[mirror_ubuntu-jammy-kernel.git] / net / openvswitch / flow.h

/* SPDX-License-Identifier: GPL-2.0-only */
/*
 * Copyright (c) 2007-2017 Nicira, Inc.
 */

#ifndef FLOW_H
#define FLOW_H 1

#include <linux/cache.h>
#include <linux/kernel.h>
#include <linux/netlink.h>
#include <linux/openvswitch.h>
#include <linux/spinlock.h>
#include <linux/types.h>
#include <linux/rcupdate.h>
#include <linux/if_ether.h>
#include <linux/in6.h>
#include <linux/jiffies.h>
#include <linux/time.h>
#include <linux/cpumask.h>
#include <net/inet_ecn.h>
#include <net/ip_tunnels.h>
#include <net/dst_metadata.h>
#include <net/nsh.h>

struct sk_buff;

enum sw_flow_mac_proto {
	MAC_PROTO_NONE = 0,
	MAC_PROTO_ETHERNET,
};
#define SW_FLOW_KEY_INVALID	0x80
#define MPLS_LABEL_DEPTH       3

/* Store options at the end of the array if they are less than the
 * maximum size. This allows us to get the benefits of variable length
 * matching for small options.
 */
#define TUN_METADATA_OFFSET(opt_len) \
	(sizeof_field(struct sw_flow_key, tun_opts) - opt_len)
#define TUN_METADATA_OPTS(flow_key, opt_len) \
	((void *)((flow_key)->tun_opts + TUN_METADATA_OFFSET(opt_len)))

struct ovs_tunnel_info {
	struct metadata_dst	*tun_dst;
};

struct vlan_head {
	__be16 tpid; /* Vlan type. Generally 802.1q or 802.1ad.*/
	__be16 tci;  /* 0 if no VLAN, VLAN_CFI_MASK set otherwise. */
};

#define OVS_SW_FLOW_KEY_METADATA_SIZE			\
	(offsetof(struct sw_flow_key, recirc_id) +	\
	sizeof_field(struct sw_flow_key, recirc_id))

struct ovs_key_nsh {
	struct ovs_nsh_key_base base;
	__be32 context[NSH_MD1_CONTEXT_SIZE];
};

struct sw_flow_key {
	u8 tun_opts[IP_TUNNEL_OPTS_MAX];
	u8 tun_opts_len;
	struct ip_tunnel_key tun_key;	/* Encapsulating tunnel key. */
	struct {
		u32	priority;	/* Packet QoS priority. */
		u32	skb_mark;	/* SKB mark. */
		u16	in_port;	/* Input switch port (or DP_MAX_PORTS). */
	} __packed phy; /* Safe when right after 'tun_key'. */
	u8 mac_proto;			/* MAC layer protocol (e.g. Ethernet). */
	u8 tun_proto;			/* Protocol of encapsulating tunnel. */
	u32 ovs_flow_hash;		/* Datapath computed hash value.  */
	u32 recirc_id;			/* Recirculation ID.  */
	struct {
		u8     src[ETH_ALEN];	/* Ethernet source address. */
		u8     dst[ETH_ALEN];	/* Ethernet destination address. */
		struct vlan_head vlan;
		struct vlan_head cvlan;
		__be16 type;		/* Ethernet frame type. */
	} eth;
	/* Filling a hole of two bytes. */
	u8 ct_state;
	u8 ct_orig_proto;		/* CT original direction tuple IP
					 * protocol.
					 */
	union {
		struct {
			u8     proto;	/* IP protocol or lower 8 bits of ARP opcode. */
			u8     tos;	    /* IP ToS. */
			u8     ttl;	    /* IP TTL/hop limit. */
			u8     frag;	/* One of OVS_FRAG_TYPE_*. */
		} ip;
	};
	u16 ct_zone;			/* Conntrack zone. */
	struct {
		__be16 src;		/* TCP/UDP/SCTP source port. */
		__be16 dst;		/* TCP/UDP/SCTP destination port. */
		__be16 flags;		/* TCP flags. */
	} tp;
	union {
		struct {
			struct {
				__be32 src;	/* IP source address. */
				__be32 dst;	/* IP destination address. */
			} addr;
			union {
				struct {
					__be32 src;
					__be32 dst;
				} ct_orig;	/* Conntrack original direction fields. */
				struct {
					u8 sha[ETH_ALEN];	/* ARP source hardware address. */
					u8 tha[ETH_ALEN];	/* ARP target hardware address. */
				} arp;
			};
		} ipv4;
		struct {
			struct {
				struct in6_addr src;	/* IPv6 source address. */
				struct in6_addr dst;	/* IPv6 destination address. */
			} addr;
			__be32 label;			/* IPv6 flow label. */
			union {
				struct {
					struct in6_addr src;
					struct in6_addr dst;
				} ct_orig;	/* Conntrack original direction fields. */
				struct {
					struct in6_addr target;	/* ND target address. */
					u8 sll[ETH_ALEN];	/* ND source link layer address. */
					u8 tll[ETH_ALEN];	/* ND target link layer address. */
				} nd;
			};
		} ipv6;
		struct {
			u32 num_labels_mask;    /* labels present bitmap of effective length MPLS_LABEL_DEPTH */
			__be32 lse[MPLS_LABEL_DEPTH];     /* label stack entry  */
		} mpls;

		struct ovs_key_nsh nsh;         /* network service header */
	};
	struct {
		/* Connection tracking fields not packed above. */
		struct {
			__be16 src;	/* CT orig tuple tp src port. */
			__be16 dst;	/* CT orig tuple tp dst port. */
		} orig_tp;
		u32 mark;
		struct ovs_key_ct_labels labels;
	} ct;

} __aligned(BITS_PER_LONG/8); /* Ensure that we can do comparisons as longs. */

static inline bool sw_flow_key_is_nd(const struct sw_flow_key *key)
{
	return key->eth.type == htons(ETH_P_IPV6) &&
		key->ip.proto == NEXTHDR_ICMP &&
		key->tp.dst == 0 &&
		(key->tp.src == htons(NDISC_NEIGHBOUR_SOLICITATION) ||
		 key->tp.src == htons(NDISC_NEIGHBOUR_ADVERTISEMENT));
}

struct sw_flow_key_range {
	unsigned short int start;
	unsigned short int end;
};

struct sw_flow_mask {
	int ref_count;
	struct rcu_head rcu;
	struct sw_flow_key_range range;
	struct sw_flow_key key;
};

struct sw_flow_match {
	struct sw_flow_key *key;
	struct sw_flow_key_range range;
	struct sw_flow_mask *mask;
};

#define MAX_UFID_LENGTH 16 /* 128 bits */

struct sw_flow_id {
	u32 ufid_len;
	union {
		u32 ufid[MAX_UFID_LENGTH / 4];
		struct sw_flow_key *unmasked_key;
	};
};

struct sw_flow_actions {
	struct rcu_head rcu;
	size_t orig_len;	/* From flow_cmd_new netlink actions size */
	u32 actions_len;
	struct nlattr actions[];
};

struct sw_flow_stats {
	u64 packet_count;		/* Number of packets matched. */
	u64 byte_count;			/* Number of bytes matched. */
	unsigned long used;		/* Last used time (in jiffies). */
	spinlock_t lock;		/* Lock for atomic stats update. */
	__be16 tcp_flags;		/* Union of seen TCP flags. */
};

struct sw_flow {
	struct rcu_head rcu;
	struct {
		struct hlist_node node[2];
		u32 hash;
	} flow_table, ufid_table;
	int stats_last_writer;		/* CPU id of the last writer on
					 * 'stats[0]'.
					 */
	struct sw_flow_key key;
	struct sw_flow_id id;
	struct cpumask cpu_used_mask;
	struct sw_flow_mask *mask;
	struct sw_flow_actions __rcu *sf_acts;
	struct sw_flow_stats __rcu *stats[]; /* One for each CPU.  First one
					   * is allocated at flow creation time,
					   * the rest are allocated on demand
					   * while holding the 'stats[0].lock'.
					   */
};

struct arp_eth_header {
	__be16      ar_hrd;	/* format of hardware address   */
	__be16      ar_pro;	/* format of protocol address   */
	unsigned char   ar_hln;	/* length of hardware address   */
	unsigned char   ar_pln;	/* length of protocol address   */
	__be16      ar_op;	/* ARP opcode (command)     */

	/* Ethernet+IPv4 specific members. */
	unsigned char       ar_sha[ETH_ALEN];	/* sender hardware address  */
	unsigned char       ar_sip[4];		/* sender IP address        */
	unsigned char       ar_tha[ETH_ALEN];	/* target hardware address  */
	unsigned char       ar_tip[4];		/* target IP address        */
} __packed;

static inline u8 ovs_key_mac_proto(const struct sw_flow_key *key)
{
	return key->mac_proto & ~SW_FLOW_KEY_INVALID;
}

static inline u16 __ovs_mac_header_len(u8 mac_proto)
{
	return mac_proto == MAC_PROTO_ETHERNET ? ETH_HLEN : 0;
}

static inline u16 ovs_mac_header_len(const struct sw_flow_key *key)
{
	return __ovs_mac_header_len(ovs_key_mac_proto(key));
}

static inline bool ovs_identifier_is_ufid(const struct sw_flow_id *sfid)
{
	return sfid->ufid_len;
}

static inline bool ovs_identifier_is_key(const struct sw_flow_id *sfid)
{
	return !ovs_identifier_is_ufid(sfid);
}

void ovs_flow_stats_update(struct sw_flow *, __be16 tcp_flags,
			   const struct sk_buff *);
void ovs_flow_stats_get(const struct sw_flow *, struct ovs_flow_stats *,
			unsigned long *used, __be16 *tcp_flags);
void ovs_flow_stats_clear(struct sw_flow *);
u64 ovs_flow_used_time(unsigned long flow_jiffies);

int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key);
int ovs_flow_key_update_l3l4(struct sk_buff *skb, struct sw_flow_key *key);
int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
			 struct sk_buff *skb,
			 struct sw_flow_key *key);
/* Extract key from packet coming from userspace. */
int ovs_flow_key_extract_userspace(struct net *net, const struct nlattr *attr,
				   struct sk_buff *skb,
				   struct sw_flow_key *key, bool log);

#endif /* flow.h */
Commit	Line	Data
c9422999	1	/* SPDX-License-Identifier: GPL-2.0-only */
ccb1352e	2	/*
9dd7f890	3	* Copyright (c) 2007-2017 Nicira, Inc.
ccb1352e JG	4	*/
	5
	6	#ifndef FLOW_H
	7	#define FLOW_H 1
	8
e298e505	9	#include <linux/cache.h>
ccb1352e JG	10	#include <linux/kernel.h>
	11	#include <linux/netlink.h>
	12	#include <linux/openvswitch.h>
	13	#include <linux/spinlock.h>
	14	#include <linux/types.h>
	15	#include <linux/rcupdate.h>
	16	#include <linux/if_ether.h>
	17	#include <linux/in6.h>
	18	#include <linux/jiffies.h>
	19	#include <linux/time.h>
c4b2bf6b	20	#include <linux/cpumask.h>
ccb1352e	21	#include <net/inet_ecn.h>
1d8fff90	22	#include <net/ip_tunnels.h>
34ae932a	23	#include <net/dst_metadata.h>
b2d0f5d5	24	#include <net/nsh.h>
ccb1352e JG	25
ccb1352e JG	26	struct sk_buff;
ccb1352e	27
329f45bc JB	28	enum sw_flow_mac_proto {
	29	MAC_PROTO_NONE = 0,
	30	MAC_PROTO_ETHERNET,
	31	};
	32	#define SW_FLOW_KEY_INVALID 0x80
fbdcdd78	33	#define MPLS_LABEL_DEPTH 3
329f45bc	34
f5796684 JG	35	/* Store options at the end of the array if they are less than the
	36	* maximum size. This allows us to get the benefits of variable length
	37	* matching for small options.
	38	*/
d91641d9	39	#define TUN_METADATA_OFFSET(opt_len) \
c593642c	40	(sizeof_field(struct sw_flow_key, tun_opts) - opt_len)
d91641d9 TG	41	#define TUN_METADATA_OPTS(flow_key, opt_len) \
d91641d9 TG	42	((void *)((flow_key)->tun_opts + TUN_METADATA_OFFSET(opt_len)))
f5796684	43
34ae932a TG	44	struct ovs_tunnel_info {
	45	struct metadata_dst *tun_dst;
	46	};
	47
018c1dda EG	48	struct vlan_head {
018c1dda EG	49	__be16 tpid; /* Vlan type. Generally 802.1q or 802.1ad.*/
9df46aef	50	__be16 tci; /* 0 if no VLAN, VLAN_CFI_MASK set otherwise. */
018c1dda EG	51	};
018c1dda EG	52
8f0aad6f WZ	53	#define OVS_SW_FLOW_KEY_METADATA_SIZE \
8f0aad6f WZ	54	(offsetof(struct sw_flow_key, recirc_id) + \
c593642c	55	sizeof_field(struct sw_flow_key, recirc_id))
8f0aad6f	56
b2d0f5d5 YY	57	struct ovs_key_nsh {
	58	struct ovs_nsh_key_base base;
	59	__be32 context[NSH_MD1_CONTEXT_SIZE];
	60	};
	61
ccb1352e	62	struct sw_flow_key {
fca5fdf6	63	u8 tun_opts[IP_TUNNEL_OPTS_MAX];
f5796684	64	u8 tun_opts_len;
1d8fff90	65	struct ip_tunnel_key tun_key; /* Encapsulating tunnel key. */
ccb1352e JG	66	struct {
ccb1352e JG	67	u32 priority; /* Packet QoS priority. */
39c7caeb	68	u32 skb_mark; /* SKB mark. */
15eac2a7	69	u16 in_port; /* Input switch port (or DP_MAX_PORTS). */
1139e241	70	} __packed phy; /* Safe when right after 'tun_key'. */
329f45bc	71	u8 mac_proto; /* MAC layer protocol (e.g. Ethernet). */
00a93bab	72	u8 tun_proto; /* Protocol of encapsulating tunnel. */
971427f3 AZ	73	u32 ovs_flow_hash; /* Datapath computed hash value. */
971427f3 AZ	74	u32 recirc_id; /* Recirculation ID. */
ccb1352e JG	75	struct {
	76	u8 src[ETH_ALEN]; /* Ethernet source address. */
	77	u8 dst[ETH_ALEN]; /* Ethernet destination address. */
018c1dda EG	78	struct vlan_head vlan;
018c1dda EG	79	struct vlan_head cvlan;
ccb1352e JG	80	__be16 type; /* Ethernet frame type. */
ccb1352e JG	81	} eth;
316d4d78 JR	82	/* Filling a hole of two bytes. */
	83	u8 ct_state;
	84	u8 ct_orig_proto; /* CT original direction tuple IP
	85	* protocol.
	86	*/
25cd9ba0	87	union {
25cd9ba0 SH	88	struct {
	89	u8 proto; /* IP protocol or lower 8 bits of ARP opcode. */
	90	u8 tos; /* IP ToS. */
	91	u8 ttl; /* IP TTL/hop limit. */
	92	u8 frag; /* One of OVS_FRAG_TYPE_. /
	93	} ip;
	94	};
316d4d78	95	u16 ct_zone; /* Conntrack zone. */
1139e241 JR	96	struct {
	97	__be16 src; /* TCP/UDP/SCTP source port. */
	98	__be16 dst; /* TCP/UDP/SCTP destination port. */
	99	__be16 flags; /* TCP flags. */
	100	} tp;
ccb1352e JG	101	union {
	102	struct {
	103	struct {
	104	__be32 src; /* IP source address. */
	105	__be32 dst; /* IP destination address. */
	106	} addr;
9dd7f890 JR	107	union {
	108	struct {
	109	__be32 src;
	110	__be32 dst;
	111	} ct_orig; /* Conntrack original direction fields. */
	112	struct {
	113	u8 sha[ETH_ALEN]; /* ARP source hardware address. */
	114	u8 tha[ETH_ALEN]; /* ARP target hardware address. */
	115	} arp;
	116	};
ccb1352e JG	117	} ipv4;
	118	struct {
	119	struct {
	120	struct in6_addr src; /* IPv6 source address. */
	121	struct in6_addr dst; /* IPv6 destination address. */
	122	} addr;
	123	__be32 label; /* IPv6 flow label. */
9dd7f890 JR	124	union {
	125	struct {
	126	struct in6_addr src;
	127	struct in6_addr dst;
	128	} ct_orig; /* Conntrack original direction fields. */
	129	struct {
	130	struct in6_addr target; /* ND target address. */
	131	u8 sll[ETH_ALEN]; /* ND source link layer address. */
	132	u8 tll[ETH_ALEN]; /* ND target link layer address. */
	133	} nd;
	134	};
ccb1352e	135	} ipv6;
fbdcdd78 MV	136	struct {
	137	u32 num_labels_mask; /* labels present bitmap of effective length MPLS_LABEL_DEPTH */
	138	__be32 lse[MPLS_LABEL_DEPTH]; /* label stack entry */
	139	} mpls;
	140
b2d0f5d5	141	struct ovs_key_nsh nsh; /* network service header */
ccb1352e	142	};
7f8a436e	143	struct {
316d4d78	144	/* Connection tracking fields not packed above. */
9dd7f890 JR	145	struct {
	146	__be16 src; /* CT orig tuple tp src port. */
	147	__be16 dst; /* CT orig tuple tp dst port. */
	148	} orig_tp;
316d4d78	149	u32 mark;
33db4125	150	struct ovs_key_ct_labels labels;
7f8a436e JS	151	} ct;
7f8a436e JS	152
0d40f75b	153	} __aligned(BITS_PER_LONG/8); /* Ensure that we can do comparisons as longs. */
ccb1352e	154
9dd7f890 JR	155	static inline bool sw_flow_key_is_nd(const struct sw_flow_key *key)
	156	{
	157	return key->eth.type == htons(ETH_P_IPV6) &&
	158	key->ip.proto == NEXTHDR_ICMP &&
	159	key->tp.dst == 0 &&
	160	(key->tp.src == htons(NDISC_NEIGHBOUR_SOLICITATION) \|\|
	161	key->tp.src == htons(NDISC_NEIGHBOUR_ADVERTISEMENT));
	162	}
	163
e6445719	164	struct sw_flow_key_range {
8f49ce11 BP	165	unsigned short int start;
8f49ce11 BP	166	unsigned short int end;
e6445719 PS	167	};
	168
	169	struct sw_flow_mask {
	170	int ref_count;
	171	struct rcu_head rcu;
e6445719 PS	172	struct sw_flow_key_range range;
	173	struct sw_flow_key key;
	174	};
	175
	176	struct sw_flow_match {
	177	struct sw_flow_key *key;
	178	struct sw_flow_key_range range;
	179	struct sw_flow_mask *mask;
	180	};
	181
74ed7ab9 JS	182	#define MAX_UFID_LENGTH 16 /* 128 bits */
	183
	184	struct sw_flow_id {
	185	u32 ufid_len;
	186	union {
	187	u32 ufid[MAX_UFID_LENGTH / 4];
	188	struct sw_flow_key *unmasked_key;
	189	};
	190	};
	191
e6445719 PS	192	struct sw_flow_actions {
e6445719 PS	193	struct rcu_head rcu;
8e2fed1c	194	size_t orig_len; /* From flow_cmd_new netlink actions size */
e6445719 PS	195	u32 actions_len;
	196	struct nlattr actions[];
	197	};
	198
aef833c5	199	struct sw_flow_stats {
e298e505 PS	200	u64 packet_count; /* Number of packets matched. */
	201	u64 byte_count; /* Number of bytes matched. */
	202	unsigned long used; /* Last used time (in jiffies). */
	203	spinlock_t lock; /* Lock for atomic stats update. */
	204	__be16 tcp_flags; /* Union of seen TCP flags. */
	205	};
	206
ccb1352e JG	207	struct sw_flow {
ccb1352e JG	208	struct rcu_head rcu;
74ed7ab9 JS	209	struct {
	210	struct hlist_node node[2];
	211	u32 hash;
	212	} flow_table, ufid_table;
db74a333	213	int stats_last_writer; /* CPU id of the last writer on
63e7959c JR	214	* 'stats[0]'.
63e7959c JR	215	*/
ccb1352e	216	struct sw_flow_key key;
74ed7ab9	217	struct sw_flow_id id;
c4b2bf6b	218	struct cpumask cpu_used_mask;
03f0d916	219	struct sw_flow_mask *mask;
ccb1352e	220	struct sw_flow_actions __rcu *sf_acts;
aef833c5	221	struct sw_flow_stats __rcu stats[]; / One for each CPU. First one
63e7959c JR	222	* is allocated at flow creation time,
	223	* the rest are allocated on demand
	224	* while holding the 'stats[0].lock'.
	225	*/
ccb1352e JG	226	};
	227
	228	struct arp_eth_header {
	229	__be16 ar_hrd; /* format of hardware address */
	230	__be16 ar_pro; /* format of protocol address */
	231	unsigned char ar_hln; /* length of hardware address */
	232	unsigned char ar_pln; /* length of protocol address */
	233	__be16 ar_op; /* ARP opcode (command) */
	234
	235	/* Ethernet+IPv4 specific members. */
	236	unsigned char ar_sha[ETH_ALEN]; /* sender hardware address */
	237	unsigned char ar_sip[4]; /* sender IP address */
	238	unsigned char ar_tha[ETH_ALEN]; /* target hardware address */
	239	unsigned char ar_tip[4]; /* target IP address */
	240	} __packed;
	241
329f45bc JB	242	static inline u8 ovs_key_mac_proto(const struct sw_flow_key *key)
	243	{
	244	return key->mac_proto & ~SW_FLOW_KEY_INVALID;
	245	}
	246
	247	static inline u16 __ovs_mac_header_len(u8 mac_proto)
	248	{
	249	return mac_proto == MAC_PROTO_ETHERNET ? ETH_HLEN : 0;
	250	}
	251
	252	static inline u16 ovs_mac_header_len(const struct sw_flow_key *key)
	253	{
	254	return __ovs_mac_header_len(ovs_key_mac_proto(key));
	255	}
	256
74ed7ab9 JS	257	static inline bool ovs_identifier_is_ufid(const struct sw_flow_id *sfid)
	258	{
	259	return sfid->ufid_len;
	260	}
	261
	262	static inline bool ovs_identifier_is_key(const struct sw_flow_id *sfid)
	263	{
	264	return !ovs_identifier_is_ufid(sfid);
	265	}
	266
ad552007	267	void ovs_flow_stats_update(struct sw_flow *, __be16 tcp_flags,
12eb18f7	268	const struct sk_buff *);
86ec8dba	269	void ovs_flow_stats_get(const struct sw_flow , struct ovs_flow_stats ,
e298e505	270	unsigned long used, __be16 tcp_flags);
86ec8dba	271	void ovs_flow_stats_clear(struct sw_flow *);
ccb1352e	272	u64 ovs_flow_used_time(unsigned long flow_jiffies);
ccb1352e	273
971427f3	274	int ovs_flow_key_update(struct sk_buff skb, struct sw_flow_key key);
ad06a566	275	int ovs_flow_key_update_l3l4(struct sk_buff skb, struct sw_flow_key key);
1d8fff90	276	int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
12eb18f7	277	struct sk_buff *skb,
f0b128c1	278	struct sw_flow_key *key);
83c8df26	279	/* Extract key from packet coming from userspace. */
c2ac6673	280	int ovs_flow_key_extract_userspace(struct net net, const struct nlattr attr,
83c8df26	281	struct sk_buff *skb,
05da5898	282	struct sw_flow_key *key, bool log);
03f0d916	283
ccb1352e	284	#endif /* flow.h */