From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934513Ab3CMX6x (ORCPT ); Wed, 13 Mar 2013 19:58:53 -0400 Received: from mail-qc0-f176.google.com ([209.85.216.176]:57516 "EHLO mail-qc0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934446Ab3CMX6a (ORCPT ); Wed, 13 Mar 2013 19:58:30 -0400 From: Tejun Heo To: linux-kernel@vger.kernel.org Cc: laijs@cn.fujitsu.com, Tejun Heo Subject: [PATCH 3/6] workqueue: fix max_active handling in init_and_link_pwq() Date: Wed, 13 Mar 2013 16:58:15 -0700 Message-Id: <1363219098-16083-4-git-send-email-tj@kernel.org> X-Mailer: git-send-email 1.8.1.4 In-Reply-To: <1363219098-16083-1-git-send-email-tj@kernel.org> References: <1363219098-16083-1-git-send-email-tj@kernel.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Since 9e8cd2f589 ("workqueue: implement apply_workqueue_attrs()"), init_and_link_pwq() may be called to initialize a new pool_workqueue for a workqueue which is already online, but the function was setting pwq->max_active to wq->saved_max_active without proper synchronization. Fix it by calling pwq_adjust_max_active() under proper locking instead of manually setting max_active. Signed-off-by: Tejun Heo --- kernel/workqueue.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 9e2ec4c..7567614 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3771,21 +3771,25 @@ static void init_and_link_pwq(struct pool_workqueue *pwq, pwq->wq = wq; pwq->flush_color = -1; pwq->refcnt = 1; - pwq->max_active = wq->saved_max_active; INIT_LIST_HEAD(&pwq->delayed_works); INIT_LIST_HEAD(&pwq->mayday_node); INIT_WORK(&pwq->unbound_release_work, pwq_unbound_release_workfn); - /* - * Link @pwq and set the matching work_color. This is synchronized - * with flush_mutex to avoid confusing flush_workqueue(). - */ mutex_lock(&wq->flush_mutex); spin_lock_irq(&workqueue_lock); + /* + * Set the matching work_color. This is synchronized with + * flush_mutex to avoid confusing flush_workqueue(). + */ if (p_last_pwq) *p_last_pwq = first_pwq(wq); pwq->work_color = wq->work_color; + + /* sync max_active to the current setting */ + pwq_adjust_max_active(pwq); + + /* link in @pwq */ list_add_rcu(&pwq->pwqs_node, &wq->pwqs); spin_unlock_irq(&workqueue_lock); -- 1.8.1.4