From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752230AbeCWTsI (ORCPT ); Fri, 23 Mar 2018 15:48:08 -0400 Received: from mail-yw0-f193.google.com ([209.85.161.193]:44029 "EHLO mail-yw0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752122AbeCWTr4 (ORCPT ); Fri, 23 Mar 2018 15:47:56 -0400 X-Google-Smtp-Source: AG47ELvM8OSlq08S/gxmHqERSPGirbxq0vFA5fNhgKtYJPKvmU/1g4W4/DKduPRMrDcidms0hDKQdA== Date: Fri, 23 Mar 2018 12:47:53 -0700 From: Tejun Heo To: Li Zefan , Johannes Weiner Cc: cgroups@vger.kernel.org, kernel-team@fb.com, linux-kernel@vger.kernel.org Subject: [PATCH cgroup/for-4.17 2/2] Subject: cgroup: Limit event generation frequency Message-ID: <20180323194753.GF2149215@devbig577.frc2.facebook.com> References: <20180323194722.GE2149215@devbig577.frc2.facebook.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180323194722.GE2149215@devbig577.frc2.facebook.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org ".events" files generate file modified event to notify userland of possible new events. Some of the events can be quite bursty (e.g. memory high event) and generating notification each time is costly and pointless. This patch implements a event rate limit mechanism. If a new notification is requested before 10ms has passed since the previous notification, the new notification is delayed till then. As this only delays from the second notification on in a given close cluster of notifications, userland reactions to notifications shouldn't be delayed at all in most cases while avoiding notification storms. Signed-off-by: Tejun Heo --- include/linux/cgroup-defs.h | 2 ++ kernel/cgroup/cgroup.c | 25 +++++++++++++++++++++++-- 2 files changed, 25 insertions(+), 2 deletions(-) --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -105,6 +105,8 @@ enum { struct cgroup_file { /* do not access any fields from outside cgroup core */ struct kernfs_node *kn; + unsigned long notified_at; + struct timer_list notify_timer; }; /* --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -61,6 +61,8 @@ #define CGROUP_FILE_NAME_MAX (MAX_CGROUP_TYPE_NAMELEN + \ MAX_CFTYPE_NAME + 2) +/* let's not notify more than 100 times per second */ +#define CGROUP_FILE_NOTIFY_MIN_INTV DIV_ROUND_UP(HZ, 100) /* * cgroup_mutex is the master lock. Any modification to cgroup or its @@ -1554,6 +1556,8 @@ static void cgroup_rm_file(struct cgroup spin_lock_irq(&cgroup_file_kn_lock); cfile->kn = NULL; spin_unlock_irq(&cgroup_file_kn_lock); + + del_timer_sync(&cfile->notify_timer); } kernfs_remove_by_name(cgrp->kn, cgroup_file_name(cgrp, cft, name)); @@ -3532,6 +3536,12 @@ static int cgroup_kn_set_ugid(struct ker return kernfs_setattr(kn, &iattr); } +static void cgroup_file_notify_timer(struct timer_list *timer) +{ + cgroup_file_notify(container_of(timer, struct cgroup_file, + notify_timer)); +} + static int cgroup_add_file(struct cgroup_subsys_state *css, struct cgroup *cgrp, struct cftype *cft) { @@ -3558,6 +3568,8 @@ static int cgroup_add_file(struct cgroup if (cft->file_offset) { struct cgroup_file *cfile = (void *)css + cft->file_offset; + timer_setup(&cfile->notify_timer, cgroup_file_notify_timer, 0); + spin_lock_irq(&cgroup_file_kn_lock); cfile->kn = kn; spin_unlock_irq(&cgroup_file_kn_lock); @@ -3807,8 +3819,17 @@ void cgroup_file_notify(struct cgroup_fi unsigned long flags; spin_lock_irqsave(&cgroup_file_kn_lock, flags); - if (cfile->kn) - kernfs_notify(cfile->kn); + if (cfile->kn) { + unsigned long last = cfile->notified_at; + unsigned long next = last + CGROUP_FILE_NOTIFY_MIN_INTV; + + if (time_in_range(jiffies, last, next)) { + timer_reduce(&cfile->notify_timer, next); + } else { + kernfs_notify(cfile->kn); + cfile->notified_at = jiffies; + } + } spin_unlock_irqrestore(&cgroup_file_kn_lock, flags); }