From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1751326AbcGMU5T (ORCPT <rfc822;w@1wt.eu>);
	Wed, 13 Jul 2016 16:57:19 -0400
Received: from mail-oi0-f43.google.com ([209.85.218.43]:34133 "EHLO
	mail-oi0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1750897AbcGMU5I (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Wed, 13 Jul 2016 16:57:08 -0400
MIME-Version: 1.0
In-Reply-To: <20160713203944.GC29670@mtj.duckdns.org>
References: <CALAqxLV2H06M2AxP2kK7pGHjLgQAaGdHB9HJEGZjH_Rouy4WhQ@mail.gmail.com>
 <20160713182102.GJ4065@mtj.duckdns.org> <20160713183347.GK4065@mtj.duckdns.org>
 <CALAqxLU76kVm8b5GNnSk25-Xk6D1efiWtwNJbY1th_pcN9Ks=g@mail.gmail.com>
 <20160713201823.GB29670@mtj.duckdns.org> <20160713202657.GW30154@twins.programming.kicks-ass.net>
 <20160713203944.GC29670@mtj.duckdns.org>
From: John Stultz <john.stultz@linaro.org>
Date: Wed, 13 Jul 2016 13:57:06 -0700
Message-ID: <CALAqxLUZ90ZLKKnJhh6Pafm89Ph_syGG4dgTPTVHU-ba_tUXMg@mail.gmail.com>
Subject: Re: Severe performance regression w/ 4.4+ on Android due to cgroup
 locking changes
To: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>,
        lkml <linux-kernel@vger.kernel.org>,
        Dmitry Shmidt <dimitrysh@google.com>,
        Rom Lemarchand <romlem@google.com>, Colin Cross <ccross@google.com>,
        Todd Kjos <tkjos@google.com>, Oleg Nesterov <oleg@redhat.com>,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Wed, Jul 13, 2016 at 1:39 PM, Tejun Heo <tj@kernel.org> wrote:
> Hello,
>
> On Wed, Jul 13, 2016 at 10:26:57PM +0200, Peter Zijlstra wrote:
>> > So, it's a percpu rwsem issue then.  I haven't really followed the
>> > perpcpu rwsem changes closely.  Oleg, are multi-milisec delay expected
>> > on down write expected with the current implementation of
>> > percpu_rwsem?
>>
>> There is a synchronize_sched() in there, so sorta. That thing is heavily
>> geared towards readers, as is the only 'sane' choice for global locks.
>
> It used to use the expedited variant until 001dac627ff3
> ("locking/percpu-rwsem: Make use of the rcu_sync infrastructure"), so
> it might have been okay before then.

Just as a datapoint, just reverting 001dac627ff3
("locking/percpu-rwsem: Make use of the rcu_sync infrastructure"),
doesn't seem to help as much as your prior patch going to a normal
rwsem.

I'm consistently seeing values in the 100-550us range (with occasional
spikes around 6ms), with a max value at 36ms.

thanks
-john