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 S1754584AbcAEVLv (ORCPT <rfc822;w@1wt.eu>);
	Tue, 5 Jan 2016 16:11:51 -0500
Received: from mx2.suse.de ([195.135.220.15]:34796 "EHLO mx2.suse.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1752826AbcAEVLs (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Tue, 5 Jan 2016 16:11:48 -0500
Date: Tue, 5 Jan 2016 13:11:40 -0800
From: Davidlohr Bueso <dave@stgolabs.net>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Mel Gorman <mgorman@techsingularity.net>,
        Thomas Gleixner <tglx@linutronix.de>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Chris Mason <clm@fb.com>, Darren Hart <dvhart@linux.intel.com>,
        linux-kernel@vger.kernel.org, Davidlohr Bueso <dbueso@suse.de>
Subject: Re: [PATCH] futex: Reduce the scope of lock_page, aka lockless
 futex_get_key()
Message-ID: <20160105211140.GA15919@linux-uzut.site>
References: <1452025435-15355-1-git-send-email-dave@stgolabs.net>
 <20160105203317.GQ6344@twins.programming.kicks-ass.net>
 <20160105204342.GA15465@linux-uzut.site>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
In-Reply-To: <20160105204342.GA15465@linux-uzut.site>
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 Tue, 05 Jan 2016, Bueso wrote:

>On Tue, 05 Jan 2016, Peter Zijlstra wrote:
>
>>On Tue, Jan 05, 2016 at 12:23:55PM -0800, Davidlohr Bueso wrote:
>>>+	if (unlikely(!mapping)) {
>>>+		int shmem_swizzled;
>>>+
>>>+		/*
>>>+		 * Page lock is required to identify which special case above
>>>+		 * applies. If this is really a shmem page then the page lock
>>>+		 * will prevent unexpected transitions.
>>>+		 */
>>>+		lock_page(page);
>>>+		shmem_swizzled = PageSwapCache(page);
>>> 		unlock_page(page);
>>> 		put_page(page);
>>>+		WARN_ON_ONCE(mapping);
>>
>>We've not re-loaded mapping, so how could this possibly be?
>
>Yep, this wants to be page->mapping.

Actually under compound_head() afaict, along with the rest of the page->mapping
loads.