dev: advertise the new ifindex when the netns iface changes

The goal is to let the user follow an interface that moves to another
netns.

CC: Jiri Benc <jbenc@redhat.com>
CC: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 36 insertions, 20 deletions

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 62d508b31f56..0514cc36ac34 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -19,10 +19,11 @@ extern int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst,
 
 void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change, gfp_t flags);
 void rtmsg_ifinfo_newnet(int type, struct net_device *dev, unsigned int change,
-                         gfp_t flags, int *new_nsid);
+                         gfp_t flags, int *new_nsid, int new_ifindex);
 struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
                                        unsigned change, u32 event,
-                                       gfp_t flags, int *new_nsid);
+                                       gfp_t flags, int *new_nsid,
+                                       int new_ifindex);
 void rtmsg_ifinfo_send(struct sk_buff *skb, struct net_device *dev,
                        gfp_t flags);
 
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 8616131e2c61..6d9447700e18 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -163,6 +163,7 @@ enum {
         IFLA_IF_NETNSID,
         IFLA_CARRIER_UP_COUNT,
         IFLA_CARRIER_DOWN_COUNT,
+        IFLA_NEW_IFINDEX,
         __IFLA_MAX
 };
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 59987eb6511a..858501b12869 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7360,7 +7360,7 @@ static void rollback_registered_many(struct list_head *head)
                 if (!dev->rtnl_link_ops ||
                     dev->rtnl_link_state == RTNL_LINK_INITIALIZED)
                         skb = rtmsg_ifinfo_build_skb(RTM_DELLINK, dev, ~0U, 0,
-                                                     GFP_KERNEL, NULL);
+                                                     GFP_KERNEL, NULL, 0);
 
                 /*
                  *      Flush the unicast and multicast chains
@@ -8473,7 +8473,7 @@ EXPORT_SYMBOL(unregister_netdev);
 
 int dev_change_net_namespace(struct net_device *dev, struct net *net, const char *pat)
 {
-        int err, new_nsid;
+        int err, new_nsid, new_ifindex;
 
         ASSERT_RTNL();
 
@@ -8529,8 +8529,16 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
         call_netdevice_notifiers(NETDEV_UNREGISTER, dev);
         rcu_barrier();
         call_netdevice_notifiers(NETDEV_UNREGISTER_FINAL, dev);
+
         new_nsid = peernet2id_alloc(dev_net(dev), net);
-        rtmsg_ifinfo_newnet(RTM_DELLINK, dev, ~0U, GFP_KERNEL, &new_nsid);
+        /* If there is an ifindex conflict assign a new one */
+        if (__dev_get_by_index(net, dev->ifindex))
+                new_ifindex = dev_new_index(net);
+        else
+                new_ifindex = dev->ifindex;
+
+        rtmsg_ifinfo_newnet(RTM_DELLINK, dev, ~0U, GFP_KERNEL, &new_nsid,
+                            new_ifindex);
 
         /*
          *      Flush the unicast and multicast chains
@@ -8544,10 +8552,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
 
         /* Actually switch the network namespace */
         dev_net_set(dev, net);
-
-        /* If there is an ifindex conflict assign a new one */
-        if (__dev_get_by_index(net, dev->ifindex))
-                dev->ifindex = dev_new_index(net);
+        dev->ifindex = new_ifindex;
 
         /* Send a netdev-add uevent to the new namespace */
         kobject_uevent(&dev->dev.kobj, KOBJ_ADD);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index f111557958bb..e04af7b7f448 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -988,6 +988,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev,
                + rtnl_xdp_size() /* IFLA_XDP */
                + nla_total_size(4)  /* IFLA_EVENT */
                + nla_total_size(4)  /* IFLA_NEW_NETNSID */
+               + nla_total_size(4)  /* IFLA_NEW_IFINDEX */
                + nla_total_size(1)  /* IFLA_PROTO_DOWN */
                + nla_total_size(4)  /* IFLA_IF_NETNSID */
                + nla_total_size(4)  /* IFLA_CARRIER_UP_COUNT */
@@ -1511,7 +1512,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb,
                             struct net_device *dev, struct net *src_net,
                             int type, u32 pid, u32 seq, u32 change,
                             unsigned int flags, u32 ext_filter_mask,
-                            u32 event, int *new_nsid, int tgt_netnsid)
+                            u32 event, int *new_nsid, int new_ifindex,
+                            int tgt_netnsid)
 {
         struct ifinfomsg *ifm;
         struct nlmsghdr *nlh;
@@ -1608,6 +1610,10 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb,
         if (new_nsid &&
             nla_put_s32(skb, IFLA_NEW_NETNSID, *new_nsid) < 0)
                 goto nla_put_failure;
+        if (new_ifindex &&
+            nla_put_s32(skb, IFLA_NEW_IFINDEX, new_ifindex) < 0)
+                goto nla_put_failure;
+
 
         rcu_read_lock();
         if (rtnl_fill_link_af(skb, dev, ext_filter_mask))
@@ -1853,7 +1859,7 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
                                                NETLINK_CB(cb->skb).portid,
                                                cb->nlh->nlmsg_seq, 0,
                                                flags,
-                                               ext_filter_mask, 0, NULL,
+                                               ext_filter_mask, 0, NULL, 0,
                                                netnsid);
 
                         if (err < 0) {
@@ -3088,7 +3094,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
         err = rtnl_fill_ifinfo(nskb, dev, net,
                                RTM_NEWLINK, NETLINK_CB(skb).portid,
                                nlh->nlmsg_seq, 0, 0, ext_filter_mask,
-                               0, NULL, netnsid);
+                               0, NULL, 0, netnsid);
         if (err < 0) {
                 /* -EMSGSIZE implies BUG in if_nlmsg_size */
                 WARN_ON(err == -EMSGSIZE);
@@ -3184,7 +3190,8 @@ static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
 
 struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
                                        unsigned int change,
-                                       u32 event, gfp_t flags, int *new_nsid)
+                                       u32 event, gfp_t flags, int *new_nsid,
+                                       int new_ifindex)
 {
         struct net *net = dev_net(dev);
         struct sk_buff *skb;
@@ -3197,7 +3204,7 @@ struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev,
 
         err = rtnl_fill_ifinfo(skb, dev, dev_net(dev),
                                type, 0, 0, change, 0, 0, event,
-                               new_nsid, -1);
+                               new_nsid, new_ifindex, -1);
         if (err < 0) {
                 /* -EMSGSIZE implies BUG in if_nlmsg_size() */
                 WARN_ON(err == -EMSGSIZE);
@@ -3220,14 +3227,15 @@ void rtmsg_ifinfo_send(struct sk_buff *skb, struct net_device *dev, gfp_t flags)
 
 static void rtmsg_ifinfo_event(int type, struct net_device *dev,
                                unsigned int change, u32 event,
-                               gfp_t flags, int *new_nsid)
+                               gfp_t flags, int *new_nsid, int new_ifindex)
 {
         struct sk_buff *skb;
 
         if (dev->reg_state != NETREG_REGISTERED)
                 return;
 
-        skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags, new_nsid);
+        skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags, new_nsid,
+                                     new_ifindex);
         if (skb)
                 rtmsg_ifinfo_send(skb, dev, flags);
 }
@@ -3235,14 +3243,15 @@ static void rtmsg_ifinfo_event(int type, struct net_device *dev,
 void rtmsg_ifinfo(int type, struct net_device *dev, unsigned int change,
                   gfp_t flags)
 {
-        rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags, NULL);
+        rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags,
+                           NULL, 0);
 }
 
 void rtmsg_ifinfo_newnet(int type, struct net_device *dev, unsigned int change,
-                         gfp_t flags, int *new_nsid)
+                         gfp_t flags, int *new_nsid, int new_ifindex)
 {
         rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags,
-                           new_nsid);
+                           new_nsid, new_ifindex);
 }
 
 static int nlmsg_populate_fdb_fill(struct sk_buff *skb,
@@ -4642,7 +4651,7 @@ static int rtnetlink_event(struct notifier_block *this, unsigned long event, voi
         case NETDEV_CHANGELOWERSTATE:
         case NETDEV_CHANGE_TX_QUEUE_LEN:
                 rtmsg_ifinfo_event(RTM_NEWLINK, dev, 0, rtnl_get_event(event),
-                                   GFP_KERNEL, NULL);
+                                   GFP_KERNEL, NULL, 0);
                 break;
         default:
                 break;