePD2/aZ6r8ZeHzyc4zD+DT2I1L2nOwtCxIMj cg4XSq/qQ4xi5WCuXh8oaMLqQnUlEROB3Pd8rH6ZTd+XHr6MkH7d1OzkgC5BRw== DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020e; t=1600339722; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: references:references; bh=F4LX5M8oToHpRgZAS6ecPSJjSmNxKQ0e3pPx4ZO3+sE=; b=tVOqmwi8xRJIfqSCK9pzAWz5tpZS5fivTlEzMb7EPaVgXJ3q4MUXxXfJd9igkOwHWCrMzO 0kEs50VV5uRR4eAA== Date: Thu, 17 Sep 2020 11:42:04 +0200 From: Thomas Gleixner To: LKML Cc: Sebastian Siewior , Qais Yousef , "Peter Zijlstra (Intel)" , Scott Wood , Valentin Schneider , Ingo Molnar , Peter Zijlstra , Juri Lelli , Vincent Guittot , Dietmar Eggemann , Steven Rostedt , Ben Segall , Mel Gorman , Daniel Bristot de Oliveira , Vincent Donnefort Subject: [patch 02/10] sched/hotplug: Ensure only per-cpu kthreads run during hotplug References: <20200917094202.301694311@linutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-transfer-encoding: 8-bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org In preparation for migrate_disable(), make sure only per-cpu kthreads are allowed to run on !active CPUs. This is ran (as one of the very first steps) from the cpu-hotplug task which is a per-cpu kthread and completion of the hotplug operation only requires such tasks. This constraint enables the migrate_disable() implementation to wait for completion of all migrate_disable regions on this CPU at hotplug time without fear of any new ones starting. This replaces the unlikely(rq->balance_callbacks) test at the tail of context_switch with an unlikely(rq->balance_work), the fast path is not affected. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20200916121020.GP2674@hirez.programming.kicks-ass.net --- kernel/sched/core.c | 112 ++++++++++++++++++++++++++++++++++++++++++++++++++- kernel/sched/sched.h | 5 ++ 2 files changed, 115 insertions(+), 2 deletions(-) --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3513,8 +3513,10 @@ static inline struct callback_head *spli struct callback_head *head = rq->balance_callback; lockdep_assert_held(&rq->lock); - if (head) + if (head) { rq->balance_callback = NULL; + rq->balance_flags &= ~BALANCE_WORK; + } return head; } @@ -3535,6 +3537,22 @@ static inline void balance_callbacks(str } } +static bool balance_push(struct rq *rq); + +static inline void balance_switch(struct rq *rq) +{ + if (unlikely(rq->balance_flags)) { + /* + * Run the balance_callbacks, except on hotplug + * when we need to push the current task away. + */ + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || + !(rq->balance_flags & BALANCE_PUSH) || + !balance_push(rq)) + __balance_callbacks(rq); + } +} + #else static inline void __balance_callbacks(struct rq *rq) @@ -3550,6 +3568,8 @@ static inline void balance_callbacks(str { } +static inline void balance_switch(struct rq *rq) { } + #endif static inline void @@ -3577,7 +3597,7 @@ static inline void finish_lock_switch(st * prev into current: */ spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_); - __balance_callbacks(rq); + balance_switch(rq); raw_spin_unlock_irq(&rq->lock); } @@ -6833,6 +6853,89 @@ static void migrate_tasks(struct rq *dea rq->stop = stop; } + +static int __balance_push_cpu_stop(void *arg) +{ + struct task_struct *p = arg; + struct rq *rq = this_rq(); + struct rq_flags rf; + int cpu; + + raw_spin_lock_irq(&p->pi_lock); + rq_lock(rq, &rf); + + update_rq_clock(rq); + + if (task_rq(p) == rq && task_on_rq_queued(p)) { + cpu = select_fallback_rq(rq->cpu, p); + rq = __migrate_task(rq, &rf, p, cpu); + } + + rq_unlock(rq, &rf); + raw_spin_unlock_irq(&p->pi_lock); + + put_task_struct(p); + + return 0; +} + +static DEFINE_PER_CPU(struct cpu_stop_work, push_work); + +/* + * Ensure we only run per-cpu kthreads once the CPU goes !active. + */ +static bool balance_push(struct rq *rq) +{ + struct task_struct *push_task = rq->curr; + + lockdep_assert_held(&rq->lock); + SCHED_WARN_ON(rq->cpu != smp_processor_id()); + + /* + * Both the cpu-hotplug and stop task are in this case and are + * required to complete the hotplug process. + */ + if (is_per_cpu_kthread(push_task)) + return false; + + get_task_struct(push_task); + /* + * Temporarily drop rq->lock such that we can wake-up the stop task. + * Both preemption and IRQs are still disabled. + */ + raw_spin_unlock(&rq->lock); + stop_one_cpu_nowait(rq->cpu, __balance_push_cpu_stop, push_task, + this_cpu_ptr(&push_work)); + /* + * At this point need_resched() is true and we'll take the loop in + * schedule(). The next pick is obviously going to be the stop task + * which is_per_cpu_kthread() and will push this task away. + */ + raw_spin_lock(&rq->lock); + + return true; +} + +static void balance_push_set(int cpu, bool on) +{ + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; + + rq_lock_irqsave(rq, &rf); + if (on) + rq->balance_flags |= BALANCE_PUSH; + else + rq->balance_flags &= ~BALANCE_PUSH; + rq_unlock_irqrestore(rq, &rf); +} + +#else + +static inline bool balance_push(struct rq *rq) +{ + return false; +} + #endif /* CONFIG_HOTPLUG_CPU */ void set_rq_online(struct rq *rq) @@ -6918,6 +7021,8 @@ int sched_cpu_activate(unsigned int cpu) struct rq *rq = cpu_rq(cpu); struct rq_flags rf; + balance_push_set(cpu, false); + #ifdef CONFIG_SCHED_SMT /* * When going up, increment the number of cores with SMT present. @@ -6965,6 +7070,8 @@ int sched_cpu_deactivate(unsigned int cp */ synchronize_rcu(); + balance_push_set(cpu, true); + #ifdef CONFIG_SCHED_SMT /* * When going down, decrement the number of cores with SMT present. @@ -6978,6 +7085,7 @@ int sched_cpu_deactivate(unsigned int cp ret = cpuset_cpu_inactive(cpu); if (ret) { + balance_push_set(cpu, false); set_cpu_active(cpu, true); return ret; } --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -973,6 +973,7 @@ struct rq { unsigned long cpu_capacity_orig; struct callback_head *balance_callback; + unsigned char balance_flags; unsigned char nohz_idle_balance; unsigned char idle_balance; @@ -1384,6 +1385,9 @@ init_numa_balancing(unsigned long clone_ #ifdef CONFIG_SMP +#define BALANCE_WORK 0x01 +#define BALANCE_PUSH 0x02 + static inline void queue_balance_callback(struct rq *rq, struct callback_head *head, @@ -1397,6 +1401,7 @@ queue_balance_callback(struct rq *rq, head->func = (void (*)(struct callback_head *))func; head->next = rq->balance_callback; rq->balance_callback = head; + rq->balance_flags |= BALANCE_WORK; } #define rcu_dereference_check_sched_domain(p) \