From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751949AbeADBRl (ORCPT + 1 other); Wed, 3 Jan 2018 20:17:41 -0500 Received: from mga14.intel.com ([192.55.52.115]:52903 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751515AbeADBRk (ORCPT ); Wed, 3 Jan 2018 20:17:40 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.45,504,1508828400"; d="scan'208";a="192058735" From: "Huang\, Ying" To: Mel Gorman Cc: Jan Kara , Mel Gorman , Minchan Kim , Andrew Morton , , , Hugh Dickins , "Paul E . McKenney" , Johannes Weiner , Tim Chen , Shaohua Li , J???r???me Glisse , "Michal Hocko" , Andrea Arcangeli , "David Rientjes" , Rik van Riel , Dave Jiang , Aaron Lu Subject: Re: [PATCH -V4 -mm] mm, swap: Fix race between swapoff and some swap operations References: <20171220012632.26840-1-ying.huang@intel.com> <20171221021619.GA27475@bbox> <871sjopllj.fsf@yhuang-dev.intel.com> <20171221235813.GA29033@bbox> <87r2rmj1d8.fsf@yhuang-dev.intel.com> <20171223013653.GB5279@bgram> <20180102102103.mpah2ehglufwhzle@suse.de> <20180102112955.GA29170@quack2.suse.cz> <20180102132908.hv3qwxqpz7h2jyqp@techsingularity.net> <87o9mbixi0.fsf@yhuang-dev.intel.com> <20180103095408.pqxggi7voser7ia3@techsingularity.net> Date: Thu, 04 Jan 2018 09:17:36 +0800 In-Reply-To: <20180103095408.pqxggi7voser7ia3@techsingularity.net> (Mel Gorman's message of "Wed, 3 Jan 2018 09:54:08 +0000") Message-ID: <87lgheh173.fsf@yhuang-dev.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.2 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Return-Path: Mel Gorman writes: > On Wed, Jan 03, 2018 at 08:42:15AM +0800, Huang, Ying wrote: >> Mel Gorman writes: >> >> > On Tue, Jan 02, 2018 at 12:29:55PM +0100, Jan Kara wrote: >> >> On Tue 02-01-18 10:21:03, Mel Gorman wrote: >> >> > On Sat, Dec 23, 2017 at 10:36:53AM +0900, Minchan Kim wrote: >> >> > > > code path. It appears that similar situation is possible for them too. >> >> > > > >> >> > > > The file cache pages will be delete from file cache address_space before >> >> > > > address_space (embedded in inode) is freed. But they will be deleted >> >> > > > from LRU list only when its refcount dropped to zero, please take a look >> >> > > > at put_page() and release_pages(). While address_space will be freed >> >> > > > after putting reference to all file cache pages. If someone holds a >> >> > > > reference to a file cache page for quite long time, it is possible for a >> >> > > > file cache page to be in LRU list after the inode/address_space is >> >> > > > freed. >> >> > > > >> >> > > > And I found inode/address_space is freed witch call_rcu(). I don't know >> >> > > > whether this is related to page_mapping(). >> >> > > > >> >> > > > This is just my understanding. >> >> > > >> >> > > Hmm, it smells like a bug of __isolate_lru_page. >> >> > > >> >> > > Ccing Mel: >> >> > > >> >> > > What locks protects address_space destroying when race happens between >> >> > > inode trauncation and __isolate_lru_page? >> >> > > >> >> > >> >> > I'm just back online and have a lot of catching up to do so this is a rushed >> >> > answer and I didn't read the background of this. However the question is >> >> > somewhat ambiguous and the scope is broad as I'm not sure which race you >> >> > refer to. For file cache pages, I wouldnt' expect the address_space to be >> >> > destroyed specifically as long as the inode exists which is the structure >> >> > containing the address_space in this case. A page on the LRU being isolated >> >> > in __isolate_lru_page will have an elevated reference count which will >> >> > pin the inode until remove_mapping is called which holds the page lock >> >> > while inode truncation looking at a page for truncation also only checks >> >> > page_mapping under the page lock. Very broadly speaking, pages avoid being >> >> > added back to an inode being freed by checking the I_FREEING state. >> >> >> >> So I'm wondering what prevents the following: >> >> >> >> CPU1 CPU2 >> >> >> >> truncate(inode) __isolate_lru_page() >> >> ... >> >> truncate_inode_page(mapping, page); >> >> delete_from_page_cache(page) >> >> spin_lock_irqsave(&mapping->tree_lock, flags); >> >> __delete_from_page_cache(page, NULL) >> >> page_cache_tree_delete(..) >> >> ... mapping = page_mapping(page); >> >> page->mapping = NULL; >> >> ... >> >> spin_unlock_irqrestore(&mapping->tree_lock, flags); >> >> page_cache_free_page(mapping, page) >> >> put_page(page) >> >> if (put_page_testzero(page)) -> false >> >> - inode now has no pages and can be freed including embedded address_space >> >> >> >> if (mapping && !mapping->a_ops->migratepage) >> >> - we've dereferenced mapping which is potentially already free. >> >> >> > >> > Hmm, possible if unlikely. >> > >> > Before delete_from_page_cache, we called truncate_cleanup_page so the >> > page is likely to be !PageDirty or PageWriteback which gets skipped by >> > the only caller that checks the mappping in __isolate_lru_page. The race >> > is tiny but it does exist. One way of closing it is to check the mapping >> > under the page lock which will prevent races with truncation. The >> > overhead is minimal as the calling context (compaction) is quite a heavy >> > operation anyway. >> > >> >> I think another possible fix is to use call_rcu_sched() to free inode >> (and address_space). Because __isolate_lru_page() will be called with >> LRU spinlock held and IRQ disabled, call_rcu_sched() will wait >> LRU spin_unlock and IRQ enabled. >> > > Maybe, but in this particular case, I would prefer to go with something > more conventional unless there is strong evidence that it's an improvement > (which I doubt in this case given the cost of migration overall and the > corner case of migrating a dirty page). So you like page_lock() more than RCU? Is there any problem of RCU? The object to be protected isn't clear? Another way to fix this with RCU is to replace trylock_page()/unlock_page() with rcu_read_lock()/rcu_read_unlock() in your fix. JFYI, please keep your fix if you think that is more appropriate. Best Regards, Huang, Ying