/*
- * Copyright (c) 2007-2013 Nicira, Inc.
+ * Copyright (c) 2007-2014 Nicira, Inc.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public
#include <linux/if_arp.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
+#include <linux/sctp.h>
+#include <linux/smp.h>
#include <linux/tcp.h>
#include <linux/udp.h>
#include <linux/icmp.h>
#include "vlan.h"
-static struct kmem_cache *flow_cache;
-
-static void ovs_sw_flow_mask_set(struct sw_flow_mask *mask,
- struct sw_flow_key_range *range, u8 val);
-
-static void update_range__(struct sw_flow_match *match,
- size_t offset, size_t size, bool is_mask)
+u64 ovs_flow_used_time(unsigned long flow_jiffies)
{
- struct sw_flow_key_range *range = NULL;
- size_t start = offset;
- size_t end = offset + size;
-
- if (!is_mask)
- range = &match->range;
- else if (match->mask)
- range = &match->mask->range;
-
- if (!range)
- return;
-
- if (range->start == range->end) {
- range->start = start;
- range->end = end;
- return;
- }
+ struct timespec cur_ts;
+ u64 cur_ms, idle_ms;
- if (range->start > start)
- range->start = start;
+ ktime_get_ts(&cur_ts);
+ idle_ms = jiffies_to_msecs(jiffies - flow_jiffies);
+ cur_ms = (u64)cur_ts.tv_sec * MSEC_PER_SEC +
+ cur_ts.tv_nsec / NSEC_PER_MSEC;
- if (range->end < end)
- range->end = end;
+ return cur_ms - idle_ms;
}
-#define SW_FLOW_KEY_PUT(match, field, value, is_mask) \
- do { \
- update_range__(match, offsetof(struct sw_flow_key, field), \
- sizeof((match)->key->field), is_mask); \
- if (is_mask && match->mask != NULL) { \
- (match)->mask->key.field = value; \
- } else { \
- (match)->key->field = value; \
- } \
- } while (0)
-
-#define SW_FLOW_KEY_MEMCPY(match, field, value_p, len, is_mask) \
- do { \
- update_range__(match, offsetof(struct sw_flow_key, field), \
- len, is_mask); \
- if (is_mask && match->mask != NULL) { \
- memcpy(&(match)->mask->key.field, value_p, len); \
- } else { \
- memcpy(&(match)->key->field, value_p, len); \
- } \
- } while (0)
-
-void ovs_match_init(struct sw_flow_match *match,
- struct sw_flow_key *key,
- struct sw_flow_mask *mask)
-{
- memset(match, 0, sizeof(*match));
- match->key = key;
- match->mask = mask;
-
- memset(key, 0, sizeof(*key));
+#define TCP_FLAGS_BE16(tp) (*(__be16 *)&tcp_flag_word(tp) & htons(0x0FFF))
- if (mask) {
- memset(&mask->key, 0, sizeof(mask->key));
- mask->range.start = mask->range.end = 0;
- }
-}
-
-static bool ovs_match_validate(const struct sw_flow_match *match,
- u64 key_attrs, u64 mask_attrs)
+void ovs_flow_stats_update(struct sw_flow *flow, __be16 tcp_flags,
+ struct sk_buff *skb)
{
- u64 key_expected = 1ULL << OVS_KEY_ATTR_ETHERNET;
- u64 mask_allowed = key_attrs; /* At most allow all key attributes */
-
- /* The following mask attributes allowed only if they
- * pass the validation tests. */
- mask_allowed &= ~((1ULL << OVS_KEY_ATTR_IPV4)
- | (1ULL << OVS_KEY_ATTR_IPV6)
- | (1ULL << OVS_KEY_ATTR_TCP)
- | (1ULL << OVS_KEY_ATTR_UDP)
- | (1ULL << OVS_KEY_ATTR_ICMP)
- | (1ULL << OVS_KEY_ATTR_ICMPV6)
- | (1ULL << OVS_KEY_ATTR_ARP)
- | (1ULL << OVS_KEY_ATTR_ND));
-
- if (match->key->eth.type == htons(ETH_P_802_2) &&
- match->mask && (match->mask->key.eth.type == htons(0xffff)))
- mask_allowed |= (1ULL << OVS_KEY_ATTR_ETHERTYPE);
-
- /* Check key attributes. */
- if (match->key->eth.type == htons(ETH_P_ARP)
- || match->key->eth.type == htons(ETH_P_RARP)) {
- key_expected |= 1ULL << OVS_KEY_ATTR_ARP;
- if (match->mask && (match->mask->key.eth.type == htons(0xffff)))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_ARP;
- }
+ struct flow_stats *stats;
+ int node = numa_node_id();
- if (match->key->eth.type == htons(ETH_P_IP)) {
- key_expected |= 1ULL << OVS_KEY_ATTR_IPV4;
- if (match->mask && (match->mask->key.eth.type == htons(0xffff)))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_IPV4;
+ stats = rcu_dereference(flow->stats[node]);
- if (match->key->ip.frag != OVS_FRAG_TYPE_LATER) {
- if (match->key->ip.proto == IPPROTO_UDP) {
- key_expected |= 1ULL << OVS_KEY_ATTR_UDP;
- if (match->mask && (match->mask->key.ip.proto == 0xff))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_UDP;
- }
-
- if (match->key->ip.proto == IPPROTO_TCP) {
- key_expected |= 1ULL << OVS_KEY_ATTR_TCP;
- if (match->mask && (match->mask->key.ip.proto == 0xff))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_TCP;
- }
+ /* Check if already have node-specific stats. */
+ if (likely(stats)) {
+ spin_lock(&stats->lock);
+ /* Mark if we write on the pre-allocated stats. */
+ if (node == 0 && unlikely(flow->stats_last_writer != node))
+ flow->stats_last_writer = node;
+ } else {
+ stats = rcu_dereference(flow->stats[0]); /* Pre-allocated. */
+ spin_lock(&stats->lock);
- if (match->key->ip.proto == IPPROTO_ICMP) {
- key_expected |= 1ULL << OVS_KEY_ATTR_ICMP;
- if (match->mask && (match->mask->key.ip.proto == 0xff))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_ICMP;
+ /* If the current NUMA-node is the only writer on the
+ * pre-allocated stats keep using them.
+ */
+ if (unlikely(flow->stats_last_writer != node)) {
+ /* A previous locker may have already allocated the
+ * stats, so we need to check again. If node-specific
+ * stats were already allocated, we update the pre-
+ * allocated stats as we have already locked them.
+ */
+ if (likely(flow->stats_last_writer != NUMA_NO_NODE)
+ && likely(!rcu_dereference(flow->stats[node]))) {
+ /* Try to allocate node-specific stats. */
+ struct flow_stats *new_stats;
+
+ new_stats =
+ kmem_cache_alloc_node(flow_stats_cache,
+ GFP_THISNODE |
+ __GFP_NOMEMALLOC,
+ node);
+ if (likely(new_stats)) {
+ new_stats->used = jiffies;
+ new_stats->packet_count = 1;
+ new_stats->byte_count = skb->len;
+ new_stats->tcp_flags = tcp_flags;
+ spin_lock_init(&new_stats->lock);
+
+ rcu_assign_pointer(flow->stats[node],
+ new_stats);
+ goto unlock;
+ }
}
+ flow->stats_last_writer = node;
}
}
- if (match->key->eth.type == htons(ETH_P_IPV6)) {
- key_expected |= 1ULL << OVS_KEY_ATTR_IPV6;
- if (match->mask && (match->mask->key.eth.type == htons(0xffff)))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_IPV6;
+ stats->used = jiffies;
+ stats->packet_count++;
+ stats->byte_count += skb->len;
+ stats->tcp_flags |= tcp_flags;
+unlock:
+ spin_unlock(&stats->lock);
+}
- if (match->key->ip.frag != OVS_FRAG_TYPE_LATER) {
- if (match->key->ip.proto == IPPROTO_UDP) {
- key_expected |= 1ULL << OVS_KEY_ATTR_UDP;
- if (match->mask && (match->mask->key.ip.proto == 0xff))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_UDP;
- }
+/* Must be called with rcu_read_lock or ovs_mutex. */
+void ovs_flow_stats_get(const struct sw_flow *flow,
+ struct ovs_flow_stats *ovs_stats,
+ unsigned long *used, __be16 *tcp_flags)
+{
+ int node;
- if (match->key->ip.proto == IPPROTO_TCP) {
- key_expected |= 1ULL << OVS_KEY_ATTR_TCP;
- if (match->mask && (match->mask->key.ip.proto == 0xff))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_TCP;
- }
+ *used = 0;
+ *tcp_flags = 0;
+ memset(ovs_stats, 0, sizeof(*ovs_stats));
- if (match->key->ip.proto == IPPROTO_ICMPV6) {
- key_expected |= 1ULL << OVS_KEY_ATTR_ICMPV6;
- if (match->mask && (match->mask->key.ip.proto == 0xff))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_ICMPV6;
-
- if (match->key->ipv6.tp.src ==
- htons(NDISC_NEIGHBOUR_SOLICITATION) ||
- match->key->ipv6.tp.src == htons(NDISC_NEIGHBOUR_ADVERTISEMENT)) {
- key_expected |= 1ULL << OVS_KEY_ATTR_ND;
- if (match->mask && (match->mask->key.ipv6.tp.src == htons(0xffff)))
- mask_allowed |= 1ULL << OVS_KEY_ATTR_ND;
- }
- }
+ for_each_node(node) {
+ struct flow_stats *stats = rcu_dereference_ovsl(flow->stats[node]);
+
+ if (stats) {
+ /* Local CPU may write on non-local stats, so we must
+ * block bottom-halves here.
+ */
+ spin_lock_bh(&stats->lock);
+ if (!*used || time_after(stats->used, *used))
+ *used = stats->used;
+ *tcp_flags |= stats->tcp_flags;
+ ovs_stats->n_packets += stats->packet_count;
+ ovs_stats->n_bytes += stats->byte_count;
+ spin_unlock_bh(&stats->lock);
}
}
+}
- if ((key_attrs & key_expected) != key_expected)
- /* Key attributes check failed. */
- return false;
+/* Called with ovs_mutex. */
+void ovs_flow_stats_clear(struct sw_flow *flow)
+{
+ int node;
- if ((mask_attrs & mask_allowed) != mask_attrs)
- /* Mask attributes check failed. */
- return false;
+ for_each_node(node) {
+ struct flow_stats *stats = ovsl_dereference(flow->stats[node]);
- return true;
+ if (stats) {
+ spin_lock_bh(&stats->lock);
+ stats->used = 0;
+ stats->packet_count = 0;
+ stats->byte_count = 0;
+ stats->tcp_flags = 0;
+ spin_unlock_bh(&stats->lock);
+ }
+ }
}
static int check_header(struct sk_buff *skb, int len)
sizeof(struct udphdr));
}
-static bool icmphdr_ok(struct sk_buff *skb)
+static bool sctphdr_ok(struct sk_buff *skb)
{
return pskb_may_pull(skb, skb_transport_offset(skb) +
- sizeof(struct icmphdr));
+ sizeof(struct sctphdr));
}
-u64 ovs_flow_used_time(unsigned long flow_jiffies)
+static bool icmphdr_ok(struct sk_buff *skb)
{
- struct timespec cur_ts;
- u64 cur_ms, idle_ms;
-
- ktime_get_ts(&cur_ts);
- idle_ms = jiffies_to_msecs(jiffies - flow_jiffies);
- cur_ms = (u64)cur_ts.tv_sec * MSEC_PER_SEC +
- cur_ts.tv_nsec / NSEC_PER_MSEC;
-
- return cur_ms - idle_ms;
+ return pskb_may_pull(skb, skb_transport_offset(skb) +
+ sizeof(struct icmphdr));
}
static int parse_ipv6hdr(struct sk_buff *skb, struct sw_flow_key *key)
sizeof(struct icmp6hdr));
}
-static void flow_key_mask(struct sw_flow_key *dst,
- const struct sw_flow_key *src,
- const struct sw_flow_mask *mask)
-{
- u8 *m = (u8 *)&mask->key + mask->range.start;
- u8 *s = (u8 *)src + mask->range.start;
- u8 *d = (u8 *)dst + mask->range.start;
- int i;
-
- memset(dst, 0, sizeof(*dst));
- for (i = 0; i < ovs_sw_flow_mask_size_roundup(mask); i++) {
- *d = *s & *m;
- d++, s++, m++;
- }
-}
-
-#define TCP_FLAGS_OFFSET 13
-#define TCP_FLAG_MASK 0x3f
-
-void ovs_flow_used(struct sw_flow *flow, struct sk_buff *skb)
-{
- u8 tcp_flags = 0;
-
- if ((flow->key.eth.type == htons(ETH_P_IP) ||
- flow->key.eth.type == htons(ETH_P_IPV6)) &&
- flow->key.ip.proto == IPPROTO_TCP &&
- likely(skb->len >= skb_transport_offset(skb) + sizeof(struct tcphdr))) {
- u8 *tcp = (u8 *)tcp_hdr(skb);
- tcp_flags = *(tcp + TCP_FLAGS_OFFSET) & TCP_FLAG_MASK;
- }
-
- spin_lock(&flow->lock);
- flow->used = jiffies;
- flow->packet_count++;
- flow->byte_count += skb->len;
- flow->tcp_flags |= tcp_flags;
- spin_unlock(&flow->lock);
-}
-
-struct sw_flow_actions *ovs_flow_actions_alloc(int size)
-{
- struct sw_flow_actions *sfa;
-
- if (size > MAX_ACTIONS_BUFSIZE)
- return ERR_PTR(-EINVAL);
-
- sfa = kmalloc(sizeof(*sfa) + size, GFP_KERNEL);
- if (!sfa)
- return ERR_PTR(-ENOMEM);
-
- sfa->actions_len = 0;
- return sfa;
-}
-
-struct sw_flow *ovs_flow_alloc(void)
-{
- struct sw_flow *flow;
-
- flow = kmem_cache_alloc(flow_cache, GFP_KERNEL);
- if (!flow)
- return ERR_PTR(-ENOMEM);
-
- spin_lock_init(&flow->lock);
- flow->sf_acts = NULL;
- flow->mask = NULL;
-
- return flow;
-}
-
-static struct hlist_head *find_bucket(struct flow_table *table, u32 hash)
-{
- hash = jhash_1word(hash, table->hash_seed);
- return flex_array_get(table->buckets,
- (hash & (table->n_buckets - 1)));
-}
-
-static struct flex_array *alloc_buckets(unsigned int n_buckets)
-{
- struct flex_array *buckets;
- int i, err;
-
- buckets = flex_array_alloc(sizeof(struct hlist_head *),
- n_buckets, GFP_KERNEL);
- if (!buckets)
- return NULL;
-
- err = flex_array_prealloc(buckets, 0, n_buckets, GFP_KERNEL);
- if (err) {
- flex_array_free(buckets);
- return NULL;
- }
-
- for (i = 0; i < n_buckets; i++)
- INIT_HLIST_HEAD((struct hlist_head *)
- flex_array_get(buckets, i));
-
- return buckets;
-}
-
-static void free_buckets(struct flex_array *buckets)
-{
- flex_array_free(buckets);
-}
-
-struct flow_table *ovs_flow_tbl_alloc(int new_size)
-{
- struct flow_table *table = kmalloc(sizeof(*table), GFP_KERNEL);
-
- if (!table)
- return NULL;
-
- table->buckets = alloc_buckets(new_size);
-
- if (!table->buckets) {
- kfree(table);
- return NULL;
- }
- table->n_buckets = new_size;
- table->count = 0;
- table->node_ver = 0;
- table->keep_flows = false;
- get_random_bytes(&table->hash_seed, sizeof(u32));
- INIT_LIST_HEAD(&table->mask_list);
-
- return table;
-}
-
-static void __flow_tbl_destroy(struct flow_table *table)
-{
- int i;
-
- if (table->keep_flows)
- goto skip_flows;
-
- for (i = 0; i < table->n_buckets; i++) {
- struct sw_flow *flow;
- struct hlist_head *head = flex_array_get(table->buckets, i);
- struct hlist_node *n;
- int ver = table->node_ver;
-
- hlist_for_each_entry_safe(flow, n, head, hash_node[ver]) {
- hlist_del_rcu(&flow->hash_node[ver]);
- ovs_flow_free(flow, false);
- }
- }
-
-skip_flows:
- free_buckets(table->buckets);
- kfree(table);
-}
-
-static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu)
-{
- struct flow_table *table = container_of(rcu, struct flow_table, rcu);
-
- __flow_tbl_destroy(table);
-}
-
-void ovs_flow_tbl_destroy(struct flow_table *table, bool deferred)
-{
- if (!table)
- return;
-
- if (deferred)
- call_rcu(&table->rcu, flow_tbl_destroy_rcu_cb);
- else
- __flow_tbl_destroy(table);
-}
-
-struct sw_flow *ovs_flow_dump_next(struct flow_table *table, u32 *bucket, u32 *last)
-{
- struct sw_flow *flow;
- struct hlist_head *head;
- int ver;
- int i;
-
- ver = table->node_ver;
- while (*bucket < table->n_buckets) {
- i = 0;
- head = flex_array_get(table->buckets, *bucket);
- hlist_for_each_entry_rcu(flow, head, hash_node[ver]) {
- if (i < *last) {
- i++;
- continue;
- }
- *last = i + 1;
- return flow;
- }
- (*bucket)++;
- *last = 0;
- }
-
- return NULL;
-}
-
-static void __tbl_insert(struct flow_table *table, struct sw_flow *flow)
-{
- struct hlist_head *head;
-
- head = find_bucket(table, flow->hash);
- hlist_add_head_rcu(&flow->hash_node[table->node_ver], head);
-
- table->count++;
-}
-
-static void flow_table_copy_flows(struct flow_table *old, struct flow_table *new)
-{
- int old_ver;
- int i;
-
- old_ver = old->node_ver;
- new->node_ver = !old_ver;
-
- /* Insert in new table. */
- for (i = 0; i < old->n_buckets; i++) {
- struct sw_flow *flow;
- struct hlist_head *head;
-
- head = flex_array_get(old->buckets, i);
-
- hlist_for_each_entry(flow, head, hash_node[old_ver])
- __tbl_insert(new, flow);
- }
-
- new->mask_list = old->mask_list;
- old->keep_flows = true;
-}
-
-static struct flow_table *__flow_tbl_rehash(struct flow_table *table, int n_buckets)
-{
- struct flow_table *new_table;
-
- new_table = ovs_flow_tbl_alloc(n_buckets);
- if (!new_table)
- return ERR_PTR(-ENOMEM);
-
- flow_table_copy_flows(table, new_table);
-
- return new_table;
-}
-
-struct flow_table *ovs_flow_tbl_rehash(struct flow_table *table)
-{
- return __flow_tbl_rehash(table, table->n_buckets);
-}
-
-struct flow_table *ovs_flow_tbl_expand(struct flow_table *table)
-{
- return __flow_tbl_rehash(table, table->n_buckets * 2);
-}
-
-static void __flow_free(struct sw_flow *flow)
-{
- kfree((struct sf_flow_acts __force *)flow->sf_acts);
- kmem_cache_free(flow_cache, flow);
-}
-
-static void rcu_free_flow_callback(struct rcu_head *rcu)
-{
- struct sw_flow *flow = container_of(rcu, struct sw_flow, rcu);
-
- __flow_free(flow);
-}
-
-void ovs_flow_free(struct sw_flow *flow, bool deferred)
-{
- if (!flow)
- return;
-
- ovs_sw_flow_mask_del_ref((struct sw_flow_mask __force *)flow->mask,
- deferred);
-
- if (deferred)
- call_rcu(&flow->rcu, rcu_free_flow_callback);
- else
- __flow_free(flow);
-}
-
-/* RCU callback used by ovs_flow_deferred_free_acts. */
-static void rcu_free_acts_callback(struct rcu_head *rcu)
-{
- struct sw_flow_actions *sf_acts = container_of(rcu,
- struct sw_flow_actions, rcu);
- kfree(sf_acts);
-}
-
-/* Schedules 'sf_acts' to be freed after the next RCU grace period.
- * The caller must hold rcu_read_lock for this to be sensible. */
-void ovs_flow_deferred_free_acts(struct sw_flow_actions *sf_acts)
-{
- call_rcu(&sf_acts->rcu, rcu_free_acts_callback);
-}
-
static int parse_vlan(struct sk_buff *skb, struct sw_flow_key *key)
{
struct qtag_prefix {
/* 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->ipv6.tp.src = htons(icmp->icmp6_type);
- key->ipv6.tp.dst = htons(icmp->icmp6_code);
+ key->tp.src = htons(icmp->icmp6_type);
+ key->tp.dst = htons(icmp->icmp6_code);
if (icmp->icmp6_code == 0 &&
(icmp->icmp6_type == NDISC_NEIGHBOUR_SOLICITATION ||
&& opt_len == 8) {
if (unlikely(!is_zero_ether_addr(key->ipv6.nd.sll)))
goto invalid;
- memcpy(key->ipv6.nd.sll,
- &nd->opt[offset+sizeof(*nd_opt)], ETH_ALEN);
+ ether_addr_copy(key->ipv6.nd.sll,
+ &nd->opt[offset+sizeof(*nd_opt)]);
} else if (nd_opt->nd_opt_type == ND_OPT_TARGET_LL_ADDR
&& opt_len == 8) {
if (unlikely(!is_zero_ether_addr(key->ipv6.nd.tll)))
goto invalid;
- memcpy(key->ipv6.nd.tll,
- &nd->opt[offset+sizeof(*nd_opt)], ETH_ALEN);
+ ether_addr_copy(key->ipv6.nd.tll,
+ &nd->opt[offset+sizeof(*nd_opt)]);
}
icmp_len -= opt_len;
* Ethernet header
* @in_port: port number on which @skb was received.
* @key: output flow key
- * @key_lenp: length of output flow key
*
* The caller must ensure that skb->len >= ETH_HLEN.
*
if (OVS_CB(skb)->tun_key)
memcpy(&key->tun_key, OVS_CB(skb)->tun_key, sizeof(key->tun_key));
key->phy.in_port = in_port;
- key->phy.skb_mark = skb_get_mark(skb);
+ key->phy.skb_mark = skb->mark;
skb_reset_mac_header(skb);
* header in the linear data area.
*/
eth = eth_hdr(skb);
- memcpy(key->eth.src, eth->h_source, ETH_ALEN);
- memcpy(key->eth.dst, eth->h_dest, ETH_ALEN);
+ ether_addr_copy(key->eth.src, eth->h_source);
+ ether_addr_copy(key->eth.dst, eth->h_dest);
__skb_pull(skb, 2 * ETH_ALEN);
/* We are going to push all headers that we pull, so no need to
if (key->ip.proto == IPPROTO_TCP) {
if (tcphdr_ok(skb)) {
struct tcphdr *tcp = tcp_hdr(skb);
- key->ipv4.tp.src = tcp->source;
- key->ipv4.tp.dst = tcp->dest;
+ key->tp.src = tcp->source;
+ key->tp.dst = tcp->dest;
+ key->tp.flags = TCP_FLAGS_BE16(tcp);
}
} else if (key->ip.proto == IPPROTO_UDP) {
if (udphdr_ok(skb)) {
struct udphdr *udp = udp_hdr(skb);
- key->ipv4.tp.src = udp->source;
- key->ipv4.tp.dst = udp->dest;
+ key->tp.src = udp->source;
+ key->tp.dst = udp->dest;
+ }
+ } else if (key->ip.proto == IPPROTO_SCTP) {
+ if (sctphdr_ok(skb)) {
+ struct sctphdr *sctp = sctp_hdr(skb);
+ key->tp.src = sctp->source;
+ key->tp.dst = sctp->dest;
}
} else if (key->ip.proto == IPPROTO_ICMP) {
if (icmphdr_ok(skb)) {
/* The ICMP type and code fields use the 16-bit
* transport port fields, so we need to store
* them in 16-bit network byte order. */
- key->ipv4.tp.src = htons(icmp->type);
- key->ipv4.tp.dst = htons(icmp->code);
+ key->tp.src = htons(icmp->type);
+ key->tp.dst = htons(icmp->code);
}
}
key->ip.proto = ntohs(arp->ar_op);
memcpy(&key->ipv4.addr.src, arp->ar_sip, sizeof(key->ipv4.addr.src));
memcpy(&key->ipv4.addr.dst, arp->ar_tip, sizeof(key->ipv4.addr.dst));
- memcpy(key->ipv4.arp.sha, arp->ar_sha, ETH_ALEN);
- memcpy(key->ipv4.arp.tha, arp->ar_tha, ETH_ALEN);
+ ether_addr_copy(key->ipv4.arp.sha, arp->ar_sha);
+ ether_addr_copy(key->ipv4.arp.tha, arp->ar_tha);
}
} else if (key->eth.type == htons(ETH_P_IPV6)) {
int nh_len; /* IPv6 Header + Extensions */
if (key->ip.proto == NEXTHDR_TCP) {
if (tcphdr_ok(skb)) {
struct tcphdr *tcp = tcp_hdr(skb);
- key->ipv6.tp.src = tcp->source;
- key->ipv6.tp.dst = tcp->dest;
+ key->tp.src = tcp->source;
+ key->tp.dst = tcp->dest;
+ key->tp.flags = TCP_FLAGS_BE16(tcp);
}
} else if (key->ip.proto == NEXTHDR_UDP) {
if (udphdr_ok(skb)) {
struct udphdr *udp = udp_hdr(skb);
- key->ipv6.tp.src = udp->source;
- key->ipv6.tp.dst = udp->dest;
+ key->tp.src = udp->source;
+ key->tp.dst = udp->dest;
+ }
+ } else if (key->ip.proto == NEXTHDR_SCTP) {
+ if (sctphdr_ok(skb)) {
+ struct sctphdr *sctp = sctp_hdr(skb);
+ key->tp.src = sctp->source;
+ key->tp.dst = sctp->dest;
}
} else if (key->ip.proto == NEXTHDR_ICMP) {
if (icmp6hdr_ok(skb)) {
return 0;
}
-
-static u32 ovs_flow_hash(const struct sw_flow_key *key, int key_start, int key_len)
-{
- return jhash2((u32 *)((u8 *)key + key_start),
- DIV_ROUND_UP(key_len - key_start, sizeof(u32)), 0);
-}
-
-static int flow_key_start(const struct sw_flow_key *key)
-{
- if (key->tun_key.ipv4_dst)
- return 0;
- else
- return offsetof(struct sw_flow_key, phy);
-}
-
-static bool __cmp_key(const struct sw_flow_key *key1,
- const struct sw_flow_key *key2, int key_start, int key_len)
-{
- return !memcmp((u8 *)key1 + key_start,
- (u8 *)key2 + key_start, (key_len - key_start));
-}
-
-static bool __flow_cmp_key(const struct sw_flow *flow,
- const struct sw_flow_key *key, int key_start, int key_len)
-{
- return __cmp_key(&flow->key, key, key_start, key_len);
-}
-
-static bool __flow_cmp_unmasked_key(const struct sw_flow *flow,
- const struct sw_flow_key *key, int key_start, int key_len)
-{
- return __cmp_key(&flow->unmasked_key, key, key_start, key_len);
-}
-
-bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow,
- const struct sw_flow_key *key, int key_len)
-{
- int key_start;
- key_start = flow_key_start(key);
-
- return __flow_cmp_unmasked_key(flow, key, key_start, key_len);
-
-}
-
-struct sw_flow *ovs_flow_lookup_unmasked_key(struct flow_table *table,
- struct sw_flow_match *match)
-{
- struct sw_flow_key *unmasked = match->key;
- int key_len = match->range.end;
- struct sw_flow *flow;
-
- flow = ovs_flow_lookup(table, unmasked);
- if (flow && (!ovs_flow_cmp_unmasked_key(flow, unmasked, key_len)))
- flow = NULL;
-
- return flow;
-}
-
-static struct sw_flow *ovs_masked_flow_lookup(struct flow_table *table,
- const struct sw_flow_key *flow_key,
- struct sw_flow_mask *mask)
-{
- struct sw_flow *flow;
- struct hlist_head *head;
- int key_start = mask->range.start;
- int key_len = mask->range.end;
- u32 hash;
- struct sw_flow_key masked_key;
-
- flow_key_mask(&masked_key, flow_key, mask);
- hash = ovs_flow_hash(&masked_key, key_start, key_len);
- head = find_bucket(table, hash);
- hlist_for_each_entry_rcu(flow, head, hash_node[table->node_ver]) {
- if (__flow_cmp_key(flow, &masked_key, key_start, key_len))
- return flow;
- }
- return NULL;
-}
-
-struct sw_flow *ovs_flow_lookup(struct flow_table *tbl,
- const struct sw_flow_key *key)
-{
- struct sw_flow *flow = NULL;
- struct sw_flow_mask *mask;
-
- list_for_each_entry_rcu(mask, &tbl->mask_list, list) {
- flow = ovs_masked_flow_lookup(tbl, key, mask);
- if (flow) /* Found */
- break;
- }
-
- return flow;
-}
-
-
-void ovs_flow_insert(struct flow_table *table, struct sw_flow *flow,
- const struct sw_flow_key *key, int key_len)
-{
- flow->unmasked_key = *key;
- flow_key_mask(&flow->key, &flow->unmasked_key, ovsl_dereference(flow->mask));
- flow->hash = ovs_flow_hash(&flow->key,
- ovsl_dereference(flow->mask)->range.start,
- ovsl_dereference(flow->mask)->range.end);
- __tbl_insert(table, flow);
-}
-
-void ovs_flow_remove(struct flow_table *table, struct sw_flow *flow)
-{
- BUG_ON(table->count == 0);
- hlist_del_rcu(&flow->hash_node[table->node_ver]);
- table->count--;
-}
-
-/* The size of the argument for each %OVS_KEY_ATTR_* Netlink attribute. */
-const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = {
- [OVS_KEY_ATTR_ENCAP] = -1,
- [OVS_KEY_ATTR_PRIORITY] = sizeof(u32),
- [OVS_KEY_ATTR_IN_PORT] = sizeof(u32),
- [OVS_KEY_ATTR_SKB_MARK] = sizeof(u32),
- [OVS_KEY_ATTR_ETHERNET] = sizeof(struct ovs_key_ethernet),
- [OVS_KEY_ATTR_VLAN] = sizeof(__be16),
- [OVS_KEY_ATTR_ETHERTYPE] = sizeof(__be16),
- [OVS_KEY_ATTR_IPV4] = sizeof(struct ovs_key_ipv4),
- [OVS_KEY_ATTR_IPV6] = sizeof(struct ovs_key_ipv6),
- [OVS_KEY_ATTR_TCP] = sizeof(struct ovs_key_tcp),
- [OVS_KEY_ATTR_UDP] = sizeof(struct ovs_key_udp),
- [OVS_KEY_ATTR_ICMP] = sizeof(struct ovs_key_icmp),
- [OVS_KEY_ATTR_ICMPV6] = sizeof(struct ovs_key_icmpv6),
- [OVS_KEY_ATTR_ARP] = sizeof(struct ovs_key_arp),
- [OVS_KEY_ATTR_ND] = sizeof(struct ovs_key_nd),
- [OVS_KEY_ATTR_TUNNEL] = -1,
-};
-
-static bool is_all_zero(const u8 *fp, size_t size)
-{
- int i;
-
- if (!fp)
- return false;
-
- for (i = 0; i < size; i++)
- if (fp[i])
- return false;
-
- return true;
-}
-
-static int __parse_flow_nlattrs(const struct nlattr *attr,
- const struct nlattr *a[],
- u64 *attrsp, bool nz)
-{
- const struct nlattr *nla;
- u64 attrs;
- int rem;
-
- attrs = *attrsp;
- nla_for_each_nested(nla, attr, rem) {
- u16 type = nla_type(nla);
- int expected_len;
-
- if (type > OVS_KEY_ATTR_MAX || attrs & (1ULL << type))
- return -EINVAL;
-
- expected_len = ovs_key_lens[type];
- if (nla_len(nla) != expected_len && expected_len != -1)
- return -EINVAL;
-
- if (attrs & (1ULL << type))
- /* Duplicated field. */
- return -EINVAL;
-
- if (!nz || !is_all_zero(nla_data(nla), expected_len)) {
- attrs |= 1ULL << type;
- a[type] = nla;
- }
- }
- if (rem)
- return -EINVAL;
-
- *attrsp = attrs;
- return 0;
-}
-
-static int parse_flow_mask_nlattrs(const struct nlattr *attr,
- const struct nlattr *a[], u64 *attrsp)
-{
- return __parse_flow_nlattrs(attr, a, attrsp, true);
-}
-
-static int parse_flow_nlattrs(const struct nlattr *attr,
- const struct nlattr *a[], u64 *attrsp)
-{
- return __parse_flow_nlattrs(attr, a, attrsp, false);
-}
-
-int ipv4_tun_from_nlattr(const struct nlattr *attr,
- struct sw_flow_match *match, bool is_mask)
-{
- struct nlattr *a;
- int rem;
- bool ttl = false;
- __be16 tun_flags = 0;
-
- nla_for_each_nested(a, attr, rem) {
- int type = nla_type(a);
- static const u32 ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1] = {
- [OVS_TUNNEL_KEY_ATTR_ID] = sizeof(u64),
- [OVS_TUNNEL_KEY_ATTR_IPV4_SRC] = sizeof(u32),
- [OVS_TUNNEL_KEY_ATTR_IPV4_DST] = sizeof(u32),
- [OVS_TUNNEL_KEY_ATTR_TOS] = 1,
- [OVS_TUNNEL_KEY_ATTR_TTL] = 1,
- [OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT] = 0,
- [OVS_TUNNEL_KEY_ATTR_CSUM] = 0,
- };
-
- if (type > OVS_TUNNEL_KEY_ATTR_MAX ||
- ovs_tunnel_key_lens[type] != nla_len(a))
- return -EINVAL;
-
- switch (type) {
- case OVS_TUNNEL_KEY_ATTR_ID:
- SW_FLOW_KEY_PUT(match, tun_key.tun_id,
- nla_get_be64(a), is_mask);
- tun_flags |= TUNNEL_KEY;
- break;
- case OVS_TUNNEL_KEY_ATTR_IPV4_SRC:
- SW_FLOW_KEY_PUT(match, tun_key.ipv4_src,
- nla_get_be32(a), is_mask);
- break;
- case OVS_TUNNEL_KEY_ATTR_IPV4_DST:
- SW_FLOW_KEY_PUT(match, tun_key.ipv4_dst,
- nla_get_be32(a), is_mask);
- break;
- case OVS_TUNNEL_KEY_ATTR_TOS:
- SW_FLOW_KEY_PUT(match, tun_key.ipv4_tos,
- nla_get_u8(a), is_mask);
- break;
- case OVS_TUNNEL_KEY_ATTR_TTL:
- SW_FLOW_KEY_PUT(match, tun_key.ipv4_ttl,
- nla_get_u8(a), is_mask);
- ttl = true;
- break;
- case OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT:
- tun_flags |= TUNNEL_DONT_FRAGMENT;
- break;
- case OVS_TUNNEL_KEY_ATTR_CSUM:
- tun_flags |= TUNNEL_CSUM;
- break;
- default:
- return -EINVAL;
- }
- }
-
- SW_FLOW_KEY_PUT(match, tun_key.tun_flags, tun_flags, is_mask);
-
- if (rem > 0)
- return -EINVAL;
-
- if (!match->key->tun_key.ipv4_dst)
- return -EINVAL;
-
- if (!ttl)
- return -EINVAL;
-
- return 0;
-}
-
-int ipv4_tun_to_nlattr(struct sk_buff *skb,
- const struct ovs_key_ipv4_tunnel *tun_key,
- const struct ovs_key_ipv4_tunnel *output)
-{
- struct nlattr *nla;
-
- nla = nla_nest_start(skb, OVS_KEY_ATTR_TUNNEL);
- if (!nla)
- return -EMSGSIZE;
-
- if (tun_key->tun_flags & TUNNEL_KEY &&
- nla_put_be64(skb, OVS_TUNNEL_KEY_ATTR_ID, output->tun_id))
- return -EMSGSIZE;
- if (tun_key->ipv4_src &&
- nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_SRC, output->ipv4_src))
- return -EMSGSIZE;
- if (nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_DST, output->ipv4_dst))
- return -EMSGSIZE;
- if (tun_key->ipv4_tos &&
- nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TOS, output->ipv4_tos))
- return -EMSGSIZE;
- if (nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TTL, output->ipv4_ttl))
- return -EMSGSIZE;
- if ((tun_key->tun_flags & TUNNEL_DONT_FRAGMENT) &&
- nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT))
- return -EMSGSIZE;
- if ((tun_key->tun_flags & TUNNEL_CSUM) &&
- nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_CSUM))
- return -EMSGSIZE;
-
- nla_nest_end(skb, nla);
- return 0;
-}
-
-
-static int metadata_from_nlattrs(struct sw_flow_match *match, u64 *attrs,
- const struct nlattr **a, bool is_mask)
-{
- if (*attrs & (1ULL << OVS_KEY_ATTR_PRIORITY)) {
- SW_FLOW_KEY_PUT(match, phy.priority,
- nla_get_u32(a[OVS_KEY_ATTR_PRIORITY]), is_mask);
- *attrs &= ~(1ULL << OVS_KEY_ATTR_PRIORITY);
- }
-
- if (*attrs & (1ULL << OVS_KEY_ATTR_IN_PORT)) {
- u32 in_port = nla_get_u32(a[OVS_KEY_ATTR_IN_PORT]);
-
- if (!is_mask && in_port >= DP_MAX_PORTS)
- return -EINVAL;
- SW_FLOW_KEY_PUT(match, phy.in_port, in_port, is_mask);
- *attrs &= ~(1ULL << OVS_KEY_ATTR_IN_PORT);
- }
-
- if (*attrs & (1ULL << OVS_KEY_ATTR_SKB_MARK)) {
- uint32_t mark = nla_get_u32(a[OVS_KEY_ATTR_SKB_MARK]);
-#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,20) && !defined(CONFIG_NETFILTER)
- if (!is_mask && mark != 0)
- return -EINVAL;
-#endif
- SW_FLOW_KEY_PUT(match, phy.skb_mark, mark, is_mask);
- *attrs &= ~(1ULL << OVS_KEY_ATTR_SKB_MARK);
- }
- if (*attrs & (1ULL << OVS_KEY_ATTR_TUNNEL)) {
- if (ipv4_tun_from_nlattr(a[OVS_KEY_ATTR_TUNNEL], match,
- is_mask))
- return -EINVAL;
- *attrs &= ~(1ULL << OVS_KEY_ATTR_TUNNEL);
- }
- return 0;
-}
-
-static int ovs_key_from_nlattrs(struct sw_flow_match *match, u64 attrs,
- const struct nlattr **a, bool is_mask)
-{
- int err;
-
- err = metadata_from_nlattrs(match, &attrs, a, is_mask);
- if (err)
- return err;
-
- if (attrs & (1ULL << OVS_KEY_ATTR_ETHERNET)) {
- const struct ovs_key_ethernet *eth_key;
-
- eth_key = nla_data(a[OVS_KEY_ATTR_ETHERNET]);
- SW_FLOW_KEY_MEMCPY(match, eth.src,
- eth_key->eth_src, ETH_ALEN, is_mask);
- SW_FLOW_KEY_MEMCPY(match, eth.dst,
- eth_key->eth_dst, ETH_ALEN, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_ETHERNET);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_VLAN)) {
- __be16 tci;
-
- tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]);
- if (!is_mask && (tci & htons(VLAN_TAG_PRESENT)))
- return -EINVAL;
-
- SW_FLOW_KEY_PUT(match, eth.tci, tci, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_VLAN);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_ETHERTYPE)) {
- __be16 eth_type;
-
- eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
- if (!is_mask && ntohs(eth_type) < ETH_P_802_3_MIN)
- return -EINVAL;
-
- SW_FLOW_KEY_PUT(match, eth.type, eth_type, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_ETHERTYPE);
- } else if (!is_mask) {
- SW_FLOW_KEY_PUT(match, eth.type, htons(ETH_P_802_2), is_mask);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_IPV4)) {
- const struct ovs_key_ipv4 *ipv4_key;
-
- ipv4_key = nla_data(a[OVS_KEY_ATTR_IPV4]);
- if (!is_mask && ipv4_key->ipv4_frag > OVS_FRAG_TYPE_MAX)
- return -EINVAL;
- SW_FLOW_KEY_PUT(match, ip.proto,
- ipv4_key->ipv4_proto, is_mask);
- SW_FLOW_KEY_PUT(match, ip.tos,
- ipv4_key->ipv4_tos, is_mask);
- SW_FLOW_KEY_PUT(match, ip.ttl,
- ipv4_key->ipv4_ttl, is_mask);
- SW_FLOW_KEY_PUT(match, ip.frag,
- ipv4_key->ipv4_frag, is_mask);
- SW_FLOW_KEY_PUT(match, ipv4.addr.src,
- ipv4_key->ipv4_src, is_mask);
- SW_FLOW_KEY_PUT(match, ipv4.addr.dst,
- ipv4_key->ipv4_dst, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_IPV4);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_IPV6)) {
- const struct ovs_key_ipv6 *ipv6_key;
-
- ipv6_key = nla_data(a[OVS_KEY_ATTR_IPV6]);
- if (!is_mask && ipv6_key->ipv6_frag > OVS_FRAG_TYPE_MAX)
- return -EINVAL;
- SW_FLOW_KEY_PUT(match, ipv6.label,
- ipv6_key->ipv6_label, is_mask);
- SW_FLOW_KEY_PUT(match, ip.proto,
- ipv6_key->ipv6_proto, is_mask);
- SW_FLOW_KEY_PUT(match, ip.tos,
- ipv6_key->ipv6_tclass, is_mask);
- SW_FLOW_KEY_PUT(match, ip.ttl,
- ipv6_key->ipv6_hlimit, is_mask);
- SW_FLOW_KEY_PUT(match, ip.frag,
- ipv6_key->ipv6_frag, is_mask);
- SW_FLOW_KEY_MEMCPY(match, ipv6.addr.src,
- ipv6_key->ipv6_src,
- sizeof(match->key->ipv6.addr.src),
- is_mask);
- SW_FLOW_KEY_MEMCPY(match, ipv6.addr.dst,
- ipv6_key->ipv6_dst,
- sizeof(match->key->ipv6.addr.dst),
- is_mask);
-
- attrs &= ~(1ULL << OVS_KEY_ATTR_IPV6);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_ARP)) {
- const struct ovs_key_arp *arp_key;
-
- arp_key = nla_data(a[OVS_KEY_ATTR_ARP]);
- if (!is_mask && (arp_key->arp_op & htons(0xff00)))
- return -EINVAL;
-
- SW_FLOW_KEY_PUT(match, ipv4.addr.src,
- arp_key->arp_sip, is_mask);
- SW_FLOW_KEY_PUT(match, ipv4.addr.dst,
- arp_key->arp_tip, is_mask);
- SW_FLOW_KEY_PUT(match, ip.proto,
- ntohs(arp_key->arp_op), is_mask);
- SW_FLOW_KEY_MEMCPY(match, ipv4.arp.sha,
- arp_key->arp_sha, ETH_ALEN, is_mask);
- SW_FLOW_KEY_MEMCPY(match, ipv4.arp.tha,
- arp_key->arp_tha, ETH_ALEN, is_mask);
-
- attrs &= ~(1ULL << OVS_KEY_ATTR_ARP);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_TCP)) {
- const struct ovs_key_tcp *tcp_key;
-
- tcp_key = nla_data(a[OVS_KEY_ATTR_TCP]);
- SW_FLOW_KEY_PUT(match, ipv4.tp.src,
- tcp_key->tcp_src, is_mask);
- SW_FLOW_KEY_PUT(match, ipv4.tp.dst,
- tcp_key->tcp_dst, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_TCP);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_UDP)) {
- const struct ovs_key_udp *udp_key;
-
- udp_key = nla_data(a[OVS_KEY_ATTR_UDP]);
- SW_FLOW_KEY_PUT(match, ipv4.tp.src,
- udp_key->udp_src, is_mask);
- SW_FLOW_KEY_PUT(match, ipv4.tp.dst,
- udp_key->udp_dst, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_UDP);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_ICMP)) {
- const struct ovs_key_icmp *icmp_key;
-
- icmp_key = nla_data(a[OVS_KEY_ATTR_ICMP]);
- SW_FLOW_KEY_PUT(match, ipv4.tp.src,
- htons(icmp_key->icmp_type), is_mask);
- SW_FLOW_KEY_PUT(match, ipv4.tp.dst,
- htons(icmp_key->icmp_code), is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_ICMP);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_ICMPV6)) {
- const struct ovs_key_icmpv6 *icmpv6_key;
-
- icmpv6_key = nla_data(a[OVS_KEY_ATTR_ICMPV6]);
- SW_FLOW_KEY_PUT(match, ipv6.tp.src,
- htons(icmpv6_key->icmpv6_type), is_mask);
- SW_FLOW_KEY_PUT(match, ipv6.tp.dst,
- htons(icmpv6_key->icmpv6_code), is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_ICMPV6);
- }
-
- if (attrs & (1ULL << OVS_KEY_ATTR_ND)) {
- const struct ovs_key_nd *nd_key;
-
- nd_key = nla_data(a[OVS_KEY_ATTR_ND]);
- SW_FLOW_KEY_MEMCPY(match, ipv6.nd.target,
- nd_key->nd_target,
- sizeof(match->key->ipv6.nd.target),
- is_mask);
- SW_FLOW_KEY_MEMCPY(match, ipv6.nd.sll,
- nd_key->nd_sll, ETH_ALEN, is_mask);
- SW_FLOW_KEY_MEMCPY(match, ipv6.nd.tll,
- nd_key->nd_tll, ETH_ALEN, is_mask);
- attrs &= ~(1ULL << OVS_KEY_ATTR_ND);
- }
-
- if (attrs != 0)
- return -EINVAL;
-
- return 0;
-}
-
-/**
- * ovs_match_from_nlattrs - parses Netlink attributes into a flow key and
- * mask. In case the 'mask' is NULL, the flow is treated as exact match
- * flow. Otherwise, it is treated as a wildcarded flow, except the mask
- * does not include any don't care bit.
- * @match: receives the extracted flow match information.
- * @key: Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink attribute
- * sequence. The fields should of the packet that triggered the creation
- * of this flow.
- * @mask: Optional. Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink
- * attribute specifies the mask field of the wildcarded flow.
- */
-int ovs_match_from_nlattrs(struct sw_flow_match *match,
- const struct nlattr *key,
- const struct nlattr *mask)
-{
- const struct nlattr *a[OVS_KEY_ATTR_MAX + 1];
- const struct nlattr *encap;
- u64 key_attrs = 0;
- u64 mask_attrs = 0;
- bool encap_valid = false;
- int err;
-
- err = parse_flow_nlattrs(key, a, &key_attrs);
- if (err)
- return err;
-
- if (key_attrs & 1ULL << OVS_KEY_ATTR_ENCAP) {
- encap = a[OVS_KEY_ATTR_ENCAP];
- key_attrs &= ~(1ULL << OVS_KEY_ATTR_ENCAP);
- if (nla_len(encap)) {
- __be16 eth_type = 0; /* ETH_P_8021Q */
-
- if (a[OVS_KEY_ATTR_ETHERTYPE])
- eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
-
- if ((eth_type == htons(ETH_P_8021Q)) && (a[OVS_KEY_ATTR_VLAN])) {
- encap_valid = true;
- key_attrs &= ~(1ULL << OVS_KEY_ATTR_ETHERTYPE);
- err = parse_flow_nlattrs(encap, a, &key_attrs);
- } else
- err = -EINVAL;
-
- if (err)
- return err;
- }
- }
-
- err = ovs_key_from_nlattrs(match, key_attrs, a, false);
- if (err)
- return err;
-
- if (mask) {
- err = parse_flow_mask_nlattrs(mask, a, &mask_attrs);
- if (err)
- return err;
-
- if ((mask_attrs & 1ULL << OVS_KEY_ATTR_ENCAP) && encap_valid) {
- __be16 eth_type = 0;
-
- if (a[OVS_KEY_ATTR_ETHERTYPE])
- eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]);
- if (eth_type == htons(0xffff)) {
- mask_attrs &= ~(1ULL << OVS_KEY_ATTR_ETHERTYPE);
- encap = a[OVS_KEY_ATTR_ENCAP];
- err = parse_flow_mask_nlattrs(encap, a, &mask_attrs);
- } else
- err = -EINVAL;
-
- if (err)
- return err;
- }
-
- err = ovs_key_from_nlattrs(match, mask_attrs, a, true);
- if (err)
- return err;
- } else {
- /* Populate exact match flow's key mask. */
- if (match->mask)
- ovs_sw_flow_mask_set(match->mask, &match->range, 0xff);
- }
-
- if (!ovs_match_validate(match, key_attrs, mask_attrs))
- return -EINVAL;
-
- return 0;
-}
-
-/**
- * ovs_flow_metadata_from_nlattrs - parses Netlink attributes into a flow key.
- * @flow: Receives extracted in_port, priority, tun_key and skb_mark.
- * @attr: Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink attribute
- * sequence.
- *
- * This parses a series of Netlink attributes that form a flow key, which must
- * take the same form accepted by flow_from_nlattrs(), but only enough of it to
- * get the metadata, that is, the parts of the flow key that cannot be
- * extracted from the packet itself.
- */
-
-int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow,
- const struct nlattr *attr)
-{
- struct ovs_key_ipv4_tunnel *tun_key = &flow->key.tun_key;
- const struct nlattr *a[OVS_KEY_ATTR_MAX + 1];
- u64 attrs = 0;
- int err;
- struct sw_flow_match match;
-
- flow->key.phy.in_port = DP_MAX_PORTS;
- flow->key.phy.priority = 0;
- flow->key.phy.skb_mark = 0;
- memset(tun_key, 0, sizeof(flow->key.tun_key));
-
- err = parse_flow_nlattrs(attr, a, &attrs);
- if (err)
- return -EINVAL;
-
- ovs_match_init(&match, &flow->key, NULL);
-
- err = metadata_from_nlattrs(&match, &attrs, a, false);
- if (err)
- return err;
-
- return 0;
-}
-
-int ovs_flow_to_nlattrs(const struct sw_flow_key *swkey,
- const struct sw_flow_key *output, struct sk_buff *skb)
-{
- struct ovs_key_ethernet *eth_key;
- struct nlattr *nla, *encap;
-
- if (swkey->phy.priority &&
- nla_put_u32(skb, OVS_KEY_ATTR_PRIORITY, output->phy.priority))
- goto nla_put_failure;
-
- if (swkey->tun_key.ipv4_dst &&
- ipv4_tun_to_nlattr(skb, &swkey->tun_key, &output->tun_key))
- goto nla_put_failure;
-
- if (swkey->phy.in_port != DP_MAX_PORTS) {
- /* Exact match upper 16 bits. */
- u16 upper_u16;
- upper_u16 = (swkey == output) ? 0 : 0xffff;
-
- if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT,
- (upper_u16 << 16) | output->phy.in_port))
- goto nla_put_failure;
- }
-
- if (swkey->phy.skb_mark &&
- nla_put_u32(skb, OVS_KEY_ATTR_SKB_MARK, output->phy.skb_mark))
- goto nla_put_failure;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_ETHERNET, sizeof(*eth_key));
- if (!nla)
- goto nla_put_failure;
-
- eth_key = nla_data(nla);
- memcpy(eth_key->eth_src, output->eth.src, ETH_ALEN);
- memcpy(eth_key->eth_dst, output->eth.dst, ETH_ALEN);
-
- if (swkey->eth.tci || swkey->eth.type == htons(ETH_P_8021Q)) {
- __be16 eth_type;
- eth_type = (swkey == output) ? htons(ETH_P_8021Q) : htons(0xffff) ;
- if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, eth_type) ||
- nla_put_be16(skb, OVS_KEY_ATTR_VLAN, output->eth.tci))
- goto nla_put_failure;
- encap = nla_nest_start(skb, OVS_KEY_ATTR_ENCAP);
- if (!swkey->eth.tci)
- goto unencap;
- } else
- encap = NULL;
-
- if ((swkey == output) && (swkey->eth.type == htons(ETH_P_802_2)))
- goto unencap;
-
- if (output->eth.type != 0)
- if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, output->eth.type))
- goto nla_put_failure;
-
- if (swkey->eth.type == htons(ETH_P_IP)) {
- struct ovs_key_ipv4 *ipv4_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_IPV4, sizeof(*ipv4_key));
- if (!nla)
- goto nla_put_failure;
- ipv4_key = nla_data(nla);
- ipv4_key->ipv4_src = output->ipv4.addr.src;
- ipv4_key->ipv4_dst = output->ipv4.addr.dst;
- ipv4_key->ipv4_proto = output->ip.proto;
- ipv4_key->ipv4_tos = output->ip.tos;
- ipv4_key->ipv4_ttl = output->ip.ttl;
- ipv4_key->ipv4_frag = output->ip.frag;
- } else if (swkey->eth.type == htons(ETH_P_IPV6)) {
- struct ovs_key_ipv6 *ipv6_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_IPV6, sizeof(*ipv6_key));
- if (!nla)
- goto nla_put_failure;
- ipv6_key = nla_data(nla);
- memcpy(ipv6_key->ipv6_src, &output->ipv6.addr.src,
- sizeof(ipv6_key->ipv6_src));
- memcpy(ipv6_key->ipv6_dst, &output->ipv6.addr.dst,
- sizeof(ipv6_key->ipv6_dst));
- ipv6_key->ipv6_label = output->ipv6.label;
- ipv6_key->ipv6_proto = output->ip.proto;
- ipv6_key->ipv6_tclass = output->ip.tos;
- ipv6_key->ipv6_hlimit = output->ip.ttl;
- ipv6_key->ipv6_frag = output->ip.frag;
- } else if (swkey->eth.type == htons(ETH_P_ARP) ||
- swkey->eth.type == htons(ETH_P_RARP)) {
- struct ovs_key_arp *arp_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_ARP, sizeof(*arp_key));
- if (!nla)
- goto nla_put_failure;
- arp_key = nla_data(nla);
- memset(arp_key, 0, sizeof(struct ovs_key_arp));
- arp_key->arp_sip = output->ipv4.addr.src;
- arp_key->arp_tip = output->ipv4.addr.dst;
- arp_key->arp_op = htons(output->ip.proto);
- memcpy(arp_key->arp_sha, output->ipv4.arp.sha, ETH_ALEN);
- memcpy(arp_key->arp_tha, output->ipv4.arp.tha, ETH_ALEN);
- }
-
- if ((swkey->eth.type == htons(ETH_P_IP) ||
- swkey->eth.type == htons(ETH_P_IPV6)) &&
- swkey->ip.frag != OVS_FRAG_TYPE_LATER) {
-
- if (swkey->ip.proto == IPPROTO_TCP) {
- struct ovs_key_tcp *tcp_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_TCP, sizeof(*tcp_key));
- if (!nla)
- goto nla_put_failure;
- tcp_key = nla_data(nla);
- if (swkey->eth.type == htons(ETH_P_IP)) {
- tcp_key->tcp_src = output->ipv4.tp.src;
- tcp_key->tcp_dst = output->ipv4.tp.dst;
- } else if (swkey->eth.type == htons(ETH_P_IPV6)) {
- tcp_key->tcp_src = output->ipv6.tp.src;
- tcp_key->tcp_dst = output->ipv6.tp.dst;
- }
- } else if (swkey->ip.proto == IPPROTO_UDP) {
- struct ovs_key_udp *udp_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_UDP, sizeof(*udp_key));
- if (!nla)
- goto nla_put_failure;
- udp_key = nla_data(nla);
- if (swkey->eth.type == htons(ETH_P_IP)) {
- udp_key->udp_src = output->ipv4.tp.src;
- udp_key->udp_dst = output->ipv4.tp.dst;
- } else if (swkey->eth.type == htons(ETH_P_IPV6)) {
- udp_key->udp_src = output->ipv6.tp.src;
- udp_key->udp_dst = output->ipv6.tp.dst;
- }
- } else if (swkey->eth.type == htons(ETH_P_IP) &&
- swkey->ip.proto == IPPROTO_ICMP) {
- struct ovs_key_icmp *icmp_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_ICMP, sizeof(*icmp_key));
- if (!nla)
- goto nla_put_failure;
- icmp_key = nla_data(nla);
- icmp_key->icmp_type = ntohs(output->ipv4.tp.src);
- icmp_key->icmp_code = ntohs(output->ipv4.tp.dst);
- } else if (swkey->eth.type == htons(ETH_P_IPV6) &&
- swkey->ip.proto == IPPROTO_ICMPV6) {
- struct ovs_key_icmpv6 *icmpv6_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_ICMPV6,
- sizeof(*icmpv6_key));
- if (!nla)
- goto nla_put_failure;
- icmpv6_key = nla_data(nla);
- icmpv6_key->icmpv6_type = ntohs(output->ipv6.tp.src);
- icmpv6_key->icmpv6_code = ntohs(output->ipv6.tp.dst);
-
- if (icmpv6_key->icmpv6_type == NDISC_NEIGHBOUR_SOLICITATION ||
- icmpv6_key->icmpv6_type == NDISC_NEIGHBOUR_ADVERTISEMENT) {
- struct ovs_key_nd *nd_key;
-
- nla = nla_reserve(skb, OVS_KEY_ATTR_ND, sizeof(*nd_key));
- if (!nla)
- goto nla_put_failure;
- nd_key = nla_data(nla);
- memcpy(nd_key->nd_target, &output->ipv6.nd.target,
- sizeof(nd_key->nd_target));
- memcpy(nd_key->nd_sll, output->ipv6.nd.sll, ETH_ALEN);
- memcpy(nd_key->nd_tll, output->ipv6.nd.tll, ETH_ALEN);
- }
- }
- }
-
-unencap:
- if (encap)
- nla_nest_end(skb, encap);
-
- return 0;
-
-nla_put_failure:
- return -EMSGSIZE;
-}
-
-/* Initializes the flow module.
- * Returns zero if successful or a negative error code. */
-int ovs_flow_init(void)
-{
- flow_cache = kmem_cache_create("sw_flow", sizeof(struct sw_flow), 0,
- 0, NULL);
- if (flow_cache == NULL)
- return -ENOMEM;
-
- return 0;
-}
-
-/* Uninitializes the flow module. */
-void ovs_flow_exit(void)
-{
- kmem_cache_destroy(flow_cache);
-}
-
-struct sw_flow_mask *ovs_sw_flow_mask_alloc(void)
-{
- struct sw_flow_mask *mask;
-
- mask = kmalloc(sizeof(*mask), GFP_KERNEL);
- if (mask)
- mask->ref_count = 0;
-
- return mask;
-}
-
-void ovs_sw_flow_mask_add_ref(struct sw_flow_mask *mask)
-{
- mask->ref_count++;
-}
-
-static void rcu_free_sw_flow_mask_cb(struct rcu_head *rcu)
-{
- struct sw_flow_mask *mask = container_of(rcu, struct sw_flow_mask, rcu);
-
- kfree(mask);
-}
-
-void ovs_sw_flow_mask_del_ref(struct sw_flow_mask *mask, bool deferred)
-{
- if (!mask)
- return;
-
- BUG_ON(!mask->ref_count);
- mask->ref_count--;
-
- if (!mask->ref_count) {
- list_del_rcu(&mask->list);
- if (deferred)
- call_rcu(&mask->rcu, rcu_free_sw_flow_mask_cb);
- else
- kfree(mask);
- }
-}
-
-static bool ovs_sw_flow_mask_equal(const struct sw_flow_mask *a,
- const struct sw_flow_mask *b)
-{
- u8 *a_ = (u8 *)&a->key + a->range.start;
- u8 *b_ = (u8 *)&b->key + b->range.start;
-
- return (a->range.end == b->range.end)
- && (a->range.start == b->range.start)
- && (memcmp(a_, b_, ovs_sw_flow_mask_actual_size(a)) == 0);
-}
-
-struct sw_flow_mask *ovs_sw_flow_mask_find(const struct flow_table *tbl,
- const struct sw_flow_mask *mask)
-{
- struct list_head *ml;
-
- list_for_each(ml, &tbl->mask_list) {
- struct sw_flow_mask *m;
- m = container_of(ml, struct sw_flow_mask, list);
- if (ovs_sw_flow_mask_equal(mask, m))
- return m;
- }
-
- return NULL;
-}
-
-/**
- * add a new mask into the mask list.
- * The caller needs to make sure that 'mask' is not the same
- * as any masks that are already on the list.
- */
-void ovs_sw_flow_mask_insert(struct flow_table *tbl, struct sw_flow_mask *mask)
-{
- list_add_rcu(&mask->list, &tbl->mask_list);
-}
-
-/**
- * Set 'range' fields in the mask to the value of 'val'.
- */
-static void ovs_sw_flow_mask_set(struct sw_flow_mask *mask,
- struct sw_flow_key_range *range, u8 val)
-{
- u8 *m = (u8 *)&mask->key + range->start;
-
- mask->range = *range;
- memset(m, val, ovs_sw_flow_mask_size_roundup(mask));
-}
git://git.onelab.eu / sliver-openvswitch.git / blobdiff