From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755557Ab3KAJYw (ORCPT ); Fri, 1 Nov 2013 05:24:52 -0400 Received: from cantor2.suse.de ([195.135.220.15]:43427 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753987Ab3KAJYv (ORCPT ); Fri, 1 Nov 2013 05:24:51 -0400 Date: Fri, 1 Nov 2013 09:24:46 +0000 From: Mel Gorman To: Chris Mason Cc: Thomas Gleixner , Peter Zijlstra , LKML Subject: Re: [RFC PATCH] futex: Remove requirement for lock_page in get_futex_key Message-ID: <20131101092446.GU2400@suse.de> References: <20131029173814.GH2400@suse.de> <20131029184827.10719.27487@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20131029184827.10719.27487@localhost.localdomain> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Oct 29, 2013 at 02:48:27PM -0400, Chris Mason wrote: > Quoting Mel Gorman (2013-10-29 13:38:14) > > Thomas Gleixner and Peter Zijlstra discussed off-list that real-time users > > currently have a problem with the page lock being contended for unbounded > > periods of time during futex operations. The three of us discussed the > > possibiltity that the page lock is unnecessary in this case because we are > > not concerned with the usual races with reclaim and page cache updates. For > > anonymous pages, the associated futex object is the mm_struct which does > > not require the page lock. For inodes, we should be able to check under > > RCU read lock if the page mapping is still valid to take a reference to > > the inode. This just leaves one rare race that requires the page lock > > in the slow path. This patch does not completely eliminate the page lock > > but it should reduce contention in the majority of cases. > > > > Patch boots and futextest did not explode but I did no comparison > > performance tests. Thomas, do you have details of the workload that > > drove you to examine this problem? Alternatively, can you test it and > > see does it help you? I added Chris to the To list because he mentioned > > that some filesystems might already be doing tricks similar to this > > patch that are worth copying. > > Unfortunately, all the special cases I see in the filesystems either > have an inode ref or are trylocking the page to safety. > Ok, at the time of the futex call there is an implicit ref due to the mapping but it can be torn away underneath us at any time. I *think* I have the right ordering to not make a mistake in this case but more eyes the better. > XFS is a special case because they have their own inode cache, but by my > reading they are still using i_count and free by rcu. > Good, that's what I expected. > The iput in here is a little tricky: > > > > > + /* Should be impossible but lets be paranoid for now */ > > + if (WARN_ON(inode->i_mapping != mapping)) { > > + rcu_read_unlock(); > > + iput(inode); > > + put_page(page_head); > > + goto again; > > + } > > + > > Once you call iput, you add the potential to call the filesystem unlink > operation if i_nlink had gone to zero. This shouldn't be a problem > since you've dropped the rcu lock, but just for fun I'd move the > put_page up a line. > > Or, change it to a BUG_ON instead, it really should be impossible. I'll do that. It'll blow up with the RCU lock still held so the system is going to have a bad day but we're already in hell at this point. Thanks Chris -- Mel Gorman SUSE Labs