Re: [ovs-dev] [PATCH net] openvswitch: fix a possible deadlock and lockdep warning

2014-03-28 Thread David Miller
From: Flavio Leitner Date: Thu, 27 Mar 2014 09:53:56 -0300 > There are two problematic situations. > > A deadlock can happen when is_percpu is false because it can get > interrupted while holding the spinlock. Then it executes > ovs_flow_stats_update() in softirq context which tries to get > the

[ovs-dev] [PATCH net] openvswitch: fix a possible deadlock and lockdep warning

2014-03-27 Thread Flavio Leitner
There are two problematic situations. A deadlock can happen when is_percpu is false because it can get interrupted while holding the spinlock. Then it executes ovs_flow_stats_update() in softirq context which tries to get the same lock. The second sitation is that when is_percpu is true, the code

Re: [ovs-dev] [PATCH net] openvswitch: fix a possible deadlock and lockdep warning

2014-03-27 Thread Flavio Leitner
Please ignore this one. I will repost with the correct openvswitch address. Thanks On Wed, Mar 26, 2014 at 09:56:12PM -0300, Flavio Leitner wrote: > There are two problematic situations. > > A deadlock can happen when is_percpu is false because it can get > interrupted while holding the spinlock.

[ovs-dev] [PATCH net] openvswitch: fix a possible deadlock and lockdep warning

2014-03-26 Thread Flavio Leitner
There are two problematic situations. A deadlock can happen when is_percpu is false because it can get interrupted while holding the spinlock. Then it executes ovs_flow_stats_update() in softirq context which tries to get the same lock. The second sitation is that when is_percpu is true, the code