net: openvswitch: fixes crash if nf_conncount_init() fails
authorEelco Chaudron <echaudro@redhat.com>
Mon, 31 Aug 2020 09:57:57 +0000 (11:57 +0200)
committerDavid S. Miller <davem@davemloft.net>
Tue, 1 Sep 2020 20:23:23 +0000 (13:23 -0700)
If nf_conncount_init fails currently the dispatched work is not canceled,
causing problems when the timer fires. This change fixes this by not
scheduling the work until all initialization is successful.

Fixes: a65878d6f00b ("net: openvswitch: fixes potential deadlock in dp cleanup code")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/openvswitch/datapath.c

index bf701b7..00df39b 100644 (file)
@@ -2481,13 +2481,19 @@ error:
 static int __net_init ovs_init_net(struct net *net)
 {
        struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
+       int err;
 
        INIT_LIST_HEAD(&ovs_net->dps);
        INIT_WORK(&ovs_net->dp_notify_work, ovs_dp_notify_wq);
        INIT_DELAYED_WORK(&ovs_net->masks_rebalance, ovs_dp_masks_rebalance);
+
+       err = ovs_ct_init(net);
+       if (err)
+               return err;
+
        schedule_delayed_work(&ovs_net->masks_rebalance,
                              msecs_to_jiffies(DP_MASKS_REBALANCE_INTERVAL));
-       return ovs_ct_init(net);
+       return 0;
 }
 
 static void __net_exit list_vports_from_net(struct net *net, struct net *dnet,