Message ID | 159886787741.29248.5272329110875821435.stgit@ebuild |
---|---|
State | Accepted |
Delegated to: | David Miller |
Headers | show |
Series | [net-next] net: openvswitch: fixes crash if nf_conncount_init() fails | expand |
On Mon, Aug 31, 2020 at 5:58 PM Eelco Chaudron <echaudro@redhat.com> wrote: > > 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> > --- > net/openvswitch/datapath.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c > index 6e47ef7ef036..78941822119f 100644 > --- a/net/openvswitch/datapath.c > +++ b/net/openvswitch/datapath.c > @@ -2476,13 +2476,19 @@ static int __init dp_register_genl(void) > 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, > > _______________________________________________ > dev mailing list > dev@openvswitch.org > https://mail.openvswitch.org/mailman/listinfo/ovs-dev
From: Eelco Chaudron <echaudro@redhat.com> Date: Mon, 31 Aug 2020 11:57:57 +0200 > 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> Applied, thank you.
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 6e47ef7ef036..78941822119f 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -2476,13 +2476,19 @@ static int __init dp_register_genl(void) 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,
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> --- net/openvswitch/datapath.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)