From mboxrd@z Thu Jan  1 00:00:00 1970
Content-Type: multipart/mixed; boundary="===============9197688042165971430=="
MIME-Version: 1.0
From: Davidlohr Bueso <dave@stgolabs.net>
To: lkp@lists.01.org
Subject: Re: [locking/mutexes] cb4bbc457b: -40.0% unixbench.score
Date: Thu, 21 Jan 2016 18:48:01 -0800
Message-ID: <20160122024801.GA23224@linux-uzut.site>
In-Reply-To: <874me6jql0.fsf@yhuang-dev.intel.com>
List-Id: <oe-lkp.lists.linux.dev>

--===============9197688042165971430==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

On Fri, 22 Jan 2016, kernel test robot wrote:

>FYI, we noticed the below changes on
>
>https://github.com/0day-ci/linux Ding-Tianhong/locking-mutexes-don-t-spin-=
on-owner-when-wait-list-is-not-NULL/20160121-173317
>commit cb4bbc457bfed6194ffab1b10c7be73b3f16ca2d ("locking/mutexes: don't s=
pin on owner when wait list is not NULL.")

I'm not sure why this would even be reported, as this patch has not been ac=
cepted
or acked or nothin', by anyone. In this particular case that raw performanc=
e drop
is because spinning is pretty much disabled by Ding's change. Totally expec=
ted for
the kind of workload unixbench triggers.

All this does is hurt lkml-searchability.

Thanks,
Davidlohr

--===============9197688042165971430==--


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 S1752321AbcAVCsS (ORCPT <rfc822;w@1wt.eu>);
	Thu, 21 Jan 2016 21:48:18 -0500
Received: from mx2.suse.de ([195.135.220.15]:38994 "EHLO mx2.suse.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751562AbcAVCsN (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Thu, 21 Jan 2016 21:48:13 -0500
Date: Thu, 21 Jan 2016 18:48:01 -0800
From: Davidlohr Bueso <dave@stgolabs.net>
To: kernel test robot <ying.huang@linux.intel.com>
Cc: Ding Tianhong <dingtianhong@huawei.com>, lkp@01.org,
        LKML <linux-kernel@vger.kernel.org>, Waiman Long <Waiman.Long@hp.com>,
        Tim Chen <tim.c.chen@linux.intel.com>, Jason Low <jason.low2@hp.com>,
        Will Deacon <will.deacon@arm.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        "Paul E. McKenney" <paulmck@us.ibm.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>,
        0day robot <fengguang.wu@intel.com>
Subject: Re: [lkp] [locking/mutexes] cb4bbc457b: -40.0% unixbench.score
Message-ID: <20160122024801.GA23224@linux-uzut.site>
References: <874me6jql0.fsf@yhuang-dev.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
In-Reply-To: <874me6jql0.fsf@yhuang-dev.intel.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, 22 Jan 2016, kernel test robot wrote:

>FYI, we noticed the below changes on
>
>https://github.com/0day-ci/linux Ding-Tianhong/locking-mutexes-don-t-spin-on-owner-when-wait-list-is-not-NULL/20160121-173317
>commit cb4bbc457bfed6194ffab1b10c7be73b3f16ca2d ("locking/mutexes: don't spin on owner when wait list is not NULL.")

I'm not sure why this would even be reported, as this patch has not been accepted
or acked or nothin', by anyone. In this particular case that raw performance drop
is because spinning is pretty much disabled by Ding's change. Totally expected for
the kind of workload unixbench triggers.

All this does is hurt lkml-searchability.

Thanks,
Davidlohr