From: Ted Ts'o Subject: Re: [URGENT PATCH] ext4: fix potential deadlock in ext4_evict_inode() Date: Fri, 26 Aug 2011 11:52:34 -0400 Message-ID: <20110826155234.GC5176@thunk.org> References: <20110826073507.GZ3162@dastard> <20110826084403.GA3162@dastard> <4E576152.9060405@tao.ma> <20110826092426.GB3162@dastard> <4E57670B.6070205@tao.ma> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Dave Chinner , Jiaying Zhang , linux-ext4@vger.kernel.org To: Tao Ma Return-path: Received: from li9-11.members.linode.com ([67.18.176.11]:54685 "EHLO test.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753679Ab1HZPwl (ORCPT ); Fri, 26 Aug 2011 11:52:41 -0400 Content-Disposition: inline In-Reply-To: <4E57670B.6070205@tao.ma> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Fri, Aug 26, 2011 at 05:27:39PM +0800, Tao Ma wrote: > No, it doesn't mean the ext4_truncate. But another race pasted below. > > Flush inode's i_completed_io_list before calling ext4_io_wait to > prevent the following deadlock scenario: A page fault happens while > some process is writing inode A. During page fault, > shrink_icache_memory is called that in turn evicts another inode > B. Inode B has some pending io_end work so it calls ext4_ioend_wait() > that waits for inode B's i_ioend_count to become zero. However, inode > B's ioend work was queued behind some of inode A's ioend work on the > same cpu's ext4-dio-unwritten workqueue. As the ext4-dio-unwritten > thread on that cpu is processing inode A's ioend work, it tries to > grab inode A's i_mutex lock. Since the i_mutex lock of inode A is > still hold before the page fault happened, we enter a deadlock. ... but that shouldn't be a problem since we're not holding A's i_mutex at this point, right? Or am I missing something? - Ted