--- zzzz-none-000/linux-3.10.107/net/openvswitch/datapath.c 2017-06-27 09:49:32.000000000 +0000 +++ scorpion-7490-727/linux-3.10.107/net/openvswitch/datapath.c 2021-02-04 17:41:59.000000000 +0000 @@ -1,5 +1,5 @@ /* - * Copyright (c) 2007-2012 Nicira, Inc. + * Copyright (c) 2007-2014 Nicira, Inc. * * This program is free software; you can redistribute it and/or * modify it under the terms of version 2 of the GNU General Public @@ -44,32 +44,54 @@ #include #include #include -#include #include #include #include -#include #include #include #include #include "datapath.h" #include "flow.h" +#include "flow_table.h" +#include "flow_netlink.h" #include "vport-internal_dev.h" #include "vport-netdev.h" +int ovs_net_id __read_mostly; +EXPORT_SYMBOL_GPL(ovs_net_id); -#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); +static struct genl_family dp_packet_genl_family; +static struct genl_family dp_flow_genl_family; +static struct genl_family dp_datapath_genl_family; -int ovs_net_id __read_mostly; +static const struct nla_policy flow_policy[]; + +static const struct genl_multicast_group ovs_dp_flow_multicast_group = { + .name = OVS_FLOW_MCGROUP, +}; + +static const struct genl_multicast_group ovs_dp_datapath_multicast_group = { + .name = OVS_DATAPATH_MCGROUP, +}; -static void ovs_notify(struct sk_buff *skb, struct genl_info *info, - struct genl_multicast_group *grp) +static const struct genl_multicast_group ovs_dp_vport_multicast_group = { + .name = OVS_VPORT_MCGROUP, +}; + +/* Check if need to build a reply message. + * OVS userspace sets the NLM_F_ECHO flag if it needs the reply. */ +static bool ovs_must_notify(struct genl_family *family, struct genl_info *info, + unsigned int group) +{ + return info->nlhdr->nlmsg_flags & NLM_F_ECHO || + genl_has_listeners(family, genl_info_net(info), group); +} + +static void ovs_notify(struct genl_family *family, + struct sk_buff *skb, struct genl_info *info) { - genl_notify(skb, genl_info_net(info), info->snd_portid, - grp->id, info->nlhdr, GFP_KERNEL); + genl_notify(family, skb, info, 0, GFP_KERNEL); } /** @@ -109,28 +131,41 @@ else return 1; } +EXPORT_SYMBOL_GPL(lockdep_ovsl_is_held); #endif static struct vport *new_vport(const struct vport_parms *); -static int queue_gso_packets(struct net *, int dp_ifindex, struct sk_buff *, +static int queue_gso_packets(struct datapath *dp, struct sk_buff *, + const struct sw_flow_key *, const struct dp_upcall_info *); -static int queue_userspace_packet(struct net *, int dp_ifindex, - struct sk_buff *, +static int queue_userspace_packet(struct datapath *dp, struct sk_buff *, + const struct sw_flow_key *, const struct dp_upcall_info *); -/* Must be called with rcu_read_lock or ovs_mutex. */ -static struct datapath *get_dp(struct net *net, int dp_ifindex) +/* Must be called with rcu_read_lock. */ +static struct datapath *get_dp_rcu(struct net *net, int dp_ifindex) { - struct datapath *dp = NULL; - struct net_device *dev; + struct net_device *dev = dev_get_by_index_rcu(net, dp_ifindex); - rcu_read_lock(); - dev = dev_get_by_index_rcu(net, dp_ifindex); if (dev) { struct vport *vport = ovs_internal_dev_get_vport(dev); if (vport) - dp = vport->dp; + return vport->dp; } + + return NULL; +} + +/* The caller must hold either ovs_mutex or rcu_read_lock to keep the + * returned dp pointer valid. + */ +static inline struct datapath *get_dp(struct net *net, int dp_ifindex) +{ + struct datapath *dp; + + WARN_ON_ONCE(!rcu_read_lock_held() && !lockdep_ovsl_is_held()); + rcu_read_lock(); + dp = get_dp_rcu(net, dp_ifindex); rcu_read_unlock(); return dp; @@ -140,10 +175,10 @@ const char *ovs_dp_name(const struct datapath *dp) { struct vport *vport = ovs_vport_ovsl_rcu(dp, OVSP_LOCAL); - return vport->ops->get_name(vport); + return ovs_vport_name(vport); } -static int get_dpifindex(struct datapath *dp) +static int get_dpifindex(const struct datapath *dp) { struct vport *local; int ifindex; @@ -152,7 +187,7 @@ local = ovs_vport_rcu(dp, OVSP_LOCAL); if (local) - ifindex = netdev_vport_priv(local)->dev->ifindex; + ifindex = local->dev->ifindex; else ifindex = 0; @@ -165,9 +200,8 @@ { struct datapath *dp = container_of(rcu, struct datapath, rcu); - ovs_flow_tbl_destroy((__force struct flow_table *)dp->table); + ovs_flow_tbl_destroy(&dp->table); free_percpu(dp->stats_percpu); - release_net(ovs_dp_get_net(dp)); kfree(dp->ports); kfree(dp); } @@ -178,6 +212,7 @@ return &dp->ports[port_no & (DP_VPORT_HASH_BUCKETS - 1)]; } +/* Called with ovs_mutex or RCU read lock. */ struct vport *ovs_lookup_vport(const struct datapath *dp, u16 port_no) { struct vport *vport; @@ -218,68 +253,56 @@ } /* Must be called with rcu_read_lock. */ -void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb) +void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) { + const struct vport *p = OVS_CB(skb)->input_vport; struct datapath *dp = p->dp; struct sw_flow *flow; + struct sw_flow_actions *sf_acts; struct dp_stats_percpu *stats; - struct sw_flow_key key; u64 *stats_counter; - int error; - int key_len; + u32 n_mask_hit; stats = this_cpu_ptr(dp->stats_percpu); - /* Extract flow from 'skb' into 'key'. */ - error = ovs_flow_extract(skb, p->port_no, &key, &key_len); - if (unlikely(error)) { - kfree_skb(skb); - return; - } - /* Look up flow. */ - flow = ovs_flow_tbl_lookup(rcu_dereference(dp->table), &key, key_len); + flow = ovs_flow_tbl_lookup_stats(&dp->table, key, &n_mask_hit); if (unlikely(!flow)) { struct dp_upcall_info upcall; + int error; + memset(&upcall, 0, sizeof(upcall)); upcall.cmd = OVS_PACKET_CMD_MISS; - upcall.key = &key; - upcall.userdata = NULL; - upcall.portid = p->upcall_portid; - ovs_dp_upcall(dp, skb, &upcall); - consume_skb(skb); + upcall.portid = ovs_vport_find_upcall_portid(p, skb); + upcall.mru = OVS_CB(skb)->mru; + error = ovs_dp_upcall(dp, skb, key, &upcall); + if (unlikely(error)) + kfree_skb(skb); + else + consume_skb(skb); stats_counter = &stats->n_missed; goto out; } - OVS_CB(skb)->flow = flow; + ovs_flow_stats_update(flow, key->tp.flags, skb); + sf_acts = rcu_dereference(flow->sf_acts); + ovs_execute_actions(dp, skb, sf_acts, key); stats_counter = &stats->n_hit; - ovs_flow_used(OVS_CB(skb)->flow, skb); - ovs_execute_actions(dp, skb); out: /* Update datapath statistics. */ - u64_stats_update_begin(&stats->sync); + u64_stats_update_begin(&stats->syncp); (*stats_counter)++; - u64_stats_update_end(&stats->sync); + stats->n_mask_hit += n_mask_hit; + u64_stats_update_end(&stats->syncp); } -static struct genl_family dp_packet_genl_family = { - .id = GENL_ID_GENERATE, - .hdrsize = sizeof(struct ovs_header), - .name = OVS_PACKET_FAMILY, - .version = OVS_PACKET_VERSION, - .maxattr = OVS_PACKET_ATTR_MAX, - .netnsok = true, - .parallel_ops = true, -}; - int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb, + const struct sw_flow_key *key, const struct dp_upcall_info *upcall_info) { struct dp_stats_percpu *stats; - int dp_ifindex; int err; if (upcall_info->portid == 0) { @@ -287,16 +310,10 @@ goto err; } - dp_ifindex = get_dpifindex(dp); - if (!dp_ifindex) { - err = -ENODEV; - goto err; - } - if (!skb_is_gso(skb)) - err = queue_userspace_packet(ovs_dp_get_net(dp), dp_ifindex, skb, upcall_info); + err = queue_userspace_packet(dp, skb, key, upcall_info); else - err = queue_gso_packets(ovs_dp_get_net(dp), dp_ifindex, skb, upcall_info); + err = queue_gso_packets(dp, skb, key, upcall_info); if (err) goto err; @@ -305,46 +322,48 @@ err: stats = this_cpu_ptr(dp->stats_percpu); - u64_stats_update_begin(&stats->sync); + u64_stats_update_begin(&stats->syncp); stats->n_lost++; - u64_stats_update_end(&stats->sync); + u64_stats_update_end(&stats->syncp); return err; } -static int queue_gso_packets(struct net *net, int dp_ifindex, - struct sk_buff *skb, +static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, + const struct sw_flow_key *key, const struct dp_upcall_info *upcall_info) { unsigned short gso_type = skb_shinfo(skb)->gso_type; - struct dp_upcall_info later_info; struct sw_flow_key later_key; struct sk_buff *segs, *nskb; int err; - segs = __skb_gso_segment(skb, NETIF_F_SG | NETIF_F_HW_CSUM, false); + BUILD_BUG_ON(sizeof(*OVS_CB(skb)) > SKB_SGO_CB_OFFSET); + segs = __skb_gso_segment(skb, NETIF_F_SG, false); if (IS_ERR(segs)) return PTR_ERR(segs); + if (segs == NULL) + return -EINVAL; + + if (gso_type & SKB_GSO_UDP) { + /* The initial flow key extracted by ovs_flow_key_extract() + * in this case is for a first fragment, so we need to + * properly mark later fragments. + */ + later_key = *key; + later_key.ip.frag = OVS_FRAG_TYPE_LATER; + } /* Queue all of the segments. */ skb = segs; do { - err = queue_userspace_packet(net, dp_ifindex, skb, upcall_info); + if (gso_type & SKB_GSO_UDP && skb != segs) + key = &later_key; + + err = queue_userspace_packet(dp, skb, key, upcall_info); if (err) break; - if (skb == segs && gso_type & SKB_GSO_UDP) { - /* The initial flow key extracted by ovs_flow_extract() - * in this case is for a first fragment, so we need to - * properly mark later fragments. - */ - later_key = *upcall_info->key; - later_key.ip.frag = OVS_FRAG_TYPE_LATER; - - later_info = *upcall_info; - later_info.key = &later_key; - upcall_info = &later_info; - } } while ((skb = skb->next)); /* Free all of the segments. */ @@ -359,55 +378,71 @@ return err; } -static size_t key_attr_size(void) -{ - return nla_total_size(4) /* OVS_KEY_ATTR_PRIORITY */ - + nla_total_size(4) /* OVS_KEY_ATTR_IN_PORT */ - + nla_total_size(4) /* OVS_KEY_ATTR_SKB_MARK */ - + nla_total_size(12) /* OVS_KEY_ATTR_ETHERNET */ - + nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */ - + nla_total_size(4) /* OVS_KEY_ATTR_8021Q */ - + nla_total_size(0) /* OVS_KEY_ATTR_ENCAP */ - + nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */ - + nla_total_size(40) /* OVS_KEY_ATTR_IPV6 */ - + nla_total_size(2) /* OVS_KEY_ATTR_ICMPV6 */ - + nla_total_size(28); /* OVS_KEY_ATTR_ND */ -} - -static size_t upcall_msg_size(const struct sk_buff *skb, - const struct nlattr *userdata) +static size_t upcall_msg_size(const struct dp_upcall_info *upcall_info, + unsigned int hdrlen) { size_t size = NLMSG_ALIGN(sizeof(struct ovs_header)) - + nla_total_size(skb->len) /* OVS_PACKET_ATTR_PACKET */ - + nla_total_size(key_attr_size()); /* OVS_PACKET_ATTR_KEY */ + + nla_total_size(hdrlen) /* OVS_PACKET_ATTR_PACKET */ + + nla_total_size(ovs_key_attr_size()); /* OVS_PACKET_ATTR_KEY */ /* OVS_PACKET_ATTR_USERDATA */ - if (userdata) - size += NLA_ALIGN(userdata->nla_len); + if (upcall_info->userdata) + size += NLA_ALIGN(upcall_info->userdata->nla_len); + + /* OVS_PACKET_ATTR_EGRESS_TUN_KEY */ + if (upcall_info->egress_tun_info) + size += nla_total_size(ovs_tun_key_attr_size()); + + /* OVS_PACKET_ATTR_ACTIONS */ + if (upcall_info->actions_len) + size += nla_total_size(upcall_info->actions_len); + + /* OVS_PACKET_ATTR_MRU */ + if (upcall_info->mru) + size += nla_total_size(sizeof(upcall_info->mru)); return size; } -static int queue_userspace_packet(struct net *net, int dp_ifindex, - struct sk_buff *skb, +static void pad_packet(struct datapath *dp, struct sk_buff *skb) +{ + if (!(dp->user_features & OVS_DP_F_UNALIGNED)) { + size_t plen = NLA_ALIGN(skb->len) - skb->len; + + if (plen > 0) + memset(skb_put(skb, plen), 0, plen); + } +} + +static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, + const struct sw_flow_key *key, const struct dp_upcall_info *upcall_info) { struct ovs_header *upcall; struct sk_buff *nskb = NULL; - struct sk_buff *user_skb; /* to be queued to userspace */ + struct sk_buff *user_skb = NULL; /* to be queued to userspace */ struct nlattr *nla; - int err; + struct genl_info info = { + .dst_sk = ovs_dp_get_net(dp)->genl_sock, + .snd_portid = upcall_info->portid, + }; + size_t len; + unsigned int hlen; + int err, dp_ifindex; - if (vlan_tx_tag_present(skb)) { + dp_ifindex = get_dpifindex(dp); + if (!dp_ifindex) + return -ENODEV; + + if (skb_vlan_tag_present(skb)) { nskb = skb_clone(skb, GFP_ATOMIC); if (!nskb) return -ENOMEM; - nskb = __vlan_put_tag(nskb, nskb->vlan_proto, vlan_tx_tag_get(nskb)); + nskb = __vlan_hwaccel_push_inside(nskb); if (!nskb) return -ENOMEM; - nskb->vlan_tci = 0; skb = nskb; } @@ -416,7 +451,22 @@ goto out; } - user_skb = genlmsg_new(upcall_msg_size(skb, upcall_info->userdata), GFP_ATOMIC); + /* Complete checksum if needed */ + if (skb->ip_summed == CHECKSUM_PARTIAL && + (err = skb_checksum_help(skb))) + goto out; + + /* Older versions of OVS user space enforce alignment of the last + * Netlink attribute to NLA_ALIGNTO which would require extensive + * padding logic. Only perform zerocopy if padding is not required. + */ + if (dp->user_features & OVS_DP_F_UNALIGNED) + hlen = skb_zerocopy_headlen(skb); + else + hlen = skb->len; + + len = upcall_msg_size(upcall_info, hlen); + user_skb = genlmsg_new_unicast(len, &info, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; goto out; @@ -426,283 +476,86 @@ 0, upcall_info->cmd); upcall->dp_ifindex = dp_ifindex; - nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY); - ovs_flow_to_nlattrs(upcall_info->key, user_skb); - nla_nest_end(user_skb, nla); + err = ovs_nla_put_key(key, key, OVS_PACKET_ATTR_KEY, false, user_skb); + BUG_ON(err); if (upcall_info->userdata) __nla_put(user_skb, OVS_PACKET_ATTR_USERDATA, nla_len(upcall_info->userdata), nla_data(upcall_info->userdata)); - nla = __nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, skb->len); - - skb_copy_and_csum_dev(skb, nla_data(nla)); - - genlmsg_end(user_skb, upcall); - err = genlmsg_unicast(net, user_skb, upcall_info->portid); - -out: - kfree_skb(nskb); - 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_deferred_destroy(old_table); - return 0; -} - -static int validate_actions(const struct nlattr *attr, - const struct sw_flow_key *key, int depth); - -static int validate_sample(const struct nlattr *attr, - const struct sw_flow_key *key, int depth) -{ - const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1]; - const struct nlattr *probability, *actions; - const struct nlattr *a; - int rem; - - 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 (upcall_info->egress_tun_info) { + nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_EGRESS_TUN_KEY); + err = ovs_nla_put_tunnel_info(user_skb, + upcall_info->egress_tun_info); + BUG_ON(err); + nla_nest_end(user_skb, nla); } - 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; - return validate_actions(actions, key, depth + 1); -} -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; + if (upcall_info->actions_len) { + nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_ACTIONS); + err = ovs_nla_put_actions(upcall_info->actions, + upcall_info->actions_len, + user_skb); + if (!err) + nla_nest_end(user_skb, nla); + else + nla_nest_cancel(user_skb, nla); } - return -EINVAL; -} - -static int validate_set(const struct nlattr *a, - const struct sw_flow_key *flow_key) -{ - 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 || - nla_len(ovs_key) != ovs_key_lens[key_type]) - return -EINVAL; - - switch (key_type) { - const struct ovs_key_ipv4 *ipv4_key; - const struct ovs_key_ipv6 *ipv6_key; - - case OVS_KEY_ATTR_PRIORITY: - case OVS_KEY_ATTR_SKB_MARK: - case OVS_KEY_ATTR_ETHERNET: - 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); - - default: - return -EINVAL; + /* Add OVS_PACKET_ATTR_MRU */ + if (upcall_info->mru) { + if (nla_put_u16(user_skb, OVS_PACKET_ATTR_MRU, + upcall_info->mru)) { + err = -ENOBUFS; + goto out; + } + pad_packet(dp, user_skb); } - 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 validate_actions(const struct nlattr *attr, - const struct sw_flow_key *key, int depth) -{ - 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); - - if (type > OVS_ACTION_ATTR_MAX || - (action_lens[type] != nla_len(a) && - action_lens[type] != (u32)-1)) - return -EINVAL; - - 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); - if (err) - return err; - break; - - case OVS_ACTION_ATTR_SAMPLE: - err = validate_sample(a, key, depth); - if (err) - return err; - break; - - default: - return -EINVAL; - } + /* Only reserve room for attribute header, packet data is added + * in skb_zerocopy() */ + if (!(nla = nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, 0))) { + err = -ENOBUFS; + goto out; } + nla->nla_len = nla_attr_size(skb->len); - if (rem > 0) - return -EINVAL; + err = skb_zerocopy(user_skb, skb, skb->len, hlen); + if (err) + goto out; - return 0; -} + /* Pad OVS_PACKET_ATTR_PACKET if linear copy was performed */ + pad_packet(dp, user_skb); -static void clear_stats(struct sw_flow *flow) -{ - flow->used = 0; - flow->tcp_flags = 0; - flow->packet_count = 0; - flow->byte_count = 0; + ((struct nlmsghdr *) user_skb->data)->nlmsg_len = user_skb->len; + + err = genlmsg_unicast(ovs_dp_get_net(dp), user_skb, upcall_info->portid); + user_skb = NULL; +out: + if (err) + skb_tx_error(skb); + kfree_skb(user_skb); + kfree_skb(nskb); + return err; } static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) { struct ovs_header *ovs_header = info->userhdr; + struct net *net = sock_net(skb->sk); struct nlattr **a = info->attrs; struct sw_flow_actions *acts; struct sk_buff *packet; struct sw_flow *flow; + struct sw_flow_actions *sf_acts; struct datapath *dp; struct ethhdr *eth; + struct vport *input_vport; + u16 mru = 0; int len; int err; - int key_len; + bool log = !a[OVS_PACKET_ATTR_PROBE]; err = -EINVAL; if (!a[OVS_PACKET_ATTR_PACKET] || !a[OVS_PACKET_ATTR_KEY] || @@ -724,62 +577,67 @@ /* Normally, setting the skb 'protocol' field would be handled by a * call to eth_type_trans(), but it assumes there's a sending * device, which we may not have. */ - if (ntohs(eth->h_proto) >= ETH_P_802_3_MIN) + if (eth_proto_is_802_3(eth->h_proto)) packet->protocol = eth->h_proto; else packet->protocol = htons(ETH_P_802_2); + /* Set packet's mru */ + if (a[OVS_PACKET_ATTR_MRU]) { + mru = nla_get_u16(a[OVS_PACKET_ATTR_MRU]); + packet->ignore_df = 1; + } + OVS_CB(packet)->mru = mru; + /* Build an sw_flow for sending this packet. */ flow = ovs_flow_alloc(); err = PTR_ERR(flow); if (IS_ERR(flow)) goto err_kfree_skb; - err = ovs_flow_extract(packet, -1, &flow->key, &key_len); + err = ovs_flow_key_extract_userspace(net, a[OVS_PACKET_ATTR_KEY], + packet, &flow->key, log); if (err) goto err_flow_free; - err = ovs_flow_metadata_from_nlattrs(&flow->key.phy.priority, - &flow->key.phy.skb_mark, - &flow->key.phy.in_port, - a[OVS_PACKET_ATTR_KEY]); + err = ovs_nla_copy_actions(net, a[OVS_PACKET_ATTR_ACTIONS], + &flow->key, &acts, log); if (err) goto err_flow_free; - err = validate_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0); - if (err) - goto err_flow_free; - - flow->hash = ovs_flow_hash(&flow->key, key_len); - - acts = ovs_flow_actions_alloc(a[OVS_PACKET_ATTR_ACTIONS]); - err = PTR_ERR(acts); - if (IS_ERR(acts)) - goto err_flow_free; rcu_assign_pointer(flow->sf_acts, acts); - - OVS_CB(packet)->flow = flow; packet->priority = flow->key.phy.priority; packet->mark = flow->key.phy.skb_mark; rcu_read_lock(); - dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); + dp = get_dp_rcu(net, ovs_header->dp_ifindex); err = -ENODEV; if (!dp) goto err_unlock; + input_vport = ovs_vport_rcu(dp, flow->key.phy.in_port); + if (!input_vport) + input_vport = ovs_vport_rcu(dp, OVSP_LOCAL); + + if (!input_vport) + goto err_unlock; + + packet->dev = input_vport->dev; + OVS_CB(packet)->input_vport = input_vport; + sf_acts = rcu_dereference(flow->sf_acts); + local_bh_disable(); - err = ovs_execute_actions(dp, packet); + err = ovs_execute_actions(dp, packet, sf_acts, &flow->key); local_bh_enable(); rcu_read_unlock(); - ovs_flow_free(flow); + ovs_flow_free(flow, false); return err; err_unlock: rcu_read_unlock(); err_flow_free: - ovs_flow_free(flow); + ovs_flow_free(flow, false); err_kfree_skb: kfree_skb(packet); err: @@ -790,9 +648,11 @@ [OVS_PACKET_ATTR_PACKET] = { .len = ETH_HLEN }, [OVS_PACKET_ATTR_KEY] = { .type = NLA_NESTED }, [OVS_PACKET_ATTR_ACTIONS] = { .type = NLA_NESTED }, + [OVS_PACKET_ATTR_PROBE] = { .type = NLA_FLAG }, + [OVS_PACKET_ATTR_MRU] = { .type = NLA_U16 }, }; -static struct genl_ops dp_packet_genl_ops[] = { +static const struct genl_ops dp_packet_genl_ops[] = { { .cmd = OVS_PACKET_CMD_EXECUTE, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = packet_policy, @@ -800,14 +660,30 @@ } }; -static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats) +static struct genl_family dp_packet_genl_family = { + .id = GENL_ID_GENERATE, + .hdrsize = sizeof(struct ovs_header), + .name = OVS_PACKET_FAMILY, + .version = OVS_PACKET_VERSION, + .maxattr = OVS_PACKET_ATTR_MAX, + .netnsok = true, + .parallel_ops = true, + .ops = dp_packet_genl_ops, + .n_ops = ARRAY_SIZE(dp_packet_genl_ops), +}; + +static void get_dp_stats(const struct datapath *dp, struct ovs_dp_stats *stats, + struct ovs_dp_megaflow_stats *mega_stats) { int i; - struct flow_table *table = ovsl_dereference(dp->table); - 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; @@ -816,95 +692,92 @@ percpu_stats = per_cpu_ptr(dp->stats_percpu, i); do { - start = u64_stats_fetch_begin_bh(&percpu_stats->sync); + start = u64_stats_fetch_begin_irq(&percpu_stats->syncp); local_stats = *percpu_stats; - } while (u64_stats_fetch_retry_bh(&percpu_stats->sync, start)); + } while (u64_stats_fetch_retry_irq(&percpu_stats->syncp, start)); 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; } } -static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = { - [OVS_FLOW_ATTR_KEY] = { .type = NLA_NESTED }, - [OVS_FLOW_ATTR_ACTIONS] = { .type = NLA_NESTED }, - [OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG }, -}; +static bool should_fill_key(const struct sw_flow_id *sfid, uint32_t ufid_flags) +{ + return ovs_identifier_is_ufid(sfid) && + !(ufid_flags & OVS_UFID_F_OMIT_KEY); +} -static struct genl_family dp_flow_genl_family = { - .id = GENL_ID_GENERATE, - .hdrsize = sizeof(struct ovs_header), - .name = OVS_FLOW_FAMILY, - .version = OVS_FLOW_VERSION, - .maxattr = OVS_FLOW_ATTR_MAX, - .netnsok = true, - .parallel_ops = true, -}; +static bool should_fill_mask(uint32_t ufid_flags) +{ + return !(ufid_flags & OVS_UFID_F_OMIT_MASK); +} -static struct genl_multicast_group ovs_dp_flow_multicast_group = { - .name = OVS_FLOW_MCGROUP -}; +static bool should_fill_actions(uint32_t ufid_flags) +{ + return !(ufid_flags & OVS_UFID_F_OMIT_ACTIONS); +} -static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts) +static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts, + const struct sw_flow_id *sfid, + uint32_t ufid_flags) { - return NLMSG_ALIGN(sizeof(struct ovs_header)) - + nla_total_size(key_attr_size()) /* OVS_FLOW_ATTR_KEY */ + size_t len = NLMSG_ALIGN(sizeof(struct ovs_header)); + + /* OVS_FLOW_ATTR_UFID */ + if (sfid && ovs_identifier_is_ufid(sfid)) + len += nla_total_size(sfid->ufid_len); + + /* OVS_FLOW_ATTR_KEY */ + if (!sfid || should_fill_key(sfid, ufid_flags)) + len += nla_total_size(ovs_key_attr_size()); + + /* OVS_FLOW_ATTR_MASK */ + if (should_fill_mask(ufid_flags)) + len += nla_total_size(ovs_key_attr_size()); + + /* OVS_FLOW_ATTR_ACTIONS */ + if (should_fill_actions(ufid_flags)) + len += nla_total_size(acts->orig_len); + + return len + nla_total_size(sizeof(struct ovs_flow_stats)) /* OVS_FLOW_ATTR_STATS */ + nla_total_size(1) /* OVS_FLOW_ATTR_TCP_FLAGS */ - + nla_total_size(8) /* OVS_FLOW_ATTR_USED */ - + nla_total_size(acts->actions_len); /* OVS_FLOW_ATTR_ACTIONS */ + + nla_total_size(8); /* OVS_FLOW_ATTR_USED */ } -/* Called with ovs_mutex. */ -static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, - struct sk_buff *skb, u32 portid, - u32 seq, u32 flags, u8 cmd) +/* Called with ovs_mutex or RCU read lock. */ +static int ovs_flow_cmd_fill_stats(const struct sw_flow *flow, + struct sk_buff *skb) { - const int skb_orig_len = skb->len; - const struct sw_flow_actions *sf_acts; struct ovs_flow_stats stats; - struct ovs_header *ovs_header; - struct nlattr *nla; + __be16 tcp_flags; unsigned long used; - u8 tcp_flags; - int err; - - sf_acts = ovsl_dereference(flow->sf_acts); - - ovs_header = genlmsg_put(skb, portid, seq, &dp_flow_genl_family, flags, cmd); - if (!ovs_header) - return -EMSGSIZE; - ovs_header->dp_ifindex = get_dpifindex(dp); - - nla = nla_nest_start(skb, OVS_FLOW_ATTR_KEY); - if (!nla) - goto nla_put_failure; - err = ovs_flow_to_nlattrs(&flow->key, skb); - if (err) - goto error; - nla_nest_end(skb, nla); - - spin_lock_bh(&flow->lock); - used = flow->used; - stats.n_packets = flow->packet_count; - stats.n_bytes = flow->byte_count; - tcp_flags = flow->tcp_flags; - spin_unlock_bh(&flow->lock); + ovs_flow_stats_get(flow, &stats, &used, &tcp_flags); if (used && nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(used))) - goto nla_put_failure; + return -EMSGSIZE; if (stats.n_packets && - nla_put(skb, OVS_FLOW_ATTR_STATS, - sizeof(struct ovs_flow_stats), &stats)) - goto nla_put_failure; + nla_put(skb, OVS_FLOW_ATTR_STATS, sizeof(struct ovs_flow_stats), &stats)) + return -EMSGSIZE; - if (tcp_flags && - nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS, tcp_flags)) - goto nla_put_failure; + if ((u8)ntohs(tcp_flags) && + nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS, (u8)ntohs(tcp_flags))) + return -EMSGSIZE; + + return 0; +} + +/* Called with ovs_mutex or RCU read lock. */ +static int ovs_flow_cmd_fill_actions(const struct sw_flow *flow, + struct sk_buff *skb, int skb_orig_len) +{ + struct nlattr *start; + int err; /* If OVS_FLOW_ATTR_ACTIONS doesn't fit, skip dumping the actions if * this is the first flow to be dumped into 'skb'. This is unusual for @@ -916,130 +789,221 @@ * This can only fail for dump operations because the skb is always * properly sized for single flows. */ - err = nla_put(skb, OVS_FLOW_ATTR_ACTIONS, sf_acts->actions_len, - sf_acts->actions); - if (err < 0 && skb_orig_len) + start = nla_nest_start(skb, OVS_FLOW_ATTR_ACTIONS); + if (start) { + const struct sw_flow_actions *sf_acts; + + sf_acts = rcu_dereference_ovsl(flow->sf_acts); + err = ovs_nla_put_actions(sf_acts->actions, + sf_acts->actions_len, skb); + + if (!err) + nla_nest_end(skb, start); + else { + if (skb_orig_len) + return err; + + nla_nest_cancel(skb, start); + } + } else if (skb_orig_len) { + return -EMSGSIZE; + } + + return 0; +} + +/* Called with ovs_mutex or RCU read lock. */ +static int ovs_flow_cmd_fill_info(const struct sw_flow *flow, int dp_ifindex, + struct sk_buff *skb, u32 portid, + u32 seq, u32 flags, u8 cmd, u32 ufid_flags) +{ + const int skb_orig_len = skb->len; + struct ovs_header *ovs_header; + int err; + + ovs_header = genlmsg_put(skb, portid, seq, &dp_flow_genl_family, + flags, cmd); + if (!ovs_header) + return -EMSGSIZE; + + ovs_header->dp_ifindex = dp_ifindex; + + err = ovs_nla_put_identifier(flow, skb); + if (err) goto error; - return genlmsg_end(skb, ovs_header); + if (should_fill_key(&flow->id, ufid_flags)) { + err = ovs_nla_put_masked_key(flow, skb); + if (err) + goto error; + } + + if (should_fill_mask(ufid_flags)) { + err = ovs_nla_put_mask(flow, skb); + if (err) + goto error; + } + + err = ovs_flow_cmd_fill_stats(flow, skb); + if (err) + goto error; + + if (should_fill_actions(ufid_flags)) { + err = ovs_flow_cmd_fill_actions(flow, skb, skb_orig_len); + if (err) + goto error; + } + + genlmsg_end(skb, ovs_header); + return 0; -nla_put_failure: - err = -EMSGSIZE; error: genlmsg_cancel(skb, ovs_header); return err; } -static struct sk_buff *ovs_flow_cmd_alloc_info(struct sw_flow *flow) +/* May not be called with RCU read lock. */ +static struct sk_buff *ovs_flow_cmd_alloc_info(const struct sw_flow_actions *acts, + const struct sw_flow_id *sfid, + struct genl_info *info, + bool always, + uint32_t ufid_flags) { - const struct sw_flow_actions *sf_acts; + struct sk_buff *skb; + size_t len; - sf_acts = ovsl_dereference(flow->sf_acts); + if (!always && !ovs_must_notify(&dp_flow_genl_family, info, 0)) + return NULL; - return genlmsg_new(ovs_flow_cmd_msg_size(sf_acts), GFP_KERNEL); + len = ovs_flow_cmd_msg_size(acts, sfid, ufid_flags); + skb = genlmsg_new_unicast(len, info, GFP_KERNEL); + if (!skb) + return ERR_PTR(-ENOMEM); + + return skb; } -static struct sk_buff *ovs_flow_cmd_build_info(struct sw_flow *flow, - struct datapath *dp, - u32 portid, u32 seq, u8 cmd) +/* Called with ovs_mutex. */ +static struct sk_buff *ovs_flow_cmd_build_info(const struct sw_flow *flow, + int dp_ifindex, + struct genl_info *info, u8 cmd, + bool always, u32 ufid_flags) { struct sk_buff *skb; int retval; - skb = ovs_flow_cmd_alloc_info(flow); - if (!skb) - return ERR_PTR(-ENOMEM); - - retval = ovs_flow_cmd_fill_info(flow, dp, skb, portid, seq, 0, cmd); + skb = ovs_flow_cmd_alloc_info(ovsl_dereference(flow->sf_acts), + &flow->id, info, always, ufid_flags); + if (IS_ERR_OR_NULL(skb)) + return skb; + + retval = ovs_flow_cmd_fill_info(flow, dp_ifindex, skb, + info->snd_portid, info->snd_seq, 0, + cmd, ufid_flags); BUG_ON(retval < 0); return skb; } -static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) +static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info) { + struct net *net = sock_net(skb->sk); struct nlattr **a = info->attrs; struct ovs_header *ovs_header = info->userhdr; - struct sw_flow_key key; - struct sw_flow *flow; + struct sw_flow *flow = NULL, *new_flow; + struct sw_flow_mask mask; struct sk_buff *reply; struct datapath *dp; - struct flow_table *table; + struct sw_flow_key key; + struct sw_flow_actions *acts; + struct sw_flow_match match; + u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]); int error; - int key_len; + bool log = !a[OVS_FLOW_ATTR_PROBE]; - /* Extract key. */ + /* Must have key and actions. */ error = -EINVAL; - if (!a[OVS_FLOW_ATTR_KEY]) + if (!a[OVS_FLOW_ATTR_KEY]) { + OVS_NLERR(log, "Flow key attr not present in new flow."); goto error; - error = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); - if (error) + } + if (!a[OVS_FLOW_ATTR_ACTIONS]) { + OVS_NLERR(log, "Flow actions attr not present in new flow."); goto error; + } - /* Validate actions. */ - if (a[OVS_FLOW_ATTR_ACTIONS]) { - error = validate_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, 0); - if (error) - goto error; - } else if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW) { - error = -EINVAL; + /* Most of the time we need to allocate a new flow, do it before + * locking. + */ + new_flow = ovs_flow_alloc(); + if (IS_ERR(new_flow)) { + error = PTR_ERR(new_flow); goto error; } - ovs_lock(); - dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); - error = -ENODEV; - if (!dp) - goto err_unlock_ovs; + /* Extract key. */ + ovs_match_init(&match, &key, &mask); + error = ovs_nla_get_match(net, &match, a[OVS_FLOW_ATTR_KEY], + a[OVS_FLOW_ATTR_MASK], log); + if (error) + goto err_kfree_flow; - table = ovsl_dereference(dp->table); - flow = ovs_flow_tbl_lookup(table, &key, key_len); - if (!flow) { - struct sw_flow_actions *acts; + ovs_flow_mask_key(&new_flow->key, &key, true, &mask); - /* 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; + /* Extract flow identifier. */ + error = ovs_nla_get_identifier(&new_flow->id, a[OVS_FLOW_ATTR_UFID], + &key, log); + if (error) + goto err_kfree_flow; - /* 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_deferred_destroy(table); - table = ovsl_dereference(dp->table); - } - } + /* Validate actions. */ + error = ovs_nla_copy_actions(net, a[OVS_FLOW_ATTR_ACTIONS], + &new_flow->key, &acts, log); + if (error) { + OVS_NLERR(log, "Flow actions may not be safe on all matching packets."); + goto err_kfree_flow; + } - /* Allocate flow. */ - flow = ovs_flow_alloc(); - if (IS_ERR(flow)) { - error = PTR_ERR(flow); - goto err_unlock_ovs; - } - flow->key = key; - clear_stats(flow); + reply = ovs_flow_cmd_alloc_info(acts, &new_flow->id, info, false, + ufid_flags); + if (IS_ERR(reply)) { + error = PTR_ERR(reply); + goto err_kfree_acts; + } - /* Obtain actions. */ - acts = ovs_flow_actions_alloc(a[OVS_FLOW_ATTR_ACTIONS]); - error = PTR_ERR(acts); - if (IS_ERR(acts)) - goto error_free_flow; - rcu_assign_pointer(flow->sf_acts, acts); + ovs_lock(); + dp = get_dp(net, ovs_header->dp_ifindex); + if (unlikely(!dp)) { + error = -ENODEV; + goto err_unlock_ovs; + } + + /* Check if this is a duplicate flow */ + if (ovs_identifier_is_ufid(&new_flow->id)) + flow = ovs_flow_tbl_lookup_ufid(&dp->table, &new_flow->id); + if (!flow) + flow = ovs_flow_tbl_lookup(&dp->table, &key); + if (likely(!flow)) { + rcu_assign_pointer(new_flow->sf_acts, acts); /* Put flow in bucket. */ - flow->hash = ovs_flow_hash(&key, key_len); - ovs_flow_tbl_insert(table, flow); + error = ovs_flow_tbl_insert(&dp->table, new_flow, &mask); + if (unlikely(error)) { + acts = NULL; + goto err_unlock_ovs; + } - reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, - info->snd_seq, - OVS_FLOW_CMD_NEW); + if (unlikely(reply)) { + error = ovs_flow_cmd_fill_info(new_flow, + ovs_header->dp_ifindex, + reply, info->snd_portid, + info->snd_seq, 0, + OVS_FLOW_CMD_NEW, + ufid_flags); + BUG_ON(error < 0); + } + ovs_unlock(); } else { - /* We found a matching flow. */ struct sw_flow_actions *old_acts; - struct nlattr *acts_attrs; /* Bail out if we're not allowed to modify an existing flow. * We accept NLM_F_CREATE in place of the intended NLM_F_EXCL @@ -1047,52 +1011,190 @@ * request. We also accept NLM_F_EXCL in case that bug ever * gets fixed. */ - error = -EEXIST; - if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW && - info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL)) + if (unlikely(info->nlhdr->nlmsg_flags & (NLM_F_CREATE + | NLM_F_EXCL))) { + error = -EEXIST; goto err_unlock_ovs; - + } + /* The flow identifier has to be the same for flow updates. + * Look for any overlapping flow. + */ + if (unlikely(!ovs_flow_cmp(flow, &match))) { + if (ovs_identifier_is_key(&flow->id)) + flow = ovs_flow_tbl_lookup_exact(&dp->table, + &match); + else /* UFID matches but key is different */ + flow = NULL; + if (!flow) { + error = -ENOENT; + goto err_unlock_ovs; + } + } /* Update actions. */ old_acts = ovsl_dereference(flow->sf_acts); - acts_attrs = a[OVS_FLOW_ATTR_ACTIONS]; - if (acts_attrs && - (old_acts->actions_len != nla_len(acts_attrs) || - memcmp(old_acts->actions, nla_data(acts_attrs), - old_acts->actions_len))) { - struct sw_flow_actions *new_acts; - - new_acts = ovs_flow_actions_alloc(acts_attrs); - error = PTR_ERR(new_acts); - if (IS_ERR(new_acts)) - goto err_unlock_ovs; + rcu_assign_pointer(flow->sf_acts, acts); - rcu_assign_pointer(flow->sf_acts, new_acts); - ovs_flow_deferred_free_acts(old_acts); + if (unlikely(reply)) { + error = ovs_flow_cmd_fill_info(flow, + ovs_header->dp_ifindex, + reply, info->snd_portid, + info->snd_seq, 0, + OVS_FLOW_CMD_NEW, + ufid_flags); + BUG_ON(error < 0); } + ovs_unlock(); - reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, - info->snd_seq, OVS_FLOW_CMD_NEW); + ovs_nla_free_flow_actions_rcu(old_acts); + ovs_flow_free(new_flow, false); + } + + if (reply) + ovs_notify(&dp_flow_genl_family, reply, info); + return 0; - /* Clear stats. */ - if (a[OVS_FLOW_ATTR_CLEAR]) { - spin_lock_bh(&flow->lock); - clear_stats(flow); - spin_unlock_bh(&flow->lock); +err_unlock_ovs: + ovs_unlock(); + kfree_skb(reply); +err_kfree_acts: + ovs_nla_free_flow_actions(acts); +err_kfree_flow: + ovs_flow_free(new_flow, false); +error: + return error; +} + +/* Factor out action copy to avoid "Wframe-larger-than=1024" warning. */ +static struct sw_flow_actions *get_flow_actions(struct net *net, + const struct nlattr *a, + const struct sw_flow_key *key, + const struct sw_flow_mask *mask, + bool log) +{ + struct sw_flow_actions *acts; + struct sw_flow_key masked_key; + int error; + + ovs_flow_mask_key(&masked_key, key, true, mask); + error = ovs_nla_copy_actions(net, a, &masked_key, &acts, log); + if (error) { + OVS_NLERR(log, + "Actions may not be safe on all matching packets"); + return ERR_PTR(error); + } + + return acts; +} + +static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = sock_net(skb->sk); + struct nlattr **a = info->attrs; + struct ovs_header *ovs_header = info->userhdr; + struct sw_flow_key key; + struct sw_flow *flow; + struct sw_flow_mask mask; + struct sk_buff *reply = NULL; + struct datapath *dp; + struct sw_flow_actions *old_acts = NULL, *acts = NULL; + struct sw_flow_match match; + struct sw_flow_id sfid; + u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]); + int error; + bool log = !a[OVS_FLOW_ATTR_PROBE]; + bool ufid_present; + + /* Extract key. */ + error = -EINVAL; + if (!a[OVS_FLOW_ATTR_KEY]) { + OVS_NLERR(log, "Flow key attribute not present in set flow."); + goto error; + } + + ufid_present = ovs_nla_get_ufid(&sfid, a[OVS_FLOW_ATTR_UFID], log); + ovs_match_init(&match, &key, &mask); + error = ovs_nla_get_match(net, &match, a[OVS_FLOW_ATTR_KEY], + a[OVS_FLOW_ATTR_MASK], log); + if (error) + goto error; + + /* Validate actions. */ + if (a[OVS_FLOW_ATTR_ACTIONS]) { + acts = get_flow_actions(net, a[OVS_FLOW_ATTR_ACTIONS], &key, + &mask, log); + if (IS_ERR(acts)) { + error = PTR_ERR(acts); + goto error; + } + + /* Can allocate before locking if have acts. */ + reply = ovs_flow_cmd_alloc_info(acts, &sfid, info, false, + ufid_flags); + if (IS_ERR(reply)) { + error = PTR_ERR(reply); + goto err_kfree_acts; } } - ovs_unlock(); - if (!IS_ERR(reply)) - ovs_notify(reply, info, &ovs_dp_flow_multicast_group); + ovs_lock(); + dp = get_dp(net, ovs_header->dp_ifindex); + if (unlikely(!dp)) { + error = -ENODEV; + goto err_unlock_ovs; + } + /* Check that the flow exists. */ + if (ufid_present) + flow = ovs_flow_tbl_lookup_ufid(&dp->table, &sfid); else - netlink_set_err(sock_net(skb->sk)->genl_sock, 0, - ovs_dp_flow_multicast_group.id, PTR_ERR(reply)); + flow = ovs_flow_tbl_lookup_exact(&dp->table, &match); + if (unlikely(!flow)) { + error = -ENOENT; + goto err_unlock_ovs; + } + + /* Update actions, if present. */ + if (likely(acts)) { + old_acts = ovsl_dereference(flow->sf_acts); + rcu_assign_pointer(flow->sf_acts, acts); + + if (unlikely(reply)) { + error = ovs_flow_cmd_fill_info(flow, + ovs_header->dp_ifindex, + reply, info->snd_portid, + info->snd_seq, 0, + OVS_FLOW_CMD_NEW, + ufid_flags); + BUG_ON(error < 0); + } + } else { + /* Could not alloc without acts before locking. */ + reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex, + info, OVS_FLOW_CMD_NEW, false, + ufid_flags); + + if (IS_ERR(reply)) { + error = PTR_ERR(reply); + goto err_unlock_ovs; + } + } + + /* Clear stats. */ + if (a[OVS_FLOW_ATTR_CLEAR]) + ovs_flow_stats_clear(flow); + ovs_unlock(); + + if (reply) + ovs_notify(&dp_flow_genl_family, reply, info); + if (old_acts) + ovs_nla_free_flow_actions_rcu(old_acts); + return 0; -error_free_flow: - ovs_flow_free(flow); err_unlock_ovs: ovs_unlock(); + kfree_skb(reply); +err_kfree_acts: + ovs_nla_free_flow_actions(acts); error: return error; } @@ -1101,17 +1203,28 @@ { struct nlattr **a = info->attrs; struct ovs_header *ovs_header = info->userhdr; + struct net *net = sock_net(skb->sk); struct sw_flow_key key; struct sk_buff *reply; struct sw_flow *flow; struct datapath *dp; - struct flow_table *table; - int err; - int key_len; - - if (!a[OVS_FLOW_ATTR_KEY]) - return -EINVAL; - err = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); + struct sw_flow_match match; + struct sw_flow_id ufid; + u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]); + int err = 0; + bool log = !a[OVS_FLOW_ATTR_PROBE]; + bool ufid_present; + + ufid_present = ovs_nla_get_ufid(&ufid, a[OVS_FLOW_ATTR_UFID], log); + if (a[OVS_FLOW_ATTR_KEY]) { + ovs_match_init(&match, &key, NULL); + err = ovs_nla_get_match(net, &match, a[OVS_FLOW_ATTR_KEY], NULL, + log); + } else if (!ufid_present) { + OVS_NLERR(log, + "Flow get message rejected, Key attribute missing."); + err = -EINVAL; + } if (err) return err; @@ -1122,15 +1235,17 @@ goto unlock; } - table = ovsl_dereference(dp->table); - flow = ovs_flow_tbl_lookup(table, &key, key_len); + if (ufid_present) + flow = ovs_flow_tbl_lookup_ufid(&dp->table, &ufid); + else + flow = ovs_flow_tbl_lookup_exact(&dp->table, &match); if (!flow) { err = -ENOENT; goto unlock; } - reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, - info->snd_seq, OVS_FLOW_CMD_NEW); + reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex, info, + OVS_FLOW_CMD_NEW, true, ufid_flags); if (IS_ERR(reply)) { err = PTR_ERR(reply); goto unlock; @@ -1147,52 +1262,71 @@ { struct nlattr **a = info->attrs; struct ovs_header *ovs_header = info->userhdr; + struct net *net = sock_net(skb->sk); struct sw_flow_key key; struct sk_buff *reply; - struct sw_flow *flow; + struct sw_flow *flow = NULL; struct datapath *dp; - struct flow_table *table; + struct sw_flow_match match; + struct sw_flow_id ufid; + u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]); int err; - int key_len; + bool log = !a[OVS_FLOW_ATTR_PROBE]; + bool ufid_present; + + ufid_present = ovs_nla_get_ufid(&ufid, a[OVS_FLOW_ATTR_UFID], log); + if (a[OVS_FLOW_ATTR_KEY]) { + ovs_match_init(&match, &key, NULL); + err = ovs_nla_get_match(net, &match, a[OVS_FLOW_ATTR_KEY], + NULL, log); + if (unlikely(err)) + return err; + } ovs_lock(); dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); - if (!dp) { + if (unlikely(!dp)) { err = -ENODEV; goto unlock; } - if (!a[OVS_FLOW_ATTR_KEY]) { - err = flush_flows(dp); + if (unlikely(!a[OVS_FLOW_ATTR_KEY] && !ufid_present)) { + err = ovs_flow_tbl_flush(&dp->table); goto unlock; } - err = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); - if (err) - goto unlock; - table = ovsl_dereference(dp->table); - flow = ovs_flow_tbl_lookup(table, &key, key_len); - if (!flow) { + if (ufid_present) + flow = ovs_flow_tbl_lookup_ufid(&dp->table, &ufid); + else + flow = ovs_flow_tbl_lookup_exact(&dp->table, &match); + if (unlikely(!flow)) { err = -ENOENT; goto unlock; } - reply = ovs_flow_cmd_alloc_info(flow); - if (!reply) { - err = -ENOMEM; - goto unlock; - } - - ovs_flow_tbl_remove(table, flow); - - err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_portid, - info->snd_seq, 0, OVS_FLOW_CMD_DEL); - BUG_ON(err < 0); - - ovs_flow_deferred_free(flow); + ovs_flow_tbl_remove(&dp->table, flow); ovs_unlock(); - ovs_notify(reply, info, &ovs_dp_flow_multicast_group); + reply = ovs_flow_cmd_alloc_info((const struct sw_flow_actions __force *) flow->sf_acts, + &flow->id, info, false, ufid_flags); + if (likely(reply)) { + if (likely(!IS_ERR(reply))) { + rcu_read_lock(); /*To keep RCU checker happy. */ + err = ovs_flow_cmd_fill_info(flow, ovs_header->dp_ifindex, + reply, info->snd_portid, + info->snd_seq, 0, + OVS_FLOW_CMD_DEL, + ufid_flags); + rcu_read_unlock(); + BUG_ON(err < 0); + + ovs_notify(&dp_flow_genl_family, reply, info); + } else { + netlink_set_err(sock_net(skb->sk)->genl_sock, 0, 0, PTR_ERR(reply)); + } + } + + ovs_flow_free(flow, true); return 0; unlock: ovs_unlock(); @@ -1201,47 +1335,65 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) { + struct nlattr *a[__OVS_FLOW_ATTR_MAX]; struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh)); + struct table_instance *ti; struct datapath *dp; - struct flow_table *table; + u32 ufid_flags; + int err; - ovs_lock(); - dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); + err = genlmsg_parse(cb->nlh, &dp_flow_genl_family, a, + OVS_FLOW_ATTR_MAX, flow_policy); + if (err) + return err; + ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]); + + rcu_read_lock(); + dp = get_dp_rcu(sock_net(skb->sk), ovs_header->dp_ifindex); if (!dp) { - ovs_unlock(); + rcu_read_unlock(); return -ENODEV; } - table = ovsl_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_tbl_next(table, &bucket, &obj); + flow = ovs_flow_tbl_dump_next(ti, &bucket, &obj); if (!flow) break; - if (ovs_flow_cmd_fill_info(flow, dp, skb, + if (ovs_flow_cmd_fill_info(flow, ovs_header->dp_ifindex, skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, NLM_F_MULTI, - OVS_FLOW_CMD_NEW) < 0) + OVS_FLOW_CMD_NEW, ufid_flags) < 0) break; cb->args[0] = bucket; cb->args[1] = obj; } - ovs_unlock(); + rcu_read_unlock(); return skb->len; } -static struct genl_ops dp_flow_genl_ops[] = { +static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = { + [OVS_FLOW_ATTR_KEY] = { .type = NLA_NESTED }, + [OVS_FLOW_ATTR_MASK] = { .type = NLA_NESTED }, + [OVS_FLOW_ATTR_ACTIONS] = { .type = NLA_NESTED }, + [OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG }, + [OVS_FLOW_ATTR_PROBE] = { .type = NLA_FLAG }, + [OVS_FLOW_ATTR_UFID] = { .type = NLA_UNSPEC, .len = 1 }, + [OVS_FLOW_ATTR_UFID_FLAGS] = { .type = NLA_U32 }, +}; + +static const struct genl_ops dp_flow_genl_ops[] = { { .cmd = OVS_FLOW_CMD_NEW, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = flow_policy, - .doit = ovs_flow_cmd_new_or_set + .doit = ovs_flow_cmd_new }, { .cmd = OVS_FLOW_CMD_DEL, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ @@ -1257,27 +1409,22 @@ { .cmd = OVS_FLOW_CMD_SET, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = flow_policy, - .doit = ovs_flow_cmd_new_or_set, + .doit = ovs_flow_cmd_set, }, }; -static const struct nla_policy datapath_policy[OVS_DP_ATTR_MAX + 1] = { - [OVS_DP_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, - [OVS_DP_ATTR_UPCALL_PID] = { .type = NLA_U32 }, -}; - -static struct genl_family dp_datapath_genl_family = { +static struct genl_family dp_flow_genl_family = { .id = GENL_ID_GENERATE, .hdrsize = sizeof(struct ovs_header), - .name = OVS_DATAPATH_FAMILY, - .version = OVS_DATAPATH_VERSION, - .maxattr = OVS_DP_ATTR_MAX, + .name = OVS_FLOW_FAMILY, + .version = OVS_FLOW_VERSION, + .maxattr = OVS_FLOW_ATTR_MAX, .netnsok = true, .parallel_ops = true, -}; - -static struct genl_multicast_group ovs_dp_datapath_multicast_group = { - .name = OVS_DATAPATH_MCGROUP + .ops = dp_flow_genl_ops, + .n_ops = ARRAY_SIZE(dp_flow_genl_ops), + .mcgrps = &ovs_dp_flow_multicast_group, + .n_mcgrps = 1, }; static size_t ovs_dp_cmd_msg_size(void) @@ -1286,15 +1433,19 @@ msgsize += nla_total_size(IFNAMSIZ); msgsize += nla_total_size(sizeof(struct ovs_dp_stats)); + msgsize += nla_total_size(sizeof(struct ovs_dp_megaflow_stats)); + msgsize += nla_total_size(sizeof(u32)); /* OVS_DP_ATTR_USER_FEATURES */ return msgsize; } +/* Called with ovs_mutex. */ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb, u32 portid, u32 seq, u32 flags, u8 cmd) { 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, @@ -1304,17 +1455,25 @@ ovs_header->dp_ifindex = get_dpifindex(dp); - rcu_read_lock(); err = nla_put_string(skb, OVS_DP_ATTR_NAME, ovs_dp_name(dp)); - rcu_read_unlock(); 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; + + if (nla_put_u32(skb, OVS_DP_ATTR_USER_FEATURES, dp->user_features)) goto nla_put_failure; - return genlmsg_end(skb, ovs_header); + genlmsg_end(skb, ovs_header); + return 0; nla_put_failure: genlmsg_cancel(skb, ovs_header); @@ -1322,27 +1481,14 @@ return -EMSGSIZE; } -static struct sk_buff *ovs_dp_cmd_build_info(struct datapath *dp, u32 portid, - u32 seq, u8 cmd) +static struct sk_buff *ovs_dp_cmd_alloc_info(struct genl_info *info) { - struct sk_buff *skb; - int retval; - - skb = genlmsg_new(ovs_dp_cmd_msg_size(), GFP_KERNEL); - if (!skb) - return ERR_PTR(-ENOMEM); - - retval = ovs_dp_cmd_fill_info(dp, skb, portid, seq, 0, cmd); - if (retval < 0) { - kfree_skb(skb); - return ERR_PTR(retval); - } - return skb; + return genlmsg_new_unicast(ovs_dp_cmd_msg_size(), info, GFP_KERNEL); } -/* Called with ovs_mutex. */ +/* Called with rcu_read_lock or ovs_mutex. */ static struct datapath *lookup_datapath(struct net *net, - struct ovs_header *ovs_header, + const struct ovs_header *ovs_header, struct nlattr *a[OVS_DP_ATTR_MAX + 1]) { struct datapath *dp; @@ -1352,14 +1498,30 @@ else { struct vport *vport; - rcu_read_lock(); vport = ovs_vport_locate(net, nla_data(a[OVS_DP_ATTR_NAME])); dp = vport && vport->port_no == OVSP_LOCAL ? vport->dp : NULL; - rcu_read_unlock(); } return dp ? dp : ERR_PTR(-ENODEV); } +static void ovs_dp_reset_user_features(struct sk_buff *skb, struct genl_info *info) +{ + struct datapath *dp; + + dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); + if (IS_ERR(dp)) + return; + + WARN(dp->user_features, "Dropping previously announced user features\n"); + dp->user_features = 0; +} + +static void ovs_dp_change(struct datapath *dp, struct nlattr *a[]) +{ + if (a[OVS_DP_ATTR_USER_FEATURES]) + dp->user_features = nla_get_u32(a[OVS_DP_ATTR_USER_FEATURES]); +} + static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) { struct nlattr **a = info->attrs; @@ -1374,29 +1536,30 @@ if (!a[OVS_DP_ATTR_NAME] || !a[OVS_DP_ATTR_UPCALL_PID]) goto err; - ovs_lock(); + reply = ovs_dp_cmd_alloc_info(info); + if (!reply) + return -ENOMEM; err = -ENOMEM; dp = kzalloc(sizeof(*dp), GFP_KERNEL); if (dp == NULL) - goto err_unlock_ovs; + goto err_free_reply; - ovs_dp_set_net(dp, hold_net(sock_net(skb->sk))); + ovs_dp_set_net(dp, 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); + dp->stats_percpu = netdev_alloc_pcpu_stats(struct dp_stats_percpu); if (!dp->stats_percpu) { err = -ENOMEM; goto err_destroy_table; } dp->ports = kmalloc(DP_VPORT_HASH_BUCKETS * sizeof(struct hlist_head), - GFP_KERNEL); + GFP_KERNEL); if (!dp->ports) { err = -ENOMEM; goto err_destroy_percpu; @@ -1411,7 +1574,12 @@ parms.options = NULL; parms.dp = dp; parms.port_no = OVSP_LOCAL; - parms.upcall_portid = nla_get_u32(a[OVS_DP_ATTR_UPCALL_PID]); + parms.upcall_portids = a[OVS_DP_ATTR_UPCALL_PID]; + + ovs_dp_change(dp, a); + + /* So far only local changes have been made, now need the lock. */ + ovs_lock(); vport = new_vport(&parms); if (IS_ERR(vport)) { @@ -1419,36 +1587,41 @@ if (err == -EBUSY) err = -EEXIST; + if (err == -EEXIST) { + /* An outdated user space instance that does not understand + * the concept of user_features has attempted to create a new + * datapath and is likely to reuse it. Drop all user features. + */ + if (info->genlhdr->version < OVS_DP_VER_FEATURES) + ovs_dp_reset_user_features(skb, info); + } + goto err_destroy_ports_array; } - reply = ovs_dp_cmd_build_info(dp, info->snd_portid, - info->snd_seq, OVS_DP_CMD_NEW); - err = PTR_ERR(reply); - if (IS_ERR(reply)) - goto err_destroy_local_port; + err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, + info->snd_seq, 0, OVS_DP_CMD_NEW); + BUG_ON(err < 0); ovs_net = net_generic(ovs_dp_get_net(dp), ovs_net_id); - list_add_tail(&dp->list_node, &ovs_net->dps); + list_add_tail_rcu(&dp->list_node, &ovs_net->dps); ovs_unlock(); - ovs_notify(reply, info, &ovs_dp_datapath_multicast_group); + ovs_notify(&dp_datapath_genl_family, reply, info); return 0; -err_destroy_local_port: - ovs_dp_detach_port(ovs_vport_ovsl(dp, OVSP_LOCAL)); err_destroy_ports_array: + ovs_unlock(); kfree(dp->ports); err_destroy_percpu: free_percpu(dp->stats_percpu); err_destroy_table: - ovs_flow_tbl_destroy(ovsl_dereference(dp->table)); + ovs_flow_tbl_destroy(&dp->table); err_free_dp: - release_net(ovs_dp_get_net(dp)); kfree(dp); -err_unlock_ovs: - ovs_unlock(); +err_free_reply: + kfree_skb(reply); err: return err; } @@ -1467,13 +1640,14 @@ ovs_dp_detach_port(vport); } - list_del(&dp->list_node); + list_del_rcu(&dp->list_node); /* OVSP_LOCAL is datapath internal port. We need to make sure that - * all port in datapath are destroyed first before freeing datapath. + * all ports in datapath are destroyed first before freeing datapath. */ ovs_dp_detach_port(ovs_vport_ovsl(dp, OVSP_LOCAL)); + /* RCU destroy the flow table */ call_rcu(&dp->rcu, destroy_dp_rcu); } @@ -1483,26 +1657,30 @@ struct datapath *dp; int err; + reply = ovs_dp_cmd_alloc_info(info); + if (!reply) + return -ENOMEM; + ovs_lock(); dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); err = PTR_ERR(dp); if (IS_ERR(dp)) - goto unlock; + goto err_unlock_free; - reply = ovs_dp_cmd_build_info(dp, info->snd_portid, - info->snd_seq, OVS_DP_CMD_DEL); - err = PTR_ERR(reply); - if (IS_ERR(reply)) - goto unlock; + err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, + info->snd_seq, 0, OVS_DP_CMD_DEL); + BUG_ON(err < 0); __dp_destroy(dp); ovs_unlock(); - ovs_notify(reply, info, &ovs_dp_datapath_multicast_group); + ovs_notify(&dp_datapath_genl_family, reply, info); return 0; -unlock: + +err_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1512,28 +1690,30 @@ struct datapath *dp; int err; + reply = ovs_dp_cmd_alloc_info(info); + if (!reply) + return -ENOMEM; + ovs_lock(); dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); err = PTR_ERR(dp); if (IS_ERR(dp)) - goto unlock; + goto err_unlock_free; - reply = ovs_dp_cmd_build_info(dp, info->snd_portid, - info->snd_seq, OVS_DP_CMD_NEW); - if (IS_ERR(reply)) { - err = PTR_ERR(reply); - netlink_set_err(sock_net(skb->sk)->genl_sock, 0, - ovs_dp_datapath_multicast_group.id, err); - err = 0; - goto unlock; - } + ovs_dp_change(dp, info->attrs); + + err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, + info->snd_seq, 0, OVS_DP_CMD_NEW); + BUG_ON(err < 0); ovs_unlock(); - ovs_notify(reply, info, &ovs_dp_datapath_multicast_group); + ovs_notify(&dp_datapath_genl_family, reply, info); return 0; -unlock: + +err_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1543,25 +1723,26 @@ struct datapath *dp; int err; + reply = ovs_dp_cmd_alloc_info(info); + if (!reply) + return -ENOMEM; + ovs_lock(); dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); if (IS_ERR(dp)) { err = PTR_ERR(dp); - goto unlock; + goto err_unlock_free; } - - reply = ovs_dp_cmd_build_info(dp, info->snd_portid, - info->snd_seq, OVS_DP_CMD_NEW); - if (IS_ERR(reply)) { - err = PTR_ERR(reply); - goto unlock; - } - + err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, + info->snd_seq, 0, OVS_DP_CMD_NEW); + BUG_ON(err < 0); ovs_unlock(); + return genlmsg_reply(reply, info); -unlock: +err_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1588,7 +1769,13 @@ return skb->len; } -static struct genl_ops dp_datapath_genl_ops[] = { +static const struct nla_policy datapath_policy[OVS_DP_ATTR_MAX + 1] = { + [OVS_DP_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, + [OVS_DP_ATTR_UPCALL_PID] = { .type = NLA_U32 }, + [OVS_DP_ATTR_USER_FEATURES] = { .type = NLA_U32 }, +}; + +static const struct genl_ops dp_datapath_genl_ops[] = { { .cmd = OVS_DP_CMD_NEW, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = datapath_policy, @@ -1612,27 +1799,18 @@ }, }; -static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = { - [OVS_VPORT_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, - [OVS_VPORT_ATTR_STATS] = { .len = sizeof(struct ovs_vport_stats) }, - [OVS_VPORT_ATTR_PORT_NO] = { .type = NLA_U32 }, - [OVS_VPORT_ATTR_TYPE] = { .type = NLA_U32 }, - [OVS_VPORT_ATTR_UPCALL_PID] = { .type = NLA_U32 }, - [OVS_VPORT_ATTR_OPTIONS] = { .type = NLA_NESTED }, -}; - -static struct genl_family dp_vport_genl_family = { +static struct genl_family dp_datapath_genl_family = { .id = GENL_ID_GENERATE, .hdrsize = sizeof(struct ovs_header), - .name = OVS_VPORT_FAMILY, - .version = OVS_VPORT_VERSION, - .maxattr = OVS_VPORT_ATTR_MAX, + .name = OVS_DATAPATH_FAMILY, + .version = OVS_DATAPATH_VERSION, + .maxattr = OVS_DP_ATTR_MAX, .netnsok = true, .parallel_ops = true, -}; - -struct genl_multicast_group ovs_dp_vport_multicast_group = { - .name = OVS_VPORT_MCGROUP + .ops = dp_datapath_genl_ops, + .n_ops = ARRAY_SIZE(dp_datapath_genl_ops), + .mcgrps = &ovs_dp_datapath_multicast_group, + .n_mcgrps = 1, }; /* Called with ovs_mutex or RCU read lock. */ @@ -1652,8 +1830,8 @@ if (nla_put_u32(skb, OVS_VPORT_ATTR_PORT_NO, vport->port_no) || nla_put_u32(skb, OVS_VPORT_ATTR_TYPE, vport->ops->type) || - nla_put_string(skb, OVS_VPORT_ATTR_NAME, vport->ops->get_name(vport)) || - nla_put_u32(skb, OVS_VPORT_ATTR_UPCALL_PID, vport->upcall_portid)) + nla_put_string(skb, OVS_VPORT_ATTR_NAME, + ovs_vport_name(vport))) goto nla_put_failure; ovs_vport_get_stats(vport, &vport_stats); @@ -1661,11 +1839,15 @@ &vport_stats)) goto nla_put_failure; + if (ovs_vport_get_upcall_portids(vport, skb)) + goto nla_put_failure; + err = ovs_vport_get_options(vport, skb); if (err == -EMSGSIZE) goto error; - return genlmsg_end(skb, ovs_header); + genlmsg_end(skb, ovs_header); + return 0; nla_put_failure: err = -EMSGSIZE; @@ -1674,7 +1856,12 @@ return err; } -/* Called with ovs_mutex or RCU read lock. */ +static struct sk_buff *ovs_vport_cmd_alloc_info(void) +{ + return nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); +} + +/* Called with ovs_mutex, only via ovs_dp_notify_wq(). */ struct sk_buff *ovs_vport_cmd_build_info(struct vport *vport, u32 portid, u32 seq, u8 cmd) { @@ -1693,7 +1880,7 @@ /* Called with ovs_mutex or RCU read lock. */ static struct vport *lookup_vport(struct net *net, - struct ovs_header *ovs_header, + const struct ovs_header *ovs_header, struct nlattr *a[OVS_VPORT_ATTR_MAX + 1]) { struct datapath *dp; @@ -1736,33 +1923,36 @@ u32 port_no; int err; - err = -EINVAL; if (!a[OVS_VPORT_ATTR_NAME] || !a[OVS_VPORT_ATTR_TYPE] || !a[OVS_VPORT_ATTR_UPCALL_PID]) - goto exit; + return -EINVAL; + + port_no = a[OVS_VPORT_ATTR_PORT_NO] + ? nla_get_u32(a[OVS_VPORT_ATTR_PORT_NO]) : 0; + if (port_no >= DP_MAX_PORTS) + return -EFBIG; + + reply = ovs_vport_cmd_alloc_info(); + if (!reply) + return -ENOMEM; ovs_lock(); +restart: dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); err = -ENODEV; if (!dp) - goto exit_unlock; - - if (a[OVS_VPORT_ATTR_PORT_NO]) { - port_no = nla_get_u32(a[OVS_VPORT_ATTR_PORT_NO]); - - err = -EFBIG; - if (port_no >= DP_MAX_PORTS) - goto exit_unlock; + goto exit_unlock_free; + if (port_no) { vport = ovs_vport_ovsl(dp, port_no); err = -EBUSY; if (vport) - goto exit_unlock; + goto exit_unlock_free; } else { for (port_no = 1; ; port_no++) { if (port_no >= DP_MAX_PORTS) { err = -EFBIG; - goto exit_unlock; + goto exit_unlock_free; } vport = ovs_vport_ovsl(dp, port_no); if (!vport) @@ -1775,27 +1965,27 @@ parms.options = a[OVS_VPORT_ATTR_OPTIONS]; parms.dp = dp; parms.port_no = port_no; - parms.upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]); + parms.upcall_portids = a[OVS_VPORT_ATTR_UPCALL_PID]; vport = new_vport(&parms); err = PTR_ERR(vport); - if (IS_ERR(vport)) - goto exit_unlock; - - err = 0; - reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq, - OVS_VPORT_CMD_NEW); - if (IS_ERR(reply)) { - err = PTR_ERR(reply); - ovs_dp_detach_port(vport); - goto exit_unlock; + if (IS_ERR(vport)) { + if (err == -EAGAIN) + goto restart; + goto exit_unlock_free; } - ovs_notify(reply, info, &ovs_dp_vport_multicast_group); + err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid, + info->snd_seq, 0, OVS_VPORT_CMD_NEW); + BUG_ON(err < 0); + ovs_unlock(); + + ovs_notify(&dp_vport_genl_family, reply, info); + return 0; -exit_unlock: +exit_unlock_free: ovs_unlock(); -exit: + kfree_skb(reply); return err; } @@ -1806,46 +1996,48 @@ struct vport *vport; int err; + reply = ovs_vport_cmd_alloc_info(); + if (!reply) + return -ENOMEM; + ovs_lock(); vport = lookup_vport(sock_net(skb->sk), info->userhdr, a); err = PTR_ERR(vport); if (IS_ERR(vport)) - goto exit_unlock; + goto exit_unlock_free; - err = 0; if (a[OVS_VPORT_ATTR_TYPE] && - nla_get_u32(a[OVS_VPORT_ATTR_TYPE]) != vport->ops->type) + nla_get_u32(a[OVS_VPORT_ATTR_TYPE]) != vport->ops->type) { err = -EINVAL; - - reply = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!reply) { - err = -ENOMEM; - goto exit_unlock; + goto exit_unlock_free; } - if (!err && a[OVS_VPORT_ATTR_OPTIONS]) + if (a[OVS_VPORT_ATTR_OPTIONS]) { err = ovs_vport_set_options(vport, a[OVS_VPORT_ATTR_OPTIONS]); - if (err) - goto exit_free; + if (err) + goto exit_unlock_free; + } + - if (a[OVS_VPORT_ATTR_UPCALL_PID]) - vport->upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]); + if (a[OVS_VPORT_ATTR_UPCALL_PID]) { + struct nlattr *ids = a[OVS_VPORT_ATTR_UPCALL_PID]; + + err = ovs_vport_set_upcall_portids(vport, ids); + if (err) + goto exit_unlock_free; + } err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid, info->snd_seq, 0, OVS_VPORT_CMD_NEW); BUG_ON(err < 0); ovs_unlock(); - ovs_notify(reply, info, &ovs_dp_vport_multicast_group); + ovs_notify(&dp_vport_genl_family, reply, info); return 0; - rtnl_unlock(); - return 0; - -exit_free: - kfree_skb(reply); -exit_unlock: +exit_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1856,30 +2048,33 @@ struct vport *vport; int err; + reply = ovs_vport_cmd_alloc_info(); + if (!reply) + return -ENOMEM; + ovs_lock(); vport = lookup_vport(sock_net(skb->sk), info->userhdr, a); err = PTR_ERR(vport); if (IS_ERR(vport)) - goto exit_unlock; + goto exit_unlock_free; if (vport->port_no == OVSP_LOCAL) { err = -EINVAL; - goto exit_unlock; + goto exit_unlock_free; } - reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq, - OVS_VPORT_CMD_DEL); - err = PTR_ERR(reply); - if (IS_ERR(reply)) - goto exit_unlock; - - err = 0; + err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid, + info->snd_seq, 0, OVS_VPORT_CMD_DEL); + BUG_ON(err < 0); ovs_dp_detach_port(vport); + ovs_unlock(); - ovs_notify(reply, info, &ovs_dp_vport_multicast_group); + ovs_notify(&dp_vport_genl_family, reply, info); + return 0; -exit_unlock: +exit_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1891,24 +2086,25 @@ struct vport *vport; int err; + reply = ovs_vport_cmd_alloc_info(); + if (!reply) + return -ENOMEM; + rcu_read_lock(); vport = lookup_vport(sock_net(skb->sk), ovs_header, a); err = PTR_ERR(vport); if (IS_ERR(vport)) - goto exit_unlock; - - reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq, - OVS_VPORT_CMD_NEW); - err = PTR_ERR(reply); - if (IS_ERR(reply)) - goto exit_unlock; - + goto exit_unlock_free; + err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid, + info->snd_seq, 0, OVS_VPORT_CMD_NEW); + BUG_ON(err < 0); rcu_read_unlock(); return genlmsg_reply(reply, info); -exit_unlock: +exit_unlock_free: rcu_read_unlock(); + kfree_skb(reply); return err; } @@ -1919,11 +2115,12 @@ int bucket = cb->args[0], skip = cb->args[1]; int i, j = 0; - dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); - if (!dp) - return -ENODEV; - rcu_read_lock(); + dp = get_dp_rcu(sock_net(skb->sk), ovs_header->dp_ifindex); + if (!dp) { + rcu_read_unlock(); + return -ENODEV; + } for (i = bucket; i < DP_VPORT_HASH_BUCKETS; i++) { struct vport *vport; @@ -1950,7 +2147,16 @@ return skb->len; } -static struct genl_ops dp_vport_genl_ops[] = { +static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = { + [OVS_VPORT_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, + [OVS_VPORT_ATTR_STATS] = { .len = sizeof(struct ovs_vport_stats) }, + [OVS_VPORT_ATTR_PORT_NO] = { .type = NLA_U32 }, + [OVS_VPORT_ATTR_TYPE] = { .type = NLA_U32 }, + [OVS_VPORT_ATTR_UPCALL_PID] = { .type = NLA_U32 }, + [OVS_VPORT_ATTR_OPTIONS] = { .type = NLA_NESTED }, +}; + +static const struct genl_ops dp_vport_genl_ops[] = { { .cmd = OVS_VPORT_CMD_NEW, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = vport_policy, @@ -1974,26 +2180,25 @@ }, }; -struct genl_family_and_ops { - struct genl_family *family; - struct genl_ops *ops; - int n_ops; - struct genl_multicast_group *group; +struct genl_family dp_vport_genl_family = { + .id = GENL_ID_GENERATE, + .hdrsize = sizeof(struct ovs_header), + .name = OVS_VPORT_FAMILY, + .version = OVS_VPORT_VERSION, + .maxattr = OVS_VPORT_ATTR_MAX, + .netnsok = true, + .parallel_ops = true, + .ops = dp_vport_genl_ops, + .n_ops = ARRAY_SIZE(dp_vport_genl_ops), + .mcgrps = &ovs_dp_vport_multicast_group, + .n_mcgrps = 1, }; -static const struct genl_family_and_ops dp_genl_families[] = { - { &dp_datapath_genl_family, - dp_datapath_genl_ops, ARRAY_SIZE(dp_datapath_genl_ops), - &ovs_dp_datapath_multicast_group }, - { &dp_vport_genl_family, - dp_vport_genl_ops, ARRAY_SIZE(dp_vport_genl_ops), - &ovs_dp_vport_multicast_group }, - { &dp_flow_genl_family, - dp_flow_genl_ops, ARRAY_SIZE(dp_flow_genl_ops), - &ovs_dp_flow_multicast_group }, - { &dp_packet_genl_family, - dp_packet_genl_ops, ARRAY_SIZE(dp_packet_genl_ops), - NULL }, +static struct genl_family * const dp_genl_families[] = { + &dp_datapath_genl_family, + &dp_vport_genl_family, + &dp_flow_genl_family, + &dp_packet_genl_family, }; static void dp_unregister_genl(int n_families) @@ -2001,82 +2206,85 @@ int i; for (i = 0; i < n_families; i++) - genl_unregister_family(dp_genl_families[i].family); + genl_unregister_family(dp_genl_families[i]); } static int dp_register_genl(void) { - int n_registered; int err; int i; - n_registered = 0; for (i = 0; i < ARRAY_SIZE(dp_genl_families); i++) { - const struct genl_family_and_ops *f = &dp_genl_families[i]; - err = genl_register_family_with_ops(f->family, f->ops, - f->n_ops); + err = genl_register_family(dp_genl_families[i]); if (err) goto error; - n_registered++; - - if (f->group) { - err = genl_register_mc_group(f->family, f->group); - if (err) - goto error; - } } return 0; error: - dp_unregister_genl(n_registered); + dp_unregister_genl(i); 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_deferred_destroy(old_table); - } - } - } - 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); INIT_LIST_HEAD(&ovs_net->dps); INIT_WORK(&ovs_net->dp_notify_work, ovs_dp_notify_wq); + ovs_ct_init(net); return 0; } -static void __net_exit ovs_exit_net(struct net *net) +static void __net_exit list_vports_from_net(struct net *net, struct net *dnet, + struct list_head *head) { - struct datapath *dp, *dp_next; struct ovs_net *ovs_net = net_generic(net, ovs_net_id); + struct datapath *dp; + + list_for_each_entry(dp, &ovs_net->dps, list_node) { + int i; + + for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++) { + struct vport *vport; + + hlist_for_each_entry(vport, &dp->ports[i], dp_hash_node) { + if (vport->ops->type != OVS_VPORT_TYPE_INTERNAL) + continue; + + if (dev_net(vport->dev) == dnet) + list_add(&vport->detach_list, head); + } + } + } +} +static void __net_exit ovs_exit_net(struct net *dnet) +{ + struct datapath *dp, *dp_next; + struct ovs_net *ovs_net = net_generic(dnet, ovs_net_id); + struct vport *vport, *vport_next; + struct net *net; + LIST_HEAD(head); + + ovs_ct_exit(dnet); ovs_lock(); list_for_each_entry_safe(dp, dp_next, &ovs_net->dps, list_node) __dp_destroy(dp); + + rtnl_lock(); + for_each_net(net) + list_vports_from_net(net, dnet, &head); + rtnl_unlock(); + + /* Detach all vports from given namespace. */ + list_for_each_entry_safe(vport, vport_next, &head, detach_list) { + list_del(&vport->detach_list); + ovs_dp_detach_port(vport); + } + ovs_unlock(); cancel_work_sync(&ovs_net->dp_notify_work); @@ -2097,10 +2305,18 @@ pr_info("Open vSwitch switching datapath\n"); - err = ovs_flow_init(); + err = action_fifos_init(); if (err) goto error; + err = ovs_internal_dev_rtnl_link_register(); + if (err) + goto error_action_fifos_exit; + + err = ovs_flow_init(); + if (err) + goto error_unreg_rtnl_link; + err = ovs_vport_init(); if (err) goto error_flow_exit; @@ -2113,14 +2329,18 @@ if (err) goto error_netns_exit; - err = dp_register_genl(); - if (err < 0) + err = ovs_netdev_init(); + if (err) goto error_unreg_notifier; - schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL); + err = dp_register_genl(); + if (err < 0) + goto error_unreg_netdev; return 0; +error_unreg_netdev: + ovs_netdev_exit(); error_unreg_notifier: unregister_netdevice_notifier(&ovs_dp_device_notifier); error_netns_exit: @@ -2129,19 +2349,25 @@ ovs_vport_exit(); error_flow_exit: ovs_flow_exit(); +error_unreg_rtnl_link: + ovs_internal_dev_rtnl_link_unregister(); +error_action_fifos_exit: + action_fifos_exit(); error: return err; } static void dp_cleanup(void) { - cancel_delayed_work_sync(&rehash_flow_wq); dp_unregister_genl(ARRAY_SIZE(dp_genl_families)); + ovs_netdev_exit(); unregister_netdevice_notifier(&ovs_dp_device_notifier); unregister_pernet_device(&ovs_net_ops); rcu_barrier(); ovs_vport_exit(); ovs_flow_exit(); + ovs_internal_dev_rtnl_link_unregister(); + action_fifos_exit(); } module_init(dp_init);