[IPV6] SIT: Add PRL management for ISATAP.

This patch updates the Linux the Intra-Site Automatic Tunnel Addressing
Protocol (ISATAP) implementation. It places the ISATAP potential router
list (PRL) in the kernel and adds three new private ioctls for PRL
management.

[Add several changes of structure name, constant names etc. - yoshfuji]

Signed-off-by: Fred L. Templin <fred.l.templin@boeing.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>

7 files changed, 199 insertions, 50 deletions

diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h
index 228eb4eb3129..f20c224d544c 100644
--- a/include/linux/if_tunnel.h
+++ b/include/linux/if_tunnel.h
@@ -7,6 +7,9 @@
 #define SIOCADDTUNNEL   (SIOCDEVPRIVATE + 1)
 #define SIOCDELTUNNEL   (SIOCDEVPRIVATE + 2)
 #define SIOCCHGTUNNEL   (SIOCDEVPRIVATE + 3)
+#define SIOCADDPRL      (SIOCDEVPRIVATE + 5)
+#define SIOCDELPRL      (SIOCDEVPRIVATE + 6)
+#define SIOCCHGPRL      (SIOCDEVPRIVATE + 7)
 
 #define GRE_CSUM        __constant_htons(0x8000)
 #define GRE_ROUTING     __constant_htons(0x4000)
@@ -17,9 +20,6 @@
 #define GRE_FLAGS       __constant_htons(0x00F8)
 #define GRE_VERSION     __constant_htons(0x0007)
 
-/* i_flags values for SIT mode */
-#define SIT_ISATAP      0x0001
-
 struct ip_tunnel_parm
 {
         char                    name[IFNAMSIZ];
@@ -31,4 +31,16 @@ struct ip_tunnel_parm
         struct iphdr            iph;
 };
 
+/* SIT-mode i_flags */
+#define SIT_ISATAP      0x0001
+
+struct ip_tunnel_prl {
+        __be32                  addr;
+        __u16                   flags;
+        __u16                   __reserved;
+};
+
+/* PRL flags */
+#define PRL_DEFAULT             0x0001
+
 #endif /* _IF_TUNNEL_H_ */
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index ff72145d5d9e..e10e55c9b081 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -313,7 +313,8 @@ struct sk_buff {
         __u16                   tc_verd;        /* traffic control verdict */
 #endif
 #endif
-        /* 2 byte hole */
+        __u8                    ndisc_nodetype:2;
+        /* 14 bit hole */
 
 #ifdef CONFIG_NET_DMA
         dma_cookie_t            dma_cookie;
diff --git a/include/net/ipip.h b/include/net/ipip.h
index 549e132bca9c..205536a014e8 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -24,6 +24,13 @@ struct ip_tunnel
         int                     mlink;
 
         struct ip_tunnel_parm   parms;
+        struct ip_tunnel_prl_entry      *prl;           /* potential router list */
+};
+
+struct ip_tunnel_prl_entry
+{
+        struct ip_tunnel_prl_entry      *next;
+        struct ip_tunnel_prl            entry;
 };
 
 #define IPTUNNEL_XMIT() do {                                            \
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 5aedf324de66..9f2bae68d28c 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -12,6 +12,15 @@
 #define NDISC_REDIRECT                  137
 
 /*
+ * Router type: cross-layer information from link-layer to
+ * IPv6 layer reported by certain link types (e.g., RFC4214).
+ */
+#define NDISC_NODETYPE_UNSPEC           0       /* unspecified (default) */
+#define NDISC_NODETYPE_HOST             1       /* host or unauthorized router */
+#define NDISC_NODETYPE_NODEFAULT        2       /* non-default router */
+#define NDISC_NODETYPE_DEFAULT          3       /* default router */
+
+/*
  *      ndisc options
  */
 
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 510aa747a404..53b546019fd5 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1092,6 +1092,12 @@ static void ndisc_router_discovery(struct sk_buff *skb)
                 return;
         }
 
+        if (skb->ndisc_nodetype == NDISC_NODETYPE_HOST) {
+                ND_PRINTK2(KERN_WARNING
+                           "ICMPv6 RA: from host or unauthorized router\n");
+                return;
+        }
+
         /*
          *      set the RA_RECV flag in the interface
          */
@@ -1115,6 +1121,10 @@ static void ndisc_router_discovery(struct sk_buff *skb)
                 return;
         }
 
+        /* skip link-specific parameters from interior routers */
+        if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT)
+                goto skip_linkparms;
+
         if (in6_dev->if_flags & IF_RS_SENT) {
                 /*
                  *      flag that an RA was received after an RS was sent
@@ -1229,6 +1239,8 @@ skip_defrtr:
                 }
         }
 
+skip_linkparms:
+
         /*
          *      Process options.
          */
@@ -1268,6 +1280,10 @@ skip_defrtr:
         }
 #endif
 
+        /* skip link-specific ndopts from interior routers */
+        if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT)
+                goto out;
+
         if (in6_dev->cnf.accept_ra_pinfo && ndopts.nd_opts_pi) {
                 struct nd_opt_hdr *p;
                 for (p = ndopts.nd_opts_pi;
@@ -1331,6 +1347,14 @@ static void ndisc_redirect_rcv(struct sk_buff *skb)
         int optlen;
         u8 *lladdr = NULL;
 
+        switch (skb->ndisc_nodetype) {
+        case NDISC_NODETYPE_HOST:
+        case NDISC_NODETYPE_NODEFAULT:
+                ND_PRINTK2(KERN_WARNING
+                           "ICMPv6 Redirect: from host or unauthorized router\n");
+                return;
+        }
+
         if (!(ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL)) {
                 ND_PRINTK2(KERN_WARNING
                            "ICMPv6 Redirect: source address is not link-local.\n");
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index cd82b6db35ff..f17b2f61891e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1699,8 +1699,6 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
         return rt;
 }
 
-EXPORT_SYMBOL(rt6_get_dflt_router);
-
 struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
                                      struct net_device *dev,
                                      unsigned int pref)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 1b8196c8d145..4786419ade0e 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -16,7 +16,7 @@
  *      Changes:
  * Roger Venning <r.venning@telstra.com>:       6to4 support
  * Nate Thompson <nate@thebog.net>:             6to4 support
- * Fred L. Templin <fltemplin@acm.org>:         isatap support
+ * Fred Templin <fred.l.templin@boeing.com>:    isatap support
  */
 
 #include <linux/module.h>
@@ -197,6 +197,119 @@ failed:
         return NULL;
 }
 
+static struct ip_tunnel_prl_entry *
+ipip6_tunnel_locate_prl(struct ip_tunnel *t, __be32 addr)
+{
+        struct ip_tunnel_prl_entry *p = (struct ip_tunnel_prl_entry *)NULL;
+
+        for (p = t->prl; p; p = p->next)
+                if (p->entry.addr == addr)
+                        break;
+        return p;
+
+}
+
+static int
+ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
+{
+        struct ip_tunnel_prl_entry *p;
+
+        for (p = t->prl; p; p = p->next) {
+                if (p->entry.addr == a->addr) {
+                        if (chg) {
+                                p->entry = *a;
+                                return 0;
+                        }
+                        return -EEXIST;
+                }
+        }
+
+        if (chg)
+                return -ENXIO;
+
+        p = kzalloc(sizeof(struct ip_tunnel_prl_entry), GFP_KERNEL);
+        if (!p)
+                return -ENOBUFS;
+
+        p->entry = *a;
+        p->next = t->prl;
+        t->prl = p;
+        return 0;
+}
+
+static int
+ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
+{
+        struct ip_tunnel_prl_entry *x, **p;
+
+        if (a) {
+                for (p = &t->prl; *p; p = &(*p)->next) {
+                        if ((*p)->entry.addr == a->addr) {
+                                x = *p;
+                                *p = x->next;
+                                kfree(x);
+                                return 0;
+                        }
+                }
+                return -ENXIO;
+        } else {
+                while (t->prl) {
+                        x = t->prl;
+                        t->prl = t->prl->next;
+                        kfree(x);
+                }
+        }
+        return 0;
+}
+
+/* copied directly from anycast.c */
+static int
+ipip6_onlink(struct in6_addr *addr, struct net_device *dev)
+{
+        struct inet6_dev        *idev;
+        struct inet6_ifaddr     *ifa;
+        int     onlink;
+
+        onlink = 0;
+        rcu_read_lock();
+        idev = __in6_dev_get(dev);
+        if (idev) {
+                read_lock_bh(&idev->lock);
+                for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) {
+                        onlink = ipv6_prefix_equal(addr, &ifa->addr,
+                                                   ifa->prefix_len);
+                        if (onlink)
+                                break;
+                }
+                read_unlock_bh(&idev->lock);
+        }
+        rcu_read_unlock();
+        return onlink;
+}
+
+static int
+isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t)
+{
+        struct ip_tunnel_prl_entry *p = ipip6_tunnel_locate_prl(t, iph->saddr);
+        int ok = 1;
+
+        if (p) {
+                if (p->entry.flags & PRL_DEFAULT)
+                        skb->ndisc_nodetype = NDISC_NODETYPE_DEFAULT;
+                else
+                        skb->ndisc_nodetype = NDISC_NODETYPE_NODEFAULT;
+        } else {
+                struct in6_addr *addr6 = &ipv6_hdr(skb)->saddr;
+                if (ipv6_addr_is_isatap(addr6) &&
+                    (addr6->s6_addr32[3] == iph->saddr) &&
+                    ipip6_onlink(addr6, t->dev))
+                        skb->ndisc_nodetype = NDISC_NODETYPE_HOST;
+                else
+                        ok = 0;
+        }
+        return ok;
+}
+
 static void ipip6_tunnel_uninit(struct net_device *dev)
 {
         if (dev == ipip6_fb_tunnel_dev) {
@@ -206,6 +319,7 @@ static void ipip6_tunnel_uninit(struct net_device *dev)
                 dev_put(dev);
         } else {
                 ipip6_tunnel_unlink(netdev_priv(dev));
+                ipip6_tunnel_del_prl(netdev_priv(dev), 0);
                 dev_put(dev);
         }
 }
@@ -365,48 +479,6 @@ static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb)
                 IP6_ECN_set_ce(ipv6_hdr(skb));
 }
 
-/* ISATAP (RFC4214) - check source address */
-static int
-isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev)
-{
-        struct neighbour *neigh;
-        struct dst_entry *dst;
-        struct rt6_info *rt;
-        struct flowi fl;
-        struct in6_addr *addr6;
-        struct in6_addr rtr;
-        struct ipv6hdr *iph6;
-        int ok = 0;
-
-        /* from onlink default router */
-        ipv6_addr_set(&rtr,  htonl(0xFE800000), 0, 0, 0);
-        ipv6_isatap_eui64(rtr.s6_addr + 8, iph->saddr);
-        if ((rt = rt6_get_dflt_router(&rtr, dev))) {
-                dst_release(&rt->u.dst);
-                return 1;
-        }
-
-        iph6 = ipv6_hdr(skb);
-        memset(&fl, 0, sizeof(fl));
-        fl.proto = iph6->nexthdr;
-        ipv6_addr_copy(&fl.fl6_dst, &iph6->saddr);
-        fl.oif = dev->ifindex;
-        security_skb_classify_flow(skb, &fl);
-
-        dst = ip6_route_output(&init_net, NULL, &fl);
-        if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) {
-
-                addr6 = (struct in6_addr*)&neigh->primary_key;
-
-                /* from correct previous hop */
-                if (ipv6_addr_is_isatap(addr6) &&
-                    (addr6->s6_addr32[3] == iph->saddr))
-                        ok = 1;
-        }
-        dst_release(dst);
-        return ok;
-}
-
 static int ipip6_rcv(struct sk_buff *skb)
 {
         struct iphdr *iph;
@@ -427,7 +499,7 @@ static int ipip6_rcv(struct sk_buff *skb)
                 skb->pkt_type = PACKET_HOST;
 
                 if ((tunnel->dev->priv_flags & IFF_ISATAP) &&
-                    !isatap_srcok(skb, iph, tunnel->dev)) {
+                    !isatap_chksrc(skb, iph, tunnel)) {
                         tunnel->stat.rx_errors++;
                         read_unlock(&ipip6_lock);
                         kfree_skb(skb);
@@ -707,6 +779,7 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 {
         int err = 0;
         struct ip_tunnel_parm p;
+        struct ip_tunnel_prl prl;
         struct ip_tunnel *t;
 
         switch (cmd) {
@@ -806,6 +879,31 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
                 err = 0;
                 break;
 
+        case SIOCADDPRL:
+        case SIOCDELPRL:
+        case SIOCCHGPRL:
+                err = -EPERM;
+                if (!capable(CAP_NET_ADMIN))
+                        goto done;
+                err = -EINVAL;
+                if (dev == ipip6_fb_tunnel_dev)
+                        goto done;
+                err = -EFAULT;
+                if (copy_from_user(&prl, ifr->ifr_ifru.ifru_data, sizeof(prl)))
+                        goto done;
+                err = -ENOENT;
+                if (!(t = netdev_priv(dev)))
+                        goto done;
+
+                ipip6_tunnel_unlink(t);
+                if (cmd == SIOCDELPRL)
+                        err = ipip6_tunnel_del_prl(t, &prl);
+                else
+                        err = ipip6_tunnel_add_prl(t, &prl, cmd == SIOCCHGPRL);
+                ipip6_tunnel_link(t);
+                netdev_state_change(dev);
+                break;
+
         default:
                 err = -EINVAL;
         }