From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756917AbZBPKfB (ORCPT ); Mon, 16 Feb 2009 05:35:01 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754482AbZBPKex (ORCPT ); Mon, 16 Feb 2009 05:34:53 -0500 Received: from casper.infradead.org ([85.118.1.10]:56283 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754405AbZBPKew (ORCPT ); Mon, 16 Feb 2009 05:34:52 -0500 Subject: Re: [locking bug] inotify: inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. From: Peter Zijlstra To: Ingo Molnar Cc: Nick Piggin , linux-kernel@vger.kernel.org, Andrew Morton In-Reply-To: <20090216081304.GA18980@elte.hu> References: <20090216081304.GA18980@elte.hu> Content-Type: text/plain Date: Mon, 16 Feb 2009 11:34:40 +0100 Message-Id: <1234780480.4703.13.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.25.90 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2009-02-16 at 09:13 +0100, Ingo Molnar wrote: > From 3023a3e27b909b3b309edac4bc9beb8103f127fa Mon Sep 17 00:00:00 2001 > From: Ingo Molnar > Date: Mon, 16 Feb 2009 09:10:32 +0100 > Subject: [PATCH] inotify: fix GFP_KERNEL related deadlock > > enhanced lockdep coverage of __GFP_NOFS turned up this new > lockdep assert: > > [ 1093.677775] > [ 1093.677781] ================================= > [ 1093.680031] [ INFO: inconsistent lock state ] > [ 1093.680031] 2.6.29-rc5-tip-01504-gb49eca1-dirty #1 > [ 1093.680031] --------------------------------- > [ 1093.680031] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. > [ 1093.680031] kswapd0/308 [HC0[0]:SC0[0]:HE1:SE1] takes: > [ 1093.680031] (&inode->inotify_mutex){+.+.?.}, at: [] inotify_inode_is_dead+0x20/0x80 > [ 1093.680031] {RECLAIM_FS-ON-W} state was registered at: > [ 1093.680031] [] mark_held_locks+0x43/0x5b > [ 1093.680031] [] lockdep_trace_alloc+0x6c/0x6e > [ 1093.680031] [] kmem_cache_alloc+0x20/0x150 > [ 1093.680031] [] idr_pre_get+0x27/0x6c > [ 1093.680031] [] inotify_handle_get_wd+0x25/0xad > [ 1093.680031] [] inotify_add_watch+0x7a/0x129 > [ 1093.680031] [] sys_inotify_add_watch+0x20f/0x250 > [ 1093.680031] [] sysenter_do_call+0x12/0x35 > [ 1093.680031] [] 0xffffffff > [ 1093.680031] irq event stamp: 60417 > [ 1093.680031] hardirqs last enabled at (60417): [] call_rcu+0x53/0x59 > [ 1093.680031] hardirqs last disabled at (60416): [] call_rcu+0x17/0x59 > [ 1093.680031] softirqs last enabled at (59656): [] __do_softirq+0x157/0x16b > [ 1093.680031] softirqs last disabled at (59651): [] do_softirq+0x74/0x15d > [ 1093.680031] > [ 1093.680031] other info that might help us debug this: > [ 1093.680031] 2 locks held by kswapd0/308: > [ 1093.680031] #0: (shrinker_rwsem){++++..}, at: [] shrink_slab+0x36/0x189 > [ 1093.680031] #1: (&type->s_umount_key#4){+++++.}, at: [] shrink_dcache_memory+0x110/0x1fb > [ 1093.680031] > [ 1093.680031] stack backtrace: > [ 1093.680031] Pid: 308, comm: kswapd0 Not tainted 2.6.29-rc5-tip-01504-gb49eca1-dirty #1 > [ 1093.680031] Call Trace: > [ 1093.680031] [] valid_state+0x12a/0x13d > [ 1093.680031] [] mark_lock+0xc1/0x1e9 > [ 1093.680031] [] ? check_usage_forwards+0x0/0x3f > [ 1093.680031] [] __lock_acquire+0x2c6/0xac8 > [ 1093.680031] [] ? register_lock_class+0x17/0x228 > [ 1093.680031] [] lock_acquire+0x5d/0x7a > [ 1093.680031] [] ? inotify_inode_is_dead+0x20/0x80 > [ 1093.680031] [] __mutex_lock_common+0x3a/0x4cb > [ 1093.680031] [] ? inotify_inode_is_dead+0x20/0x80 > [ 1093.680031] [] mutex_lock_nested+0x2e/0x36 > [ 1093.680031] [] ? inotify_inode_is_dead+0x20/0x80 > [ 1093.680031] [] inotify_inode_is_dead+0x20/0x80 > [ 1093.680031] [] dentry_iput+0x90/0xc2 > [ 1093.680031] [] d_kill+0x21/0x45 > [ 1093.680031] [] __shrink_dcache_sb+0x27f/0x355 > [ 1093.680031] [] shrink_dcache_memory+0x15e/0x1fb > [ 1093.680031] [] shrink_slab+0x121/0x189 > [ 1093.680031] [] kswapd+0x39f/0x561 > [ 1093.680031] [] ? isolate_pages_global+0x0/0x233 > [ 1093.680031] [] ? autoremove_wake_function+0x0/0x43 > [ 1093.680031] [] ? kswapd+0x0/0x561 > [ 1093.680031] [] kthread+0x41/0x82 > [ 1093.680031] [] ? kthread+0x0/0x82 > [ 1093.680031] [] kernel_thread_helper+0x7/0x10 > > inotify_handle_get_wd() does idr_pre_get() which does a > kmem_cache_alloc() without __GFP_FS - and is hence deadlockable under > extreme MM pressure. > > Signed-off-by: Ingo Molnar Looks like a good catch, Acked-by: Peter Zijlstra > --- > fs/notify/inotify/inotify.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/fs/notify/inotify/inotify.c b/fs/notify/inotify/inotify.c > index dae3f28..331f2e8 100644 > --- a/fs/notify/inotify/inotify.c > +++ b/fs/notify/inotify/inotify.c > @@ -156,7 +156,7 @@ static int inotify_handle_get_wd(struct inotify_handle *ih, > int ret; > > do { > - if (unlikely(!idr_pre_get(&ih->idr, GFP_KERNEL))) > + if (unlikely(!idr_pre_get(&ih->idr, GFP_NOFS))) > return -ENOSPC; > ret = idr_get_new_above(&ih->idr, watch, ih->last_wd+1, &watch->wd); > } while (ret == -EAGAIN);