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 S1752120AbcFFNqM (ORCPT <rfc822;w@1wt.eu>);
	Mon, 6 Jun 2016 09:46:12 -0400
Received: from mail-wm0-f67.google.com ([74.125.82.67]:36238 "EHLO
	mail-wm0-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751345AbcFFNqL (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Mon, 6 Jun 2016 09:46:11 -0400
Message-ID: <1465220766.4236.12.camel@gmail.com>
Subject: Re: [patch] futex: Fix v4.6+ ltp futex_wait04 regression
From: Mike Galbraith <umgwanakikbuti@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>, lkml <linux-kernel@vger.kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        kirill@shutemov.name
Date: Mon, 06 Jun 2016 15:46:06 +0200
In-Reply-To: <1465220056.4236.5.camel@gmail.com>
References: <1465206041.6397.14.camel@gmail.com>
	 <20160606130719.GA30935@twins.programming.kicks-ass.net>
	 <1465220056.4236.5.camel@gmail.com>
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.16.5 
Mime-Version: 1.0
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 Mon, 2016-06-06 at 15:34 +0200, Mike Galbraith wrote:
> On Mon, 2016-06-06 at 15:07 +0200, Peter Zijlstra wrote:
> 
> > > @@ -654,12 +666,14 @@ get_futex_key(u32 __user *uaddr, int fsh
> > >  
> > >  > > 	> > 	> > key->both.offset |= FUT_OFF_INODE;
> > > /* inode-based key */
> > >  > > 	> > 	> > key->shared.inode = inode;
> > > -> > 	> > 	> > key->shared.pgoff =
> > > basepage_index(page);
> > > +> > 	> > 	> > key->shared.pgoff =
> > > basepage_index(pinned);
> > 
> > But this seems to be the actual fix;
> 
> Yup, the three 'pinned' lines alone work just fine.

I was going by what the code used to look like.

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
...
#else
        page_head = compound_head(page);
        if (page != page_head) {
                get_page(page_head);
                put_page(page);
        }
#endif

That all went away, but the pinned page that we used in the end was
stable until Mel's optimization landed, leaving me thinking I needed to
do both.  Hohum, close but no banana.

	-Mike