Basic VRF support

Add basic VRF (virtual routing and forwarding) support. Protocols can be
associated with VRFs, such protocols will be restricted to interfaces
assigned to the VRF (as reported by Linux kernel) and will use sockets
bound to the VRF. E.g., different multihop BGP instances can use diffent
kernel routing tables to handle BGP TCP connections.

The VRF support is preliminary, currently there are several limitations:

- Recent Linux kernels (4.11) do not handle correctly sockets bound
to interaces that are part of VRF, so most protocols other than multihop
BGP do not work. This will be fixed by future kernel versions.

- Neighbor cache ignores VRFs. Breaks config with the same prefix on
local interfaces in different VRFs. Not much problem as single hop
protocols do not work anyways.

- Olock code ignores VRFs. Breaks config with multiple BGP peers with the
same IP address in different VRFs.

- Incoming BGP connections are not dispatched according to VRFs.
Breaks config with multiple BGP peers with the same IP address in
different VRFs. Perhaps we would need some kernel API to read VRF of
incoming connection? Or probably use multiple listening sockets in
int-new branch.

- We should handle master VRF interface up/down events and perhaps
disable associated protocols when VRF goes down. Or at least disable
associated interfaces.

- Also we should check if the master iface is really VRF iface and
not some other kind of master iface.

- BFD session request dispatch should be aware of VRFs.

- Perhaps kernel protocol should read default kernel table ID from VRF
iface so it is not necessary to configure it.

- Perhaps we should have per-VRF default table.

2 files changed, 40 insertions, 1 deletions

diff --git a/sysdep/linux/netlink.c b/sysdep/linux/netlink.c
index 3658c46b..4802897b 100644
--- a/sysdep/linux/netlink.c
+++ b/sysdep/linux/netlink.c
@@ -300,6 +300,7 @@ struct nl_want_attrs {
 static struct nl_want_attrs ifla_attr_want[BIRD_IFLA_MAX] = {
   [IFLA_IFNAME]	  = { 1, 0, 0 },
   [IFLA_MTU]	  = { 1, 1, sizeof(u32) },
+  [IFLA_MASTER]	  = { 1, 1, sizeof(u32) },
   [IFLA_WIRELESS] = { 1, 0, 0 },
 };
 
@@ -618,7 +619,7 @@ nl_parse_link(struct nlmsghdr *h, int scan)
   struct iface f = {};
   struct iface *ifi;
   char *name;
-  u32 mtu;
+  u32 mtu, master = 0;
   uint fl;
 
   if (!(i = nl_checkin(h, sizeof(*i))) || !nl_parse_attrs(IFLA_RTA(i), ifla_attr_want, a, sizeof(a)))
@@ -641,6 +642,9 @@ nl_parse_link(struct nlmsghdr *h, int scan)
   name = RTA_DATA(a[IFLA_IFNAME]);
   mtu = rta_get_u32(a[IFLA_MTU]);
 
+  if (a[IFLA_MASTER])
+    master = rta_get_u32(a[IFLA_MASTER]);
+
   ifi = if_find_by_index(i->ifi_index);
   if (!new)
     {
@@ -660,6 +664,9 @@ nl_parse_link(struct nlmsghdr *h, int scan)
       f.index = i->ifi_index;
       f.mtu = mtu;
 
+      f.master_index = master;
+      f.master = if_find_by_index(master);
+
       fl = i->ifi_flags;
       if (fl & IFF_UP)
 	f.flags |= IF_ADMIN_UP;
@@ -835,6 +842,26 @@ kif_do_scan(struct kif_proto *p UNUSED)
     else
       log(L_DEBUG "nl_scan_ifaces: Unknown packet received (type=%d)", h->nlmsg_type);
 
+  /* Re-resolve master interface for slaves */
+  struct iface *i;
+  WALK_LIST(i, iface_list)
+    if (i->master_index)
+    {
+      struct iface f = {
+	.flags = i->flags,
+	.mtu = i->mtu,
+	.index = i->index,
+	.master_index = i->master_index,
+	.master = if_find_by_index(i->master_index)
+      };
+
+      if (f.master != i->master)
+      {
+	memcpy(f.name, i->name, sizeof(f.name));
+	if_update(&f);
+      }
+    }
+
   nl_request_dump(BIRD_AF, RTM_GETADDR);
   while (h = nl_get_scan())
     if (h->nlmsg_type == RTM_NEWADDR || h->nlmsg_type == RTM_DELADDR)
diff --git a/sysdep/unix/io.c b/sysdep/unix/io.c
index 561d4dea..6d6a0990 100644
--- a/sysdep/unix/io.c
+++ b/sysdep/unix/io.c
@@ -1211,6 +1211,18 @@ sk_setup(sock *s)
   }
 #endif
 
+  if (s->vrf && !s->iface)
+  {
+    /* Bind socket to associated VRF interface.
+       This is Linux-specific, but so is SO_BINDTODEVICE. */
+#ifdef SO_BINDTODEVICE
+    struct ifreq ifr = {};
+    strcpy(ifr.ifr_name, s->vrf->name);
+    if (setsockopt(s->fd, SOL_SOCKET, SO_BINDTODEVICE, &ifr, sizeof(ifr)) < 0)
+      ERR("SO_BINDTODEVICE");
+#endif
+  }
+
   if (s->iface)
   {
 #ifdef SO_BINDTODEVICE