2009-04-07 01:35:21

by Grissiom

[permalink] [raw]
Subject: [ INFO: inconsistent lock state ] in least git

Hi all,

>From 2.6.29-git1, I constantly got this:

[ 329.464922]
[ 329.464929] =================================
[ 329.464940] [ INFO: inconsistent lock state ]
[ 329.464949] 2.6.29-g14-rfkill-nophy-ledon-09854-gd508afb #93
[ 329.464956] ---------------------------------
[ 329.464963] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W}
usage.
[ 329.464972] kswapd0/32 [HC0[0]:SC0[0]:HE1:SE1] takes:
[ 329.464979] (&inode->inotify_mutex){+.+.?.}, at: [<c01de4eb>]
inotify_inode_is_dead+0x1b/0x80
[ 329.465012] {RECLAIM_FS-ON-W} state was registered at:
[ 329.465019] [<c01546e4>] mark_held_locks+0x64/0x90
[ 329.465035] [<c01547e5>] lockdep_trace_alloc+0xd5/0xe0
[ 329.465048] [<c01adc60>] __kmalloc+0x40/0x220
[ 329.465059] [<c01df465>] kernel_event+0x85/0x130
[ 329.465071] [<c01df58f>] inotify_dev_queue_event+0x7f/0x180
[ 329.465082] [<c01de403>] inotify_inode_queue_event+0xa3/0xe0
[ 329.465093] [<c01bb282>] vfs_create+0xb2/0x100
[ 329.465106] [<c01be365>] do_filp_open+0x6e5/0x820
[ 329.465117] [<c01b04d9>] do_sys_open+0x59/0xf0
[ 329.465127] [<c01b05d9>] sys_open+0x29/0x40
[ 329.465136] [<c0102fa5>] syscall_call+0x7/0xb
[ 329.465148] [<ffffffff>] 0xffffffff
[ 329.465160] irq event stamp: 134535
[ 329.465166] hardirqs last enabled at (134535): [<c017558a>]
call_rcu+0x4a/0x80
[ 329.465184] hardirqs last disabled at (134534): [<c0175563>]
call_rcu+0x23/0x80
[ 329.465196] softirqs last enabled at (133954): [<c0133476>]
__do_softirq+0x156/0x1d0
[ 329.465211] softirqs last disabled at (133947): [<c013354d>]
do_softirq+0x5d/0x70
[ 329.465223]
[ 329.465226] other info that might help us debug this:
[ 329.465233] 2 locks held by kswapd0/32:
[ 329.465239] #0: (shrinker_rwsem){++++..}, at: [<c018f1f4>]
shrink_slab+0x24/0x1b0
[ 329.465260] #1: (&type->s_umount_key#17){++++..}, at:
[<c01c276c>] shrink_dcache_memory+0xfc/0x1b0
[ 329.465283]
[ 329.465284] stack backtrace:
[ 329.465293] Pid: 32, comm: kswapd0 Not tainted
2.6.29-g14-rfkill-nophy-ledon-09854-gd508afb #93
[ 329.465303] Call Trace:
[ 329.465316] [<c0559680>] ? printk+0x18/0x1a
[ 329.465327] [<c0153fd1>] print_usage_bug+0x161/0x1c0
[ 329.465339] [<c0154488>] mark_lock+0x458/0x650
[ 329.465350] [<c0154e20>] ? check_usage_forwards+0x0/0xb0
[ 329.465362] [<c0155b8f>] __lock_acquire+0x3ef/0x18e0
[ 329.465374] [<c01546e4>] ? mark_held_locks+0x64/0x90
[ 329.465386] [<c01de4eb>] ? inotify_inode_is_dead+0x1b/0x80
[ 329.465398] [<c0157148>] lock_acquire+0xc8/0x100
[ 329.465409] [<c01de4eb>] ? inotify_inode_is_dead+0x1b/0x80
[ 329.465421] [<c01de4eb>] ? inotify_inode_is_dead+0x1b/0x80
[ 329.465432] [<c055b1c9>] mutex_lock_nested+0x59/0x370
[ 329.465445] [<c01de4eb>] ? inotify_inode_is_dead+0x1b/0x80
[ 329.465456] [<c01de4eb>] inotify_inode_is_dead+0x1b/0x80
[ 329.465470] [<c01c225c>] dentry_iput+0x9c/0xc0
[ 329.465480] [<c01c2365>] d_kill+0x35/0x60
[ 329.465490] [<c01c259d>] __shrink_dcache_sb+0x20d/0x2e0
[ 329.465502] [<c01c279e>] shrink_dcache_memory+0x12e/0x1b0
[ 329.465514] [<c018f2d5>] shrink_slab+0x105/0x1b0
[ 329.465526] [<c018f88b>] kswapd+0x50b/0x670
[ 329.465540] [<c018d1f0>] ? isolate_pages_global+0x0/0x200
[ 329.465552] [<c0143910>] ? autoremove_wake_function+0x0/0x40
[ 329.465564] [<c018f380>] ? kswapd+0x0/0x670
[ 329.465573] [<c014354c>] kthread+0x3c/0x70
[ 329.465582] [<c0143510>] ? kthread+0x0/0x70
[ 329.465593] [<c01039a7>] kernel_thread_helper+0x7/0x10

What's the root cause of this? Thanks.

--
Cheers,
Grissiom