[  403.596588] ======================================================
[  403.596618] [ INFO: possible circular locking dependency detected ]
[  403.596649] 3.5.3-00283-g190cb64-dirty #83 Not tainted
[  403.596679] -------------------------------------------------------
[  403.596679] kworker/u:4/482 is trying to acquire lock:
[  403.596710]  (&wl->mutex){+.+.+.}, at: [<bf1718cc>] wl1271_connection_loss_work+0x20/0x7c [wlcore]
[  403.596923]
but task is already holding lock:
[  403.596954]  ((&(&wl->connection_loss_work)->work)){+.+...}, at: [<c0042aa4>] process_one_work+0x1d8/0x438
[  403.597045]
which lock already depends on the new lock.

[  403.597045]
the existing dependency chain (in reverse order) is:
[  403.597076]
-> #1 ((&(&wl->connection_loss_work)->work)){+.+...}:
[  403.597137]        [<c00619f0>] lock_acquire+0x60/0x74
[  403.597167]        [<c0043980>] wait_on_work+0x3c/0xdc
[  403.597198]        [<c0043ad4>] __cancel_work_timer+0xb4/0x100
[  403.597259]        [<bf173178>] wl1271_op_bss_info_changed+0x620/0xb8c [wlcore]
[  403.597381]        [<bf11d4c0>] ieee80211_bss_info_change_notify+0x150/0x170 [mac80211]
[  403.597656]        [<bf143334>] ieee80211_assoc_success+0x3b4/0x568 [mac80211]
[  403.597839]
-> #0 (&wl->mutex){+.+.+.}:
[  403.597900]        [<c0060de4>] __lock_acquire+0xf3c/0x16f0
[  403.597930]        [<c00619f0>] lock_acquire+0x60/0x74
[  403.597961]        [<c048f3a0>] mutex_lock_nested+0x44/0x2fc
[  403.598022]        [<bf1718cc>] wl1271_connection_loss_work+0x20/0x7c [wlcore]
[  403.598144]        [<c0042b28>] process_one_work+0x25c/0x438
[  403.598175]        [<c0042ed8>] worker_thread+0x1a8/0x2dc
[  403.598205]        [<c00475ec>] kthread+0x80/0x90
[  403.598266]        [<c000e10c>] kernel_thread_exit+0x0/0x8
[  403.598297]
other info that might help us debug this:

[  403.598327]  Possible unsafe locking scenario:

[  403.598358]        CPU0                    CPU1
[  403.598358]        ----                    ----
[  403.598388]   lock((&(&wl->connection_loss_work)->work));
[  403.598419]                                lock(&wl->mutex);
[  403.598449]                                lock((&(&wl->connection_loss_work)->work));
[  403.598480]   lock(&wl->mutex);
[  403.598510]
 *** DEADLOCK ***

[  403.598541] 2 locks held by kworker/u:4/482:
[  403.598571]  #0:  (wiphy_name(local->hw.wiphy)){.+.+.+}, at: [<c0042aa4>] process_one_work+0x1d8/0x438
[  403.598632]  #1:  ((&(&wl->connection_loss_work)->work)){+.+...}, at: [<c0042aa4>] process_one_work+0x1d8/0x438
[  403.598693]
stack backtrace:
[  403.598754] [<c00124bc>] (unwind_backtrace+0x0/0xe0) from [<c0487a44>] (print_circular_bug+0x260/0x2ac)
[  403.598815] [<c0487a44>] (print_circular_bug+0x260/0x2ac) from [<c0060de4>] (__lock_acquire+0xf3c/0x16f0)
[  403.598846] [<c0060de4>] (__lock_acquire+0xf3c/0x16f0) from [<c00619f0>] (lock_acquire+0x60/0x74)
[  403.598907] [<c00619f0>] (lock_acquire+0x60/0x74) from [<c048f3a0>] (mutex_lock_nested+0x44/0x2fc)
[  403.599029] [<c048f3a0>] (mutex_lock_nested+0x44/0x2fc) from [<bf1718cc>] (wl1271_connection_loss_work+0x20/0x7c [wlcore])
[  403.599151] [<bf1718cc>] (wl1271_connection_loss_work+0x20/0x7c [wlcore]) from [<c0042b28>] (process_one_work+0x25c/0x438)
[  403.599212] [<c0042b28>] (process_one_work+0x25c/0x438) from [<c0042ed8>] (worker_thread+0x1a8/0x2dc)
[  403.599273] [<c0042ed8>] (worker_thread+0x1a8/0x2dc) from [<c00475ec>] (kthread+0x80/0x90)
[  403.599304] [<c00475ec>] (kthread+0x80/0x90) from [<c000e10c>] (kernel_thread_exit+0x0/0x8)