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 S1752563AbcHKJRl (ORCPT <rfc822;w@1wt.eu>);
	Thu, 11 Aug 2016 05:17:41 -0400
Received: from mga04.intel.com ([192.55.52.120]:52881 "EHLO mga04.intel.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1752434AbcHKJQx (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Thu, 11 Aug 2016 05:16:53 -0400
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.28,504,1464678000"; 
   d="scan'208";a="1033697615"
Subject: Re: [PATCH] x86/irq: do not substract irq_tlb_count from
 irq_call_count
To: Ingo Molnar <mingo@kernel.org>
References: <20160811074430.GA18163@aaronlu.sh.intel.com>
 <20160811091347.GA10990@gmail.com>
Cc: x86@kernel.org, linux-kernel@vger.kernel.org,
        Alex Shi <alex.shi@linaro.org>,
        Tomoki Sekiyama <tomoki.sekiyama.qu@hitachi.com>,
        Davidlohr Bueso <dave@stgolabs.net>, Huang Ying <ying.huang@intel.com>,
        Ingo Molnar <mingo@redhat.com>, Thomas Gleixner <tglx@linutronix.de>,
        "H. Peter Anvin" <hpa@zytor.com>
From: Aaron Lu <aaron.lu@intel.com>
Message-ID: <b750f917-3049-7bb9-11b6-0d9980f8f1e6@intel.com>
Date: Thu, 11 Aug 2016 17:16:49 +0800
MIME-Version: 1.0
In-Reply-To: <20160811091347.GA10990@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 08/11/2016 05:13 PM, Ingo Molnar wrote:
> 
> * Aaron Lu <aaron.lu@intel.com> wrote:
> 
>> This is found by LKP's cyclic performance regression tracking recently
>> with the vm-scalability test suite. I have bisected to commit
>> 0a7ce4b5a632 ("mm/rmap: share the i_mmap_rwsem"). This commit didn't do
>> anything wrong but revealed the irq_call_count problem. IIUC, the commit
>> makes rwc->remap_one in rmap_walk_file concurrent with multiple threads.
>> When remap_one is try_to_unmap_one, then multiple threads could queue
>> flush tlb to the same CPU but only one IPI will be sent.
> 
> Note, for some reason the commit ID you used is wrong, the real one is:

Oops, I looked at a test branch where I cherry-picked that commit, sorry.

> 
>   3dec0ba0be6a ("mm/rmap: share the i_mmap_rwsem")
> 
> I have fixed this in the changelog.

Thanks!