From mboxrd@z Thu Jan 1 00:00:00 1970 From: Valdis.Kletnieks@vt.edu Subject: Re: mmotm 2011-06-22 - inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage. Date: Thu, 23 Jun 2011 12:02:11 -0400 Message-ID: <47718.1308844931@turing-police.cc.vt.edu> References: <201106222042.p5MKgiEe025352@imap1.linux-foundation.org> <6189.1308794391@turing-police.cc.vt.edu> <1308843516.1022.139.camel@twins> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="==_Exmh_1308844931_4530P"; micalg=pgp-sha1; protocol="application/pgp-signature" Content-Transfer-Encoding: 7bit Return-path: Received: from lennier.cc.vt.edu ([198.82.162.213]:43470 "EHLO lennier.cc.vt.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753858Ab1FWQJr (ORCPT ); Thu, 23 Jun 2011 12:09:47 -0400 In-Reply-To: Your message of "Thu, 23 Jun 2011 17:38:36 +0200." <1308843516.1022.139.camel@twins> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Peter Zijlstra , Jan Kara , stable@kernel.org Cc: Andrew Morton , Thomas Gleixner , Jeff Garzik , IngIngo Molnar , linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org --==_Exmh_1308844931_4530P Content-Type: text/plain; charset=us-ascii On Thu, 23 Jun 2011 17:38:36 +0200, Peter Zijlstra said: > On Wed, 2011-06-22 at 21:59 -0400, Valdis.Kletnieks@vt.edu wrote: > > [ 20.254275] {SOFTIRQ-ON-W} state was registered at: > > [ 20.254275] [] mark_irqflags+0xf2/0x13e > > [ 20.254275] [] __lock_acquire+0x22b/0x3e2 > > [ 20.254275] [] lock_acquire+0x103/0x153 > > [ 20.254275] [] _raw_spin_lock+0x36/0x45 > > [ 20.254275] [] end_writeback+0x33/0x103 > > [ 20.254275] [] bdev_evict_inode+0x3e/0xbe > > [ 20.254275] [] evict+0xb0/0x173 > > [ 20.254275] [] iput_final+0x171/0x17a > > [ 20.254275] [] iput+0x4e/0x53 > > [ 20.254275] [] __blkdev_put+0x1c0/0x1eb > > [ 20.254275] [] blkdev_put+0x1e6/0x1f5 > > [ 20.254275] [] register_disk+0xea/0x13c > > [ 20.254275] [] add_disk+0x12f/0x1a4 > > [ 20.254275] [] sd_probe_async+0x115/0x1b5 > > [ 20.254275] [] async_run_entry_fn+0x99/0x12a > > [ 20.254275] [] process_one_work+0x25d/0x467 > > [ 20.254275] [] worker_thread+0x152/0x206 > > [ 20.254275] [] kthread+0x7f/0x87 > > [ 20.254275] [] kernel_thread_helper+0x4/0x10=20 > > That looks broken. Not having -mm, is there a git tree some place?, I > cannot quite see how end_writeback() is taking mapping->tree_lock as my > version looks like: > > void end_writeback(struct inode *inode) > { > might_sleep(); > BUG_ON(inode->i_data.nrpages); > BUG_ON(!list_empty(&inode->i_data.private_list)); mm-fix-assertion-mapping-nrpages-==-0-in-end_writeback.patch does this: diff -puN fs/inode.c~mm-fix-assertion-mapping-nrpages-==-0-in-end_writeback fs/inode.c --- a/fs/inode.c~mm-fix-assertion-mapping-nrpages-==-0-in-end_writeback +++ a/fs/inode.c @@ -423,7 +423,14 @@ EXPORT_SYMBOL(remove_inode_hash); void end_writeback(struct inode *inode) { might_sleep(); + /* + * We have to cycle tree_lock here because reclaim can be still in the + * process of removing the last page (in __delete_from_page_cache()) + * and we must not free mapping under it. + */ + spin_lock(&inode->i_data.tree_lock); BUG_ON(inode->i_data.nrpages); + spin_unlock(&inode->i_data.tree_lock); BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); Adding Jan Kara to the list, and stable@kernel.org because the patch was cc'ed to there... --==_Exmh_1308844931_4530P Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Exmh version 2.5 07/13/2001 iD8DBQFOA2ODcC3lWbTT17ARArNGAJoCGWCf5FztlYZwlY1/X1a1ix6dFwCfe0Tc TuL0/wzrpQkHW/hfmivGH3U= =4a9D -----END PGP SIGNATURE----- --==_Exmh_1308844931_4530P--