[ovs-dev] [PATCH v3 net-next] openvswitch: fix vport-netdev unregister

2013-10-13 Thread Alexei Starovoitov
The combination of two commits: commit 8e4e1713e4 ("openvswitch: Simplify datapath locking.") commit 2537b4dd0a ("openvswitch:: link upper device for port devices") introduced a bug where upper_dev wasn't unlinked upon netdev_unregister notification The following steps: modprobe openvswitch

Re: [ovs-dev] [PATCH v2 net-next] openvswitch: fix vport-netdev unregister

2013-10-13 Thread Alexei Starovoitov
On Sun, Oct 13, 2013 at 2:22 PM, Cong Wang wrote: > On Sat, 12 Oct 2013 at 01:12 GMT, Alexei Starovoitov > wrote: >> @@ -87,7 +81,11 @@ static int dp_device_event(struct notifier_block *unused, >> unsigned long event, >> if (!vport) >> return NOTIFY_DONE; >> >> - if (eve

Re: [ovs-dev] [PATCH v2 net-next] openvswitch: fix vport-netdev unregister

2013-10-13 Thread Cong Wang
On Sat, 12 Oct 2013 at 01:12 GMT, Alexei Starovoitov wrote: > @@ -87,7 +81,11 @@ static int dp_device_event(struct notifier_block *unused, > unsigned long event, > if (!vport) > return NOTIFY_DONE; > > - if (event == NETDEV_UNREGISTER) { > + if (event == NETDEV_UNREG

Re: [ovs-dev] [PATCH] monitor: Fix improper use of stubs.

2013-10-13 Thread Alex Wang
Thanks Simon for the information, Yes, I think I have fixed this bug. I have tried in on both physical machine and vm On Sun, Oct 13, 2013 at 12:43 AM, Simon Horman wrote: > To clarify, > > I was running the test-suite with a concurrency level of 2 > and the -e flag set. The first two tests f

Re: [ovs-dev] [PATCH] monitor: Fix improper use of stubs.

2013-10-13 Thread Simon Horman
To clarify, I was running the test-suite with a concurrency level of 2 and the -e flag set. The first two tests failed and then the test-run exited. I am happy to run test suite to answer your question about the other tests but a subsequent email from you seems to indicate that you may have found