diff options
author | David S. Miller <davem@davemloft.net> | 2013-11-04 16:25:04 -0500 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2013-11-04 16:25:04 -0500 |
commit | 6fcf018ae4491dc11b080892fa9f3dbd928fdbb9 (patch) | |
tree | f28592676e295183581d59149fd1eac913b229a2 /net/openvswitch | |
parent | 5a6e55c461db3364aa5be919101db972bc859133 (diff) | |
parent | 8ddd094675cfd453fc9838caa46ea108a4107183 (diff) |
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch
Jesse Gross says:
====================
Open vSwitch
A set of updates for net-next/3.13. Major changes are:
* Restructure flow handling code to be more logically organized and
easier to read.
* Rehashing of the flow table is moved from a workqueue to flow
installation time. Before, heavy load could block the workqueue for
excessive periods of time.
* Additional debugging information is provided to help diagnose megaflows.
* It's now possible to match on TCP flags.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/openvswitch')
-rw-r--r-- | net/openvswitch/Makefile | 2 | ||||
-rw-r--r-- | net/openvswitch/datapath.c | 668 | ||||
-rw-r--r-- | net/openvswitch/datapath.h | 9 | ||||
-rw-r--r-- | net/openvswitch/flow.c | 1605 | ||||
-rw-r--r-- | net/openvswitch/flow.h | 132 | ||||
-rw-r--r-- | net/openvswitch/flow_netlink.c | 1630 | ||||
-rw-r--r-- | net/openvswitch/flow_netlink.h | 60 | ||||
-rw-r--r-- | net/openvswitch/flow_table.c | 592 | ||||
-rw-r--r-- | net/openvswitch/flow_table.h | 81 | ||||
-rw-r--r-- | net/openvswitch/vport-gre.c | 2 | ||||
-rw-r--r-- | net/openvswitch/vport-internal_dev.c | 2 | ||||
-rw-r--r-- | net/openvswitch/vport-vxlan.c | 1 |
12 files changed, 2496 insertions, 2288 deletions
diff --git a/net/openvswitch/Makefile b/net/openvswitch/Makefile index ea36e99089a..3591cb5dae9 100644 --- a/net/openvswitch/Makefile +++ b/net/openvswitch/Makefile @@ -9,6 +9,8 @@ openvswitch-y := \ datapath.o \ dp_notify.o \ flow.o \ + flow_netlink.o \ + flow_table.o \ vport.o \ vport-internal_dev.o \ vport-netdev.o diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 2aa13bd7f2b..1408adc2a2a 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -55,14 +55,10 @@ #include "datapath.h" #include "flow.h" +#include "flow_netlink.h" #include "vport-internal_dev.h" #include "vport-netdev.h" - -#define REHASH_FLOW_INTERVAL (10 * 60 * HZ) -static void rehash_flow_table(struct work_struct *work); -static DECLARE_DELAYED_WORK(rehash_flow_wq, rehash_flow_table); - int ovs_net_id __read_mostly; static void ovs_notify(struct sk_buff *skb, struct genl_info *info, @@ -165,7 +161,7 @@ static void destroy_dp_rcu(struct rcu_head *rcu) { struct datapath *dp = container_of(rcu, struct datapath, rcu); - ovs_flow_tbl_destroy((__force struct flow_table *)dp->table, false); + ovs_flow_tbl_destroy(&dp->table); free_percpu(dp->stats_percpu); release_net(ovs_dp_get_net(dp)); kfree(dp->ports); @@ -225,6 +221,7 @@ void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb) struct dp_stats_percpu *stats; struct sw_flow_key key; u64 *stats_counter; + u32 n_mask_hit; int error; stats = this_cpu_ptr(dp->stats_percpu); @@ -237,7 +234,7 @@ void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb) } /* Look up flow. */ - flow = ovs_flow_lookup(rcu_dereference(dp->table), &key); + flow = ovs_flow_tbl_lookup(&dp->table, &key, &n_mask_hit); if (unlikely(!flow)) { struct dp_upcall_info upcall; @@ -262,6 +259,7 @@ out: /* Update datapath statistics. */ u64_stats_update_begin(&stats->sync); (*stats_counter)++; + stats->n_mask_hit += n_mask_hit; u64_stats_update_end(&stats->sync); } @@ -435,7 +433,7 @@ static int queue_userspace_packet(struct net *net, int dp_ifindex, upcall->dp_ifindex = dp_ifindex; nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY); - ovs_flow_to_nlattrs(upcall_info->key, upcall_info->key, user_skb); + ovs_nla_put_flow(upcall_info->key, upcall_info->key, user_skb); nla_nest_end(user_skb, nla); if (upcall_info->userdata) @@ -455,398 +453,6 @@ out: return err; } -/* Called with ovs_mutex. */ -static int flush_flows(struct datapath *dp) -{ - struct flow_table *old_table; - struct flow_table *new_table; - - old_table = ovsl_dereference(dp->table); - new_table = ovs_flow_tbl_alloc(TBL_MIN_BUCKETS); - if (!new_table) - return -ENOMEM; - - rcu_assign_pointer(dp->table, new_table); - - ovs_flow_tbl_destroy(old_table, true); - return 0; -} - -static struct nlattr *reserve_sfa_size(struct sw_flow_actions **sfa, int attr_len) -{ - - struct sw_flow_actions *acts; - int new_acts_size; - int req_size = NLA_ALIGN(attr_len); - int next_offset = offsetof(struct sw_flow_actions, actions) + - (*sfa)->actions_len; - - if (req_size <= (ksize(*sfa) - next_offset)) - goto out; - - new_acts_size = ksize(*sfa) * 2; - - if (new_acts_size > MAX_ACTIONS_BUFSIZE) { - if ((MAX_ACTIONS_BUFSIZE - next_offset) < req_size) - return ERR_PTR(-EMSGSIZE); - new_acts_size = MAX_ACTIONS_BUFSIZE; - } - - acts = ovs_flow_actions_alloc(new_acts_size); - if (IS_ERR(acts)) - return (void *)acts; - - memcpy(acts->actions, (*sfa)->actions, (*sfa)->actions_len); - acts->actions_len = (*sfa)->actions_len; - kfree(*sfa); - *sfa = acts; - -out: - (*sfa)->actions_len += req_size; - return (struct nlattr *) ((unsigned char *)(*sfa) + next_offset); -} - -static int add_action(struct sw_flow_actions **sfa, int attrtype, void *data, int len) -{ - struct nlattr *a; - - a = reserve_sfa_size(sfa, nla_attr_size(len)); - if (IS_ERR(a)) - return PTR_ERR(a); - - a->nla_type = attrtype; - a->nla_len = nla_attr_size(len); - - if (data) - memcpy(nla_data(a), data, len); - memset((unsigned char *) a + a->nla_len, 0, nla_padlen(len)); - - return 0; -} - -static inline int add_nested_action_start(struct sw_flow_actions **sfa, int attrtype) -{ - int used = (*sfa)->actions_len; - int err; - - err = add_action(sfa, attrtype, NULL, 0); - if (err) - return err; - - return used; -} - -static inline void add_nested_action_end(struct sw_flow_actions *sfa, int st_offset) -{ - struct nlattr *a = (struct nlattr *) ((unsigned char *)sfa->actions + st_offset); - - a->nla_len = sfa->actions_len - st_offset; -} - -static int validate_and_copy_actions(const struct nlattr *attr, - const struct sw_flow_key *key, int depth, - struct sw_flow_actions **sfa); - -static int validate_and_copy_sample(const struct nlattr *attr, - const struct sw_flow_key *key, int depth, - struct sw_flow_actions **sfa) -{ - const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1]; - const struct nlattr *probability, *actions; - const struct nlattr *a; - int rem, start, err, st_acts; - - memset(attrs, 0, sizeof(attrs)); - nla_for_each_nested(a, attr, rem) { - int type = nla_type(a); - if (!type || type > OVS_SAMPLE_ATTR_MAX || attrs[type]) - return -EINVAL; - attrs[type] = a; - } - if (rem) - return -EINVAL; - - probability = attrs[OVS_SAMPLE_ATTR_PROBABILITY]; - if (!probability || nla_len(probability) != sizeof(u32)) - return -EINVAL; - - actions = attrs[OVS_SAMPLE_ATTR_ACTIONS]; - if (!actions || (nla_len(actions) && nla_len(actions) < NLA_HDRLEN)) - return -EINVAL; - - /* validation done, copy sample action. */ - start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SAMPLE); - if (start < 0) - return start; - err = add_action(sfa, OVS_SAMPLE_ATTR_PROBABILITY, nla_data(probability), sizeof(u32)); - if (err) - return err; - st_acts = add_nested_action_start(sfa, OVS_SAMPLE_ATTR_ACTIONS); - if (st_acts < 0) - return st_acts; - - err = validate_and_copy_actions(actions, key, depth + 1, sfa); - if (err) - return err; - - add_nested_action_end(*sfa, st_acts); - add_nested_action_end(*sfa, start); - - return 0; -} - -static int validate_tp_port(const struct sw_flow_key *flow_key) -{ - if (flow_key->eth.type == htons(ETH_P_IP)) { - if (flow_key->ipv4.tp.src || flow_key->ipv4.tp.dst) - return 0; - } else if (flow_key->eth.type == htons(ETH_P_IPV6)) { - if (flow_key->ipv6.tp.src || flow_key->ipv6.tp.dst) - return 0; - } - - return -EINVAL; -} - -static int validate_and_copy_set_tun(const struct nlattr *attr, - struct sw_flow_actions **sfa) -{ - struct sw_flow_match match; - struct sw_flow_key key; - int err, start; - - ovs_match_init(&match, &key, NULL); - err = ovs_ipv4_tun_from_nlattr(nla_data(attr), &match, false); - if (err) - return err; - - start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SET); - if (start < 0) - return start; - - err = add_action(sfa, OVS_KEY_ATTR_IPV4_TUNNEL, &match.key->tun_key, - sizeof(match.key->tun_key)); - add_nested_action_end(*sfa, start); - - return err; -} - -static int validate_set(const struct nlattr *a, - const struct sw_flow_key *flow_key, - struct sw_flow_actions **sfa, - bool *set_tun) -{ - const struct nlattr *ovs_key = nla_data(a); - int key_type = nla_type(ovs_key); - - /* There can be only one key in a action */ - if (nla_total_size(nla_len(ovs_key)) != nla_len(a)) - return -EINVAL; - - if (key_type > OVS_KEY_ATTR_MAX || - (ovs_key_lens[key_type] != nla_len(ovs_key) && - ovs_key_lens[key_type] != -1)) - return -EINVAL; - - switch (key_type) { - const struct ovs_key_ipv4 *ipv4_key; - const struct ovs_key_ipv6 *ipv6_key; - int err; - - case OVS_KEY_ATTR_PRIORITY: - case OVS_KEY_ATTR_SKB_MARK: - case OVS_KEY_ATTR_ETHERNET: - break; - - case OVS_KEY_ATTR_TUNNEL: - *set_tun = true; - err = validate_and_copy_set_tun(a, sfa); - if (err) - return err; - break; - - case OVS_KEY_ATTR_IPV4: - if (flow_key->eth.type != htons(ETH_P_IP)) - return -EINVAL; - - if (!flow_key->ip.proto) - return -EINVAL; - - ipv4_key = nla_data(ovs_key); - if (ipv4_key->ipv4_proto != flow_key->ip.proto) - return -EINVAL; - - if (ipv4_key->ipv4_frag != flow_key->ip.frag) - return -EINVAL; - - break; - - case OVS_KEY_ATTR_IPV6: - if (flow_key->eth.type != htons(ETH_P_IPV6)) - return -EINVAL; - - if (!flow_key->ip.proto) - return -EINVAL; - - ipv6_key = nla_data(ovs_key); - if (ipv6_key->ipv6_proto != flow_key->ip.proto) - return -EINVAL; - - if (ipv6_key->ipv6_frag != flow_key->ip.frag) - return -EINVAL; - - if (ntohl(ipv6_key->ipv6_label) & 0xFFF00000) - return -EINVAL; - - break; - - case OVS_KEY_ATTR_TCP: - if (flow_key->ip.proto != IPPROTO_TCP) - return -EINVAL; - - return validate_tp_port(flow_key); - - case OVS_KEY_ATTR_UDP: - if (flow_key->ip.proto != IPPROTO_UDP) - return -EINVAL; - - return validate_tp_port(flow_key); - - case OVS_KEY_ATTR_SCTP: - if (flow_key->ip.proto != IPPROTO_SCTP) - return -EINVAL; - - return validate_tp_port(flow_key); - - default: - return -EINVAL; - } - - return 0; -} - -static int validate_userspace(const struct nlattr *attr) -{ - static const struct nla_policy userspace_policy[OVS_USERSPACE_ATTR_MAX + 1] = { - [OVS_USERSPACE_ATTR_PID] = {.type = NLA_U32 }, - [OVS_USERSPACE_ATTR_USERDATA] = {.type = NLA_UNSPEC }, - }; - struct nlattr *a[OVS_USERSPACE_ATTR_MAX + 1]; - int error; - - error = nla_parse_nested(a, OVS_USERSPACE_ATTR_MAX, - attr, userspace_policy); - if (error) - return error; - - if (!a[OVS_USERSPACE_ATTR_PID] || - !nla_get_u32(a[OVS_USERSPACE_ATTR_PID])) - return -EINVAL; - - return 0; -} - -static int copy_action(const struct nlattr *from, - struct sw_flow_actions **sfa) -{ - int totlen = NLA_ALIGN(from->nla_len); - struct nlattr *to; - - to = reserve_sfa_size(sfa, from->nla_len); - if (IS_ERR(to)) - return PTR_ERR(to); - - memcpy(to, from, totlen); - return 0; -} - -static int validate_and_copy_actions(const struct nlattr *attr, - const struct sw_flow_key *key, - int depth, - struct sw_flow_actions **sfa) -{ - const struct nlattr *a; - int rem, err; - - if (depth >= SAMPLE_ACTION_DEPTH) - return -EOVERFLOW; - - nla_for_each_nested(a, attr, rem) { - /* Expected argument lengths, (u32)-1 for variable length. */ - static const u32 action_lens[OVS_ACTION_ATTR_MAX + 1] = { - [OVS_ACTION_ATTR_OUTPUT] = sizeof(u32), - [OVS_ACTION_ATTR_USERSPACE] = (u32)-1, - [OVS_ACTION_ATTR_PUSH_VLAN] = sizeof(struct ovs_action_push_vlan), - [OVS_ACTION_ATTR_POP_VLAN] = 0, - [OVS_ACTION_ATTR_SET] = (u32)-1, - [OVS_ACTION_ATTR_SAMPLE] = (u32)-1 - }; - const struct ovs_action_push_vlan *vlan; - int type = nla_type(a); - bool skip_copy; - - if (type > OVS_ACTION_ATTR_MAX || - (action_lens[type] != nla_len(a) && - action_lens[type] != (u32)-1)) - return -EINVAL; - - skip_copy = false; - switch (type) { - case OVS_ACTION_ATTR_UNSPEC: - return -EINVAL; - - case OVS_ACTION_ATTR_USERSPACE: - err = validate_userspace(a); - if (err) - return err; - break; - - case OVS_ACTION_ATTR_OUTPUT: - if (nla_get_u32(a) >= DP_MAX_PORTS) - return -EINVAL; - break; - - - case OVS_ACTION_ATTR_POP_VLAN: - break; - - case OVS_ACTION_ATTR_PUSH_VLAN: - vlan = nla_data(a); - if (vlan->vlan_tpid != htons(ETH_P_8021Q)) - return -EINVAL; - if (!(vlan->vlan_tci & htons(VLAN_TAG_PRESENT))) - return -EINVAL; - break; - - case OVS_ACTION_ATTR_SET: - err = validate_set(a, key, sfa, &skip_copy); - if (err) - return err; - break; - - case OVS_ACTION_ATTR_SAMPLE: - err = validate_and_copy_sample(a, key, depth, sfa); - if (err) - return err; - skip_copy = true; - break; - - default: - return -EINVAL; - } - if (!skip_copy) { - err = copy_action(a, sfa); - if (err) - return err; - } - } - - if (rem > 0) - return -EINVAL; - - return 0; -} - static void clear_stats(struct sw_flow *flow) { flow->used = 0; @@ -902,15 +508,16 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) if (err) goto err_flow_free; - err = ovs_flow_metadata_from_nlattrs(flow, a[OVS_PACKET_ATTR_KEY]); + err = ovs_nla_get_flow_metadata(flow, a[OVS_PACKET_ATTR_KEY]); if (err) goto err_flow_free; - acts = ovs_flow_actions_alloc(nla_len(a[OVS_PACKET_ATTR_ACTIONS])); + acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_PACKET_ATTR_ACTIONS])); err = PTR_ERR(acts); if (IS_ERR(acts)) goto err_flow_free; - err = validate_and_copy_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0, &acts); + err = ovs_nla_copy_actions(a[OVS_PACKET_ATTR_ACTIONS], + &flow->key, 0, &acts); rcu_assign_pointer(flow->sf_acts, acts); if (err) goto err_flow_free; @@ -958,15 +565,18 @@ static struct genl_ops dp_packet_genl_ops[] = { } }; -static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats) +static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats, + struct ovs_dp_megaflow_stats *mega_stats) { - struct flow_table *table; int i; - table = rcu_dereference_check(dp->table, lockdep_ovsl_is_held()); - stats->n_flows = ovs_flow_tbl_count(table); + memset(mega_stats, 0, sizeof(*mega_stats)); + + stats->n_flows = ovs_flow_tbl_count(&dp->table); + mega_stats->n_masks = ovs_flow_tbl_num_masks(&dp->table); stats->n_hit = stats->n_missed = stats->n_lost = 0; + for_each_possible_cpu(i) { const struct dp_stats_percpu *percpu_stats; struct dp_stats_percpu local_stats; @@ -982,6 +592,7 @@ static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats) stats->n_hit += local_stats.n_hit; stats->n_missed += local_stats.n_missed; stats->n_lost += local_stats.n_lost; + mega_stats->n_mask_hit += local_stats.n_mask_hit; } } @@ -1005,100 +616,6 @@ static struct genl_multicast_group ovs_dp_flow_multicast_group = { .name = OVS_FLOW_MCGROUP }; -static int actions_to_attr(const struct nlattr *attr, int len, struct sk_buff *skb); -static int sample_action_to_attr(const struct nlattr *attr, struct sk_buff *skb) -{ - const struct nlattr *a; - struct nlattr *start; - int err = 0, rem; - - start = nla_nest_start(skb, OVS_ACTION_ATTR_SAMPLE); - if (!start) - return -EMSGSIZE; - - nla_for_each_nested(a, attr, rem) { - int type = nla_type(a); - struct nlattr *st_sample; - - switch (type) { - case OVS_SAMPLE_ATTR_PROBABILITY: - if (nla_put(skb, OVS_SAMPLE_ATTR_PROBABILITY, sizeof(u32), nla_data(a))) - return -EMSGSIZE; - break; - case OVS_SAMPLE_ATTR_ACTIONS: - st_sample = nla_nest_start(skb, OVS_SAMPLE_ATTR_ACTIONS); - if (!st_sample) - return -EMSGSIZE; - err = actions_to_attr(nla_data(a), nla_len(a), skb); - if (err) - return err; - nla_nest_end(skb, st_sample); - break; - } - } - - nla_nest_end(skb, start); - return err; -} - -static int set_action_to_attr(const struct nlattr *a, struct sk_buff *skb) -{ - const struct nlattr *ovs_key = nla_data(a); - int key_type = nla_type(ovs_key); - struct nlattr *start; - int err; - - switch (key_type) { - case OVS_KEY_ATTR_IPV4_TUNNEL: - start = nla_nest_start(skb, OVS_ACTION_ATTR_SET); - if (!start) - return -EMSGSIZE; - - err = ovs_ipv4_tun_to_nlattr(skb, nla_data(ovs_key), - nla_data(ovs_key)); - if (err) - return err; - nla_nest_end(skb, start); - break; - default: - if (nla_put(skb, OVS_ACTION_ATTR_SET, nla_len(a), ovs_key)) - return -EMSGSIZE; - break; - } - - return 0; -} - -static int actions_to_attr(const struct nlattr *attr, int len, struct sk_buff *skb) -{ - const struct nlattr *a; - int rem, err; - - nla_for_each_attr(a, attr, len, rem) { - int type = nla_type(a); - - switch (type) { - case OVS_ACTION_ATTR_SET: - err = set_action_to_attr(a, skb); - if (err) - return err; - break; - - case OVS_ACTION_ATTR_SAMPLE: - err = sample_action_to_attr(a, skb); - if (err) - return err; - break; - default: - if (nla_put(skb, type, nla_len(a), nla_data(a))) - return -EMSGSIZE; - break; - } - } - - return 0; -} - static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts) { return NLMSG_ALIGN(sizeof(struct ovs_header)) @@ -1135,8 +652,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, if (!nla) goto nla_put_failure; - err = ovs_flow_to_nlattrs(&flow->unmasked_key, - &flow->unmasked_key, skb); + err = ovs_nla_put_flow(&flow->unmasked_key, &flow->unmasked_key, skb); if (err) goto error; nla_nest_end(skb, nla); @@ -1145,7 +661,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, if (!nla) goto nla_put_failure; - err = ovs_flow_to_nlattrs(&flow->key, &flow->mask->key, skb); + err = ovs_nla_put_flow(&flow->key, &flow->mask->key, skb); if (err) goto error; @@ -1155,7 +671,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, used = flow->used; stats.n_packets = flow->packet_count; stats.n_bytes = flow->byte_count; - tcp_flags = flow->tcp_flags; + tcp_flags = (u8)ntohs(flow->tcp_flags); spin_unlock_bh(&flow->lock); if (used && @@ -1188,7 +704,8 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, sf_acts = rcu_dereference_check(flow->sf_acts, lockdep_ovsl_is_held()); - err = actions_to_attr(sf_acts->actions, sf_acts->actions_len, skb); + err = ovs_nla_put_actions(sf_acts->actions, + sf_acts->actions_len, skb); if (!err) nla_nest_end(skb, start); else { @@ -1234,6 +751,14 @@ static struct sk_buff *ovs_flow_cmd_build_info(struct sw_flow *flow, return skb; } +static struct sw_flow *__ovs_flow_tbl_lookup(struct flow_table *tbl, + const struct sw_flow_key *key) +{ + u32 __always_unused n_mask_hit; + + return ovs_flow_tbl_lookup(tbl, key, &n_mask_hit); +} + static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) { struct nlattr **a = info->attrs; @@ -1243,7 +768,6 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) struct sw_flow_mask mask; struct sk_buff *reply; struct datapath *dp; - struct flow_table *table; struct sw_flow_actions *acts = NULL; struct sw_flow_match match; int error; @@ -1254,21 +778,21 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) goto error; ovs_match_init(&match, &key, &mask); - error = ovs_match_from_nlattrs(&match, - a[OVS_FLOW_ATTR_KEY], a[OVS_FLOW_ATTR_MASK]); + error = ovs_nla_get_match(&match, + a[OVS_FLOW_ATTR_KEY], a[OVS_FLOW_ATTR_MASK]); if (error) goto error; /* Validate actions. */ if (a[OVS_FLOW_ATTR_ACTIONS]) { - acts = ovs_flow_actions_alloc(nla_len(a[OVS_FLOW_ATTR_ACTIONS])); + acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_FLOW_ATTR_ACTIONS])); error = PTR_ERR(acts); if (IS_ERR(acts)) goto error; - ovs_flow_key_mask(&masked_key, &key, &mask); - error = validate_and_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], - &masked_key, 0, &acts); + ovs_flow_mask_key(&masked_key, &key, &mask); + error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], + &masked_key, 0, &acts); if (error) { OVS_NLERR("Flow actions may not be safe on all matching packets.\n"); goto err_kfree; @@ -1284,29 +808,14 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) if (!dp) goto err_unlock_ovs; - table = ovsl_dereference(dp->table); - /* Check if this is a duplicate flow */ - flow = ovs_flow_lookup(table, &key); + flow = __ovs_flow_tbl_lookup(&dp->table, &key); if (!flow) { - struct sw_flow_mask *mask_p; /* Bail out if we're not allowed to create a new flow. */ error = -ENOENT; if (info->genlhdr->cmd == OVS_FLOW_CMD_SET) goto err_unlock_ovs; - /* Expand table, if necessary, to make room. */ - if (ovs_flow_tbl_need_to_expand(table)) { - struct flow_table *new_table; - - new_table = ovs_flow_tbl_expand(table); - if (!IS_ERR(new_table)) { - rcu_assign_pointer(dp->table, new_table); - ovs_flow_tbl_destroy(table, true); - table = ovsl_dereference(dp->table); - } - } - /* Allocate flow. */ flow = ovs_flow_alloc(); if (IS_ERR(flow)) { @@ -1317,25 +826,14 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) flow->key = masked_key; flow->unmasked_key = key; - - /* Make sure mask is unique in the system */ - mask_p = ovs_sw_flow_mask_find(table, &mask); - if (!mask_p) { - /* Allocate a new mask if none exsits. */ - mask_p = ovs_sw_flow_mask_alloc(); - if (!mask_p) - goto err_flow_free; - mask_p->key = mask.key; - mask_p->range = mask.range; - ovs_sw_flow_mask_insert(table, mask_p); - } - - ovs_sw_flow_mask_add_ref(mask_p); - flow->mask = mask_p; rcu_assign_pointer(flow->sf_acts, acts); /* Put flow in bucket. */ - ovs_flow_insert(table, flow); + error = ovs_flow_tbl_insert(&dp->table, flow, &mask); + if (error) { + acts = NULL; + goto err_flow_free; + } reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, info->snd_seq, OVS_FLOW_CMD_NEW); @@ -1356,7 +854,7 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) /* The unmasked key has to be the same for flow updates. */ error = -EINVAL; - if (!ovs_flow_cmp_unmasked_key(flow, &key, match.range.end)) { + if (!ovs_flow_cmp_unmasked_key(flow, &match)) { OVS_NLERR("Flow modification message rejected, unmasked key does not match.\n"); goto err_unlock_ovs; } @@ -1364,7 +862,7 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) /* Update actions. */ old_acts = ovsl_dereference(flow->sf_acts); rcu_assign_pointer(flow->sf_acts, acts); - ovs_flow_deferred_free_acts(old_acts); + ovs_nla_free_flow_actions(old_acts); reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, info->snd_seq, OVS_FLOW_CMD_NEW); @@ -1403,7 +901,6 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info) struct sk_buff *reply; struct sw_flow *flow; struct datapath *dp; - struct flow_table *table; struct sw_flow_match match; int err; @@ -1413,7 +910,7 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info) } ovs_match_init(&match, &key, NULL); - err = ovs_match_from_nlattrs(&match, a[OVS_FLOW_ATTR_KEY], NULL); + err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL); if (err) return err; @@ -1424,9 +921,8 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info) goto unlock; } - table = ovsl_dereference(dp->table); - flow = ovs_flow_lookup_unmasked_key(table, &match); - if (!flow) { + flow = __ovs_flow_tbl_lookup(&dp->table, &key); + if (!flow || !ovs_flow_cmp_unmasked_key(flow, &match)) { err = -ENOENT; goto unlock; } @@ -1453,7 +949,6 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info) struct sk_buff *reply; struct sw_flow *flow; struct datapath *dp; - struct flow_table *table; struct sw_flow_match match; int err; @@ -1465,18 +960,17 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info) } if (!a[OVS_FLOW_ATTR_KEY]) { - err = flush_flows(dp); + err = ovs_flow_tbl_flush(&dp->table); goto unlock; } ovs_match_init(&match, &key, NULL); - err = ovs_match_from_nlattrs(&match, a[OVS_FLOW_ATTR_KEY], NULL); + err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL); if (err) goto unlock; - table = ovsl_dereference(dp->table); - flow = ovs_flow_lookup_unmasked_key(table, &match); - if (!flow) { + flow = __ovs_flow_tbl_lookup(&dp->table, &key); + if (!flow || !ovs_flow_cmp_unmasked_key(flow, &match)) { err = -ENOENT; goto unlock; } @@ -1487,7 +981,7 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info) goto unlock; } - ovs_flow_remove(table, flow); + ovs_flow_tbl_remove(&dp->table, flow); err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_portid, info->snd_seq, 0, OVS_FLOW_CMD_DEL); @@ -1506,8 +1000,8 @@ unlock: static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) { struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh)); + struct table_instance *ti; struct datapath *dp; - struct flow_table *table; rcu_read_lock(); dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); @@ -1516,14 +1010,14 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) return -ENODEV; } - table = rcu_dereference(dp->table); + ti = rcu_dereference(dp->table.ti); for (;;) { struct sw_flow *flow; u32 bucket, obj; bucket = cb->args[0]; obj = cb->args[1]; - flow = ovs_flow_dump_next(table, &bucket, &obj); + flow = ovs_flow_tbl_dump_next(ti, &bucket, &obj); if (!flow) break; @@ -1589,6 +1083,7 @@ static size_t ovs_dp_cmd_msg_size(void) msgsize += nla_total_size(IFNAMSIZ); msgsize += nla_total_size(sizeof(struct ovs_dp_stats)); + msgsize += nla_total_size(sizeof(struct ovs_dp_megaflow_stats)); return msgsize; } @@ -1598,6 +1093,7 @@ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb, { struct ovs_header *ovs_header; struct ovs_dp_stats dp_stats; + struct ovs_dp_megaflow_stats dp_megaflow_stats; int err; ovs_header = genlmsg_put(skb, portid, seq, &dp_datapath_genl_family, @@ -1613,8 +1109,14 @@ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb, if (err) goto nla_put_failure; - get_dp_stats(dp, &dp_stats); - if (nla_put(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats), &dp_stats)) + get_dp_stats(dp, &dp_stats, &dp_megaflow_stats); + if (nla_put(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats), + &dp_stats)) + goto nla_put_failure; + + if (nla_put(skb, OVS_DP_ATTR_MEGAFLOW_STATS, + sizeof(struct ovs_dp_megaflow_stats), + &dp_megaflow_stats)) goto nla_put_failure; return genlmsg_end(skb, ovs_header); @@ -1687,9 +1189,8 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) ovs_dp_set_net(dp, hold_net(sock_net(skb->sk))); /* Allocate table. */ - err = -ENOMEM; - rcu_assign_pointer(dp->table, ovs_flow_tbl_alloc(TBL_MIN_BUCKETS)); - if (!dp->table) + err = ovs_flow_tbl_init(&dp->table); + if (err) goto err_free_dp; dp->stats_percpu = alloc_percpu(struct dp_stats_percpu); @@ -1699,7 +1200,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) } dp->ports = kmalloc(DP_VPORT_HASH_BUCKETS * sizeof(struct hlist_head), - GFP_KERNEL); + GFP_KERNEL); if (!dp->ports) { err = -ENOMEM; goto err_destroy_percpu; @@ -1746,7 +1247,7 @@ err_destroy_ports_array: err_destroy_percpu: free_percpu(dp->stats_percpu); err_destroy_table: - ovs_flow_tbl_destroy(ovsl_dereference(dp->table), false); + ovs_flow_tbl_destroy(&dp->table); err_free_dp: release_net(ovs_dp_get_net(dp)); kfree(dp); @@ -2336,32 +1837,6 @@ error: return err; } -static void rehash_flow_table(struct work_struct *work) -{ - struct datapath *dp; - struct net *net; - - ovs_lock(); - rtnl_lock(); - for_each_net(net) { - struct ovs_net *ovs_net = net_generic(net, ovs_net_id); - - list_for_each_entry(dp, &ovs_net->dps, list_node) { - struct flow_table *old_table = ovsl_dereference(dp->table); - struct flow_table *new_table; - - new_table = ovs_flow_tbl_rehash(old_table); - if (!IS_ERR(new_table)) { - rcu_assign_pointer(dp->table, new_table); - ovs_flow_tbl_destroy(old_table, true); - } - } - } - rtnl_unlock(); - ovs_unlock(); - schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL); -} - static int __net_init ovs_init_net(struct net *net) { struct ovs_net *ovs_net = net_generic(net, ovs_net_id); @@ -2419,8 +1894,6 @@ static int __init dp_init(void) if (err < 0) goto error_unreg_notifier; - schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL); - return 0; error_unreg_notifier: @@ -2437,7 +1910,6 @@ error: static void dp_cleanup(void) { - cancel_delayed_work_sync(&rehash_flow_wq); dp_unregister_genl(ARRAY_SIZE(dp_genl_families)); unregister_netdevice_notifier(&ovs_dp_device_notifier); unregister_pernet_device(&ovs_net_ops); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 4d109c176ef..d3d14a58aa9 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -27,6 +27,7 @@ #include <linux/u64_stats_sync.h> #include "flow.h" +#include "flow_table.h" #include "vport.h" #define DP_MAX_PORTS USHRT_MAX @@ -45,11 +46,15 @@ * @n_lost: Number of received packets that had no matching flow in the flow * table that could not be sent to userspace (normally due to an overflow in * one of the datapath's queues). + * @n_mask_hit: Number of masks looked up for flow match. + * @n_mask_hit / (@n_hit + @n_missed) will be the average masks looked + * up per packet. */ struct dp_stats_percpu { u64 n_hit; u64 n_missed; u64 n_lost; + u64 n_mask_hit; struct u64_stats_sync sync; }; @@ -57,7 +62,7 @@ struct dp_stats_percpu { * struct datapath - datapath for flow-based packet switching * @rcu: RCU callback head for deferred destruction. * @list_node: Element in global 'dps' list. - * @table: Current flow table. Protected by ovs_mutex and RCU. + * @table: flow table. * @ports: Hash table for ports. %OVSP_LOCAL port always exists. Protected by * ovs_mutex and RCU. * @stats_percpu: Per-CPU datapath statistics. @@ -71,7 +76,7 @@ struct datapath { struct list_head list_node; /* Flow table. */ - struct flow_table __rcu *table; + struct flow_table table; /* Switch ports. */ struct hlist_head *ports; diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 410db90db73..b409f527960 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -45,202 +45,38 @@ #include <net/ipv6.h> #include <net/ndisc.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 = rounddown(offset, sizeof(long)); - size_t end = roundup(offset + size, sizeof(long)); - - 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; - } - - if (range->start > start) - range->start = start; + struct timespec cur_ts; + u64 cur_ms, idle_ms; - if (range->end < end) - range->end = end; -} + 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; -#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) { \ - if ((match)->mask) \ - (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) { \ - if ((match)->mask) \ - memcpy(&(match)->mask->key.field, value_p, len);\ - } else { \ - memcpy(&(match)->key->field, value_p, len); \ - } \ - } while (0) - -static u16 range_n_bytes(const struct sw_flow_key_range *range) -{ - return range->end - range->start; + return cur_ms - idle_ms; } -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; +#define TCP_FLAGS_BE16(tp) (*(__be16 *)&tcp_flag_word(tp) & htons(0x0FFF)) - memset(key, 0, sizeof(*key)); - - 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_used(struct sw_flow *flow, struct sk_buff *skb) { - u64 key_expected = 1 << 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 &= ~((1 << OVS_KEY_ATTR_IPV4) - | (1 << OVS_KEY_ATTR_IPV6) - | (1 << OVS_KEY_ATTR_TCP) - | (1 << OVS_KEY_ATTR_UDP) - | (1 << OVS_KEY_ATTR_SCTP) - | (1 << OVS_KEY_ATTR_ICMP) - | (1 << OVS_KEY_ATTR_ICMPV6) - | (1 << OVS_KEY_ATTR_ARP) - | (1 << OVS_KEY_ATTR_ND)); - - /* Always allowed mask fields. */ - mask_allowed |= ((1 << OVS_KEY_ATTR_TUNNEL) - | (1 << OVS_KEY_ATTR_IN_PORT) - | (1 << 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 |= 1 << OVS_KEY_ATTR_ARP; - if (match->mask && (match->mask->key.eth.type == htons(0xffff))) - mask_allowed |= 1 << OVS_KEY_ATTR_ARP; - } + __be16 tcp_flags = 0; - if (match->key->eth.type == htons(ETH_P_IP)) { - key_expected |= 1 << OVS_KEY_ATTR_IPV4; - if (match->mask && (match->mask->key.eth.type == htons(0xffff))) - mask_allowed |= 1 << OVS_KEY_ATTR_IPV4; - - if (match->key->ip.frag != OVS_FRAG_TYPE_LATER) { - if (match->key->ip.proto == IPPROTO_UDP) { - key_expected |= 1 << OVS_KEY_ATTR_UDP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_UDP; - } - - if (match->key->ip.proto == IPPROTO_SCTP) { - key_expected |= 1 << OVS_KEY_ATTR_SCTP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_SCTP; - } - - if (match->key->ip.proto == IPPROTO_TCP) { - key_expected |= 1 << OVS_KEY_ATTR_TCP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_TCP; - } - - if (match->key->ip.proto == IPPROTO_ICMP) { - key_expected |= 1 << OVS_KEY_ATTR_ICMP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_ICMP; - } - } - } - - if (match->key->eth.type == htons(ETH_P_IPV6)) { - key_expected |= 1 << OVS_KEY_ATTR_IPV6; - if (match->mask && (match->mask->key.eth.type == htons(0xffff))) - mask_allowed |= 1 << OVS_KEY_ATTR_IPV6; - - if (match->key->ip.frag != OVS_FRAG_TYPE_LATER) { - if (match->key->ip.proto == IPPROTO_UDP) { - key_expected |= 1 << OVS_KEY_ATTR_UDP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_UDP; - } - - if (match->key->ip.proto == IPPROTO_SCTP) { - key_expected |= 1 << OVS_KEY_ATTR_SCTP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_SCTP; - } - - if (match->key->ip.proto == IPPROTO_TCP) { - key_expected |= 1 << OVS_KEY_ATTR_TCP; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << OVS_KEY_ATTR_TCP; - } - - if (match->key->ip.proto == IPPROTO_ICMPV6) { - key_expected |= 1 << OVS_KEY_ATTR_ICMPV6; - if (match->mask && (match->mask->key.ip.proto == 0xff)) - mask_allowed |= 1 << 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 |= 1 << OVS_KEY_ATTR_ND; - if (match->mask && (match->mask->key.ipv6.tp.src == htons(0xffff))) - mask_allowed |= 1 << OVS_KEY_ATTR_ND; - } - } - } - } - - if ((key_attrs & key_expected) != key_expected) { - /* Key attributes check failed. */ - OVS_NLERR("Missing expected key attributes (key_attrs=%llx, expected=%llx).\n", - key_attrs, key_expected); - return false; - } - - if ((mask_attrs & mask_allowed) != mask_attrs) { - /* Mask attributes check failed. */ - OVS_NLERR("Contain more than allowed mask fields (mask_attrs=%llx, mask_allowed=%llx).\n", - mask_attrs, mask_allowed); - return false; + 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))) { + tcp_flags = TCP_FLAGS_BE16(tcp_hdr(skb)); } - return true; + spin_lock(&flow->lock); + flow->used = jiffies; + flow->packet_count++; + flow->byte_count += skb->len; + flow->tcp_flags |= tcp_flags; + spin_unlock(&flow->lock); } static int check_header(struct sk_buff *skb, int len) @@ -311,19 +147,6 @@ static bool icmphdr_ok(struct sk_buff *skb) sizeof(struct icmphdr)); } -u64 ovs_flow_used_time(unsigned long flow_jiffies) -{ - 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; -} - static int parse_ipv6hdr(struct sk_buff *skb, struct sw_flow_key *key) { unsigned int nh_ofs = skb_network_offset(skb); @@ -372,311 +195,6 @@ static bool icmp6hdr_ok(struct sk_buff *skb) sizeof(struct icmp6hdr)); } -void ovs_flow_key_mask(struct sw_flow_key *dst, const struct sw_flow_key *src, - const struct sw_flow_mask *mask) -{ - const long *m = (long *)((u8 *)&mask->key + mask->range.start); - const long *s = (long *)((u8 *)src + mask->range.start); - long *d = (long *)((u8 *)dst + mask->range.start); - int i; - - /* The memory outside of the 'mask->range' are not set since - * further operations on 'dst' only uses contents within - * 'mask->range'. - */ - for (i = 0; i < range_n_bytes(&mask->range); i += sizeof(long)) - *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); -} - -static struct flow_table *__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)); - table->mask_list = NULL; - - 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(&flow->hash_node[ver]); - ovs_flow_free(flow, false); - } - } - - BUG_ON(!list_empty(table->mask_list)); - kfree(table->mask_list); - -skip_flows: - free_buckets(table->buckets); - kfree(table); -} - -struct flow_table *ovs_flow_tbl_alloc(int new_size) -{ - struct flow_table *table = __flow_tbl_alloc(new_size); - - if (!table) - return NULL; - - table->mask_list = kmalloc(sizeof(struct list_head), GFP_KERNEL); - if (!table->mask_list) { - table->keep_flows = true; - __flow_tbl_destroy(table); - return NULL; - } - INIT_LIST_HEAD(table->mask_list); - - return 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 = __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(flow->mask, deferred); - - if (deferred) - call_rcu(&flow->rcu, rcu_free_flow_callback); - else - __flow_free(flow); -} - -/* 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) -{ - kfree_rcu(sf_acts, rcu); -} - static int parse_vlan(struct sk_buff *skb, struct sw_flow_key *key) { struct qtag_prefix { @@ -910,6 +428,7 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) struct tcphdr *tcp = tcp_hdr(skb); key->ipv4.tp.src = tcp->source; key->ipv4.tp.dst = tcp->dest; + key->ipv4.tp.flags = TCP_FLAGS_BE16(tcp); } } else if (key->ip.proto == IPPROTO_UDP) { if (udphdr_ok(skb)) { @@ -978,6 +497,7 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) struct tcphdr *tcp = tcp_hdr(skb); key->ipv6.tp.src = tcp->source; key->ipv6.tp.dst = tcp->dest; + key->ipv6.tp.flags = TCP_FLAGS_BE16(tcp); } } else if (key->ip.proto == NEXTHDR_UDP) { if (udphdr_ok(skb)) { @@ -1002,1080 +522,3 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) return 0; } - -static u32 ovs_flow_hash(const struct sw_flow_key *key, int key_start, - int key_end) -{ - u32 *hash_key = (u32 *)((u8 *)key + key_start); - int hash_u32s = (key_end - key_start) >> 2; - - /* Make sure number of hash bytes are multiple of u32. */ - BUILD_BUG_ON(sizeof(long) % sizeof(u32)); - - return jhash2(hash_key, hash_u32s, 0); -} - -static int flow_key_start(const struct sw_flow_key *key) -{ - if (key->tun_key.ipv4_dst) - return 0; - else - return rounddown(offsetof(struct sw_flow_key, phy), - sizeof(long)); -} - -static bool __cmp_key(const struct sw_flow_key *key1, - const struct sw_flow_key *key2, int key_start, int key_end) -{ - const long *cp1 = (long *)((u8 *)key1 + key_start); - const long *cp2 = (long *)((u8 *)key2 + key_start); - long diffs = 0; - int i; - - for (i = key_start; i < key_end; i += sizeof(long)) - diffs |= *cp1++ ^ *cp2++; - - return diffs == 0; -} - -static bool __flow_cmp_masked_key(const struct sw_flow *flow, - const struct sw_flow_key *key, int key_start, int key_end) -{ - return __cmp_key(&flow->key, key, key_start, key_end); -} - -static bool __flow_cmp_unmasked_key(const struct sw_flow *flow, - const struct sw_flow_key *key, int key_start, int key_end) -{ - return __cmp_key(&flow->unmasked_key, key, key_start, key_end); -} - -bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow, - const struct sw_flow_key *key, int key_end) -{ - int key_start; - key_start = flow_key_start(key); - - return __flow_cmp_unmasked_key(flow, key, key_start, key_end); - -} - -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_end = match->range.end; - struct sw_flow *flow; - - flow = ovs_flow_lookup(table, unmasked); - if (flow && (!ovs_flow_cmp_unmasked_key(flow, unmasked, key_end))) - flow = NULL; - - return flow; -} - -static struct sw_flow *ovs_masked_flow_lookup(struct flow_table *table, - const struct sw_flow_key *unmasked, - struct sw_flow_mask *mask) -{ - struct sw_flow *flow; - struct hlist_head *head; - int key_start = mask->range.start; - int key_end = mask->range.end; - u32 hash; - struct sw_flow_key masked_key; - - ovs_flow_key_mask(&masked_key, unmasked, mask); - hash = ovs_flow_hash(&masked_key, key_start, key_end); - head = find_bucket(table, hash); - hlist_for_each_entry_rcu(flow, head, hash_node[table->node_ver]) { - if (flow->mask == mask && - __flow_cmp_masked_key(flow, &masked_key, - key_start, key_end)) - 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) -{ - flow->hash = ovs_flow_hash(&flow->key, flow->mask->range.start, - 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_SCTP] = sizeof(struct ovs_key_sctp), - [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; - u32 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) { - OVS_NLERR("Unknown key attribute (type=%d, max=%d).\n", - type, OVS_KEY_ATTR_MAX); - return -EINVAL; - } - - if (attrs & (1 << type)) { - OVS_NLERR("Duplicate key attribute (type %d).\n", type); - return -EINVAL; - } - - expected_len = ovs_key_lens[type]; - if (nla_len(nla) != expected_len && expected_len != -1) { - OVS_NLERR("Key attribute has unexpected length (type=%d" - ", length=%d, expected=%d).\n", type, - nla_len(nla), expected_len); - return -EINVAL; - } - - if (!nz || !is_all_zero(nla_data(nla), expected_len)) { - attrs |= 1 << type; - a[type] = nla; - } - } - if (rem) { - OVS_NLERR("Message has %d unknown bytes.\n", 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 ovs_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_NLERR("Unknown IPv4 tunnel attribute (type=%d, max=%d).\n", - type, OVS_TUNNEL_KEY_ATTR_MAX); - return -EINVAL; - } - - if (ovs_tunnel_key_lens[type] != nla_len(a)) { - OVS_NLERR("IPv4 tunnel attribute type has unexpected " - " length (type=%d, length=%d, expected=%d).\n", - type, nla_len(a), ovs_tunnel_key_lens[type]); - 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) { - OVS_NLERR("IPv4 tunnel attribute has %d unknown bytes.\n", rem); - return -EINVAL; - } - - if (!is_mask) { - if (!match->key->tun_key.ipv4_dst) { - OVS_NLERR("IPv4 tunnel destination address is zero.\n"); - return -EINVAL; - } - - if (!ttl) { - OVS_NLERR("IPv4 tunnel TTL not specified.\n"); - return -EINVAL; - } - } - - return 0; -} - -int ovs_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 (output->tun_flags & TUNNEL_KEY && - nla_put_be64(skb, OVS_TUNNEL_KEY_ATTR_ID, output->tun_id)) - return -EMSGSIZE; - if (output->ipv4_src && - nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_SRC, output->ipv4_src)) - return -EMSGSIZE; - if (output->ipv4_dst && - nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_DST, output->ipv4_dst)) - return -EMSGSIZE; - if (output->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 ((output->tun_flags & TUNNEL_DONT_FRAGMENT) && - nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT)) - return -EMSGSIZE; - if ((output->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 & (1 << OVS_KEY_ATTR_PRIORITY)) { - SW_FLOW_KEY_PUT(match, phy.priority, - nla_get_u32(a[OVS_KEY_ATTR_PRIORITY]), is_mask); - *attrs &= ~(1 << OVS_KEY_ATTR_PRIORITY); - } - - if (*attrs & (1 << OVS_KEY_ATTR_IN_PORT)) { - u32 in_port = nla_get_u32(a[OVS_KEY_ATTR_IN_PORT]); - - if (is_mask) - in_port = 0xffffffff; /* Always exact match in_port. */ - else if (in_port >= DP_MAX_PORTS) - return -EINVAL; - - SW_FLOW_KEY_PUT(match, phy.in_port, in_port, is_mask); - *attrs &= ~(1 << OVS_KEY_ATTR_IN_PORT); - } else if (!is_mask) { - SW_FLOW_KEY_PUT(match, phy.in_port, DP_MAX_PORTS, is_mask); - } - - if (*attrs & (1 << OVS_KEY_ATTR_SKB_MARK)) { - uint32_t mark = nla_get_u32(a[OVS_KEY_ATTR_SKB_MARK]); - - SW_FLOW_KEY_PUT(match, phy.skb_mark, mark, is_mask); - *attrs &= ~(1 << OVS_KEY_ATTR_SKB_MARK); - } - if (*attrs & (1 << OVS_KEY_ATTR_TUNNEL)) { - if (ovs_ipv4_tun_from_nlattr(a[OVS_KEY_ATTR_TUNNEL], match, - is_mask)) - return -EINVAL; - *attrs &= ~(1 << 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; - u64 orig_attrs = attrs; - - err = metadata_from_nlattrs(match, &attrs, a, is_mask); - if (err) - return err; - - if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ETHERNET); - } - - if (attrs & (1 << OVS_KEY_ATTR_VLAN)) { - __be16 tci; - - tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]); - if (!(tci & htons(VLAN_TAG_PRESENT))) { - if (is_mask) - OVS_NLERR("VLAN TCI mask does not have exact match for VLAN_TAG_PRESENT bit.\n"); - else - OVS_NLERR("VLAN TCI does not have VLAN_TAG_PRESENT bit set.\n"); - - return -EINVAL; - } - - SW_FLOW_KEY_PUT(match, eth.tci, tci, is_mask); - attrs &= ~(1 << OVS_KEY_ATTR_VLAN); - } else if (!is_mask) - SW_FLOW_KEY_PUT(match, eth.tci, htons(0xffff), true); - - if (attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) { - __be16 eth_type; - - eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]); - if (is_mask) { - /* Always exact match EtherType. */ - eth_type = htons(0xffff); - } else if (ntohs(eth_type) < ETH_P_802_3_MIN) { - OVS_NLERR("EtherType is less than minimum (type=%x, min=%x).\n", - ntohs(eth_type), ETH_P_802_3_MIN); - return -EINVAL; - } - - SW_FLOW_KEY_PUT(match, eth.type, eth_type, is_mask); - attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE); - } else if (!is_mask) { - SW_FLOW_KEY_PUT(match, eth.type, htons(ETH_P_802_2), is_mask); - } - - if (attrs & (1 << 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) { - OVS_NLERR("Unknown IPv4 fragment type (value=%d, max=%d).\n", - 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 &= ~(1 << OVS_KEY_ATTR_IPV4); - } - - if (attrs & (1 << 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) { - OVS_NLERR("Unknown IPv6 fragment type (value=%d, max=%d).\n", - 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 &= ~(1 << OVS_KEY_ATTR_IPV6); - } - - if (attrs & (1 << 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))) { - OVS_NLERR("Unknown ARP opcode (opcode=%d).\n", - arp_key->arp_op); - 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 &= ~(1 << OVS_KEY_ATTR_ARP); - } - - if (attrs & (1 << OVS_KEY_ATTR_TCP)) { - const struct ovs_key_tcp *tcp_key; - - tcp_key = nla_data(a[OVS_KEY_ATTR_TCP]); - if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - 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); - } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.src, - tcp_key->tcp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, - tcp_key->tcp_dst, is_mask); - } - attrs &= ~(1 << OVS_KEY_ATTR_TCP); - } - - if (attrs & (1 << OVS_KEY_ATTR_UDP)) { - const struct ovs_key_udp *udp_key; - - udp_key = nla_data(a[OVS_KEY_ATTR_UDP]); - if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - 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); - } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.src, - udp_key->udp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, - udp_key->udp_dst, is_mask); - } - attrs &= ~(1 << OVS_KEY_ATTR_UDP); - } - - if (attrs & (1 << OVS_KEY_ATTR_SCTP)) { - const struct ovs_key_sctp *sctp_key; - - sctp_key = nla_data(a[OVS_KEY_ATTR_SCTP]); - if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - SW_FLOW_KEY_PUT(match, ipv4.tp.src, - sctp_key->sctp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv4.tp.dst, - sctp_key->sctp_dst, is_mask); - } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.src, - sctp_key->sctp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, - sctp_key->sctp_dst, is_mask); - } - attrs &= ~(1 << OVS_KEY_ATTR_SCTP); - } - - if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ICMP); - } - - if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ICMPV6); - } - - if (attrs & (1 << 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 &= ~(1 << 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 & (1 << OVS_KEY_ATTR_ETHERNET)) && - (key_attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) && - (nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]) == htons(ETH_P_8021Q))) { - __be16 tci; - - if (!((key_attrs & (1 << OVS_KEY_ATTR_VLAN)) && - (key_attrs & (1 << OVS_KEY_ATTR_ENCAP)))) { - OVS_NLERR("Invalid Vlan frame.\n"); - return -EINVAL; - } - - key_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE); - tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]); - encap = a[OVS_KEY_ATTR_ENCAP]; - key_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP); - encap_valid = true; - - if (tci & htons(VLAN_TAG_PRESENT)) { - err = parse_flow_nlattrs(encap, a, &key_attrs); - if (err) - return err; - } else if (!tci) { - /* Corner case for truncated 802.1Q header. */ - if (nla_len(encap)) { - OVS_NLERR("Truncated 802.1Q header has non-zero encap attribute.\n"); - return -EINVAL; - } - } else { - OVS_NLERR("Encap attribute is set for a non-VLAN frame.\n"); - return -EINVAL; - } - } - - 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) { - __be16 eth_type = 0; - __be16 tci = 0; - - if (!encap_valid) { - OVS_NLERR("Encap mask attribute is set for non-VLAN frame.\n"); - return -EINVAL; - } - - mask_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP); - if (a[OVS_KEY_ATTR_ETHERTYPE]) - eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]); - - if (eth_type == htons(0xffff)) { - mask_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE); - encap = a[OVS_KEY_ATTR_ENCAP]; - err = parse_flow_mask_nlattrs(encap, a, &mask_attrs); - } else { - OVS_NLERR("VLAN frames must have an exact match on the TPID (mask=%x).\n", - ntohs(eth_type)); - return -EINVAL; - } - - if (a[OVS_KEY_ATTR_VLAN]) - tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]); - - if (!(tci & htons(VLAN_TAG_PRESENT))) { - OVS_NLERR("VLAN tag present bit must have an exact match (tci_mask=%x).\n", ntohs(tci)); - return -EINVAL; - } - } - - 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; - - memset(&match, 0, sizeof(match)); - match.key = &flow->key; - - 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; - bool is_mask = (swkey != output); - - if (nla_put_u32(skb, OVS_KEY_ATTR_PRIORITY, output->phy.priority)) - goto nla_put_failure; - - if ((swkey->tun_key.ipv4_dst || is_mask) && - ovs_ipv4_tun_to_nlattr(skb, &swkey->tun_key, &output->tun_key)) - goto nla_put_failure; - - if (swkey->phy.in_port == DP_MAX_PORTS) { - if (is_mask && (output->phy.in_port == 0xffff)) - if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT, 0xffffffff)) - goto nla_put_failure; - } else { - u16 upper_u16; - upper_u16 = !is_mask ? 0 : 0xffff; - - if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT, - (upper_u16 << 16) | output->phy.in_port)) - goto nla_put_failure; - } - - if (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 = !is_mask ? 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->eth.type == htons(ETH_P_802_2)) { - /* - * Ethertype 802.2 is represented in the netlink with omitted - * OVS_KEY_ATTR_ETHERTYPE in the flow key attribute, and - * 0xffff in the mask attribute. Ethertype can also - * be wildcarded. - */ - if (is_mask && output->eth.type) - if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, - output->eth.type)) - goto nla_put_failure; - goto unencap; - } - - 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->ip.proto == IPPROTO_SCTP) { - struct ovs_key_sctp *sctp_key; - - nla = nla_reserve(skb, OVS_KEY_ATTR_SCTP, sizeof(*sctp_key)); - if (!nla) - goto nla_put_failure; - sctp_key = nla_data(nla); - if (swkey->eth.type == htons(ETH_P_IP)) { - sctp_key->sctp_src = swkey->ipv4.tp.src; - sctp_key->sctp_dst = swkey->ipv4.tp.dst; - } else if (swkey->eth.type == htons(ETH_P_IPV6)) { - sctp_key->sctp_src = swkey->ipv6.tp.src; - sctp_key->sctp_dst = swkey->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) -{ - BUILD_BUG_ON(__alignof__(struct sw_flow_key) % __alignof__(long)); - BUILD_BUG_ON(sizeof(struct sw_flow_key) % sizeof(long)); - - 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++; -} - -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) - kfree_rcu(mask, rcu); - 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_, range_n_bytes(&a->range)) == 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, range_n_bytes(range)); -} diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index 212fbf7510c..1510f51dbf7 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -33,14 +33,6 @@ #include <net/inet_ecn.h> struct sk_buff; -struct sw_flow_mask; -struct flow_table; - -struct sw_flow_actions { - struct rcu_head rcu; - u32 actions_len; - struct nlattr actions[]; -}; /* Used to memset ovs_key_ipv4_tunnel padding. */ #define OVS_TUNNEL_KEY_SIZE \ @@ -101,6 +93,7 @@ struct sw_flow_key { struct { __be16 src; /* TCP/UDP/SCTP source port. */ __be16 dst; /* TCP/UDP/SCTP destination port. */ + __be16 flags; /* TCP flags. */ } tp; struct { u8 sha[ETH_ALEN]; /* ARP source hardware address. */ @@ -117,6 +110,7 @@ struct sw_flow_key { struct { __be16 src; /* TCP/UDP/SCTP source port. */ __be16 dst; /* TCP/UDP/SCTP destination port. */ + __be16 flags; /* TCP flags. */ } tp; struct { struct in6_addr target; /* ND target address. */ @@ -127,6 +121,31 @@ struct sw_flow_key { }; } __aligned(BITS_PER_LONG/8); /* Ensure that we can do comparisons as longs. */ +struct sw_flow_key_range { + size_t start; + size_t end; +}; + +struct sw_flow_mask { + int ref_count; + struct rcu_head rcu; + struct list_head list; + 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; +}; + +struct sw_flow_actions { + struct rcu_head rcu; + u32 actions_len; + struct nlattr actions[]; +}; + struct sw_flow { struct rcu_head rcu; struct hlist_node hash_node[2]; @@ -141,23 +160,9 @@ struct sw_flow { unsigned long used; /* Last used time (in jiffies). */ u64 packet_count; /* Number of packets matched. */ u64 byte_count; /* Number of bytes matched. */ - u8 tcp_flags; /* Union of seen TCP flags. */ -}; - -struct sw_flow_key_range { - size_t start; - size_t end; + __be16 tcp_flags; /* Union of seen TCP flags. */ }; -struct sw_flow_match { - struct sw_flow_key *key; - struct sw_flow_key_range range; - struct sw_flow_mask *mask; -}; - -void ovs_match_init(struct sw_flow_match *match, - struct sw_flow_key *key, struct sw_flow_mask *mask); - struct arp_eth_header { __be16 ar_hrd; /* format of hardware address */ __be16 ar_pro; /* format of protocol address */ @@ -172,88 +177,9 @@ struct arp_eth_header { unsigned char ar_tip[4]; /* target IP address */ } __packed; -int ovs_flow_init(void); -void ovs_flow_exit(void); - -struct sw_flow *ovs_flow_alloc(void); -void ovs_flow_deferred_free(struct sw_flow *); -void ovs_flow_free(struct sw_flow *, bool deferred); - -struct sw_flow_actions *ovs_flow_actions_alloc(int actions_len); -void ovs_flow_deferred_free_acts(struct sw_flow_actions *); - -int ovs_flow_extract(struct sk_buff *, u16 in_port, struct sw_flow_key *); void ovs_flow_used(struct sw_flow *, struct sk_buff *); u64 ovs_flow_used_time(unsigned long flow_jiffies); -int ovs_flow_to_nlattrs(const struct sw_flow_key *, - const struct sw_flow_key *, struct sk_buff *); -int ovs_match_from_nlattrs(struct sw_flow_match *match, - const struct nlattr *, - const struct nlattr *); -int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow, - const struct nlattr *attr); -#define MAX_ACTIONS_BUFSIZE (32 * 1024) -#define TBL_MIN_BUCKETS 1024 - -struct flow_table { - struct flex_array *buckets; - unsigned int count, n_buckets; - struct rcu_head rcu; - struct list_head *mask_list; - int node_ver; - u32 hash_seed; - bool keep_flows; -}; - -static inline int ovs_flow_tbl_count(struct flow_table *table) -{ - return table->count; -} - -static inline int ovs_flow_tbl_need_to_expand(struct flow_table *table) -{ - return (table->count > table->n_buckets); -} - -struct sw_flow *ovs_flow_lookup(struct flow_table *, - const struct sw_flow_key *); -struct sw_flow *ovs_flow_lookup_unmasked_key(struct flow_table *table, - struct sw_flow_match *match); - -void ovs_flow_tbl_destroy(struct flow_table *table, bool deferred); -struct flow_table *ovs_flow_tbl_alloc(int new_size); -struct flow_table *ovs_flow_tbl_expand(struct flow_table *table); -struct flow_table *ovs_flow_tbl_rehash(struct flow_table *table); - -void ovs_flow_insert(struct flow_table *table, struct sw_flow *flow); -void ovs_flow_remove(struct flow_table *table, struct sw_flow *flow); - -struct sw_flow *ovs_flow_dump_next(struct flow_table *table, u32 *bucket, u32 *idx); -extern const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1]; -int ovs_ipv4_tun_from_nlattr(const struct nlattr *attr, - struct sw_flow_match *match, bool is_mask); -int ovs_ipv4_tun_to_nlattr(struct sk_buff *skb, - const struct ovs_key_ipv4_tunnel *tun_key, - const struct ovs_key_ipv4_tunnel *output); - -bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow, - const struct sw_flow_key *key, int key_end); - -struct sw_flow_mask { - int ref_count; - struct rcu_head rcu; - struct list_head list; - struct sw_flow_key_range range; - struct sw_flow_key key; -}; +int ovs_flow_extract(struct sk_buff *, u16 in_port, struct sw_flow_key *); -struct sw_flow_mask *ovs_sw_flow_mask_alloc(void); -void ovs_sw_flow_mask_add_ref(struct sw_flow_mask *); -void ovs_sw_flow_mask_del_ref(struct sw_flow_mask *, bool deferred); -void ovs_sw_flow_mask_insert(struct flow_table *, struct sw_flow_mask *); -struct sw_flow_mask *ovs_sw_flow_mask_find(const struct flow_table *, - const struct sw_flow_mask *); -void ovs_flow_key_mask(struct sw_flow_key *dst, const struct sw_flow_key *src, - const struct sw_flow_mask *mask); #endif /* flow.h */ diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c new file mode 100644 index 00000000000..2bc1bc1aca3 --- /dev/null +++ b/net/openvswitch/flow_netlink.c @@ -0,0 +1,1630 @@ +/* + * Copyright (c) 2007-2013 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 + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA + */ + +#include "flow.h" +#include "datapath.h" +#include <linux/uaccess.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/if_ether.h> +#include <linux/if_vlan.h> +#include <net/llc_pdu.h> +#include <linux/kernel.h> +#include <linux/jhash.h> +#include <linux/jiffies.h> +#include <linux/llc.h> +#include <linux/module.h> +#include <linux/in.h> +#include <linux/rcupdate.h> +#include <linux/if_arp.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/sctp.h> +#include <linux/tcp.h> +#include <linux/udp.h> +#include <linux/icmp.h> +#include <linux/icmpv6.h> +#include <linux/rculist.h> +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/ndisc.h> + +#include "flow_netlink.h" + +static void update_range__(struct sw_flow_match *match, + size_t offset, size_t size, bool is_mask) +{ + struct sw_flow_key_range *range = NULL; + size_t start = rounddown(offset, sizeof(long)); + size_t end = roundup(offset + size, sizeof(long)); + + 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; + } + + if (range->start > start) + range->start = start; + + if (range->end < end) + range->end = end; +} + +#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) { \ + if ((match)->mask) \ + (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) { \ + if ((match)->mask) \ + memcpy(&(match)->mask->key.field, value_p, len);\ + } else { \ + memcpy(&(match)->key->field, value_p, len); \ + } \ + } while (0) + +static u16 range_n_bytes(const struct sw_flow_key_range *range) +{ + return range->end - range->start; +} + +static bool match_validate(const struct sw_flow_match *match, + u64 key_attrs, u64 mask_attrs) +{ + u64 key_expected = 1 << 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 &= ~((1 << OVS_KEY_ATTR_IPV4) + | (1 << OVS_KEY_ATTR_IPV6) + | (1 << OVS_KEY_ATTR_TCP) + | (1 << OVS_KEY_ATTR_TCP_FLAGS) + | (1 << OVS_KEY_ATTR_UDP) + | (1 << OVS_KEY_ATTR_SCTP) + | (1 << OVS_KEY_ATTR_ICMP) + | (1 << OVS_KEY_ATTR_ICMPV6) + | (1 << OVS_KEY_ATTR_ARP) + | (1 << OVS_KEY_ATTR_ND)); + + /* Always allowed mask fields. */ + mask_allowed |= ((1 << OVS_KEY_ATTR_TUNNEL) + | (1 << OVS_KEY_ATTR_IN_PORT) + | (1 << 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 |= 1 << OVS_KEY_ATTR_ARP; + if (match->mask && (match->mask->key.eth.type == htons(0xffff))) + mask_allowed |= 1 << OVS_KEY_ATTR_ARP; + } + + if (match->key->eth.type == htons(ETH_P_IP)) { + key_expected |= 1 << OVS_KEY_ATTR_IPV4; + if (match->mask && (match->mask->key.eth.type == htons(0xffff))) + mask_allowed |= 1 << OVS_KEY_ATTR_IPV4; + + if (match->key->ip.frag != OVS_FRAG_TYPE_LATER) { + if (match->key->ip.proto == IPPROTO_UDP) { + key_expected |= 1 << OVS_KEY_ATTR_UDP; + if (match->mask && (match->mask->key.ip.proto == 0xff)) + mask_allowed |= 1 << OVS_KEY_ATTR_UDP; + } + + if (match->key->ip.proto == IPPROTO_SCTP) { + key_expected |= 1 << OVS_KEY_ATTR_SCTP; + if (match->mask && (match->mask->key.ip.proto == 0xff)) + mask_allowed |= 1 << OVS_KEY_ATTR_SCTP; + } + + if (match->key->ip.proto == IPPROTO_TCP) { + key_expected |= 1 << OVS_KEY_ATTR_TCP; + key_expected |= 1 << OVS_KEY_ATTR_TCP_FLAGS; + if (match->mask && (match->mask->key.ip.proto == 0xff)) { + mask_allowed |= 1 << OVS_KEY_ATTR_TCP; + mask_allowed |= 1 << OVS_KEY_ATTR_TCP_FLAGS; + } + } + + if (match->key->ip.proto == IPPROTO_ICMP) { + key_expected |= 1 << OVS_KEY_ATTR_ICMP; + if (match->mask && (match->mask->key.ip.proto == 0xff)) + mask_allowed |= 1 << OVS_KEY_ATTR_ICMP; + } + } + } + + if (match->key->eth.type == htons(ETH_P_IPV6)) { + key_expected |= 1 << OVS_KEY_ATTR_IPV6; + if (match->mask && (match->mask->key.eth.type == htons(0xffff))) + mask_allowed |= 1 << OVS_KEY_ATTR_IPV6; + + if (match->key->ip.frag != OVS_FRAG_TYPE_LATER) { + if (match->key->ip.proto == IPPROTO_UDP) { + key_expected |= 1 << OVS_KEY_ATTR_UDP; + if (match->mask && (match->mask->key.ip.proto == 0xff)) + mask_allowed |= 1 << OVS_KEY_ATTR_UDP; + } + + if (match->key->ip.proto == IPPROTO_SCTP) { + key_expected |= 1 << OVS_KEY_ATTR_SCTP; + if (match->mask && (match->mask->key.ip.proto == 0xff)) + mask_allowed |= 1 << OVS_KEY_ATTR_SCTP; + } + + if (match->key->ip.proto == IPPROTO_TCP) { + key_expected |= 1 << OVS_KEY_ATTR_TCP; + key_expected |= 1 << OVS_KEY_ATTR_TCP_FLAGS; + if (match->mask && (match->mask->key.ip.proto == 0xff)) { + mask_allowed |= 1 << OVS_KEY_ATTR_TCP; + mask_allowed |= 1 << OVS_KEY_ATTR_TCP_FLAGS; + } + } + + if (match->key->ip.proto == IPPROTO_ICMPV6) { + key_expected |= 1 << OVS_KEY_ATTR_ICMPV6; + if (match->mask && (match->mask->key.ip.proto == 0xff)) + mask_allowed |= 1 << 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 |= 1 << OVS_KEY_ATTR_ND; + if (match->mask && (match->mask->key.ipv6.tp.src == htons(0xffff))) + mask_allowed |= 1 << OVS_KEY_ATTR_ND; + } + } + } + } + + if ((key_attrs & key_expected) != key_expected) { + /* Key attributes check failed. */ + OVS_NLERR("Missing expected key attributes (key_attrs=%llx, expected=%llx).\n", + key_attrs, key_expected); + return false; + } + + if ((mask_attrs & mask_allowed) != mask_attrs) { + /* Mask attributes check failed. */ + OVS_NLERR("Contain more than allowed mask fields (mask_attrs=%llx, mask_allowed=%llx).\n", + mask_attrs, mask_allowed); + return false; + } + + return true; +} + +/* The size of the argument for each %OVS_KEY_ATTR_* Netlink attribute. */ +static 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_TCP_FLAGS] = sizeof(__be16), + [OVS_KEY_ATTR_UDP] = sizeof(struct ovs_key_udp), + [OVS_KEY_ATTR_SCTP] = sizeof(struct ovs_key_sctp), + [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) { + OVS_NLERR("Unknown key attribute (type=%d, max=%d).\n", + type, OVS_KEY_ATTR_MAX); + return -EINVAL; + } + + if (attrs & (1 << type)) { + OVS_NLERR("Duplicate key attribute (type %d).\n", type); + return -EINVAL; + } + + expected_len = ovs_key_lens[type]; + if (nla_len(nla) != expected_len && expected_len != -1) { + OVS_NLERR("Key attribute has unexpected length (type=%d" + ", length=%d, expected=%d).\n", type, + nla_len(nla), expected_len); + return -EINVAL; + } + + if (!nz || !is_all_zero(nla_data(nla), expected_len)) { + attrs |= 1 << type; + a[type] = nla; + } + } + if (rem) { + OVS_NLERR("Message has %d unknown bytes.\n", 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); +} + +static 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_NLERR("Unknown IPv4 tunnel attribute (type=%d, max=%d).\n", + type, OVS_TUNNEL_KEY_ATTR_MAX); + return -EINVAL; + } + + if (ovs_tunnel_key_lens[type] != nla_len(a)) { + OVS_NLERR("IPv4 tunnel attribute type has unexpected " + " length (type=%d, length=%d, expected=%d).\n", + type, nla_len(a), ovs_tunnel_key_lens[type]); + 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) { + OVS_NLERR("IPv4 tunnel attribute has %d unknown bytes.\n", rem); + return -EINVAL; + } + + if (!is_mask) { + if (!match->key->tun_key.ipv4_dst) { + OVS_NLERR("IPv4 tunnel destination address is zero.\n"); + return -EINVAL; + } + + if (!ttl) { + OVS_NLERR("IPv4 tunnel TTL not specified.\n"); + return -EINVAL; + } + } + + return 0; +} + +static 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 (output->tun_flags & TUNNEL_KEY && + nla_put_be64(skb, OVS_TUNNEL_KEY_ATTR_ID, output->tun_id)) + return -EMSGSIZE; + if (output->ipv4_src && + nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_SRC, output->ipv4_src)) + return -EMSGSIZE; + if (output->ipv4_dst && + nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_DST, output->ipv4_dst)) + return -EMSGSIZE; + if (output->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 ((output->tun_flags & TUNNEL_DONT_FRAGMENT) && + nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT)) + return -EMSGSIZE; + if ((output->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 & (1 << OVS_KEY_ATTR_PRIORITY)) { + SW_FLOW_KEY_PUT(match, phy.priority, + nla_get_u32(a[OVS_KEY_ATTR_PRIORITY]), is_mask); + *attrs &= ~(1 << OVS_KEY_ATTR_PRIORITY); + } + + if (*attrs & (1 << OVS_KEY_ATTR_IN_PORT)) { + u32 in_port = nla_get_u32(a[OVS_KEY_ATTR_IN_PORT]); + + if (is_mask) + in_port = 0xffffffff; /* Always exact match in_port. */ + else if (in_port >= DP_MAX_PORTS) + return -EINVAL; + + SW_FLOW_KEY_PUT(match, phy.in_port, in_port, is_mask); + *attrs &= ~(1 << OVS_KEY_ATTR_IN_PORT); + } else if (!is_mask) { + SW_FLOW_KEY_PUT(match, phy.in_port, DP_MAX_PORTS, is_mask); + } + + if (*attrs & (1 << OVS_KEY_ATTR_SKB_MARK)) { + uint32_t mark = nla_get_u32(a[OVS_KEY_ATTR_SKB_MARK]); + + SW_FLOW_KEY_PUT(match, phy.skb_mark, mark, is_mask); + *attrs &= ~(1 << OVS_KEY_ATTR_SKB_MARK); + } + if (*attrs & (1 << OVS_KEY_ATTR_TUNNEL)) { + if (ipv4_tun_from_nlattr(a[OVS_KEY_ATTR_TUNNEL], match, + is_mask)) + return -EINVAL; + *attrs &= ~(1 << 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; + u64 orig_attrs = attrs; + + err = metadata_from_nlattrs(match, &attrs, a, is_mask); + if (err) + return err; + + if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ETHERNET); + } + + if (attrs & (1 << OVS_KEY_ATTR_VLAN)) { + __be16 tci; + + tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]); + if (!(tci & htons(VLAN_TAG_PRESENT))) { + if (is_mask) + OVS_NLERR("VLAN TCI mask does not have exact match for VLAN_TAG_PRESENT bit.\n"); + else + OVS_NLERR("VLAN TCI does not have VLAN_TAG_PRESENT bit set.\n"); + + return -EINVAL; + } + + SW_FLOW_KEY_PUT(match, eth.tci, tci, is_mask); + attrs &= ~(1 << OVS_KEY_ATTR_VLAN); + } else if (!is_mask) + SW_FLOW_KEY_PUT(match, eth.tci, htons(0xffff), true); + + if (attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) { + __be16 eth_type; + + eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]); + if (is_mask) { + /* Always exact match EtherType. */ + eth_type = htons(0xffff); + } else if (ntohs(eth_type) < ETH_P_802_3_MIN) { + OVS_NLERR("EtherType is less than minimum (type=%x, min=%x).\n", + ntohs(eth_type), ETH_P_802_3_MIN); + return -EINVAL; + } + + SW_FLOW_KEY_PUT(match, eth.type, eth_type, is_mask); + attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE); + } else if (!is_mask) { + SW_FLOW_KEY_PUT(match, eth.type, htons(ETH_P_802_2), is_mask); + } + + if (attrs & (1 << 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) { + OVS_NLERR("Unknown IPv4 fragment type (value=%d, max=%d).\n", + 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 &= ~(1 << OVS_KEY_ATTR_IPV4); + } + + if (attrs & (1 << 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) { + OVS_NLERR("Unknown IPv6 fragment type (value=%d, max=%d).\n", + 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 &= ~(1 << OVS_KEY_ATTR_IPV6); + } + + if (attrs & (1 << 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))) { + OVS_NLERR("Unknown ARP opcode (opcode=%d).\n", + arp_key->arp_op); + 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 &= ~(1 << OVS_KEY_ATTR_ARP); + } + + if (attrs & (1 << OVS_KEY_ATTR_TCP)) { + const struct ovs_key_tcp *tcp_key; + + tcp_key = nla_data(a[OVS_KEY_ATTR_TCP]); + if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { + 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); + } else { + SW_FLOW_KEY_PUT(match, ipv6.tp.src, + tcp_key->tcp_src, is_mask); + SW_FLOW_KEY_PUT(match, ipv6.tp.dst, + tcp_key->tcp_dst, is_mask); + } + attrs &= ~(1 << OVS_KEY_ATTR_TCP); + } + + if (attrs & (1 << OVS_KEY_ATTR_TCP_FLAGS)) { + if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { + SW_FLOW_KEY_PUT(match, ipv4.tp.flags, + nla_get_be16(a[OVS_KEY_ATTR_TCP_FLAGS]), + is_mask); + } else { + SW_FLOW_KEY_PUT(match, ipv6.tp.flags, + nla_get_be16(a[OVS_KEY_ATTR_TCP_FLAGS]), + is_mask); + } + attrs &= ~(1 << OVS_KEY_ATTR_TCP_FLAGS); + } + + if (attrs & (1 << OVS_KEY_ATTR_UDP)) { + const struct ovs_key_udp *udp_key; + + udp_key = nla_data(a[OVS_KEY_ATTR_UDP]); + if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { + 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); + } else { + SW_FLOW_KEY_PUT(match, ipv6.tp.src, + udp_key->udp_src, is_mask); + SW_FLOW_KEY_PUT(match, ipv6.tp.dst, + udp_key->udp_dst, is_mask); + } + attrs &= ~(1 << OVS_KEY_ATTR_UDP); + } + + if (attrs & (1 << OVS_KEY_ATTR_SCTP)) { + const struct ovs_key_sctp *sctp_key; + + sctp_key = nla_data(a[OVS_KEY_ATTR_SCTP]); + if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { + SW_FLOW_KEY_PUT(match, ipv4.tp.src, + sctp_key->sctp_src, is_mask); + SW_FLOW_KEY_PUT(match, ipv4.tp.dst, + sctp_key->sctp_dst, is_mask); + } else { + SW_FLOW_KEY_PUT(match, ipv6.tp.src, + sctp_key->sctp_src, is_mask); + SW_FLOW_KEY_PUT(match, ipv6.tp.dst, + sctp_key->sctp_dst, is_mask); + } + attrs &= ~(1 << OVS_KEY_ATTR_SCTP); + } + + if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ICMP); + } + + if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ICMPV6); + } + + if (attrs & (1 << 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 &= ~(1 << OVS_KEY_ATTR_ND); + } + + if (attrs != 0) + return -EINVAL; + + return 0; +} + +static void 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, range_n_bytes(range)); +} + +/** + * ovs_nla_get_match - 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_nla_get_match(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 & (1 << OVS_KEY_ATTR_ETHERNET)) && + (key_attrs & (1 << OVS_KEY_ATTR_ETHERTYPE)) && + (nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]) == htons(ETH_P_8021Q))) { + __be16 tci; + + if (!((key_attrs & (1 << OVS_KEY_ATTR_VLAN)) && + (key_attrs & (1 << OVS_KEY_ATTR_ENCAP)))) { + OVS_NLERR("Invalid Vlan frame.\n"); + return -EINVAL; + } + + key_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE); + tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]); + encap = a[OVS_KEY_ATTR_ENCAP]; + key_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP); + encap_valid = true; + + if (tci & htons(VLAN_TAG_PRESENT)) { + err = parse_flow_nlattrs(encap, a, &key_attrs); + if (err) + return err; + } else if (!tci) { + /* Corner case for truncated 802.1Q header. */ + if (nla_len(encap)) { + OVS_NLERR("Truncated 802.1Q header has non-zero encap attribute.\n"); + return -EINVAL; + } + } else { + OVS_NLERR("Encap attribute is set for a non-VLAN frame.\n"); + return -EINVAL; + } + } + + 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 & 1 << OVS_KEY_ATTR_ENCAP) { + __be16 eth_type = 0; + __be16 tci = 0; + + if (!encap_valid) { + OVS_NLERR("Encap mask attribute is set for non-VLAN frame.\n"); + return -EINVAL; + } + + mask_attrs &= ~(1 << OVS_KEY_ATTR_ENCAP); + if (a[OVS_KEY_ATTR_ETHERTYPE]) + eth_type = nla_get_be16(a[OVS_KEY_ATTR_ETHERTYPE]); + + if (eth_type == htons(0xffff)) { + mask_attrs &= ~(1 << OVS_KEY_ATTR_ETHERTYPE); + encap = a[OVS_KEY_ATTR_ENCAP]; + err = parse_flow_mask_nlattrs(encap, a, &mask_attrs); + } else { + OVS_NLERR("VLAN frames must have an exact match on the TPID (mask=%x).\n", + ntohs(eth_type)); + return -EINVAL; + } + + if (a[OVS_KEY_ATTR_VLAN]) + tci = nla_get_be16(a[OVS_KEY_ATTR_VLAN]); + + if (!(tci & htons(VLAN_TAG_PRESENT))) { + OVS_NLERR("VLAN tag present bit must have an exact match (tci_mask=%x).\n", ntohs(tci)); + return -EINVAL; + } + } + + 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) + sw_flow_mask_set(match->mask, &match->range, 0xff); + } + + if (!match_validate(match, key_attrs, mask_attrs)) + return -EINVAL; + + return 0; +} + +/** + * ovs_nla_get_flow_metadata - 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_nla_get_flow_metadata(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; + + memset(&match, 0, sizeof(match)); + match.key = &flow->key; + + err = metadata_from_nlattrs(&match, &attrs, a, false); + if (err) + return err; + + return 0; +} + +int ovs_nla_put_flow(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; + bool is_mask = (swkey != output); + + if (nla_put_u32(skb, OVS_KEY_ATTR_PRIORITY, output->phy.priority)) + goto nla_put_failure; + + if ((swkey->tun_key.ipv4_dst || is_mask) && + ipv4_tun_to_nlattr(skb, &swkey->tun_key, &output->tun_key)) + goto nla_put_failure; + + if (swkey->phy.in_port == DP_MAX_PORTS) { + if (is_mask && (output->phy.in_port == 0xffff)) + if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT, 0xffffffff)) + goto nla_put_failure; + } else { + u16 upper_u16; + upper_u16 = !is_mask ? 0 : 0xffff; + + if (nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT, + (upper_u16 << 16) | output->phy.in_port)) + goto nla_put_failure; + } + + if (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 = !is_mask ? 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->eth.type == htons(ETH_P_802_2)) { + /* + * Ethertype 802.2 is represented in the netlink with omitted + * OVS_KEY_ATTR_ETHERTYPE in the flow key attribute, and + * 0xffff in the mask attribute. Ethertype can also + * be wildcarded. + */ + if (is_mask && output->eth.type) + if (nla_put_be16(skb, OVS_KEY_ATTR_ETHERTYPE, + output->eth.type)) + goto nla_put_failure; + goto unencap; + } + + 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; + if (nla_put_be16(skb, OVS_KEY_ATTR_TCP_FLAGS, + output->ipv4.tp.flags)) + goto nla_put_failure; + } 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; + if (nla_put_be16(skb, OVS_KEY_ATTR_TCP_FLAGS, + output->ipv6.tp.flags)) + goto nla_put_failure; + } + } 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->ip.proto == IPPROTO_SCTP) { + struct ovs_key_sctp *sctp_key; + + nla = nla_reserve(skb, OVS_KEY_ATTR_SCTP, sizeof(*sctp_key)); + if (!nla) + goto nla_put_failure; + sctp_key = nla_data(nla); + if (swkey->eth.type == htons(ETH_P_IP)) { + sctp_key->sctp_src = swkey->ipv4.tp.src; + sctp_key->sctp_dst = swkey->ipv4.tp.dst; + } else if (swkey->eth.type == htons(ETH_P_IPV6)) { + sctp_key->sctp_src = swkey->ipv6.tp.src; + sctp_key->sctp_dst = swkey->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; +} + +#define MAX_ACTIONS_BUFSIZE (32 * 1024) + +struct sw_flow_actions *ovs_nla_alloc_flow_actions(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; +} + +/* RCU callback used by ovs_nla_free_flow_actions. */ +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_nla_free_flow_actions(struct sw_flow_actions *sf_acts) +{ + call_rcu(&sf_acts->rcu, rcu_free_acts_callback); +} + +static struct nlattr *reserve_sfa_size(struct sw_flow_actions **sfa, + int attr_len) +{ + + struct sw_flow_actions *acts; + int new_acts_size; + int req_size = NLA_ALIGN(attr_len); + int next_offset = offsetof(struct sw_flow_actions, actions) + + (*sfa)->actions_len; + + if (req_size <= (ksize(*sfa) - next_offset)) + goto out; + + new_acts_size = ksize(*sfa) * 2; + + if (new_acts_size > MAX_ACTIONS_BUFSIZE) { + if ((MAX_ACTIONS_BUFSIZE - next_offset) < req_size) + return ERR_PTR(-EMSGSIZE); + new_acts_size = MAX_ACTIONS_BUFSIZE; + } + + acts = ovs_nla_alloc_flow_actions(new_acts_size); + if (IS_ERR(acts)) + return (void *)acts; + + memcpy(acts->actions, (*sfa)->actions, (*sfa)->actions_len); + acts->actions_len = (*sfa)->actions_len; + kfree(*sfa); + *sfa = acts; + +out: + (*sfa)->actions_len += req_size; + return (struct nlattr *) ((unsigned char *)(*sfa) + next_offset); +} + +static int add_action(struct sw_flow_actions **sfa, int attrtype, void *data, int len) +{ + struct nlattr *a; + + a = reserve_sfa_size(sfa, nla_attr_size(len)); + if (IS_ERR(a)) + return PTR_ERR(a); + + a->nla_type = attrtype; + a->nla_len = nla_attr_size(len); + + if (data) + memcpy(nla_data(a), data, len); + memset((unsigned char *) a + a->nla_len, 0, nla_padlen(len)); + + return 0; +} + +static inline int add_nested_action_start(struct sw_flow_actions **sfa, + int attrtype) +{ + int used = (*sfa)->actions_len; + int err; + + err = add_action(sfa, attrtype, NULL, 0); + if (err) + return err; + + return used; +} + +static inline void add_nested_action_end(struct sw_flow_actions *sfa, + int st_offset) +{ + struct nlattr *a = (struct nlattr *) ((unsigned char *)sfa->actions + + st_offset); + + a->nla_len = sfa->actions_len - st_offset; +} + +static int validate_and_copy_sample(const struct nlattr *attr, + const struct sw_flow_key *key, int depth, + struct sw_flow_actions **sfa) +{ + const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1]; + const struct nlattr *probability, *actions; + const struct nlattr *a; + int rem, start, err, st_acts; + + memset(attrs, 0, sizeof(attrs)); + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + if (!type || type > OVS_SAMPLE_ATTR_MAX || attrs[type]) + return -EINVAL; + attrs[type] = a; + } + if (rem) + return -EINVAL; + + probability = attrs[OVS_SAMPLE_ATTR_PROBABILITY]; + if (!probability || nla_len(probability) != sizeof(u32)) + return -EINVAL; + + actions = attrs[OVS_SAMPLE_ATTR_ACTIONS]; + if (!actions || (nla_len(actions) && nla_len(actions) < NLA_HDRLEN)) + return -EINVAL; + + /* validation done, copy sample action. */ + start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SAMPLE); + if (start < 0) + return start; + err = add_action(sfa, OVS_SAMPLE_ATTR_PROBABILITY, + nla_data(probability), sizeof(u32)); + if (err) + return err; + st_acts = add_nested_action_start(sfa, OVS_SAMPLE_ATTR_ACTIONS); + if (st_acts < 0) + return st_acts; + + err = ovs_nla_copy_actions(actions, key, depth + 1, sfa); + if (err) + return err; + + add_nested_action_end(*sfa, st_acts); + add_nested_action_end(*sfa, start); + + return 0; +} + +static int validate_tp_port(const struct sw_flow_key *flow_key) +{ + if (flow_key->eth.type == htons(ETH_P_IP)) { + if (flow_key->ipv4.tp.src || flow_key->ipv4.tp.dst) + return 0; + } else if (flow_key->eth.type == htons(ETH_P_IPV6)) { + if (flow_key->ipv6.tp.src || flow_key->ipv6.tp.dst) + return 0; + } + + return -EINVAL; +} + +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)); + + if (mask) { + memset(&mask->key, 0, sizeof(mask->key)); + mask->range.start = mask->range.end = 0; + } +} + +static int validate_and_copy_set_tun(const struct nlattr *attr, + struct sw_flow_actions **sfa) +{ + struct sw_flow_match match; + struct sw_flow_key key; + int err, start; + + ovs_match_init(&match, &key, NULL); + err = ipv4_tun_from_nlattr(nla_data(attr), &match, false); + if (err) + return err; + + start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SET); + if (start < 0) + return start; + + err = add_action(sfa, OVS_KEY_ATTR_IPV4_TUNNEL, &match.key->tun_key, + sizeof(match.key->tun_key)); + add_nested_action_end(*sfa, start); + + return err; +} + +static int validate_set(const struct nlattr *a, + const struct sw_flow_key *flow_key, + struct sw_flow_actions **sfa, + bool *set_tun) +{ + const struct nlattr *ovs_key = nla_data(a); + int key_type = nla_type(ovs_key); + + /* There can be only one key in a action */ + if (nla_total_size(nla_len(ovs_key)) != nla_len(a)) + return -EINVAL; + + if (key_type > OVS_KEY_ATTR_MAX || + (ovs_key_lens[key_type] != nla_len(ovs_key) && + ovs_key_lens[key_type] != -1)) + return -EINVAL; + + switch (key_type) { + const struct ovs_key_ipv4 *ipv4_key; + const struct ovs_key_ipv6 *ipv6_key; + int err; + + case OVS_KEY_ATTR_PRIORITY: + case OVS_KEY_ATTR_SKB_MARK: + case OVS_KEY_ATTR_ETHERNET: + break; + + case OVS_KEY_ATTR_TUNNEL: + *set_tun = true; + err = validate_and_copy_set_tun(a, sfa); + if (err) + return err; + break; + + case OVS_KEY_ATTR_IPV4: + if (flow_key->eth.type != htons(ETH_P_IP)) + return -EINVAL; + + if (!flow_key->ip.proto) + return -EINVAL; + + ipv4_key = nla_data(ovs_key); + if (ipv4_key->ipv4_proto != flow_key->ip.proto) + return -EINVAL; + + if (ipv4_key->ipv4_frag != flow_key->ip.frag) + return -EINVAL; + + break; + + case OVS_KEY_ATTR_IPV6: + if (flow_key->eth.type != htons(ETH_P_IPV6)) + return -EINVAL; + + if (!flow_key->ip.proto) + return -EINVAL; + + ipv6_key = nla_data(ovs_key); + if (ipv6_key->ipv6_proto != flow_key->ip.proto) + return -EINVAL; + + if (ipv6_key->ipv6_frag != flow_key->ip.frag) + return -EINVAL; + + if (ntohl(ipv6_key->ipv6_label) & 0xFFF00000) + return -EINVAL; + + break; + + case OVS_KEY_ATTR_TCP: + if (flow_key->ip.proto != IPPROTO_TCP) + return -EINVAL; + + return validate_tp_port(flow_key); + + case OVS_KEY_ATTR_UDP: + if (flow_key->ip.proto != IPPROTO_UDP) + return -EINVAL; + + return validate_tp_port(flow_key); + + case OVS_KEY_ATTR_SCTP: + if (flow_key->ip.proto != IPPROTO_SCTP) + return -EINVAL; + + return validate_tp_port(flow_key); + + default: + return -EINVAL; + } + + return 0; +} + +static int validate_userspace(const struct nlattr *attr) +{ + static const struct nla_policy userspace_policy[OVS_USERSPACE_ATTR_MAX + 1] = { + [OVS_USERSPACE_ATTR_PID] = {.type = NLA_U32 }, + [OVS_USERSPACE_ATTR_USERDATA] = {.type = NLA_UNSPEC }, + }; + struct nlattr *a[OVS_USERSPACE_ATTR_MAX + 1]; + int error; + + error = nla_parse_nested(a, OVS_USERSPACE_ATTR_MAX, + attr, userspace_policy); + if (error) + return error; + + if (!a[OVS_USERSPACE_ATTR_PID] || + !nla_get_u32(a[OVS_USERSPACE_ATTR_PID])) + return -EINVAL; + + return 0; +} + +static int copy_action(const struct nlattr *from, + struct sw_flow_actions **sfa) +{ + int totlen = NLA_ALIGN(from->nla_len); + struct nlattr *to; + + to = reserve_sfa_size(sfa, from->nla_len); + if (IS_ERR(to)) + return PTR_ERR(to); + + memcpy(to, from, totlen); + return 0; +} + +int ovs_nla_copy_actions(const struct nlattr *attr, + const struct sw_flow_key *key, + int depth, + struct sw_flow_actions **sfa) +{ + const struct nlattr *a; + int rem, err; + + if (depth >= SAMPLE_ACTION_DEPTH) + return -EOVERFLOW; + + nla_for_each_nested(a, attr, rem) { + /* Expected argument lengths, (u32)-1 for variable length. */ + static const u32 action_lens[OVS_ACTION_ATTR_MAX + 1] = { + [OVS_ACTION_ATTR_OUTPUT] = sizeof(u32), + [OVS_ACTION_ATTR_USERSPACE] = (u32)-1, + [OVS_ACTION_ATTR_PUSH_VLAN] = sizeof(struct ovs_action_push_vlan), + [OVS_ACTION_ATTR_POP_VLAN] = 0, + [OVS_ACTION_ATTR_SET] = (u32)-1, + [OVS_ACTION_ATTR_SAMPLE] = (u32)-1 + }; + const struct ovs_action_push_vlan *vlan; + int type = nla_type(a); + bool skip_copy; + + if (type > OVS_ACTION_ATTR_MAX || + (action_lens[type] != nla_len(a) && + action_lens[type] != (u32)-1)) + return -EINVAL; + + skip_copy = false; + switch (type) { + case OVS_ACTION_ATTR_UNSPEC: + return -EINVAL; + + case OVS_ACTION_ATTR_USERSPACE: + err = validate_userspace(a); + if (err) + return err; + break; + + case OVS_ACTION_ATTR_OUTPUT: + if (nla_get_u32(a) >= DP_MAX_PORTS) + return -EINVAL; + break; + + + case OVS_ACTION_ATTR_POP_VLAN: + break; + + case OVS_ACTION_ATTR_PUSH_VLAN: + vlan = nla_data(a); + if (vlan->vlan_tpid != htons(ETH_P_8021Q)) + return -EINVAL; + if (!(vlan->vlan_tci & htons(VLAN_TAG_PRESENT))) + return -EINVAL; + break; + + case OVS_ACTION_ATTR_SET: + err = validate_set(a, key, sfa, &skip_copy); + if (err) + return err; + break; + + case OVS_ACTION_ATTR_SAMPLE: + err = validate_and_copy_sample(a, key, depth, sfa); + if (err) + return err; + skip_copy = true; + break; + + default: + return -EINVAL; + } + if (!skip_copy) { + err = copy_action(a, sfa); + if (err) + return err; + } + } + + if (rem > 0) + return -EINVAL; + + return 0; +} + +static int sample_action_to_attr(const struct nlattr *attr, struct sk_buff *skb) +{ + const struct nlattr *a; + struct nlattr *start; + int err = 0, rem; + + start = nla_nest_start(skb, OVS_ACTION_ATTR_SAMPLE); + if (!start) + return -EMSGSIZE; + + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + struct nlattr *st_sample; + + switch (type) { + case OVS_SAMPLE_ATTR_PROBABILITY: + if (nla_put(skb, OVS_SAMPLE_ATTR_PROBABILITY, + sizeof(u32), nla_data(a))) + return -EMSGSIZE; + break; + case OVS_SAMPLE_ATTR_ACTIONS: + st_sample = nla_nest_start(skb, OVS_SAMPLE_ATTR_ACTIONS); + if (!st_sample) + return -EMSGSIZE; + err = ovs_nla_put_actions(nla_data(a), nla_len(a), skb); + if (err) + return err; + nla_nest_end(skb, st_sample); + break; + } + } + + nla_nest_end(skb, start); + return err; +} + +static int set_action_to_attr(const struct nlattr *a, struct sk_buff *skb) +{ + const struct nlattr *ovs_key = nla_data(a); + int key_type = nla_type(ovs_key); + struct nlattr *start; + int err; + + switch (key_type) { + case OVS_KEY_ATTR_IPV4_TUNNEL: + start = nla_nest_start(skb, OVS_ACTION_ATTR_SET); + if (!start) + return -EMSGSIZE; + + err = ipv4_tun_to_nlattr(skb, nla_data(ovs_key), + nla_data(ovs_key)); + if (err) + return err; + nla_nest_end(skb, start); + break; + default: + if (nla_put(skb, OVS_ACTION_ATTR_SET, nla_len(a), ovs_key)) + return -EMSGSIZE; + break; + } + + return 0; +} + +int ovs_nla_put_actions(const struct nlattr *attr, int len, struct sk_buff *skb) +{ + const struct nlattr *a; + int rem, err; + + nla_for_each_attr(a, attr, len, rem) { + int type = nla_type(a); + + switch (type) { + case OVS_ACTION_ATTR_SET: + err = set_action_to_attr(a, skb); + if (err) + return err; + break; + + case OVS_ACTION_ATTR_SAMPLE: + err = sample_action_to_attr(a, skb); + if (err) + return err; + break; + default: + if (nla_put(skb, type, nla_len(a), nla_data(a))) + return -EMSGSIZE; + break; + } + } + + return 0; +} diff --git a/net/openvswitch/flow_netlink.h b/net/openvswitch/flow_netlink.h new file mode 100644 index 00000000000..440151045d3 --- /dev/null +++ b/net/openvswitch/flow_netlink.h @@ -0,0 +1,60 @@ +/* + * Copyright (c) 2007-2013 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 + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA + */ + + +#ifndef FLOW_NETLINK_H +#define FLOW_NETLINK_H 1 + +#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/flex_array.h> + +#include <net/inet_ecn.h> +#include <net/ip_tunnels.h> + +#include "flow.h" + +void ovs_match_init(struct sw_flow_match *match, + struct sw_flow_key *key, struct sw_flow_mask *mask); + +int ovs_nla_put_flow(const struct sw_flow_key *, + const struct sw_flow_key *, struct sk_buff *); +int ovs_nla_get_flow_metadata(struct sw_flow *flow, + const struct nlattr *attr); +int ovs_nla_get_match(struct sw_flow_match *match, + const struct nlattr *, + const struct nlattr *); + +int ovs_nla_copy_actions(const struct nlattr *attr, + const struct sw_flow_key *key, int depth, + struct sw_flow_actions **sfa); +int ovs_nla_put_actions(const struct nlattr *attr, + int len, struct sk_buff *skb); + +struct sw_flow_actions *ovs_nla_alloc_flow_actions(int actions_len); +void ovs_nla_free_flow_actions(struct sw_flow_actions *); + +#endif /* flow_netlink.h */ diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c new file mode 100644 index 00000000000..e4254270608 --- /dev/null +++ b/net/openvswitch/flow_table.c @@ -0,0 +1,592 @@ +/* + * Copyright (c) 2007-2013 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 + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA + */ + +#include "flow.h" +#include "datapath.h" +#include <linux/uaccess.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/if_ether.h> +#include <linux/if_vlan.h> +#include <net/llc_pdu.h> +#include <linux/kernel.h> +#include <linux/jhash.h> +#include <linux/jiffies.h> +#include <linux/llc.h> +#include <linux/module.h> +#include <linux/in.h> +#include <linux/rcupdate.h> +#include <linux/if_arp.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/sctp.h> +#include <linux/tcp.h> +#include <linux/udp.h> +#include <linux/icmp.h> +#include <linux/icmpv6.h> +#include <linux/rculist.h> +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/ndisc.h> + +#include "datapath.h" + +#define TBL_MIN_BUCKETS 1024 +#define REHASH_INTERVAL (10 * 60 * HZ) + +static struct kmem_cache *flow_cache; + +static u16 range_n_bytes(const struct sw_flow_key_range *range) +{ + return range->end - range->start; +} + +void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src, + const struct sw_flow_mask *mask) +{ + const long *m = (long *)((u8 *)&mask->key + mask->range.start); + const long *s = (long *)((u8 *)src + mask->range.start); + long *d = (long *)((u8 *)dst + mask->range.start); + int i; + + /* The memory outside of the 'mask->range' are not set since + * further operations on 'dst' only uses contents within + * 'mask->range'. + */ + for (i = 0; i < range_n_bytes(&mask->range); i += sizeof(long)) + *d++ = *s++ & *m++; +} + +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; +} + +int ovs_flow_tbl_count(struct flow_table *table) +{ + return table->count; +} + +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 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); +} + +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); +} + +static void 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); + } +} + +void ovs_flow_free(struct sw_flow *flow, bool deferred) +{ + if (!flow) + return; + + flow_mask_del_ref(flow->mask, deferred); + + if (deferred) + call_rcu(&flow->rcu, rcu_free_flow_callback); + else + flow_free(flow); +} + +static void free_buckets(struct flex_array *buckets) +{ + flex_array_free(buckets); +} + +static void __table_instance_destroy(struct table_instance *ti) +{ + int i; + + if (ti->keep_flows) + goto skip_flows; + + for (i = 0; i < ti->n_buckets; i++) { + struct sw_flow *flow; + struct hlist_head *head = flex_array_get(ti->buckets, i); + struct hlist_node *n; + int ver = ti->node_ver; + + hlist_for_each_entry_safe(flow, n, head, hash_node[ver]) { + hlist_del(&flow->hash_node[ver]); + ovs_flow_free(flow, false); + } + } + +skip_flows: + free_buckets(ti->buckets); + kfree(ti); +} + +static struct table_instance *table_instance_alloc(int new_size) +{ + struct table_instance *ti = kmalloc(sizeof(*ti), GFP_KERNEL); + + if (!ti) + return NULL; + + ti->buckets = alloc_buckets(new_size); + + if (!ti->buckets) { + kfree(ti); + return NULL; + } + ti->n_buckets = new_size; + ti->node_ver = 0; + ti->keep_flows = false; + get_random_bytes(&ti->hash_seed, sizeof(u32)); + + return ti; +} + +int ovs_flow_tbl_init(struct flow_table *table) +{ + struct table_instance *ti; + + ti = table_instance_alloc(TBL_MIN_BUCKETS); + + if (!ti) + return -ENOMEM; + + rcu_assign_pointer(table->ti, ti); + INIT_LIST_HEAD(&table->mask_list); + table->last_rehash = jiffies; + table->count = 0; + return 0; +} + +static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu) +{ + struct table_instance *ti = container_of(rcu, struct table_instance, rcu); + + __table_instance_destroy(ti); +} + +static void table_instance_destroy(struct table_instance *ti, bool deferred) +{ + if (!ti) + return; + + if (deferred) + call_rcu(&ti->rcu, flow_tbl_destroy_rcu_cb); + else + __table_instance_destroy(ti); +} + +void ovs_flow_tbl_destroy(struct flow_table *table) +{ + struct table_instance *ti = ovsl_dereference(table->ti); + + table_instance_destroy(ti, false); +} + +struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *ti, + u32 *bucket, u32 *last) +{ + struct sw_flow *flow; + struct hlist_head *head; + int ver; + int i; + + ver = ti->node_ver; + while (*bucket < ti->n_buckets) { + i = 0; + head = flex_array_get(ti->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 struct hlist_head *find_bucket(struct table_instance *ti, u32 hash) +{ + hash = jhash_1word(hash, ti->hash_seed); + return flex_array_get(ti->buckets, + (hash & (ti->n_buckets - 1))); +} + +static void table_instance_insert(struct table_instance *ti, struct sw_flow *flow) +{ + struct hlist_head *head; + + head = find_bucket(ti, flow->hash); + hlist_add_head_rcu(&flow->hash_node[ti->node_ver], head); +} + +static void flow_table_copy_flows(struct table_instance *old, + struct table_instance *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]) + table_instance_insert(new, flow); + } + + old->keep_flows = true; +} + +static struct table_instance *table_instance_rehash(struct table_instance *ti, + int n_buckets) +{ + struct table_instance *new_ti; + + new_ti = table_instance_alloc(n_buckets); + if (!new_ti) + return NULL; + + flow_table_copy_flows(ti, new_ti); + + return new_ti; +} + +int ovs_flow_tbl_flush(struct flow_table *flow_table) +{ + struct table_instance *old_ti; + struct table_instance *new_ti; + + old_ti = ovsl_dereference(flow_table->ti); + new_ti = table_instance_alloc(TBL_MIN_BUCKETS); + if (!new_ti) + return -ENOMEM; + + rcu_assign_pointer(flow_table->ti, new_ti); + flow_table->last_rehash = jiffies; + flow_table->count = 0; + + table_instance_destroy(old_ti, true); + return 0; +} + +static u32 flow_hash(const struct sw_flow_key *key, int key_start, + int key_end) +{ + u32 *hash_key = (u32 *)((u8 *)key + key_start); + int hash_u32s = (key_end - key_start) >> 2; + + /* Make sure number of hash bytes are multiple of u32. */ + BUILD_BUG_ON(sizeof(long) % sizeof(u32)); + + return jhash2(hash_key, hash_u32s, 0); +} + +static int flow_key_start(const struct sw_flow_key *key) +{ + if (key->tun_key.ipv4_dst) + return 0; + else + return rounddown(offsetof(struct sw_flow_key, phy), + sizeof(long)); +} + +static bool cmp_key(const struct sw_flow_key *key1, + const struct sw_flow_key *key2, + int key_start, int key_end) +{ + const long *cp1 = (long *)((u8 *)key1 + key_start); + const long *cp2 = (long *)((u8 *)key2 + key_start); + long diffs = 0; + int i; + + for (i = key_start; i < key_end; i += sizeof(long)) + diffs |= *cp1++ ^ *cp2++; + + return diffs == 0; +} + +static bool flow_cmp_masked_key(const struct sw_flow *flow, + const struct sw_flow_key *key, + int key_start, int key_end) +{ + return cmp_key(&flow->key, key, key_start, key_end); +} + +bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow, + struct sw_flow_match *match) +{ + struct sw_flow_key *key = match->key; + int key_start = flow_key_start(key); + int key_end = match->range.end; + + return cmp_key(&flow->unmasked_key, key, key_start, key_end); +} + +static struct sw_flow *masked_flow_lookup(struct table_instance *ti, + const struct sw_flow_key *unmasked, + struct sw_flow_mask *mask) +{ + struct sw_flow *flow; + struct hlist_head *head; + int key_start = mask->range.start; + int key_end = mask->range.end; + u32 hash; + struct sw_flow_key masked_key; + + ovs_flow_mask_key(&masked_key, unmasked, mask); + hash = flow_hash(&masked_key, key_start, key_end); + head = find_bucket(ti, hash); + hlist_for_each_entry_rcu(flow, head, hash_node[ti->node_ver]) { + if (flow->mask == mask && flow->hash == hash && + flow_cmp_masked_key(flow, &masked_key, + key_start, key_end)) + return flow; + } + return NULL; +} + +struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *tbl, + const struct sw_flow_key *key, + u32 *n_mask_hit) +{ + struct table_instance *ti = rcu_dereference(tbl->ti); + struct sw_flow_mask *mask; + struct sw_flow *flow; + + *n_mask_hit = 0; + list_for_each_entry_rcu(mask, &tbl->mask_list, list) { + (*n_mask_hit)++; + flow = masked_flow_lookup(ti, key, mask); + if (flow) /* Found */ + return flow; + } + return NULL; +} + +int ovs_flow_tbl_num_masks(const struct flow_table *table) +{ + struct sw_flow_mask *mask; + int num = 0; + + list_for_each_entry(mask, &table->mask_list, list) + num++; + + return num; +} + +static struct table_instance *table_instance_expand(struct table_instance *ti) +{ + return table_instance_rehash(ti, ti->n_buckets * 2); +} + +void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow) +{ + struct table_instance *ti = ovsl_dereference(table->ti); + + BUG_ON(table->count == 0); + hlist_del_rcu(&flow->hash_node[ti->node_ver]); + table->count--; +} + +static struct sw_flow_mask *mask_alloc(void) +{ + struct sw_flow_mask *mask; + + mask = kmalloc(sizeof(*mask), GFP_KERNEL); + if (mask) + mask->ref_count = 0; + + return mask; +} + +static void mask_add_ref(struct sw_flow_mask *mask) +{ + mask->ref_count++; +} + +static bool 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_, range_n_bytes(&a->range)) == 0); +} + +static struct sw_flow_mask *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 (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. + */ +static int flow_mask_insert(struct flow_table *tbl, struct sw_flow *flow, + struct sw_flow_mask *new) +{ + struct sw_flow_mask *mask; + mask = flow_mask_find(tbl, new); + if (!mask) { + /* Allocate a new mask if none exsits. */ + mask = mask_alloc(); + if (!mask) + return -ENOMEM; + mask->key = new->key; + mask->range = new->range; + list_add_rcu(&mask->list, &tbl->mask_list); + } + + mask_add_ref(mask); + flow->mask = mask; + return 0; +} + +int ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow, + struct sw_flow_mask *mask) +{ + struct table_instance *new_ti = NULL; + struct table_instance *ti; + int err; + + err = flow_mask_insert(table, flow, mask); + if (err) + return err; + + flow->hash = flow_hash(&flow->key, flow->mask->range.start, + flow->mask->range.end); + ti = ovsl_dereference(table->ti); + table_instance_insert(ti, flow); + table->count++; + + /* Expand table, if necessary, to make room. */ + if (table->count > ti->n_buckets) + new_ti = table_instance_expand(ti); + else if (time_after(jiffies, table->last_rehash + REHASH_INTERVAL)) + new_ti = table_instance_rehash(ti, ti->n_buckets); + + if (new_ti) { + rcu_assign_pointer(table->ti, new_ti); + table_instance_destroy(ti, true); + table->last_rehash = jiffies; + } + return 0; +} + +/* Initializes the flow module. + * Returns zero if successful or a negative error code. */ +int ovs_flow_init(void) +{ + BUILD_BUG_ON(__alignof__(struct sw_flow_key) % __alignof__(long)); + BUILD_BUG_ON(sizeof(struct sw_flow_key) % sizeof(long)); + + 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); +} diff --git a/net/openvswitch/flow_table.h b/net/openvswitch/flow_table.h new file mode 100644 index 00000000000..fbe45d5ad07 --- /dev/null +++ b/net/openvswitch/flow_table.h @@ -0,0 +1,81 @@ +/* + * Copyright (c) 2007-2013 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 + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA + */ + +#ifndef FLOW_TABLE_H +#define FLOW_TABLE_H 1 + +#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/flex_array.h> + +#include <net/inet_ecn.h> +#include <net/ip_tunnels.h> + +#include "flow.h" + +struct table_instance { + struct flex_array *buckets; + unsigned int n_buckets; + struct rcu_head rcu; + int node_ver; + u32 hash_seed; + bool keep_flows; +}; + +struct flow_table { + struct table_instance __rcu *ti; + struct list_head mask_list; + unsigned long last_rehash; + unsigned int count; +}; + +int ovs_flow_init(void); +void ovs_flow_exit(void); + +struct sw_flow *ovs_flow_alloc(void); +void ovs_flow_free(struct sw_flow *, bool deferred); + +int ovs_flow_tbl_init(struct flow_table *); +int ovs_flow_tbl_count(struct flow_table *table); +void ovs_flow_tbl_destroy(struct flow_table *table); +int ovs_flow_tbl_flush(struct flow_table *flow_table); + +int ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow, + struct sw_flow_mask *mask); +void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow); +int ovs_flow_tbl_num_masks(const struct flow_table *table); +struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *table, + u32 *bucket, u32 *idx); +struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *, + const struct sw_flow_key *, + u32 *n_mask_hit); + +bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow, + struct sw_flow_match *match); + +void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src, + const struct sw_flow_mask *mask); +#endif /* flow_table.h */ diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c index c99dea543d6..a3d6951602d 100644 --- a/net/openvswitch/vport-gre.c +++ b/net/openvswitch/vport-gre.c @@ -24,8 +24,6 @@ #include <linux/if_tunnel.h> #include <linux/if_vlan.h> #include <linux/in.h> -#include <linux/if_vlan.h> -#include <linux/in.h> #include <linux/in_route.h> #include <linux/inetdevice.h> #include <linux/jhash.h> diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c index 98d3edbbc23..729c68763fe 100644 --- a/net/openvswitch/vport-internal_dev.c +++ b/net/openvswitch/vport-internal_dev.c @@ -134,7 +134,7 @@ static void do_setup(struct net_device *netdev) netdev->tx_queue_len = 0; netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST | - NETIF_F_HIGHDMA | NETIF_F_HW_CSUM | NETIF_F_TSO; + NETIF_F_HIGHDMA | NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE; netdev->vlan_features = netdev->features; netdev->features |= NETIF_F_HW_VLAN_CTAG_TX; diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c index 56e22b74cf9..e797a50ac2b 100644 --- a/net/openvswitch/vport-vxlan.c +++ b/net/openvswitch/vport-vxlan.c @@ -29,7 +29,6 @@ #include <net/ip.h> #include <net/udp.h> #include <net/ip_tunnels.h> -#include <net/udp.h> #include <net/rtnetlink.h> #include <net/route.h> #include <net/dsfield.h> |