Bug#754354: Something holds dentry-related mutex forever in Wheezy amd64 kernel

[Mike Crowe <mac@mcrowe.com>]

I compiled my own version of the Debian 3.2.60-1+deb7u3 kernel with
CONFIG_LOCKDEP and panic on hung task enabled.

>From the crash dump:

[25202.156175] INFO: task nfsd:3247 blocked for more than 900 seconds.
[25202.162565] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[25202.170432] nfsd            D ffff88080aa0eca8     0  3247      2 0x00000000
[25202.170444]  ffff88080a8e19f0 0000000000000046 0000000000000006 ffff880800000000
[25202.170458]  ffff88080aa0e9c0 ffff88080a8e1fd8 ffff88080a8e1fd8 00000000001d4040
[25202.170472]  ffff88040e9926c0 ffff88080aa0e9c0 ffffffff8138d6da 00000001a04c47dd
[25202.170488] Call Trace:
[25202.170504]  [<ffffffff8138d6da>] ? __mutex_lock_common+0x236/0x379
[25202.170531]  [<ffffffffa04c47dd>] ? fh_lock_nested+0x4d/0x61 [nfsd]
[25202.170542]  [<ffffffff8138cda2>] schedule+0x55/0x57
[25202.170552]  [<ffffffff8138d6e7>] __mutex_lock_common+0x243/0x379
[25202.170569]  [<ffffffffa04c47dd>] ? fh_lock_nested+0x4d/0x61 [nfsd]
[25202.170581]  [<ffffffff8138d8dc>] mutex_lock_nested+0x2a/0x31
[25202.170598]  [<ffffffffa04c47dd>] fh_lock_nested+0x4d/0x61 [nfsd]
[25202.170610]  [<ffffffff810140f5>] ? sched_clock+0x9/0xd
[25202.170626]  [<ffffffffa04c50fe>] nfsd_lookup_dentry+0x196/0x227 [nfsd]
[25202.170646]  [<ffffffffa04cef7f>] nfsd4_secinfo.part.15+0x26/0x9e [nfsd]
[25202.170666]  [<ffffffffa04cf044>] nfsd4_secinfo+0x4d/0x5b [nfsd]
[25202.170688]  [<ffffffffa04ce105>] nfsd4_proc_compound+0x265/0x43e [nfsd]
[25202.170703]  [<ffffffffa04c181d>] nfsd_dispatch+0xe2/0x1c8 [nfsd]
[25202.170734]  [<ffffffffa03759c1>] svc_process_common+0x2cf/0x4d0 [sunrpc]
[25202.170759]  [<ffffffffa0375de0>] svc_process+0x118/0x136 [sunrpc]
[25202.170773]  [<ffffffffa04c10eb>] nfsd+0xeb/0x131 [nfsd]
[25202.170796]  [<ffffffffa04c1000>] ? 0xffffffffa04c0fff
[25202.170806]  [<ffffffff81065c75>] kthread+0xa3/0xab
[25202.170815]  [<ffffffff81396584>] kernel_thread_helper+0x4/0x10
[25202.170823]  [<ffffffff8138f074>] ? retint_restore_args+0x13/0x13
[25202.170830]  [<ffffffff81065bd2>] ? __init_kthread_worker+0x53/0x53
[25202.170837]  [<ffffffff81396580>] ? gs_change+0x13/0x13
[25202.170842] 1 lock held by nfsd/3247:
[25202.170845]  #0:  (&sb->s_type->i_mutex_key#13){+.+.+.}, at: [<ffffffffa04c47dd>] fh_lock_nested+0x4d/0x61 [nfsd]
[25202.170870] Kernel panic - not syncing: hung_task: blocked tasks

I'm no expert at interpreting lockdep output but I think this is saying
that nfsd is taking a nested lock and then deadlocks trying to take it
again (which presumably shouldn't happen because it is nested.)

Mike.