X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=datapath%2Ftunnel.h;h=e853146090e6d60d6bbee056d32a0bb6775deb29;hb=077996afd9aabcbd29a5ca72629b01dcc2fb1793;hp=951a6f1f7fceab0657be94c23cd6c62daa94549d;hpb=7c690f462fec1af60d76c67fac226fc648ea9149;p=sliver-openvswitch.git diff --git a/datapath/tunnel.h b/datapath/tunnel.h index 951a6f1f7..e85314609 100644 --- a/datapath/tunnel.h +++ b/datapath/tunnel.h @@ -24,265 +24,42 @@ #include #include "flow.h" -#include "openvswitch/tunnel.h" #include "vport.h" -/* - * The absolute minimum fragment size. Note that there are many other - * definitions of the minimum MTU. - */ -#define IP_MIN_MTU 68 - -/* - * One of these goes in struct tnl_ops and in tnl_find_port(). - * These values are in the same namespace as other TNL_T_* values, so - * only the least significant 10 bits are available to define protocol - * identifiers. - */ -#define TNL_T_PROTO_GRE 0 -#define TNL_T_PROTO_GRE64 1 -#define TNL_T_PROTO_CAPWAP 2 - -/* These flags are only needed when calling tnl_find_port(). */ -#define TNL_T_KEY_EXACT (1 << 10) -#define TNL_T_KEY_MATCH (1 << 11) - -/* Private flags not exposed to userspace in this form. */ -#define TNL_F_IN_KEY_MATCH (1 << 16) /* Store the key in tun_id to - * match in flow table. */ -#define TNL_F_OUT_KEY_ACTION (1 << 17) /* Get the key from a SET_TUNNEL - * action. */ - -/* All public tunnel flags. */ -#define TNL_F_PUBLIC (TNL_F_CSUM | TNL_F_TOS_INHERIT | TNL_F_TTL_INHERIT | \ - TNL_F_DF_INHERIT | TNL_F_DF_DEFAULT | TNL_F_PMTUD | \ - TNL_F_HDR_CACHE | TNL_F_IPSEC) - -/** - * struct port_lookup_key - Tunnel port key, used as hash table key. - * @in_key: Key to match on input, 0 for wildcard. - * @net: Network namespace of the port. - * @saddr: IPv4 source address to match, 0 to accept any source address. - * @daddr: IPv4 destination of tunnel. - * @tunnel_type: Set of TNL_T_* flags that define lookup. - */ -struct port_lookup_key { - __be64 in_key; -#ifdef CONFIG_NET_NS - struct net *net; -#endif - __be32 saddr; - __be32 daddr; - u32 tunnel_type; -}; - -#define PORT_KEY_LEN (offsetof(struct port_lookup_key, tunnel_type) + \ - FIELD_SIZEOF(struct port_lookup_key, tunnel_type)) - -static inline struct net *port_key_get_net(const struct port_lookup_key *key) -{ - return read_pnet(&key->net); -} - -static inline void port_key_set_net(struct port_lookup_key *key, struct net *net) -{ - write_pnet(&key->net, net); -} - -/** - * struct tnl_mutable_config - modifiable configuration for a tunnel. - * @key: Used as key for tunnel port. Configured via OVS_TUNNEL_ATTR_* - * attributes. - * @rcu: RCU callback head for deferred destruction. - * @seq: Sequence number for distinguishing configuration versions. - * @tunnel_hlen: Tunnel header length. - * @eth_addr: Source address for packets generated by tunnel itself - * (e.g. ICMP fragmentation needed messages). - * @out_key: Key to use on output, 0 if this tunnel has no fixed output key. - * @flags: TNL_F_* flags. - * @tos: IPv4 TOS value to use for tunnel, 0 if no fixed TOS. - * @ttl: IPv4 TTL value to use for tunnel, 0 if no fixed TTL. - */ -struct tnl_mutable_config { - struct port_lookup_key key; - struct rcu_head rcu; - - unsigned seq; - - unsigned char eth_addr[ETH_ALEN]; - - /* Configured via OVS_TUNNEL_ATTR_* attributes. */ - __be64 out_key; - u32 flags; - u8 tos; - u8 ttl; - - /* Multicast configuration. */ - int mlink; -}; - struct tnl_ops { - u32 tunnel_type; /* Put the TNL_T_PROTO_* type in here. */ u8 ipproto; /* The IP protocol for the tunnel. */ /* * Returns the length of the tunnel header that will be added in - * build_header() (i.e. excludes the IP header). Returns a negative - * error code if the configuration is invalid. - */ - int (*hdr_len)(const struct tnl_mutable_config *, - const struct ovs_key_ipv4_tunnel *); - - /* - * Builds the static portion of the tunnel header, which is stored in - * the header cache. In general the performance of this function is - * not too important as we try to only call it when building the cache - * so it is preferable to shift as much work as possible here. However, - * in some circumstances caching is disabled and this function will be - * called for every packet, so try not to make it too slow. + * build_header() (i.e. excludes the IP header). */ - void (*build_header)(const struct vport *, - const struct tnl_mutable_config *, - const struct ovs_key_ipv4_tunnel *, void *header); - + int (*hdr_len)(const struct ovs_key_ipv4_tunnel *); /* - * Updates the cached header of a packet to match the actual packet - * data. Typical things that might need to be updated are length, - * checksum, etc. The IP header will have already been updated and this - * is the final step before transmission. Returns a linked list of - * completed SKBs (multiple packets may be generated in the event - * of fragmentation). - */ - struct sk_buff *(*update_header)(const struct vport *, - const struct tnl_mutable_config *, - struct dst_entry *, struct sk_buff *, - int tunnel_hlen); -}; - -#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,20) -/* - * On these kernels we have a fast mechanism to tell if the ARP cache for a - * particular destination has changed. - */ -#define HAVE_HH_SEQ -#endif -#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,27) -/* - * On these kernels we have a fast mechanism to tell if the routing table - * has changed. - */ -#define HAVE_RT_GENID -#endif -#if !defined(HAVE_HH_SEQ) || !defined(HAVE_RT_GENID) -/* If we can't detect all system changes directly we need to use a timeout. */ -#define NEED_CACHE_TIMEOUT -#endif -struct tnl_cache { - struct rcu_head rcu; - - int len; /* Length of data to be memcpy'd from cache. */ - int hh_len; /* Hardware hdr length, cached from hh_cache. */ - - /* Sequence number of mutable->seq from which this cache was - * generated. */ - unsigned mutable_seq; - -#ifdef HAVE_HH_SEQ - /* - * The sequence number from the seqlock protecting the hardware header - * cache (in the ARP cache). Since every write increments the counter - * this gives us an easy way to tell if it has changed. - */ - unsigned hh_seq; -#endif - -#ifdef NEED_CACHE_TIMEOUT - /* - * If we don't have direct mechanisms to detect all important changes in - * the system fall back to an expiration time. This expiration time - * can be relatively short since at high rates there will be millions of - * packets per second, so we'll still get plenty of benefit from the - * cache. Note that if something changes we may blackhole packets - * until the expiration time (depending on what changed and the kernel - * version we may be able to detect the change sooner). Expiration is - * expressed as a time in jiffies. - */ - unsigned long expiration; -#endif - - /* - * The routing table entry that is the result of looking up the tunnel - * endpoints. It also contains a sequence number (called a generation - * ID) that can be compared to a global sequence to tell if the routing - * table has changed (and therefore there is a potential that this - * cached route has been invalidated). - */ - struct rtable *rt; - - /* - * If the output device for tunnel traffic is an OVS internal device, - * the flow of that datapath. Since all tunnel traffic will have the - * same headers this allows us to cache the flow lookup. NULL if the - * output device is not OVS or if there is no flow installed. - */ - struct sw_flow *flow; - - /* The cached header follows after padding for alignment. */ + * Builds header for given SKB. Space will have already been + * allocated at the start of the packet equal + * to sizeof(struct iphdr) + value returned by hdr_len(). + */ + void (*build_header)(const struct vport *, struct sk_buff *, + int tunnel_hlen); }; struct tnl_vport { struct rcu_head rcu; - struct hlist_node hash_node; + __be16 dst_port; char name[IFNAMSIZ]; const struct tnl_ops *tnl_ops; - - struct tnl_mutable_config __rcu *mutable; - - /* - * ID of last fragment sent (for tunnel protocols with direct support - * fragmentation). If the protocol relies on IP fragmentation then - * this is not needed. - */ - atomic_t frag_id; - - spinlock_t cache_lock; - struct tnl_cache __rcu *cache; /* Protected by RCU/cache_lock. */ - -#ifdef NEED_CACHE_TIMEOUT - /* - * If we must rely on expiration time to invalidate the cache, this is - * the interval. It is randomized within a range (defined by - * MAX_CACHE_EXP in tunnel.c) to avoid synchronized expirations caused - * by creation of a large number of tunnels at a one time. - */ - unsigned long cache_exp_interval; -#endif }; struct vport *ovs_tnl_create(const struct vport_parms *, const struct vport_ops *, const struct tnl_ops *); void ovs_tnl_destroy(struct vport *); -int ovs_tnl_set_options(struct vport *, struct nlattr *); -int ovs_tnl_get_options(const struct vport *, struct sk_buff *); - -int ovs_tnl_set_addr(struct vport *vport, const unsigned char *addr); const char *ovs_tnl_get_name(const struct vport *vport); -const unsigned char *ovs_tnl_get_addr(const struct vport *vport); int ovs_tnl_send(struct vport *vport, struct sk_buff *skb); void ovs_tnl_rcv(struct vport *vport, struct sk_buff *skb); +u16 ovs_tnl_get_src_port(struct sk_buff *skb); -struct vport *ovs_tnl_find_port(struct net *net, __be32 saddr, __be32 daddr, - __be64 key, int tunnel_type, - const struct tnl_mutable_config **mutable); -bool ovs_tnl_frag_needed(struct vport *vport, - const struct tnl_mutable_config *mutable, - struct sk_buff *skb, unsigned int mtu); -void ovs_tnl_free_linked_skbs(struct sk_buff *skb); - -int ovs_tnl_init(void); -void ovs_tnl_exit(void); static inline struct tnl_vport *tnl_vport_priv(const struct vport *vport) { return vport_priv(vport); @@ -297,6 +74,10 @@ static inline void tnl_tun_key_init(struct ovs_key_ipv4_tunnel *tun_key, tun_key->ipv4_tos = iph->tos; tun_key->ipv4_ttl = iph->ttl; tun_key->tun_flags = tun_flags; + + /* clear struct padding. */ + memset((unsigned char*) tun_key + OVS_TUNNEL_KEY_SIZE, 0, + sizeof(*tun_key) - OVS_TUNNEL_KEY_SIZE); } #endif /* tunnel.h */