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 S1754144AbbHCPFg (ORCPT <rfc822;w@1wt.eu>);
	Mon, 3 Aug 2015 11:05:36 -0400
Received: from mail-pa0-f43.google.com ([209.85.220.43]:36521 "EHLO
	mail-pa0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752403AbbHCPFe (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Mon, 3 Aug 2015 11:05:34 -0400
Date: Mon, 3 Aug 2015 23:05:23 +0800
From: Minfei Huang <mnfhuang@gmail.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: yalin wang <yalin.wang2010@gmail.com>, tj@kernel.org, mingo@redhat.com,
        linux-kernel@vger.kernel.org, mhuang@redhat.com
Subject: Re: [PATCH] workqueue: Add the allocation flags to function
 schedule_on_each_cpu_gfp
Message-ID: <20150803150523.GA3116@t440s.lenovo>
References: <1438590425-30307-1-git-send-email-mnfhuang@gmail.com>
 <4F413CE0-A857-4350-AE04-176AA8ADC44E@gmail.com>
 <20150803100449.52f4dadb@gandalf.local.home>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150803100449.52f4dadb@gandalf.local.home>
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 08/03/15 at 10:04am, Steven Rostedt wrote:
> On Mon, 3 Aug 2015 17:15:53 +0800
> yalin wang <yalin.wang2010@gmail.com> wrote:
> 
> > better to also provide a wrapper function with name schedule_on_each_cpu(), 
> > as this function is used frequently .
> > 
> > #define schedule_on_each_cpu(f)  schedule_on_each_cpu_gfp(f, GFP_KERNEL) 
> 
> I was about to say pretty much the same thing. But please make it an
> inline function:
> 
> static inline int schedule_on_each_cpu(work_func_t func)
> {
> 	return schedule_on_each_cpu_gfp(func, GFP_KERNEL);
> }
> 
> Otherwise, NACK to the patch to the ftrace code.

Hi, Steve.

The main reason I posted this patch is to fix the data race bug, when
ftrace tries to free the ops->trampoline in arch x86.

Function schedule_on_each_cpu may fail to alloc percpu work to
synchronise each online cpu. In such situation, trying to free the
trampoline may casue the kernel crash, because one cpu may be executing
the trampoline at this moment.

So I add a new wrapper function to fix it. 

Thanks
Minfei