From mboxrd@z Thu Jan 1 00:00:00 1970 From: Theodore Ts'o Subject: Re: [PATCH] ext4: fix potential deadlock in ext4_nonda_switch() Date: Thu, 27 Sep 2012 14:09:49 -0400 Message-ID: <20120927180949.GA6862@thunk.org> References: <1348111390-23083-1-git-send-email-tytso@mit.edu> <505CE635.6060101@redhat.com> <20120921235912.GA27207@thunk.org> <87ipaziduh.fsf@openvz.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Eric Sandeen , Ext4 Developers List , stable@vger.kernel.org To: Dmitry Monakhov Return-path: Received: from li9-11.members.linode.com ([67.18.176.11]:52723 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752270Ab2I0SJy (ORCPT ); Thu, 27 Sep 2012 14:09:54 -0400 Content-Disposition: inline In-Reply-To: <87ipaziduh.fsf@openvz.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Thu, Sep 27, 2012 at 09:18:30PM +0400, Dmitry Monakhov wrote: > Hmm... even this '!' patch is still not good. I've got that complain > WARNING: at fs/fs-writeback.c:1316 writeback_inodes_sb_nr+0x1a3/0x200() This WARN_ON checking to make sure the s_umount mutex is grabbed. What kernel are you applying this patch against? Commit 14da92001: "fs: Protect write paths by sb_start_write - sb_end_write" adds a call to sb_start_write() in generic_file_aio_write() which means by the time we call ext4_nonda_switch(), s_umount is grabbed, which probably explains why I'm not seeing this --- 14da92001 was added to mainline as of 3.6-rc1, and I'm guessing you're using an older kernel? This is something we need to consider, though, since "ext4: fix potential deadlock in ext4_nonda_switch()" is currently marked for stable, but commit 14da92001 is not marked for stable. So if this commit gets backported w/o 14da92001, stable kernel users will see this warning. - Ted