nicira-ext: Support matching IPv6 traffic.
[sliver-openvswitch.git] / datapath / flow.c
index d83c17d..4365e22 100644 (file)
 #include <linux/if_arp.h>
 #include <linux/if_ether.h>
 #include <linux/ip.h>
+#include <linux/ipv6.h>
 #include <linux/tcp.h>
 #include <linux/udp.h>
 #include <linux/icmp.h>
+#include <linux/icmpv6.h>
 #include <net/inet_ecn.h>
 #include <net/ip.h>
+#include <net/ipv6.h>
 
 static struct kmem_cache *flow_cache;
 static unsigned int hash_seed __read_mostly;
@@ -95,6 +98,63 @@ u64 flow_used_time(unsigned long flow_jiffies)
        return cur_ms - idle_ms;
 }
 
+static int parse_ipv6hdr(struct sk_buff *skb, struct sw_flow_key *key)
+{
+       unsigned int nh_ofs = skb_network_offset(skb);
+       unsigned int nh_len;
+       int payload_ofs;
+       int payload_len;
+       struct ipv6hdr *nh;
+       uint8_t nexthdr;
+
+       if (unlikely(skb->len < nh_ofs + sizeof(*nh)))
+               return -EINVAL;
+
+       nh = ipv6_hdr(skb);
+       nexthdr = nh->nexthdr;
+       payload_ofs = (u8 *)(nh + 1) - skb->data;
+       payload_len = ntohs(nh->payload_len);
+
+       memcpy(key->ipv6_src, nh->saddr.in6_u.u6_addr8, sizeof(key->ipv6_src));
+       memcpy(key->ipv6_dst, nh->daddr.in6_u.u6_addr8, sizeof(key->ipv6_dst));
+       key->nw_tos = ipv6_get_dsfield(nh) & ~INET_ECN_MASK;
+       key->nw_proto = NEXTHDR_NONE;
+
+       /* We don't process jumbograms. */
+       if (!payload_len)
+               return -EINVAL;
+
+       if (unlikely(skb->len < nh_ofs + sizeof(*nh) + payload_len))
+               return -EINVAL;
+
+       payload_ofs = ipv6_skip_exthdr(skb, payload_ofs, &nexthdr);
+       if (payload_ofs < 0) {
+               return -EINVAL;
+       }
+       nh_len = payload_ofs - nh_ofs;
+
+       /* Ensure that the payload length claimed is at least large enough
+        * for the headers we've already processed. */
+       if (payload_len < nh_len - sizeof(*nh))
+               return -EINVAL;
+
+       /* Pull enough header bytes to account for the IP header plus the
+        * longest transport header that we parse, currently 20 bytes for TCP.
+        * To dig deeper than the transport header, transport parsers may need
+        * to pull more header bytes.
+        */
+       if (unlikely(!pskb_may_pull(skb, min(nh_ofs + nh_len + 20, skb->len))))
+               return -ENOMEM;
+
+       skb_set_transport_header(skb, nh_ofs + nh_len);
+       key->nw_proto = nexthdr;
+       return nh_len;
+}
+
+static bool icmp6hdr_ok(struct sk_buff *skb)
+{
+       return skb->len >= skb_transport_offset(skb) + sizeof(struct icmp6hdr);
+}
 
 #define TCP_FLAGS_OFFSET 13
 #define TCP_FLAG_MASK 0x3f
@@ -274,10 +334,10 @@ static __be16 parse_ethertype(struct sk_buff *skb)
  *    - skb->network_header: just past the Ethernet header, or just past the
  *      VLAN header, to the first byte of the Ethernet payload.
  *
- *    - skb->transport_header: If key->dl_type is ETH_P_IP on output, then just
- *      past the IPv4 header, if one is present and of a correct length,
- *      otherwise the same as skb->network_header.  For other key->dl_type
- *      values it is left untouched.
+ *    - skb->transport_header: If key->dl_type is ETH_P_IP or ETH_P_IPV6
+ *      on output, then just past the IP header, if one is present and
+ *      of a correct length, otherwise the same as skb->network_header.
+ *      For other key->dl_type values it is left untouched.
  */
 int flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key,
                 bool *is_frag)
@@ -291,7 +351,8 @@ int flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key,
 
        /*
         * We would really like to pull as many bytes as we could possibly
-        * want to parse into the linear data area.  Currently that is:
+        * want to parse into the linear data area.  Currently, for IPv4,
+        * that is:
         *
         *    14     Ethernet header
         *     4     VLAN header
@@ -339,8 +400,8 @@ int flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key,
                }
 
                nh = ip_hdr(skb);
-               key->nw_src = nh->saddr;
-               key->nw_dst = nh->daddr;
+               key->ipv4_src = nh->saddr;
+               key->ipv4_dst = nh->daddr;
                key->nw_tos = nh->tos & ~INET_ECN_MASK;
                key->nw_proto = nh->protocol;
 
@@ -388,12 +449,47 @@ int flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key,
 
                        if (key->nw_proto == ARPOP_REQUEST
                                        || key->nw_proto == ARPOP_REPLY) {
-                               memcpy(&key->nw_src, arp->ar_sip, sizeof(key->nw_src));
-                               memcpy(&key->nw_dst, arp->ar_tip, sizeof(key->nw_dst));
+                               memcpy(&key->ipv4_src, arp->ar_sip, sizeof(key->ipv4_src));
+                               memcpy(&key->ipv4_dst, arp->ar_tip, sizeof(key->ipv4_dst));
                                memcpy(key->arp_sha, arp->ar_sha, ETH_ALEN);
                                memcpy(key->arp_tha, arp->ar_tha, ETH_ALEN);
                        }
                }
+       } else if (key->dl_type == htons(ETH_P_IPV6)) {
+               int nh_len;             /* IPv6 Header + Extensions */
+
+               nh_len = parse_ipv6hdr(skb, key);
+               if (unlikely(nh_len < 0)) {
+                       if (nh_len == -EINVAL) {
+                               skb->transport_header = skb->network_header;
+                               return 0;
+                       }
+                       return nh_len;
+               }
+
+               /* Transport layer. */
+               if (key->nw_proto == NEXTHDR_TCP) {
+                       if (tcphdr_ok(skb)) {
+                               struct tcphdr *tcp = tcp_hdr(skb);
+                               key->tp_src = tcp->source;
+                               key->tp_dst = tcp->dest;
+                       }
+               } else if (key->nw_proto == NEXTHDR_UDP) {
+                       if (udphdr_ok(skb)) {
+                               struct udphdr *udp = udp_hdr(skb);
+                               key->tp_src = udp->source;
+                               key->tp_dst = udp->dest;
+                       }
+               } else if (key->nw_proto == NEXTHDR_ICMP) {
+                       if (icmp6hdr_ok(skb)) {
+                               struct icmp6hdr *icmp = icmp6_hdr(skb);
+                               /* The ICMPv6 type and code fields use the 16-bit
+                                * transport port fields, so we need to store them
+                                * in 16-bit network byte order. */
+                               key->tp_src = htons(icmp->icmp6_type);
+                               key->tp_dst = htons(icmp->icmp6_code);
+                       }
+               }
        }
        return 0;
 }
@@ -420,7 +516,8 @@ int flow_cmp(const struct tbl_node *node, void *key2_)
  * This state machine accepts the following forms, with [] for optional
  * elements and | for alternatives:
  *
- * [tun_id] in_port ethernet [8021q] [ethertype [IP [TCP|UDP|ICMP] | ARP]
+ * [tun_id] in_port ethernet [8021q] [ethertype \
+ *              [IPv4 [TCP|UDP|ICMP] | IPv6 [TCP|UDP|ICMPv6] | ARP]]
  */
 int flow_from_nlattrs(struct sw_flow_key *swkey, const struct nlattr *attr)
 {
@@ -440,18 +537,22 @@ int flow_from_nlattrs(struct sw_flow_key *swkey, const struct nlattr *attr)
                        [ODP_KEY_ATTR_8021Q] = sizeof(struct odp_key_8021q),
                        [ODP_KEY_ATTR_ETHERTYPE] = 2,
                        [ODP_KEY_ATTR_IPV4] = sizeof(struct odp_key_ipv4),
+                       [ODP_KEY_ATTR_IPV6] = sizeof(struct odp_key_ipv6),
                        [ODP_KEY_ATTR_TCP] = sizeof(struct odp_key_tcp),
                        [ODP_KEY_ATTR_UDP] = sizeof(struct odp_key_udp),
                        [ODP_KEY_ATTR_ICMP] = sizeof(struct odp_key_icmp),
+                       [ODP_KEY_ATTR_ICMPV6] = sizeof(struct odp_key_icmpv6),
                        [ODP_KEY_ATTR_ARP] = sizeof(struct odp_key_arp),
                };
 
                const struct odp_key_ethernet *eth_key;
                const struct odp_key_8021q *q_key;
                const struct odp_key_ipv4 *ipv4_key;
+               const struct odp_key_ipv6 *ipv6_key;
                const struct odp_key_tcp *tcp_key;
                const struct odp_key_udp *udp_key;
                const struct odp_key_icmp *icmp_key;
+               const struct odp_key_icmpv6 *icmpv6_key;
                const struct odp_key_arp *arp_key;
 
                 int type = nla_type(nla);
@@ -499,15 +600,30 @@ int flow_from_nlattrs(struct sw_flow_key *swkey, const struct nlattr *attr)
                        if (swkey->dl_type != htons(ETH_P_IP))
                                return -EINVAL;
                        ipv4_key = nla_data(nla);
-                       swkey->nw_src = ipv4_key->ipv4_src;
-                       swkey->nw_dst = ipv4_key->ipv4_dst;
+                       swkey->ipv4_src = ipv4_key->ipv4_src;
+                       swkey->ipv4_dst = ipv4_key->ipv4_dst;
                        swkey->nw_proto = ipv4_key->ipv4_proto;
                        swkey->nw_tos = ipv4_key->ipv4_tos;
                        if (swkey->nw_tos & INET_ECN_MASK)
                                return -EINVAL;
                        break;
 
+               case TRANSITION(ODP_KEY_ATTR_ETHERTYPE, ODP_KEY_ATTR_IPV6):
+                       if (swkey->dl_type != htons(ETH_P_IPV6))
+                               return -EINVAL;
+                       ipv6_key = nla_data(nla);
+                       memcpy(swkey->ipv6_src, ipv6_key->ipv6_src,
+                                       sizeof(swkey->ipv6_src));
+                       memcpy(swkey->ipv6_dst, ipv6_key->ipv6_dst,
+                                       sizeof(swkey->ipv6_dst));
+                       swkey->nw_proto = ipv6_key->ipv6_proto;
+                       swkey->nw_tos = ipv6_key->ipv6_tos;
+                       if (swkey->nw_tos & INET_ECN_MASK)
+                               return -EINVAL;
+                       break;
+
                case TRANSITION(ODP_KEY_ATTR_IPV4, ODP_KEY_ATTR_TCP):
+               case TRANSITION(ODP_KEY_ATTR_IPV6, ODP_KEY_ATTR_TCP):
                        if (swkey->nw_proto != IPPROTO_TCP)
                                return -EINVAL;
                        tcp_key = nla_data(nla);
@@ -516,6 +632,7 @@ int flow_from_nlattrs(struct sw_flow_key *swkey, const struct nlattr *attr)
                        break;
 
                case TRANSITION(ODP_KEY_ATTR_IPV4, ODP_KEY_ATTR_UDP):
+               case TRANSITION(ODP_KEY_ATTR_IPV6, ODP_KEY_ATTR_UDP):
                        if (swkey->nw_proto != IPPROTO_UDP)
                                return -EINVAL;
                        udp_key = nla_data(nla);
@@ -531,12 +648,20 @@ int flow_from_nlattrs(struct sw_flow_key *swkey, const struct nlattr *attr)
                        swkey->tp_dst = htons(icmp_key->icmp_code);
                        break;
 
+               case TRANSITION(ODP_KEY_ATTR_IPV6, ODP_KEY_ATTR_ICMPV6):
+                       if (swkey->nw_proto != IPPROTO_ICMPV6)
+                               return -EINVAL;
+                       icmpv6_key = nla_data(nla);
+                       swkey->tp_src = htons(icmpv6_key->icmpv6_type);
+                       swkey->tp_dst = htons(icmpv6_key->icmpv6_code);
+                       break;
+
                case TRANSITION(ODP_KEY_ATTR_ETHERTYPE, ODP_KEY_ATTR_ARP):
                        if (swkey->dl_type != htons(ETH_P_ARP))
                                return -EINVAL;
                        arp_key = nla_data(nla);
-                       swkey->nw_src = arp_key->arp_sip;
-                       swkey->nw_dst = arp_key->arp_tip;
+                       swkey->ipv4_src = arp_key->arp_sip;
+                       swkey->ipv4_dst = arp_key->arp_tip;
                        if (arp_key->arp_op & htons(0xff00))
                                return -EINVAL;
                        swkey->nw_proto = ntohs(arp_key->arp_op);
@@ -578,9 +703,17 @@ int flow_from_nlattrs(struct sw_flow_key *swkey, const struct nlattr *attr)
                        return -EINVAL;
                return 0;
 
+       case ODP_KEY_ATTR_IPV6:
+               if (swkey->nw_proto == IPPROTO_TCP ||
+                   swkey->nw_proto == IPPROTO_UDP ||
+                   swkey->nw_proto == IPPROTO_ICMPV6)
+                       return -EINVAL;
+               return 0;
+
        case ODP_KEY_ATTR_TCP:
        case ODP_KEY_ATTR_UDP:
        case ODP_KEY_ATTR_ICMP:
+       case ODP_KEY_ATTR_ICMPV6:
        case ODP_KEY_ATTR_ARP:
                return 0;
        }
@@ -626,10 +759,39 @@ int flow_to_nlattrs(const struct sw_flow_key *swkey, struct sk_buff *skb)
                if (!nla)
                        goto nla_put_failure;
                ipv4_key = nla_data(nla);
-               ipv4_key->ipv4_src = swkey->nw_src;
-               ipv4_key->ipv4_dst = swkey->nw_dst;
+               ipv4_key->ipv4_src = swkey->ipv4_src;
+               ipv4_key->ipv4_dst = swkey->ipv4_dst;
                ipv4_key->ipv4_proto = swkey->nw_proto;
                ipv4_key->ipv4_tos = swkey->nw_tos;
+       } else if (swkey->dl_type == htons(ETH_P_IPV6)) {
+               struct odp_key_ipv6 *ipv6_key;
+
+               nla = nla_reserve(skb, ODP_KEY_ATTR_IPV6, sizeof(*ipv6_key));
+               if (!nla)
+                       goto nla_put_failure;
+               ipv6_key = nla_data(nla);
+               memcpy(ipv6_key->ipv6_src, swkey->ipv6_src,
+                               sizeof(ipv6_key->ipv6_src));
+               memcpy(ipv6_key->ipv6_dst, swkey->ipv6_dst,
+                               sizeof(ipv6_key->ipv6_dst));
+               ipv6_key->ipv6_proto = swkey->nw_proto;
+               ipv6_key->ipv6_tos = swkey->nw_tos;
+       } else if (swkey->dl_type == htons(ETH_P_ARP)) {
+               struct odp_key_arp *arp_key;
+
+               nla = nla_reserve(skb, ODP_KEY_ATTR_ARP, sizeof(*arp_key));
+               if (!nla)
+                       goto nla_put_failure;
+               arp_key = nla_data(nla);
+               arp_key->arp_sip = swkey->ipv4_src;
+               arp_key->arp_tip = swkey->ipv4_dst;
+               arp_key->arp_op = htons(swkey->nw_proto);
+               memcpy(arp_key->arp_sha, swkey->arp_sha, ETH_ALEN);
+               memcpy(arp_key->arp_tha, swkey->arp_tha, ETH_ALEN);
+       }
+
+       if (swkey->dl_type == htons(ETH_P_IP)
+                       || swkey->dl_type == htons(ETH_P_IPV6)) {
 
                if (swkey->nw_proto == IPPROTO_TCP) {
                        struct odp_key_tcp *tcp_key;
@@ -649,7 +811,8 @@ int flow_to_nlattrs(const struct sw_flow_key *swkey, struct sk_buff *skb)
                        udp_key = nla_data(nla);
                        udp_key->udp_src = swkey->tp_src;
                        udp_key->udp_dst = swkey->tp_dst;
-               } else if (swkey->nw_proto == IPPROTO_ICMP) {
+               } else if (swkey->dl_type == htons(ETH_P_IP)
+                               && swkey->nw_proto == IPPROTO_ICMP) {
                        struct odp_key_icmp *icmp_key;
 
                        nla = nla_reserve(skb, ODP_KEY_ATTR_ICMP, sizeof(*icmp_key));
@@ -658,19 +821,17 @@ int flow_to_nlattrs(const struct sw_flow_key *swkey, struct sk_buff *skb)
                        icmp_key = nla_data(nla);
                        icmp_key->icmp_type = ntohs(swkey->tp_src);
                        icmp_key->icmp_code = ntohs(swkey->tp_dst);
-               }
-       } else if (swkey->dl_type == htons(ETH_P_ARP)) {
-               struct odp_key_arp *arp_key;
+               } else if (swkey->dl_type == htons(ETH_P_IPV6)
+                               && swkey->nw_proto == IPPROTO_ICMPV6) {
+                       struct odp_key_icmpv6 *icmpv6_key;
 
-               nla = nla_reserve(skb, ODP_KEY_ATTR_ARP, sizeof(*arp_key));
-               if (!nla)
-                       goto nla_put_failure;
-               arp_key = nla_data(nla);
-               arp_key->arp_sip = swkey->nw_src;
-               arp_key->arp_tip = swkey->nw_dst;
-               arp_key->arp_op = htons(swkey->nw_proto);
-               memcpy(arp_key->arp_sha, swkey->arp_sha, ETH_ALEN);
-               memcpy(arp_key->arp_tha, swkey->arp_tha, ETH_ALEN);
+                       nla = nla_reserve(skb, ODP_KEY_ATTR_ICMPV6, sizeof(*icmpv6_key));
+                       if (!nla)
+                               goto nla_put_failure;
+                       icmpv6_key = nla_data(nla);
+                       icmpv6_key->icmpv6_type = ntohs(swkey->tp_src);
+                       icmpv6_key->icmpv6_code = ntohs(swkey->tp_dst);
+               }
        }
 
        return 0;