Re: [Xen-devel] [PATCH V6 net-next 0/5] xen-net{back, front}: Multiple transmit and receive queues

[Sander Eikelenboom <linux@xxxxxxxxxxxxxx>]

Friday, March 14, 2014, 5:06:15 PM, you wrote:

> On Thu, 2014-03-06 at 17:52 +0100, Sander Eikelenboom wrote:
>> Hi Andrew,
>> 
>> Just tried your series but i ran into this lockdep warning:

> In the guest (so netfront), correct?

Erhmm yes correct, sorry for not mentioning that.

>> 
>> [    0.932289]
>> [    0.932293] =============================================
>> [    0.932297] [ INFO: possible recursive locking detected ]
>> [    0.932302] 3.14.0-rc5-20140306-xennext-netnext-bennie+ #1 Not tainted
>> [    0.932306] ---------------------------------------------
>> [    0.932311] xenwatch/26 is trying to acquire lock:
>> [    0.932315]  (&(&queue->rx_lock)->rlock){+.....}, at: 
>> [<ffffffff817b30f4>] netback_changed+0xc84/0xea0
>> [    0.932328]
>> [    0.932328] but task is already holding lock:
>> [    0.932333]  (&(&queue->rx_lock)->rlock){+.....}, at: 
>> [<ffffffff817b30f4>] netback_changed+0xc84/0xea0
>> [    0.932343]
>> [    0.932343] other info that might help us debug this:
>> [    0.932348]  Possible unsafe locking scenario:
>> [    0.932348]
>> [    0.932353]        CPU0
>> [    0.932355]        ----
>> [    0.932358]   lock(&(&queue->rx_lock)->rlock);
>> [    0.932363]   lock(&(&queue->rx_lock)->rlock);
>> [    0.932367]
>> [    0.932367]  *** DEADLOCK ***
>> [    0.932367]
>> [    0.932372]  May be due to missing lock nesting notation
>> [    0.932372]
>> [    0.932378] 3 locks held by xenwatch/26:
>> [    0.935540]  #0:  (xenwatch_mutex){+.+.+.}, at: [<ffffffff81581d96>] 
>> xenwatch_thread+0x86/0x130
>> [    0.935540]  #1:  (&(&queue->rx_lock)->rlock){+.....}, at: 
>> [<ffffffff817b30f4>] netback_changed+0xc84/0xea0
>> [    0.935540]  #2:  (&(&queue->tx_lock)->rlock){......}, at: 
>> [<ffffffff817b3101>] netback_changed+0xc91/0xea0
>> [    0.935540]
>> [    0.935540] stack backtrace:
>> [    0.935540] CPU: 1 PID: 26 Comm: xenwatch Not tainted 
>> 3.14.0-rc5-20140306-xennext-netnext-bennie+ #1
>> [    0.935540]  ffffffff82766230 ffff88001eac3b98 ffffffff81b83684 
>> ffff88001e97d870
>> [    0.935540]  ffffffff82766230 ffff88001eac3c68 ffffffff81115b7e 
>> 00000000000233a0
>> [    0.935540]  ffffffff00000003 ffffffff82766230 ffffffff82ca7ec0 
>> 5001f47aeae10000
>> [    0.935540] Call Trace:
>> [    0.935540]  [<ffffffff81b83684>] dump_stack+0x46/0x58
>> [    0.935540]  [<ffffffff81115b7e>] __lock_acquire+0x86e/0x2220
>> [    0.935540]  [<ffffffff811e40be>] ? kfree+0x1ee/0x200
>> [    0.935540]  [<ffffffff81117b9d>] lock_acquire+0xbd/0x150
>> [    0.935540]  [<ffffffff817b30f4>] ? netback_changed+0xc84/0xea0
>> [    0.935540]  [<ffffffff81b8c4fe>] ? mutex_unlock+0xe/0x10
>> [    0.935540]  [<ffffffff817b00f4>] ? xennet_release_tx_bufs+0x104/0x110
>> [    0.935540]  [<ffffffff81b8d7cf>] _raw_spin_lock_bh+0x3f/0x50
>> [    0.935540]  [<ffffffff817b30f4>] ? netback_changed+0xc84/0xea0
>> [    0.935540]  [<ffffffff817b30f4>] netback_changed+0xc84/0xea0
>> [    0.935540]  [<ffffffff815835f0>] xenbus_otherend_changed+0xb0/0xc0
>> [    0.935540]  [<ffffffff81581d10>] ? xs_watch+0x60/0x60
>> [    0.935540]  [<ffffffff815851d3>] backend_changed+0x13/0x20
>> [    0.935540]  [<ffffffff81581d55>] xenwatch_thread+0x45/0x130
>> [    0.935540]  [<ffffffff8110d590>] ? __init_waitqueue_head+0x60/0x60
>> [    0.935540]  [<ffffffff810ee394>] kthread+0xe4/0x100
>> [    0.935540]  [<ffffffff81b8ddb0>] ? _raw_spin_unlock_irq+0x30/0x50
>> [    0.935540]  [<ffffffff810ee2b0>] ? __init_kthread_worker+0x70/0x70
>> [    0.935540]  [<ffffffff81b8efbc>] ret_from_fork+0x7c/0xb0
>> [    0.935540]  [<ffffffff810ee2b0>] ? __init_kthread_worker+0x70/0x70
>> 
>> 
>> 
>> --
>> Sander
>> 





_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel